Inscribed in Clay Provenance Study of TH

INSCRIBED IN CLAY
PROVENANCE STUDY OF THE AMARNA TABLETS

AND OTHER ANCIENT NEAR EASTERN TEXTS
TEL AVIV UNIVERSITY

SONIA AND MARCO NADLER INSTITUTE OF ARCHAEOLOGY
MONOGRAPH SERIES
NUMBER 23
Executive Editor:
Editorial Board:
Jak Yakar
Israel Finkelstein
Avi Gopher
Zeev Herzog
David Ussishkin
Manuscript & Production Editor: Shirley Gassner
INSCRIBED IN CLAY
PROVENANCE STUDY OF THE AMARNA TABLETS
AND OTHER ANCIENT NEAR EASTERN TEXTS
YUVAL GOREN, ISRAEL FINKELSTEIN AND NADAV NAAMAN
Contributions by
Michal Artzy, Frank Asaro, Shlomo Bunimovitz, Vladimir Chikatunov,
Lydia Grossowicz, Moshe Inbar, Simcha Lev-Yadun, Hans Mommsen and Joseph Yellin
EMERY AND CLAIRE YASS PUBLICATIONS IN ARCHAEOLOGY

TEL AVIV 2004
Under the auspices of the

Friends of the Institute of Archaeology of Tel Aviv University
Published by the Emery and Claire Yass Publications in Archaeology

(Bequeathed by the Yass Estate, Sydney, Australia)
of the Institute of Archaeology, Tel Aviv University
ISBN 965-266-020-5
Copyright 2004
All rights reserved
Printed in Israel by Top
Print
CONTENTS
Preface
xvii
Chapter 1
INTRODUCTION
Chapter 2
METHOD AND THEORY

THEORETICAL BACKGROUND
METHOD SELECTION
SAMPLING PROCEDURES
ELEMENTAL ANALYSIS
DATA MANAGEMENT
TERMINOLOGY AND DESCRIPTION OF THE PETROGRAPHIC DATA
QUANTITATIVE ANALYSIS OF THE ELEMENTAL DATA
SOURCES FOR PROVENANCE INTERPRETATION OF THE TABLETS
Chapter 3
LETTERS OF THE GREAT POWERS
I. EGYPT
Ceramic ecology and petrographic references
EA 1 (BM 29784), from the King of Egypt to the King of Babylonia
EA 14 (VAT 1651+2711+Ash 1891.1-41), inventory of gifts for the
Babylonian king
EA 162 (VAT 347), from the King of Egypt to Aziru of Amurru
EA 163 (VAT 1885), from the King of Egypt to a Canaanite ruler (?)
EA 190 (Ash 1893.1-41:411), from Egypt to Etakkama of Qidshu
EA 367 (AO 7095), from the King of Egypt to Endaruta of Akshapa
EA 370 (BM 134870), from the King of Egypt to Yidia of Ashkelon
EA 382 (BM 58364), small fragment of a letter from an Egyptian ofcial (?)
EA 339 (VAT 1887), Canaanite letter written in Egypt
Chemical analyses of selected Egyptian letters
CONCLUSION: THE EGYPTIAN TEXTS
1
4
4
9
11
13
14
15
18
20
23
23
24
24
25
II. ATTI
25
26
27
27
27
28
29
29
29
31

EA 42 (VAT 1655), from atti (?)
EA 44 (VAT 1656), from a Hittite prince to the King of Egypt
KUB XIX 20, a draft letter from Shuppiluliuma of atti to the King of Egypt
Chemical analyses of the Hittite tablets
CONCLUSION: THE HITTITE CORRESPONDENCE
III. BABYLONIA
EA 2 (VAT 148+2706), from the King of Babylonia to the King of Egypt
EA 4 (VAT 1657), from the King of Babylonia to the King of Egypt
EA 11 (VAT 151 + 1878), from the King of Babylonia to the King of Egypt
EA 12 (VAT 1605), from a Babylonian princess (?)
EA 13 (VAT 1717), an inventory from Babylonia
Chemical analysis of selected Babylonian tablets
CONCLUSION: THE BABYLONIAN CORRESPONDENCE
31
31
31
32
32
32
33
33
34
34
35
35
35
36
36
37
37
v
IV.
MITANNI

EA 17 (BM 29792), from the King of Mitanni to the King of Egypt
EA 18 (VAT 1880 + 1879), from Mitanni (?)
EA 20 (VAT 191), from the King of Mitanni to the King of Egypt
EA 25 (VAT 340 + fragments 2191 a-c, no. 2), inventory of gifts from the
King of Mitanni
EA 26 (BM 29794), from the King of Mitanni to the Queen of Egypt
EA 27 (VAT 233 + 2197, no. 1; 2193), from the King of Mitanni to the
King of Egypt
EA 30 (BM 29841), from the King of Mitanni to the mayors of Canaan
Chemical analyses of two Mitannian tablets
CONCLUSION: THE MITANNIAN TABLETS AND THE LOCATION OF WAUKANNI
V.
ARZAWA
EA 32 (VAT 342), from the King of Arzawa to the King of Egypt
PETROGRAPHIC ANALYSIS
NEUTRON ACTIVATION ANALYSIS OF EA 32
Michal Artzy, Hans Mommsen and Frank Asaro
CONCLUSION: THE ARZAWA CORRESPONDENCE
VI. ALASHIYA
Yuval Goren, Shlomo Bunimovitz, Israel Finkelstein and Nadav Naaman
Catalogue: The Alashiya letters from Amarna
EA 37 (BM 29790), from the King of Alashiya to the King of Egypt
EA 33 (VAT 1654), from the King of Alashiya to the King of Egypt
EA 38 (VAT153), from the King of Alashiya to the King of Egypt
Comparative study of Cypro-Minoan texts from Enkomi
E-1687, a Cypro-Minoan tablet
H-1885, a Cypro-Minoan tablet
E-1610, a Cypro-Minoan inscribed cylinder
E-1193 + Enk. F.E20.01, a Cypro-Minoan tablet
AM 2336, a Cypro-Minoan tablet
Comparative study of Cypro-Minoan texts from Kalavasos Ayios Dhimitrios
K-AD 82, 38, a Cypro-Minoan inscribed cylinder
A letter of the King of Alashiya from Ugarit
RS L.1 (AO 21087), from the King of Alashiya to the King of Ugarit
Excursus: Letter RS L.1 and Carchemish
RS 8.333 (AO 19.955, PRU III, 7-8), from the King of Carchemish to the
King of Ugarit
vi
38
38
39
40
40
40
41
41
41
41
42
42
42
43
43
43
43
44
45
45
45
45
48
49
49
50
50
51
51
51
52
52
53
53
54
54
54
54
55
55
55
56
57
Interpretation of the petrographic data

The general provenance of the Alashiya letters
The provenance of the Alashiya letters within Cyprus
Chemical analyses of the Alashiya letters and selected Cypriote clays
CONCLUSION: THE ALASHIYA TABLETS AND LATE BRONZE AGE CYPRUS
The government of Cyprus in the Late Bronze Age: an overview
The political-administrative centre of Alashiya during the 14th-13th
centuries BCE according to the analytical results
The government of Cyprus in the Late Bronze Age: Alternative models
VII. INTERNATIONAL LETTER OF UNKNOWN PROVENANCE
EA 172 (VAT 1877)
Chapter 4
SCHOLARLY TEXTS
EA 340 (VAT 1583), a historical tale or a letter fragment
EA 341 (Ash. 1893 1-41: 415), the story of Kei
EA 342 (Ash. 1893 1-41: 414), an exercise in letter writing?
EA 343 (Ash. 1893 1-41: 427), an exercise
EA 344 (Ash. 1893 1-41: 417), an exercise (?)
EA 347 (Ash. 1893 1-41: 422), a lexical list?
EA 348 (Ash. 1893 1-41: 419) a fragment of a Sa signlist
EA 349 (Ash. 1893 1-41: 428), a fragment of a syllabary?
EA 350 (Ash. 1893 1-41: 425), a fragment of a tu-ta-ti exercise
EA 352 + 353 (Ash. 1893 1-41: 413 + 421, two pieces joined), a fragment of diri
EA 354 (Ash. 1893 1-41: 418), a fragment of diri
EA 355 (Ash 1893.1-41: 416), an inscribed cylinder
EA 356 (VAT 348), the myth of Adapa and the South Wind
EA 357 (VAT 1611+1613+1614+2710; BM 29865), the myth of Nergal and
Ereshkigal
EA 358 (VAT 1612+2708), a narrative of still undetermined genre
EA 368 (Ash. 1921-1154), Egyptian-Akkadian vocabulary
A blank tablet (Ash. 1893 1-41: 429)
NEUTRON ACTIVATION ANALYSIS (NAA) OF AMARNA LITERARY TABLETS EA 356-358 AND ITS
IMPLICATIONS by Joseph Yellin
ICP analyses of the Triad (EA 356-358)
CONCLUSION: THE AMARNA SCHOLARLY TEXTS
Chapter 5
NORTH SYRIAN KINGDOMS

I. UGARIT
EA 45 (VAT 1692), from Ammishtamru to the King of Egypt
EA 46 (VAT 1694), from the King of Ugarit to the King of Egypt
EA 48 (VAT 1690), from the Queen of Ugarit to the Queen of Egypt
Chemical analyses of two Ugaritic tablets
CONCLUSION: THE UGARITIC CORRESPONDENCE
II. NUASHE
EA 51 (VAT 559), from Addu-nirari to the King of Egypt
57
57
60
63
70
70
71
73
75
75
76
76
77
77
78
78
78
79
79
80
80
80
81
81
82
82
83
83
84
84
85
85
87
88
88
88
88
89
90
90
90
91
91
91
vii
Chapter 6
III. NII (?)

EA 67 (VAT 1591), from an unknown ruler north of Canaan
92
92
SYRIAN KINGDOMS IN THE MIDDLE ORONTES AREA
94
94
94
94
95
95
96
I. QATNA
EA 55 (BM 29819), from Akizzi to an Egyptian ofcial
EA 53 (BM 29820), from Akizzi to the King of Egypt
EA 54 (VAT 1868 + 1869 + 1721), from Akizzi to the King of Egypt
EA 57 (VAT 1738), from an unknown king (in the area of Qatna?) to the
King of Egypt
II. QIDSHU
EA 189 (VAT 336), from Etakkama to the King of Egypt
III. LABANA
EA 193 (VAT 1608), from Teuwatti to the King of Egypt
IV. RUIZZI
EA 192 (VAT 1674), from Arsawuya to the King of Egypt
Chapter 7
THE KINGDOM OF AMURRU AND NEIGHBOURING POLITIES

I. AMURRU
EA 60 (VAT 343), from Abdi-Ashirta to the King of Egypt
EA 61 (Ash. 1893.1-41: 410), from Abdi-Ashirta to the King of Egypt
EA 62 (VAT 1680), from Abdi-Ashirta to Pahanata, the Commissioner of umur
EA 156 (VAT 337), from Aziru to the King of Egypt
EA 161 (BM 29818), from Aziru to the King of Egypt
EA 164 (VAT 249), from Aziru to the Egyptian ofcial Tutu
EA 166 (VAT 250), from Aziru to the Egyptian ofcial aay
EA 167 (VAT 326), from Aziru (?) to the Egyptian ofcial Tutu (?)
EA 169 (VAT 1660), from DU-Teshup or Bet-ili (?) to the Egyptian ofcial
EA 170 (VAT 327), from Baluya and Bet-ili to Aziru in Egypt
II. UMUR
EA 96 (VAT 1238), from an army commander to Rib-Hadda of Gubla
III. TUNIP
EA 59 (BM 29824), from the citizens of Tunip to the King of Egypt
IV. IRQATA
EA 100 (BM 29825), from the elders of Irqata to the King of Egypt
V. AN UNSPECIFIED CITY IN THE NEIGHBOURHOOD OF AMURRU
EA 58 (VAT 1716), from Teu-Teshup to the King of Egypt
CONCLUSION: THE AMURRU CORRESPONDENCE IN LIGHT OF THE PETROGRAPHIC STUDY
viii
97
97
97
98
98
100
100
100
101
101
103
103
105
106
106
107
107
108
108
108
111
111
112
114
115
115
116
116
116
118
118
122
122
122
122
122
123
Chapter 8
THE LEBANESE BEQA>

I. ENISHASI
EA 187 (BM 29860), from Shatiya to the King of Egypt
II. ASI
EA 185 (VAT 1725), from Mayarzana to the King of Egypt
III. GUDDASHUNA
EA 177 (VAT 1684), from Yamiuta to the King of Egypt
IV. A GROUP OF IDENTICAL LETTERS SENT BY FOUR BEQA RULERS
EA 174 (VAT 1585), from Bieri of ashabu to the King of Egypt
EA 175 (VAT 1588), from Ildayyi of asi to the King of Egypt
EA 363 (AO 7097), from Abdi-Risha of Enishasi to the King of Egypt
EA 173 (VAT 1875), from (?) to the King of Egypt
V. LETTERS OF UNSPECIFIED LOCATIONS, POSSIBLY FROM THE BEQA> VALLEY
EA 178 (VAT 1677), from ibiya to an Egyptian ofcial
EA 179 (VAT 1703), from the deposed ruler of Tubiu to the King of Egypt
CONCLUSION: THE LEBANESE BEQA> CORRESPONDENCE
Chapter 9
THE LEBANESE LITTORAL

I. BYBLOS (GUBLA)
EA 72 (VAT 1712), from Rib-Hadda to the King of Egypt
EA 69 (BM 29856), from Rib-Hadda to an Egyptian ofcial
EA 71 (VAT1632), from Rib-Hadda to the Egyptian ofcial aya
EA 74 (BM 29795), from Rib-Hadda to the King of Egypt
EA 77 (VAT 1635 + 1700), from Rib-Hadda to the Egyptian ofcial Amanappa
EA 82 (BM 37648), from Rib-Hadda to the Egyptian ofcial Amanappa
EA 87 (BM 29805), from Rib-Hadda to theEgyptian ofcial Amanappa
EA 93 (VAT 1663), from Rib-Hadda to the Egyptian ofcial Amanappa
EXCURSUS: BEETLE FRAGMENTS IN THE CLAY OF EA 95
Moshe Inbar and Vladimir Chikatunov
126
126
126
127
127
127
128
128
128
129
129
130
130
130
131
131
131
132
132
134
134
134
134
136
136
137
137
138
138
138
139
139
140
140
141
141
141
142
142
143
143
144
144
144
145
146
147
ix

EA 115 (VAT 1630), from Rib-Hadda to the King of Egypt (?)
EA 129 (VAT 1637+1638), from Rib-Hadda to the King of Egypt
EA 139 (BM 29828), from Ili-Rapi to the King of Egypt
EA 140 (VAT 1639), from Ili-Rapi to the King of Egypt
EA 362 (AO 7093), from Rib-Hadda to the King of Egypt
EA 50 (VAT 1594), from the daughter of (?) to the Queen of Egypt
CONCLUSION: THE BYBLOS CORRESPONDENCE
II. BEIRUT (BIRUTA)
EA 97 (VAT 1598), from Yappa-Hadda to Shumu-Haddi
EA 98 (VAT 1675), from Yappa-Hadda to the Egyptian commissioner Yanamu
EA 141 (BM 29809), from Ammunira to the King of Egypt
EA 143 (VAT 1584 + C 4764), from Ammunira to the King of Egypt
Comparative material: A letter from Beirut in the Ugarit archive
RS 11.730 (AO 29.507), a letter from Beirut to Ugarit
CONCLUSION: THE BEIRUT CORRESPONDENCE
III. SIDON (IDUNA)
EA 144 (VAT 323), from Zimreddi to the King of Egypt
CONCLUSION: THE SIDON CORRESPONDENCE
IV. TYRE (URRU)
EA 151 (BM 29813), from Abi-Milku to the King of Egypt
x
147
147
148
148
149
149
149
150
150
151
151
151
152
153
153
153
154
154
155
155
156
156
156
157
157
157
158
158
158
159
161
161
161
161
162
162
163
163
164
164
164
164
164
165
165
166
166
166
166

EA 152 (VAT 1719), from Abi-Milku to the King of Egypt
EA 155 (BM 29814 + VAT 1872), from Abi-Milku to the King of Egypt
EA 295 (VAT 1650), from [Ba>lu]-danu to the King of Egypt
CONCLUSION: THE TYRE CORRESPONDENCE
Chapter 10 SOUTHERN SYRIA AND THE NEIGHBOURING AREAS

I. DAMASCUS (DIMASHQU)
EA 194 (VAT 1705), from Biryawaza to the King of Egypt
EA 196 (VAT 1592 + 1710), from Biryawaza to the King of Egypt
EA 197 (BM 29826), from Biryawaza to the King of Egypt
II. MUSHIUNA
EA 183 (VAT 1595), from Shutarna to the King of Egypt
EA 184 (Ash. 1893. 1-41: 426), from Shutarna to the King of Egypt
CONCLUSION: THE MUSHIUNA LETTERS
Chapter 11 THE BASHAN AREA

I. Z/IRIBASHANI
EA 201 (VAT 338), from Artamanya to the King of Egypt
II. SHASIMI
EA 203 (VAT 330), from Abdi-Milki to the King of Egypt
III. QANU
EA 204 (VAT 328), from the ruler of Qanu to the King of Egypt
IV. TUBU
EA 205 (BM 29861), from the ruler of Tubu to the King of Egypt
V. ASHTAROTH (ASHTARTU)
EA 364 (AO 7094), from Ayyab to the King of Egypt
VI. ZURA
EA 337 (VAT 1679), from iziru to the King of Egypt
VII. SHARUNA
EA 241 (VAT 1678), from Rusmanya to the King of Egypt
VIII. TABLETS FROM UNSPECIFIED CITIES IN THE BASHAN
EA 202 (VAT 331), from Amawashe to the King of Egypt
EA 207 (VAT 1593), from Ipte to the King of Egypt
EA 208 (VAT 1699), from (?) to the King of Egypt (?)
EA 209 (AO 2036), from Zishamimi to the King of Egypt
EA 210 (VAT 1876), from Zishamimi to an Egyptian ofcial (?)
CONCLUSION: THE BASHAN CORRESPONDENCE
167
167
167
167
168
168
168
169
170
170
170
170
171
171
172
172
172
173
173
215
215
216
216
216
216
217
217
217
217
218
218
219
219
219
220
220
220
221
221
221
221
222
222
222
223
xi
Chapter 12 THE GALILEE, THE COASTAL PLAIN OF ACCO AND THE

NORTHERN VALLEYS
I. HAZOR (AURA))
EA 227 (BM 29830), from the ruler of aura to the King of Egypt
EA 228 (BM 29831), from Abdi-Tirshi to the King of Egypt
Comparative study of the Late Bronze Age cuneiform texts from Hazor
IAA 67-1188/1189, an inscribed liver model
IAA 1997-3307, an economic document
IAA 1997-3308, a short administrative document
CONCLUSION: THE HAZOR CORRESPONDENCE
II. ACHSHAPH (AKSHAPA)
EA 223 (VAT 1870), from Endaruta to the King of Egypt
III. SHIMON (SHAMUNA)
EA 224 (BM 29849), from um-Adda to the King of Egypt
IV. ACCO (AKKA)
EA 234 (VAT 1641), from Shatatna to the King of Egypt
EA 235 (BM 29815), from Shatatna to the King of Egypt
EA 232 (VAT 1640), from Surata to the King of Egypt
CONCLUSION: THE ACCO CORRESPONDENCE
V. ANAHARATH
EA 237 (VAT 1701), from Bayadi to the King of Egypt
EA 238 (VAT 1867), from Bayadi to an Egyptian ofcial
EA 239 (VAT 334), from Baduzana to the King of Egypt
CONCLUSION: THE BAYADI - BADUZANA CORRESPONDENCE
VI. MEGIDDO (MAGIDDA)
EA 242 (VAT 1670), from Biridiya to the King of Egypt
EA 243 (VAT 1669), from Biridiya to the King of Egypt,
EA 245 (BM 29855), from Biridiya to the King of Egypt
EA 365 (AO 7098), from Biridiya to the King of Egypt
Petrographic interpretation of the Megiddo letters
CONCLUSION: THE MEGIDDO CORRESPONDENCE
A letter of Yashdata sent from Megiddo
EA 248 (BM 29842), from Yashdata to the King of Egypt
VII. REHOB
EA 249 (VAT 1603), from Ba>lu-UR.SAG to the King of Egypt
VIII. [URUX-(X)-I]G-MA-TE
EA 257 (VAT 1715), from Ba>lu-meir to the King of Egypt
EA 221 (VAT 341), from Wiktasu to the King of Egypt
xii
226
226
226
228
228
229
229
230
230
231
231
232
232
233
233
234
237
237
238
238
239
239
240
240
240
241
241
243
243
244
244
245
245
245
246
246
246
246
248
249
250
250
251
251
252
255
255
IX. GINTI-KIRMIL
EA 264 (BM 29853), from Tagi to the King of Egypt
EA 265 (VAT 1697), from Tagi to the King of Egypt
Comparative study of the Tagi cylinder from Beth-Shean
Beth-Shean 050020, an inscribed clay cylinder from Tagi to Lab<ayu
X. PEHEL (PIILU)
EA 256 (BM 29847), from Mut-Ba>lu to the King of Egypt
EA 255 (VAT 333), from Mut-Ba>lu, to the King of Egypt
Chapter 13 THE CENTRAL HILL COUNTRY

I. SHECHEM (SHAKMU)
EA 252 (BM 29844), from Lab<ayu to the King of Egypt
EA 253 (VAT 1589), from Lab<ayu to the King of Egypt
CONCLUSION: THE SHECHEM CORRESPONDENCE
II. JERUSALEM (URUSALIM)
EA 286 (VAT 1642), from Abdi-eba to the King of Egypt
EA 289 (VAT 1645 + 2709), from Abdi-eba to the King of Egypt
EA 291 (VAT 1713), from the ruler of Jerusalem (?)
CONCLUSION: THE JERUSALEM CORRESPONDENCE
Chapter 14 THE SHEPHELAH AND THE SOUTHERN COASTAL PLAIN

I. GEZER (GAZRU)
EA 268 (VAT 1532), from Milkilu to the King of Egypt
EA 269 (BM 29846), from Milkilu to the King of Egypt
EA 292 (BM 37647), from Ba>lu-danu to the King of Egypt
EA 297 (BM 29834), from Yapau to the King of Egypt
EA 300 (VAT 1606), from Yapau to the King of Egypt
Other letters sent from Gezer
EA 272 (BM 29863), from Shum-[...] to the King of Egypt
EA 273 (VAT 1686), from Belit-nesheti to the King of Egypt
CONCLUSION: THE GEZER CORRESPONDENCE
II. GATH (GIMTU)
EA 279 (VAT 1647), from Shuwardatu to the King of Egypt
256
256
258
258
259
259
260
260
260
261
262
262
262
262
264
264
265
265
265
266
266
266
267
267
268
268
269
270
270
270
271
271
272
272
273
273
273
274
274
275
275
275
277
279
279
280
280
281
xiii
EA 282 (BM 29851), from Shuwardatu to the King of Egypt

EA 366 (AO 7096), from Shuwardatu to the King of Egypt
EA 63 (BM 29817), from Abdi-Ashtarti to the King of Egypt
EA 64 (BM 29816), from Abdi-dINNIN to the King of Egypt
EA 65 (VAT 1685), from Abdi-Ashtarti to the King of Egypt
EA 335 (VAT 1616 + 1708), from Abdi-Ashtarti (?) to the King of Egypt
EA 229 (VAT 1689), from Abdina (?) to the King of Egypt
CONCLUSION: THE GATH CORRESPONDENCE
III. LACHISH (LAKISHA)
EA 332 (VAT 1883), from Shipi-Ba>lu to the King of Egypt
EA 330 (BM 29848), from Shipi-Ba>lu to the King of Egypt
CONCLUSION: THE LACHISH CORRESPONDENCE
IV. A GROUP OF IDENTICAL LETTERS SENT BY TWO RULERS FROM THE SHEPHELAH
EA 275 (VAT 1682), from Yazib-Adda to the King of Egypt
EA 277 (BM 29864), from Yazib-Adda (?) to the King of Egypt
CONCLUSION: EA 275-278
V. ASHDOD
EA 296 (BM 29840), from Yatiru to the King of Egypt
EA 294 (BM 29854), from i-x-x-ni to the King of Egypt
VI. ASHKELON (AQALUNA)
EA 321 (VAT 1671), from Yidia to the King of Egypt
EA 323 (BM 29836), from Yidia to the King of Egypt
EA 302 (VAT 332), from Shubandu to the King of Egypt
EA 303 (BM 29821), from Shubandu to the King of Egypt
CONCLUSION: THE ASHKELON CORRESPONDENCE
VII. YURZA
EA 315 (BM 29839), from Pu-Ba>lu to the King of Egypt
CONCLUSION: THE YURZA CORRESPONDENCE
Chapter 15 UNIDENTIFIED CITIES IN CANAAN

I. AHIRUNA
EA 319 (VAT 1722), from Zur-Ashar to the King of Egypt
II. LETTERS OF UNPROVENANCED CANAANITE RULERS
xiv
281
282
282
282
283
283
284
285
285
286
286
287
287
287
288
288
289
289
289
290
290
291
291
292
292
293
294
294
295
295
295
296
296
297
297
298
298
299
299
299
300
301
301
302
302
302
303
303
EA 230 (BM 37646), from Yama to the King of Egypt

EA 251 (BM 29862), from (?) to the King of Egypt
EA 261 (BM 29858) from Dashru to the King of Egypt
III. LETTERS OF UNPROVENANCED RULERS, SENT FROM GAZA
EA 211 (VAT 1648), from Zitriyara to the King of Egypt
EA 213 (BM 29859), from Zitriyara to the King of Egypt
EA 215 (BM 29843), from Bayawa to the King of Egypt
EA 226 (VAT 1610), from Shipu-ria to the King of Egypt
EA 317 (VAT 1676), from Dagan-takala to the King of Egypt
EA 318 (BM 29857), from Dagan-takala to the King of Egypt
EA 312 (VAT 1886 + 1709), from (?) to the King of Egypt (?)
IV. SMALL FRAGMENTS OF LETTERS OF OTHER PROVENANCES
EA 240 (VAT 2198 + 2707)
EA 236 (Ash. 1893 1-41: 423), from (?)
Chapter 16 CONCLUSION
CONTRIBUTIONS AND LIMITATIONS OF THE ANALYTICAL METHODS
TECHNOLOGICAL ASPECTS OF THE AMARNA TABLETS AND THEIR IMPLICATIONS
THE SYSTEM OF CANAANITE CITY-STATES
THE EGYPTIAN ADMINISTRATION AND THE ORIGIN OF THE VASSAL LETTERS
APPENDIX :
RESULTS OF THE CHEMICAL ANALYSES

TABLE 1: RESULTS OF ICP ANALYSIS OF SELECTED AMARNA TABLETS
TABLE 2: ICP ANALYSIS OF SOUTHERN PALESTINIAN WORKSHOP WASTERS
TABLE 3: RESULTS OF NEUTRON ACTIVATION ANALYSIS OF SELECTED AMARNA TABLETS
Michal Artzy
304
304
305
306
306
307
307
308
308
309
309
309
310
310
311
311
311
312
312
313
313
313
314
315
315
316
316
317
320
322
326
326
329
332
SUPPLEMENT: CATALOGUE OF LATE BRONZE AGE II SITES IN THE EGYPTIAN

PROVINCE OF CANAAN
333
REFERENCES
356
INDEX OF THE PETROGRAPHIC FABRICS
xv
LIST OF FIGURES
Fig. 3.1:
Fig. 3.2:
Fig. 3.3:
Fig. 3.4:
Fig. 7.1:
Fig. 9.1:
Fig. 9.2:
Fig. 12.1:
Fig. 14.1:
The Near East during the Amarna period.

Principal component analysis of ICP data of selected international letters and scholarly texts.
General geological map of Cyprus showing the distribution of the formations and the location of the Late Bronze Age sites mentioned in the text.
Cypriote clay samples and the Alashiya letters, tree diagram.
The Land of Amurru with sites mentioned in the text.
Main city-states and Egyptian administrative centres in northern Canaan.
a) Leg fragments of the darkling beetle Akis latreillei in EA 95; b) Adult beetle A. latreillei
Main city-states and Egyptian administrative centres in cental and southern Canaan.
Distribution of loess soil in the southern coastal plain.
23
37
60
64
102
135
146
227
276
LIST OF COLOUR PLATES OF THIN SECTIONS

Pl. I:
Pl. I (contd.):
Pl. I (contd.):
Pl. I (contd.):
Pl. I (contd.):
Pl. I (contd.):
Pl. II:
Pl. II (contd.):
Pl. II (contd.):
Pl. III:
Pl. IV:
Pl. V:
Pl. V (contd.):
Pl. VI:
Pl. VII:
Pl. VII (contd.):
Pl. VII (contd.):
Pl. VII (contd.):
Pl. VII (contd.):
Pl. VIII:
Pl. IX:
Pl. IX (contd.):
Pl. X:
Pl. X (contd.):
Pl. X (contd.):
Pl. X (contd.):
Pl. XI:
Pl. XII:
Pl. XII (contd.):
Pl. XII (contd.):
Pl. XII (contd.)
Pl. XIII:
xvi
Letters of the Great Powers. I. Egypt.

Letters of the Great Powers. I. Egypt; II. atti
Letters of the Great Powers. III. Babylonia
Letters of the Great Powers. IV. Mitanni
Letters of the Great Powers. IV. Mitanni; V. Arzawa; VI. Alashiya
Letters of the Great Powers. VI. Alashiya; VII. International letter of unknown provenance
Scholarly texts
Scholarly texts
Scholarly texts
North Syrian kingdoms. I. Ugarit; II. Nuashe; III. Nii (?)
Syrian kingdoms in the Middle Orontes area. I. Qatna; II. Qidshu; III. Labana
The kingdom of Amurru and neighbouring polities. I. Amurru
The kingdom of Amurru and neighbouring polities. I. Amurru; II. umur; III. Tunip;
IV. Irqata; V. Unspecied city
The Lebanese Beqa>. I. Enishasi; II. asi; III. Guddashuna
The Lebanese littoral. I. Byblos
The Lebanese littoral. I. Byblos
The Lebanese littoral. I. Byblos; II. Beirut
The Lebanese littoral. II. Beirut; III. Sidon
The Lebanese littoral. IV. Tyre
Southern Syria and the neighbouring areas. I. Damascus; II. Mushiuna
The Bashan area.
The Bashan area.
The Galilee, the coastal plain of Acco and the northern valleys. I. Hazor;
II. Achshaph; III. Shim>on; IV. Acco
The Galilee, the coastal plain of Acco and the northern valleys. V. Anaharath; VI. Megiddo
The Galilee, the coastal plain of Acco and the northern valleys. VII. Rehob;
VIII. [URUx-(x)-i]G-ma-te
The Galilee, the coastal plain of Acco and the northern valleys. IX. Ginti-kirmil; X. Pehel
The central hill country. I. Shechem; II. Jerusalem
The Shephelah and the southern coastal plain. I. Gezer
The Shephelah and the southern coastal plain. II. Gath
The Shephelah and the southern coastal plain. III. Lachish; IV. Group of identical letters;
V. Ashdod
The Shephelah and the southern coastal plain. VI. Ashkelon; VII. Yurza
Unidentied cities in Canaan
175
176
177
178-179
180
181-182
183
184
185
186
187
188-189
190
191
192
193-197
198
199
200
201
201
202
203
204
205
206
207
208
209
210
211
212-214
PREFACE
This publication is a synthesis of the results of a study which approaches the problem of locating the
provenance of the Amarna tablets from a different angle. Through mineralogical and chemical analyses of
samples from over 300 tablets housed in museums in Berlin, London, Oxford, and Paris, the project aims
at pin-pointing their geographic origin and clarifying the geographic history of the Ancient Near East.
Mineralogical and chemical studies of clay tablets may resolve historical problems that have haunted
research for decades. In the case of the Amarna archive, the introduction of this scientic tool helps to
clear up the controversy over the location of Alashiya and Tunip; opens the way to track the territorial
expansion of the kingdom of Amurru; enables reconstruction of the territorial disposition of the Canaanite
city-states of the Late Bronze Age; and sheds light on the Egyptian administration system in Canaan.
The method described here should be applied in other cases. It may serve to establish the location
of cities and states mentioned in tablets retrieved in other areas of the ancient Near East and beyond,
and dated to other periods. It may also help to elucidate the origin of tablets of unspecified provenance
which appear in many collections.
In each case, if possible, the mineralogical data should be correlated with information provided
by the text and the archaeological evidence from the site or region involved. The combination of these
disciplines is the key to successful research of this type. It is obvious that collaboration of experts
from different backgrounds is essential in such a comprehensive, multi-disciplinary research project.
Consequently, the team includes experts in the fields of ceramic analysis, microarchaeology and
geoarchaeology (Goren), archaeology and historical geography (Finkelstein), philology, history and
historical geography (Naaman).
This study was generously supported by the Center for Collaboration between Natural Sciences
and Archaeology on behalf of the Weizmann Institute of Science, and the Fund for Internal Researches
of the Tel-Aviv University. It could not have been carried-out without the willing collaboration of the
following individuals and institutions, to all of whom we are greatly indebted.
B. Salje, Director of the Vorderasiatisches Museum in the Staatlische Museen zu Berlin, and E.
Klengel and J. Marzahn, Curators of its tablet collection, cordially allowed us to study the material in
their keeping. The sampling was done with the kind help of U. Von Eickstedt.
J. Curtis, Keeper of the Department of Western Asiatic Antiquities and S. Bowman, Director of
the Department of Scientific Research kindly permitted the study of the tablets in the British Museum.
The sampling was done with the assistance and advice of C. Walker, A. Middleton and I. Freestone of
the latter department.
P.R.S. Moorey, Curator of the Department of Antiquities of the Ashmolean Museum, Oxford, and
H. Whitehouse, Curator of the Department of Egyptian Antiquities, kindly allowed us to study the
tablets from this museum.
The study of the tablets from the Muse du Louvre was kindly permitted by A. Caubet, Curator of
the Oriental Antiquities, and B. Andre-Salvini, Keeper of the Near Eastern Texts.
We would also like to thank S. Hadjisavvas, Director of the Department of Antiquities of Cyprus,
P. Florentzos, Curator of the Cyprus Museum at Nicosia, M. Hadjicosti of the Cyprus Museum and G.
Konstantinou, Director of the Cyprus Geological Survey, for their collaboration and help. We wish to
thank V. Karageorghis from the Cyprus University at Nicosia for his useful comments.
The examination of the tablets that are stored in the Israel Museum in Jerusalem was made
possible through the permission and collaboration of O. Misch-Brandl, Curator of the Chalcolithic and
xvii
Bronze Age Antiquities. A general permit to examine the tablets was kindly given by H. Katz, Head of
the National Treasuries Division in the Israel Antiquities Authority.
A. Ben-Tor, Y. Tsafrir and A. Mazar of the Institute of Archaeology of the Hebrew University
in Jerusalem kindly allowed us to examine the cuneiform tablets from their excavations at Hazor and
Beth-Shean. The sampling and the study were carried out in collaboration with W. Horowitz, R. Bonfil
and S. Zuckerman.
A reference collection of pottery from Ras Shamra and Byblos was assembled with the gracious
permission and help of A. Caubet and S. Cluzan of the Muse du Louvre. The following scholars kindly
allowed us to use their thin-section reference collections: E. Bettles from the Institute of Archaeology,
University College London (Tyre and Sarepta); D. Griffiths from the Institute of Archaeology,
University College London (Sidon and Tell Nebi Mend). A. Middleton from the Department of
Scientific Research of the British Museum allowed us to use the thin sections belonging to the Gerald
Avery Wainwright Archive of Mesopotamian Ceramics, deposited in the British Museum, Department
of Scientific Research by Syriol Mynors. We are very grateful to all of them.
We would like to thank J-P. Thalmann of the Sorbonne University, Paris, and E. Cooper of the
University of British Columbia, for their useful comments on the clay types in Syria and Lebanon, and
S. Vaughan who read the draft concerning the Alashiya letters for his input on Cypriote clays.
Most of the petrographic examinations were carried out in the Laboratory for Comparative
Microarchaeology of the Institute of Archaeology, Tel-Aviv University. Several petrographic thin
sections were prepared in the sectioning laboratory of the Institute of Archaeology, University College
London, with the kind permission and collaboration of P. Ucko, D. Griffiths and S. Bond. We wish
to thank N. Porat, A. Shimron and A. Sendler of the Geological Survey of Israel, for their useful
advice concerning specific details in the petrographic data. Our appreciation goes also to I. Segal who
performed the ICP analyses and L. Grossowicz who made the palaeontological identifications, both
of the Geological Survey of Israel. M. Huges, A. Middleton and I. Freestone of the Department of
Scientific Research in the British Museum were most helpful regarding the statistical processing of the
chemical data.
We wish to express our gratitude to the Director of the Sonia and Marco Nadler Institute of
Archaeology of Tel Aviv University for making the services of its various departments available
to us and for supporting this publication. Our thanks also to those specialists in various associated
disciplines who contributed to this publication: Michal Artzy of the Department of Maritime Studies,
Haifa University, Hans Mommsen of the Helmholtz-Institut fr Strahlen- und Kernphysik, Universitaet
Bonn, Germany, Lydia Grossowicz of the Geological Survey of Israel, Simha Lev-Yadun and Moshe Inbar
of the Department of Biology, Haifa University at Oranim, Tivon, Vladimir Chikatunov of the Department of
Zoology, Tel Aviv University, Joseph Yellin of the Department of Physics, the Hebrew University and Shlomo
Bunimovitz of the Department of Archaeology and Ancient Near Eastern Cultures, Tel Aviv University. We
thank them for sharing their expertise with us.
The thin section were photographed by Yuval Goren. The maps were ably prepared by Judith Dekel
and Yura Smertenko and the design on the front binding was drawn by Rodika Penchas.
Finally, special thanks are due to Shirley Gassner for her highly professional editing of the
manuscript and the layout of this book.
Yuval Goren, Israel Finkelstein and Nadav Naaman
Tel Aviv March 2004
xviii
CHAPTER 1
INTRODUCTION
In the late 19th century an archive of several hundred clay tablets was discovered at el-Amarna in
Egypt, on the east bank of the Nile, about 300 km south of Cairo (Knudtzon 1915; Moran 1992). This
was the site of Akhetaten, the capital of Egypt for a brief period in the 14th century BCE. The city was
founded in the fourth year of Amenophis IV (Akhenaten) (1350 BCE) and consecrated in his seventh
year (1347 BCE). The royal court left it in the third year of Tutankhamun (1334 BCE) having served
as the capital of Egypt for a total of 13-14 years. Altogether a total of 380 tablets are known today, of
which the majority were recovered by local farmers and sold to museums. Only a small fraction of the
assemblage was unearthed in excavations. The tablets were written in Akkadian in cuneiform script. The
greater part constitutes diplomatic correspondence of the Egyptian royal court, while a small group of 32
tablets are school texts used for teaching and learning Akkadian. Most of the letters were sent to Egypt
from Western Asiatic states and vassal Canaanite city-states. A minority of 12 letters was dispatched
from Egypt to Western Asia. The correspondence of Egypt with major powers includes 44 letters sent
from (and to) kings of independent states: Babylonia, Assyria, atti, Mitanni, Alashiya, and Arzawa.
The vassal correspondence includes letters sent from (and to) rulers of city-states under direct Egyptian
domination, such as Damascus, Byblos, Hazor, Megiddo, Jerusalem, and Ashkelon, or from kingdoms
located outside the connes of the Egyptian Empire in Asia, such as Ugarit, Qidshu and Qatna.
Although representing a fairly short time-span, from about the 30th year of Amenophis III to the
3rd year of Tutankhamun (ca. 1360-1334 BCE), and small in size compared to other Syro-Mesopotamian
archives, the Amarna archive forms one of the most important collections of texts ever found in the ancient
Near East. It supplies unique and invaluable information regarding the diplomatic interaction between
Egypt and the ancient Near East and the territorial disposition and political affairs of Late Bronze Age
Canaan. The archive indicates the existence of an Egyptian administrative apparatus in Canaan through
which a system of semi-independent city-states was controlled.
Fixed epistolary rules dictated the form of the Amarna letters, and whenever Canaanite scribes
corresponded with Egypt they mentioned the name of the king and sometimes his city. Thus, the names
of many rulers and their cities are attested in the archive. However, the names of a large number of
writers, their cities or both, originally recorded on the tablets, are missing. This is due to the severe
damage that the tablets suffered in antiquity, and during the process of the archives discovery and sale
(see Knudtzon 1915:1-15).
The location of many rulers is not mentioned in their letters. Some might have governed in cities
already recorded in the archive, but there has been no way to corroborate this assumption. Some tablets
do bear the name of a city, but its precise location was disputed among scholars. Even the location of
some ancient Near Eastern kingdoms and major cities remained controversial. As a result, the number
of Canaanite city-states and their distribution has been ercely debated (Finkelstein 1996; Naaman
1997). The territorial extension of specic Canaanite city-states, as revealed by the Amarna tablets, was
discussed by Campbell (1965; 1976), Ross (1967), Rainey (1968) and Kallai and Tadmor (1969). The
rst comprehensive and detailed reconstruction of the system of Late Bronze Age polities according to
both textual and archaeological data was proposed by Naaman (1975; 1986; 1988a; 1988b; 1992; see
1
also Helck 1971:188). Bunimovitz (1989:131-161) followed Naamans basic ideas, deploying sociopolitical and geographic methods in order to produce a map of the Canaanite polities. A century of textual,
geographic-historical and archaeological research has claried some of these problems. However, many
issues remained unsettled and traditional methods seemed to have reached a stalemate in resolving them.
It was only a matter of time until scholars would turn to rapidly developing scientic methods for the
study of provenance of clay.
The idea that the Amarna tablets can be studied also by the provenance of their materials is not new.
Already at the beginning of last century, Knudtzon (1915) sought to form an opinion about the origin
of the tablets by inspection of their materials. His descriptions of each group of tablets included notes
on their colour and fabric. Technically, scholars of Knudtzons generation were capable of analyzing the
clay components of the tablets by means of petrography. H.C. Sorby had introduced this method as early
as 1849, and by the beginning of the 20th century optical analysis of the mineralogy of sediments was
already well established. In the 1860s Lartet initiated the geological mapping of Palestine, Syria and
Lebanon. Nevertheless, there was no established database for Levantine ceramic materials to perform a
thorough study.
In the years that passed since Knudtzons work, modern techniques of sourcing ceramic materials
have been adopted by archaeologists. In many cases these mineralogical and chemical methods enable
the attribution of ceramic artefacts to their original location. Since cuneiform tablets are in fact ceramic
objects, the analytical study of the clay of the Amarna tablets can clarify the provenance in which the
tablets were made and thereby the place where each tablet was written. The use of methods adopted from
natural sciences provides an independent testimony to the origin of the tablets that may be correlated
with the data extracted from the texts. Two pioneering provenance studies of several of the Alashiya and
Mitanni tablets from the Amarna archive were carried out in the past. Although preliminary and partial,
these studies demonstrated the potential of provenance examination for the study of the Amarna tablets
(Artzy et al. 1976; Dobel et al. 1977). However, perhaps due to the rather inconclusive results, no further
attempt was made to systematically investigate the source of a substantial number of tablets on the basis
of their raw materials.
Related disciplines have also made major progress in recent decades. Archaeological excavations
of some of the prominent cities which participated in the Amarna correspondence (e.g. Ashkelon,
Lachish, Tell e-a [Gath], Gezer, Jerusalem, Shechem, Ta>anach, Megiddo, Yokneam, Acco, Tell
Keisan [Achshaph], Hazor) were excavated and large-scale surveys shed light on the rural sector of the
Late Bronze Age society. These data form the base for the reconstruction of the territories of the major
city-states in the Amarna period. Systematic geological surveys conducted throughout the Near East
supplied the necessary information about the distribution and composition of clays and soils that were
relevant to pottery production processes in and around the sites. In the eld of ceramic material studies,
analytical methods were introduced; including sophisticated chemical examinations that were developed
in association with advances in geochemistry.
A research programme which would take full advantage of the great progress in scientic knowledge
in all the above-mentioned disciplines was planned. It was designed to examine all the Amarna tablets
systematically and combine the information extracted from the clay with other elds of research. The
latter includes information provided by the text itself and archaeological evidence from the site or region
involved. Research began in 1997 and continued for about ve years during which almost all the available
Amarna tablets (about 300) were analyzed. At the same time the relevant textual and archaeological data
were collected.
2
Taking into consideration the advantages and limitations of the different analytical methods for
provenance studies, petrography was selected as the method of choice. By applying this approach to the
study of the Amarna tablets, the team intended to resolve geographic and historical problems related
to Canaan of the Late Bronze Age. Additionally, we aimed to resolve some wider problems related
to the Amarna archive, such as the much-debated issues of the location of Alashiya and the location
of the capitals of Mitanni, Amurru and Tunip. To accomplish our programme, we made a systematic
provenance study of most of the Amarna tablets from the Vorderasiatisches Museum in Berlin, the
British Museum in London, the Ashmolean Museum in Oxford, and the Muse du Louvre in Paris.1
Regrettably no response was received to our request to examine the tablets housed in the Cairo Museum.
Therefore that component of the archive is not represented in this volume.
In selecting the scientic methods to be applied, we preferred a primary analytical method that
is usually considered by ceramic analysts as being intermediate, that is, between the low powered
descriptive techniques and the very accurate and fully quantitative chemical methods. In doing so, we
took the risk of being criticized by other ceramic analysts for employing an old-fashioned, low-tech
method. However, after long and careful consideration of the benets and limitations of all the research
techniques now available we came to the conclusion that this method would best t our goals. We
invested much effort in the improvement of the sampling procedures and examination processes in order
to adjust it to the specic requirements of clay cuneiform tablets.
The research strategy applied here is derived from the recognition that the best results in
analytical studies on ceramics were usually achieved when a large number of examinations were
performed by a combination of methods. The basic method applied was petrography, supported
by other micromorphological studies (micropaleontology, microarchaeobotany, and in one case
microarchaeozoology). The chemical analyses that we have undertaken were dictated by problems
that emerged in the course of the petrographic study. In the chemical analyses section the full results
of a previous study carried-out by neutron activation analysis (NAA) are included (Chapter 3.V). For
the newly sampled material inductively coupled plasma atomic emission spectrometry (ICP-AES) and
inductively coupled plasma mass spectrometry (ICP-MS) analyses was preferred.
1. A duplicate collection of thin-sections of the tablets held in the British Museum has been lodged with its Department of
Scientic Research.
CHAPTER 2
METHOD AND THEORY
THEORETICAL BACKGROUND
CERAMIC PRODUCTION SYSTEMS AND CLAY TABLETS
In assigning a provenance to ceramic artefacts, the main aspect to be considered is the availability of
ceramic resources in the proposed production location. The quality of the local resources should be
considered in any attempt to evaluate the technological characteristics of a workshop (Arnold 1985:20).
These two axiomatic rules are taken as being applicable for clay cuneiform tablets too, as at least in terms
of the starting raw materials they are indifferent from any other ceramic commodity.
When discussing the workability of ceramic raw materials, one should distinguish between their
different types. By denition, any ceramic material is composed of a plastic component (clay) that is
expected to sinter upon ring. The other component that may be added to it is a non-plastic one, commonly
termed temper, inclusions or aplastic component (Shepard 1965:6-54). Other ingredients used in
pottery production, such as paint, slip and glazing materials, are irrelevant to our discussion.
The term clay has two different denitions. Mineralogically, it refers to a certain group of
minerals characterized by particular composition (hydrous aluminum silicates, usually belonging to
the phyllosilicates group). From a sedimentological point of view this term is used to categorize the
nest particle size. In ceramic nomenclature these denitions are sometimes confused with each other.
However, in terms of ceramic materials the dual meaning of clay may reect two separate geological
and technological categories: clay minerals appearing in situ as deposits within geological formations,
or sediments containing varying proportions of clay minerals. The rst may develop from the chemical
alteration of other minerals (such as feldspars) and produce beds of pure clay that may be used for
stoneware or porcelain production. The latter are deposits originating from weathering processes of rocks
or secondary deposition of sediments, thus impure in composition and suitable only for earthenware
production. They may be geologically translocated or locally reworked, thereby being subjected to
admixture with other ingredients that are not clay minerals. Such materials were almost exclusively used
by potters in the Ancient Near East.
Many authors have dealt with the process that causes the conversion of clay into a ceramic product
(e.g. Shepard 1965:19-24; Rice 1987:80-110) hence we shall not discuss it here. Several factors dictate
the suitability of sediment for pottery production. Firstly, at least 50% of its total volume should be
composed of clay minerals. Secondly, it requires a sufcient degree of plasticity to allow it to be formed
into the desired shapes while wet. On the other hand, over-plasticity may cause shrinkage and cracking
in the process of dehydration.
The term temper refers to the coarse-grained components which are added by the potter to the
clay to make the paste to be worked. Any solid gritty material may be used for this purpose, including
sand, crushed stone or brick, chopped straw, grog, slag, hair, crushed bones or shells. Temper is added to
the clay in order to reduce its shrinkage in drying, lower the vitrication point in ring and increase the
strength of the red product. From a technical point of view, the term inclusions is preferable since this
4
does not differentiate between intentionally added grits and naturally occurring clasts within the clay.
The properties of non-plastic inclusions and impurities are thoroughly discussed in the literature (e.g.
Shepard 1965:24-31, 53-54; Rice 1987:72-75) and will not be repeated here.
Little if anything has been preserved in the historical record about the technological procedures and
preferences of scribes who produced clay tablets. In contrast to ceramic production, this technology is now
extinct and hence our knowledge cannot be supported by any ethnographic study. There are obviously
some fundamental differences between clay tablets and ceramic vessels, and consequently between scribes
and potters. Therefore, the models applied for pottery production systems are not necessarily applicable
for tablet fashioning. First and foremost, unlike pottery that is usually mass-produced for wide distribution
and for a broad spectrum of functions, each tablet is unique. They are produced for one purpose only
and usually for the very limited use of a few particular individuals. Pottery is made for all segments of
society whereas it is logical to assume that only the elite used clay tablets. Accordingly, in most cases
potters were ranked as being of a low social status, while scribes were certainly classied as a specializing
bureaucracy of high or middle rank. Undoubtedly, these differences may reect on the validity of any
technical comparison between tablets and pots, regardless of the supercial similarity in the raw materials.
Yet despite these stipulations that must always be borne in mind, some terminology and theory extracted
from ceramic studies must be adopted since this is the only related and well-understood technology that is
relevant for the present study.
When dealing with acquirement and use of raw materials it is assumed that sophisticated and skilled
potters tend to be more selective in choosing their clay and temper. Conversely, low-scale production is
usually based on random use of the raw materials at hand. Such an assumption is based on ethnographic
data (e.g. Nicklin 1979, Arnold 1985:32-57; Rice 1987:177-180). These studies also demonstrate that
potters located adjacent to outcrops of high quality clays tend to specialize and dominate the markets
more easily (Rice 1977; 1981). Therefore, expert potters tend to improve the choice and quality of their
raw materials and adjust them to the desired function of each vessel type. This observation applies to
vessels that are proposed for a specic set of functions (e.g. cooking-pots, storage jars, etc.), and are
expected to meet with special requirements, such as resistance to diffusion of liquids and thermal or
impact shocks. Such specic adjustment cannot be obtained for vessels designed for multi-purpose
use. Consequently, there should be a divergence in terms of raw material selection between highly
specialized and non-specialized ceramics. In highly specialized pottery one might expect to nd a very
selective choice of raw materials, whereas non-specialized ceramics may reect a more opportunistic
choice of clay and temper. This differentiation has an immediate effect on provenance determinations,
since a very selective choice of clay and temper may not reect the full geological diversity of the
production area. In contrast, opportunistic choice of materials may inform more about the environment
and less about the technological abilities of the potter, such as his familiarity with the properties of clay
and temper. For this reason, the examination of cooking-pots, for example, is useful for the judgment
of the technological abilities of a potter but not so much for provenance determinations. Conversely,
simple vessels such as bowls and basins may be advantageous for provenance studies but not for the
technological ranking of a workshop.
If clay tablets are considered to be another type of extremely specialized ceramic artefact, they may
be expected to illustrate a selective choice of raw materials. Hence by denition, tablets may supply more
limited information about their provenance when they were carefully made. The obvious requirements
from clay used for cuneiform tablets are that it does not contain too many grits that would hamper the
execution of clear incisions, and that it will dry hard without fracturing, crumbling or shrinking markedly
5
(Artzy et al. 1976: 179). The colour of the clay should be bright enough to highlight the script, be it
stained or incised (and thus highlighted by the shading of the slits). Therefore, the preferred clay for
tablets should be bright in colour (such as marl), with a low shrinkage rate (thus poor in clay minerals
such as smectite or montmorillonite), and include some ne temper (to enable drying without cracking)
but smooth and with no large grits or bres. Since these requirements are not necessarily requisite for
pottery vessels, one might expect a difference between production modes of pottery and cuneiform
tablets even within the same site.
CERAMIC ECOLOGY, SITE CATCHMENT ANALYSIS AND THE MEANING OF LOCAL
A large body of ethnographic literature regarding pottery production systems tends to associate the
choice of raw materials rst and foremost with the availability of clay and temper around the production
site. As Arnold (1985:20) points out, the resource feedback mechanism can be broken down into two
parts: (1) the appropriateness or quality of ceramic resources; and (2) their availability as measured in
the distance to their source locations.
A detailed survey of the ethnographic data concerning present-day traditional potters indicates that
the exploitable territory of clay and temper is usually limited by the expenditure of time and distance
to the raw material sources. Based on 111 round the world reports on clay exploitation by potters and 31
cases regarding temper use, Arnold (1985:32-60) concludes that in most cases the exploitable territory
does not extend beyond a range of 10 km from the potters location. Thus, drawing a circle with a radius
of 10 km around any given site will establish its available clay and temper resources. Within this circle,
possible clay and temper types can be determined in order to predict the Clay-Temper Factor (CTF) of
the site, namely the quality and quantity of possible clay and temper sources in the exploitation territory.
In other words, the ceramic ecology approach suggested by Matson (1965) should be applied rst and
foremost within this area. This approach is similar in principle to the concept of site-catchment analysis
namely the analysis of the morphological and economic resources of the territory around a site
(Vita-Finzi 1978:23-31, 71-88) suggested as a method for evaluating prehistoric foraging strategies.
Similarly, Christaller (1966:159-160) established the size of the lowest rank territory for his central place
theory on the basis of an hours walk from the site, by calculating distance and energy expenditure.
If we take all these models, we can suggest a more accurate de nition for the term local so
commonly used by archaeologists and ceramic analysts, frequently without any regard to its denotation.
Any artefact that has been produced of materials that are within an exploitable range of the site can be
considered local. The exploitable territory is dependent on the product type, and its radius will change
from one material type to another. Consequently, the exploitable areas for glass, pottery and metal will
differ in size. As we have seen, for ceramics we can tentatively place it within the general range of 10
km from the site. This circle can be constrained in certain directions by the presence of natural barriers
such as swamps, cliffs or steep gorges. These occurrences can be delineated today and considered in
plotting the ceramic exploitable territory of the site. In the past there could have been other factors,
invisible today, that could limit the choice of natural resources. These include political relations with
neighbouring territories, as well as ideological and religious stipulations that may have forbidden the
use of certain types of clay and temper (see for example Barley 1994 for the various taboos on pottery
production in present-day Africa).
It is possible that there were certain taboos and beliefs limiting the choice of clay used for tablets.
While we have no notion today as to the considerations that directed Near Eastern scribes to utilize
particular clay types [good] and disregard others [bad], there might have been political limitations
6
which separated a scribe from a good clay resource within the exploitable territory around his site. The
tablets themselves supply some obvious evidences for such limitations. For example, although the upper
Shephela lies in the exploitable area for Tell es-Sa (the probable site of Gimtu = biblical Gath), the
letters of Shuwardatu (EA 279-280) reveal that this area was rst threatened by groups of rebels and
later held by the ruler of Jerusalem. It is unlikely that Gaths scribes would venture into this territory
in order to acquire clay. Moreover, exploitable areas of bordering city-states could have overlapped but
been separated by a border that can no longer be reconstructed. Thus in order to evaluate the possible
exploitable sources of clay of each Canaanite city-state we must be able to dene its territorial scope.
The geographic and political situation in the ancient Near East could potentially extend the range
of exploitable territory for clay far beyond the radius of 10 km. Already in his above-mentioned model,
Arnold (1985:32-60) realized that the presence of a waterway by the workshop site could enable potters
to sail and collect their materials along it. In such cases, the exploitable area is always less than one
hours walk from the stream, but it could be more than 10 km away from the site along it. This possibility
is relevant in the case of Egypt, Babylonia, Assyria and Mitanni, but does not apply for Levantine citystates apart for a few cases, such as a city-state located along the Orontes River.
PREDICTION OF MATERIALS AVAILABILITY: CLAY-TEMPER FACTOR (CTF)
Although clay (or matrix in petrographic terms) can be described in some detail through ceramic petrology,
it is the inclusions that supply a greater part of the information. Inclusions are also more resistant than
clay to selective choices by the potters (hence also by scribes), since the requirements from them are
simpler and therefore their collection is more opportunistic. While in the clay the mineralogy, plasticity,
shrinkage and sintering rates are very important, the essential requirement from inclusions is merely to
act as an aplastic component and thus any solid matter is potentially suitable for the task. It is only when
special requirements, such as resistance to impact and repeated thermal shock are desired, that some
inclusion types are advantageous upon others (cf. Bronitsky and Hamer 1986). In such cases potters may
prefer largely sorted inclusions, such as crushed calcite or basalt, grog, slag, chopped straw and crushed
shells, all reported as being efcient in such cases (ibid.; Arnold 1985:24). These articially prepared
materials that reect technological abilities but are rather ineffective for provenance determinations, may
be dened as artefact inclusions. In contrast, inclusions that were collected from the environment and
mixed with the clay without any signicant preparation can be dened as ecofact inclusions. Only these
are effective for provenance determination.
Since the technical requirements of impact and thermal shock seem to be irrelevant to clay tablets, we
can expect to nd in them mostly ecofact inclusions. If artefact inclusions do appear, it may be suggested
that the local pottery workshops clay had been applied for tablet production (see for example EA 219).
Our research indeed indicates that these cases are extremely rare. Still, as tablets cannot be too gritty, we
may expect the inclusions to be sieved, thus preserving only the ner fraction and eliminating rock types
and minerals that appear in the coarser fraction of the sand. Since sand is common almost everywhere,
we can minimize the exploitable area for the inclusions into less than 10 km. Indeed, Arnolds model
indicates that in most cases the exploitation distance for inclusions is smaller than that for clay. When
sand is used for temper, the distance usually does not exceed three kilometers (Miksa and Heidke 1995:
134, Fig. 9.1).
In conclusion, while the denition of the exploitation territory for the clay of tablets may be
complicated, their inclusions are likely to be collected from the immediate surroundings of their
production site.
7
Applying all the above considerations, the clay and temper types that can potentially typify local
ceramics of each site under investigation should now be dened. These data can outline the unique
petrographic ngerprint of a site, thus helping to isolate its products and differentiate them from those of
other sites. Therefore, the composition of any tablet that is attributed to a given site may be matched against
the pattern of the local clay and temper in order to determine a priori whether it can be local to it. Moreover,
tablets of unknown origin can be compared with these data in order to nd corresponding sites.
By drawing the tentative exploitation territory of a site on a detailed geological map and using
the relevant geological literature concerning the area, clay formations within it can be identied and
the local lithology (the potential ecofact inclusions) de ned. Additional information can be extracted
from soil maps that may add information on the available clay resources. By listing the possible clay
and temper types that expose within the exploitation territory of the site, we can estimate the indices
of the overall available resources. The latter can be presented by the equation: CTF = NC x N T, where
NC is the number of available clay sources and N T is the number of available temper types. In theory,
the value CTF may indicate the maximal number of local petrographic groups that can appear within
the exploitable territory of a given site; in other words: the number of all the possible combinations of
clay and temper at this site. For example, a site where only one sort of clay and one potential temper
type exist, will have a CTF index of 1, since NC x N T = 1 x 1. A site where two potential clays and three
temper types exist will have a CTF index of 6. In the latter case, both NC and N T need to be indicated by
numbers. While the clay can be readily de ned and restricted geographically, the temper may appear as
various types of sand since in reality potters tend to collect available and already naturally sorted sands
in order to minimize the time-consuming task of sieving the grits. Sand can appear in many forms, i.e.,
as arkose (grains usually of granitoid igneous rocks preserving the composition of the mother rock),
as grains from a talus, as weathering products of sandstones, as detrital deposits transformed by water
channels or by the sea, etc. It is more useful to de ne the potential sand as a whole than to list the rock
types that expose within the exploitable area (see Lombard 1987; Gilead and Goren 1989; Miksa and
Heidke 1995 for further discussion on this issue). Therefore, N T is the number of available sand types
within the territory. Since in many cases the sand can be collected from the nearest riverbed, that in turn
may extract rock types that expose more than 10 km away from the site, the river sand as it appears by
the site can be considered as one category that is de ned by its compositional traits (ibid.).
In practice, certain sites can have odd CTF values. While several areas can be rich in possible clay
and temper types, others are poorer or even have only one component. For example, the Nile Delta is
very poor in potential clay deposits, as the only likely material around it is the Nile mud with its naturally
occurring sand. Therefore, the CTF of a site such as Mendes is expected to be 1. On the other hand, a site
located in an extremely heterogeneous geological environment may have very high CTF values. Such is,
for example, the situation at the site of Kalavasos in Cyprus (to be discussed in relation with the Alashiya
tablets) where the CTF value can exceed 10. In Palestine, a site such as Ashkelon can be used as an
example for CTF = 1, whereas Megiddo can represent a site with a high CTF index. In the latter, a circle
of 10 km in radius includes Taqiye shales, alluvium from the Jezreel Valley, Rendzina soil that develops
on the Eocene chalks, clays that are derived from the weathering of basaltic outcrops, and Senonian marls
of Wadi >Ara. Put together with the NT value it can easily exceed an index of CTF = 10.
This phenomenon has two important implications. Firstly, at certain sites more than one type of local
fabric can be expected. At a site such as Megiddo there may be ve or six different fabrics, all equally
local though different from one another. Secondly, in sites where CTF = 1 the same fabric may be used
continuously throughout the entire sequence of pottery production. For example, various petrographic
8
researches testify to such continuous use at Ashkelon. Loess with the local coastal sand had been used
from the 5th millennium BC, through the Chalcolithic and Early Bronze Age at Afridar (Goren 1992:
Appendix 2; Cohen-Weinberger, pers. comm.). They typify the local wares of the Middle Bronze,
Late Bronze, Iron Age and Persian period at Tel Ashkelon (Cohen-Weinberger, pers. comm.; Master
2001; Gorzalczany, pers. comm.), and continue to be present in the mass production of the Gazatian
and Ashkelonian jars during the Byzantine Period (Fabian and Goren 2002). On a larger scale, the
continuous use of Nile mud in Lower Egypt from the Predynastic period to the present may indicate a
similar occurrence. In such cases, the study of pottery production of one period can be translated into
other periods as well, at least in terms of the material composition, since no transition in the choice of raw
materials is likely to happen. For this reason, wasters of Byzantine workshops from Ashkelon, Ashdod,
Tel Jemmeh, Tel Haror and other sites in the vicinity where CTF = 1, could be used for petrographic and
chemical comparisons with the Amarna tablets. In sites where CTF > 1 this approach is inapplicable.
METHOD SELECTION
Various techniques are employed for analyzing the composition of pottery and other ceramic artefacts
(see Barclay 2001 for a brief review). They can be divided into physical and chemical methods. The
physical methods identify the minerals in the clay and temper, and dene the texture and fabric of the
sherd. The chemical methods use diverse analytical techniques to measure the concentrations of the
chemical elements present. In pottery analyses, petrography is the most commonly used physical method,
whereas Neutron Activation Analysis (NAA) is the most common chemical method. Petrographic
analysis is particularly useful for examining coarse, poorly red ceramics, whereas chemical methods
are generally considered to be more accurate for provenance determinations, being fully quantitative and
thus more precise. Additionally, chemical methods may seem to be more appropriate for the examination
of delicate items (such as tablets) due to the smaller sample required.
Naturally, every analytical method has its own advantages and limitations. It is beyond the scope
of this monograph to specify the details for each method and its suitability for studying ceramic objects
(see Rice 1987:309-445 for a summarizing review). Aside from the technical limitations of each method,
there are limitations in interpreting the analytical data. If our major goal is to assign a provenance to an
artefact, the quality of our interpretation depends heavily on the availability of comparative materials
and on the quality of our reference database. Therefore, not every item examined can be successfully
interpreted and denite answers that would resolve debated matters are often not available. In many
cases, the results can only direct the archaeological interpretations into a narrower range of options.
It is now accepted among archaeomaterial scientists that the best results are likely to be achieved by
combining several complementary methods. Usually, petrography is applied to a large number of items
and the results are used to select samples for further chemical analyses (e.g. Schubert 1986; Tite et al.
1990). This approach, so suitable for the study of pottery assemblages, is less relevant for the study of clay
tablets where the amount of the examined items is limited and each item is a particular case that poses a
unique set of questions. In addition, the examination of clay tablets calls for a somewhat different set of
methodological approaches. First and foremost, they might have been produced from clay types different
from those used for manufacturing pottery vessels, even within the site of their origin. Therefore, it is
essential to investigate this possibility prior to any attempt to use the routine methods commonly applied
for pottery sourcing, in which databases containing the elemental composition of reference pottery from
many sites are compared with the examined samples. In such a case this approach will be unsatisfactory
9
since the composition of the sample is not likely to match any known clay source that was used for the
production of ceramic vessels. In fact, our study of the Amarna tablets indeed revealed some particular
cases of this kind. For example, Egyptian tablets were usually produced of marl of the Esna Formation
that has hardly ever been used for pottery production. The letters of Biridiya from Megiddo were made
of marl of the Ghareb Formation which was never used for pottery production in the Levant. In such
cases, the standard methodology of NAA would not resolve the question of provenance. The origin of the
clays could be suggested by petrography due to their geological age (as determined by their foraminiferal
assemblages), mineralogy, and the lithology as reected by their accompanying clasts.
In this respect petrography has the advantage of being independent, in the sense that in cases when
a reference pottery database is not available, the results can be interpreted on the basis of detailed and
usually available geological maps. This is in contrast to interpretation of the chemical analyses which
rely blindly on a comparative database of trace element composition from standard samples. Hence,
although petrography does not have the accuracy of chemical analyses, it does not depend on incomplete
databases and reference groups which could have been poorly selected or unevenly spread in the area
covered by the research. Moreover, since ceramic materials are composite, petrographic analyses
supply a combination of details, each one of which can be related to a specic geological or geographic
environment. By combining the different geological features, a much better constrained geographic
source can be dened.
For these reasons petrography was selected as the primary method for this research. This decision
was triggered also by the availability of a large collection of comparative material from many major sites
in the Levant, enabling assignation of many fabrics to their geological context and thereby facilitating
provenance determination. In the last decades Goren was engaged in the examination of thousands of
Levantine pottery vessels by this method, which led to compilation of the largest available database of thin
sections for petrographic investigation in the Levant. The reference collection includes the thin sections of
other scholars (J. Glass, N. Porat) and contains over 10,000 samples from the Levant and Egypt.
Previous attempts to examine tablets from the Amarna archive by means of NAA (Artzy et al. 1976;
Dobel et al. 1977; see also Hellbing 1979:71-2; Gilbert N.D. reported to us by C. Walker) produced
indecisive results due to several methodological problems which are discussed below. Therefore,
introduction of other chemical methods was deemed advisable, and inductively coupled plasma atomic
emission spectrometry (ICP-AES) was combined with inductively coupled plasma mass spectrometry
(ICP-MS). Although these methods still lack a well-established database for Levantine ceramic
resources, they are more available, being independent and not reliant on nuclear reactors. Moreover, the
combination of the two ICP methods has a great potential for provenance studies since it supplies a wide
range of data, including major, minor, and trace elements. In any event, the petrographic results showed
that the raw materials that were used for ceramic production were not always similar to the ones that
were selected for clay tablets. This is another reason why standard elemental analyses should be used
with caution and tablets should be compared with other tablets and not with pottery, unless petrography
suggests a similarity between the two. Hence we used the ICP data mostly for examining the grouping of
tablets by their element composition.
The Amarna project required the use of comparative material on clays from the entire Near East:
Egypt, Canaan, Cyprus, Western Syria and Mesopotamia. The results of the detailed technological
research of Egyptian pottery assemblages and the increasing data on Egyptian ceramic raw materials
(cf. Tobia and Syre 1974; Allen et al. 1982; 1989; Arnold and Bourriau 1993), indicate a consistent
continuity in the use of primarily two main classes of raw materials (Nile mud and various types of
10
marl clays) throughout the periods. They also enable excellent differentiation between Canaanite and
Egyptian materials. As for the Canaanite pottery, the situation is less clear, since the examination of
Canaanite Late Bronze Age assemblages from the Levant have so far been random and inconsistent.
Furthermore, Canaanite potters had access to a greater variety of clay types and therefore Canaanite
pottery tends to be far more composite in its raw materials than Egyptian wares. This shortcoming was
however overcome by reference to the extensive database of pottery of all periods. To this end we used
the petrographic thin-section collection of the Institute of Archaeology, Tel-Aviv University, mainly for
comparisons with Palestinian site assemblages. For Syrian, Lebanese and Mesopotamian assemblages we
used the collections of the Institute of Archaeology, The University College London, and the Department
of Scientic Research of the British Museum. In the case of Cypriote materials, we used the available
petrographic data (e.g. Vaughan 1991) as well as materials that were collected and analyzed especially
for this research project (Chapter 3.VI).
SAMPLING PROCEDURES
One of the main disadvantages of petrography is the large sample size required. Therefore, although the
tablets were examined by standard petrologic means, new sampling techniques were developed due to the
importance and delicacy of these artefacts. These include SPA, Peeling, and Blocking, described below.
In the catalogue of the examined tablets the sampling method applied for each item is specied.
SCATTERED PETROGRAPHIC ANALYSIS (SPA)
The usual sampling method used in petrographic studies can be destructive for clay tablets. Therefore, in
the pilot phase of the study a new sampling and examination method was developed by Goren and named
Scattered Petrographic Analysis (henceforth SPA). This method is almost non-destructive, requiring a
sample of only a few milligrams that can be taken as tiny grains from one or several hidden or fractured
spots in the artefacts surface. The sampling process does not require drilling or slicing of any part of the
artefact, and the sample is taken entirely supercially. This method was utilized on complete or nearly
complete items that could not be sampled by the less elaborate method of peeling (below).
In principle, SPA is a petrologic method based on the examination and identication of the matter
under a polarizing microscope. However, it differs in many respects from traditional petrologic methods.
In ceramic petrography slices of the examined object are cut by a diamond saw in order to obtain an
adequate representation of the clay matrix and the inclusions (or temper). While the clay matrix can
be dened in quite small portions, the size of the sample is usually governed by the need to include
a recognizable amount of inclusions. Therefore, a slice of at least 10x5 mm (but usually far more),
should be removed from the object, depending on its homogeneity and the amount and distribution of
inclusions within it. In SPA the two components are sampled separately. First, the artefact is thoroughly
examined under a stereomicroscope using Stienstras (1986) methodology in order to dene its fabric. A
tiny ake of the clay matrix (ca. 1x1 mm) is chipped from a previously fractured surface using a scalpel.
The inclusions exposed on the surface of the object are identied under the stereomicroscope and a
representative sample dragged as single grains (usually from the edges of the tablet) using a dental tool.
In the laboratory both the clay sample and the inclusion grains are set in a small polyethylene mould
(ca. 1 cm in diameter) and dried in an oven at 600C for a few hours. Under vacuum conditions, they are
impregnated with low viscosity epoxy resin (either Hillquist thin-section epoxy type C-D or Buehler
Epo-Thin epoxy). After curing, the pellet is used for the preparation of a standard thin-section and
subjected to routine petrologic examination.
11
By applying both traditional ceramic petrology and SPA on the same sherds, the information obtained
by the two methods was compared. In most cases, SPA proved to be almost as powerful as regular ceramic
petrology, although far more time consuming in terms of sample preparation and examination.
In order to investigate the potential of SPA when applied on cuneiform clay objects, two newly
found tablets from Hazor were examined (Goren 2000a). This experiment, the rst step in the current
research, proved to be valid for indicating the origin of each letter. In a second pilot study, seven tablets
from the Amarna archive, now in the Ashmolean Museum in Oxford, were examined by SPA and by
peeling. The results proved that in this case too, the provenance of the tablets could be disclosed.
PEELING
This method was applied on tablets which had broken surfaces and thus could supply larger samples than
those retrieved by SPA without causing any damage to the artefacts. In the present study it was the most
commonly used technique.
A shallow lamina, about 5x5 mm or slightly more depending on the size of the natural fracture on
the tablet, was peeled off the broken area of the tablet with the aid of a scalpel. Since in many cases
cuneiform tablets are very lightly red, cutting such a slice was usually simple. The dust and crumbs
that fell off during this action were collected separately and kept in sealed test tubes for the elemental
analyses. The lamina was immediately impregnated in a small plastic cap with Buehler Epo-Thin
epoxy resin to prevent desegregation and crumbling due to its fragility. In order to let the sample at least
partly absorb the epoxy glue, the plastic caps with the samples and fresh glue were placed in a small glass
dissector where vacuum conditions could be created. A regular basketball hand pump equipped with an
inverted valve proved to be sufcient for this task. This process could be undertaken on the spot in the
museums where the tablets were sampled. The samples were cured on-site on a mini-hotplate. After
curing, the samples were packed and transported to the laboratory.
In the laboratory, the pellets containing the samples were used for the preparation of petrologic
thin-sections. These were made in a protracted process designed to ensure the best possible results. Each
pellet was rst dry polished to the centre of the sample on a grinding wheel with embedded 320-mesh
diamond powder, parallel to its longest axis. No cooling liquids were used at this stage because in cases
of unred tablets the clay did not completely absorb the epoxy resin and it could potentially wash away in
places. Then the pellet was impregnated again with thin-section epoxy that was heated in 600C to form a
liquid, this time under vacuum. This was intended to allow complete absorption of the glue. After curing
the sample was ground again and used for the preparation of a petrologic thin-section.
BLOCKING
This technique is more intrusive than the previous ones. It was used only in few cases in the preliminary
stages of the research, before the two above methods were developed. In this method, a dental diamond
plated disc saw is used for cutting two parallel incisions about 3-mm apart on a broken facet of the
tablet. The slice between the cuts is then broken with a scalpel and used for the preparation of the thinsection. This method is effective when well-red broken tablets are analyzed. However, it is impractical
for sampling complete or well-preserved items. In this study it was used for examining a few Middle
Bronze Age tablets from Hazor that were probably red during the burning of the room in which they
were discovered (Goren 2000a).
12
ELEMENTAL ANALYSIS
INDUCTIVELY COUPLED PLASMA SPECTROSCOPY (ICP)
Elemental analysis of selected tablets was performed in order to characterize the intergrouping of the
documents and in the case of southern Palestinian tablets, to compare them to the database of southern
Palestinian workshop wasters. In most cases the samples were collected from the sediment that crumbled
from the tablets in the process of peeling. In several instances it was found advisable to increase the size
of the sample in order to enhance the sensitivity of the chemical analysis and additional material was
taken. Still, in most cases the samples were by far smaller than the 250 mg of material that is commonly
recommended for ICP analysis of ceramic materials. Indeed, the resulting values of the analyses included
some elements with concentrations under the limits of detection, but these were a minority. In most cases
the elemental analyses supplied sufciently accurate data.
The tablets were subjected to two sets of examinations. Inductively Coupled Plasma - Atomic
Emission Spectroscopy (ICP-AES) was performed using a Jobine Yvon JY-48 polychromator. This
method was used to analyze most of the major and minor elements (Mg, Al, Si, Ca, Ti, Mn, Fe, P, S) as
well as some trace elements (V, Cr, Co, Ni, Cu, Zn, Sr, Y, Ba, Be, La). Precision of the analyses was as
follows: for major elements 1%, minors - 3% and traces - about 10%. However, due to the small size of
most samples the results of several elements were inaccurate and omitted from the list. A Perkin Elmer
Sciex Elan 6000 Inductively Coupled Plasma Mass Spectrometer (ICP-MS) was used to determine
all rare earth elements (Ce, Pr, Nd, Sm, Eu, Tb, Dy, Ho, Tm, Yb, Lu). All chemical analyses were
carried out in the Geochemistry Department of the Geological Survey of Israel. Since ICP results are
incompatible with NAA data (Porat et al. 1991), the former were not compared with the results obtained
by Artzy and her colleagues.
NEUTRON ACTIVATION ANALYSIS (NAA)
Michal Artzy
The hypothesis in the original attempt at the Lawrence Berkeley Laboratory was that Neutron Activation
analysis (NAA) was sensitive enough to handle the elements and trace elements in the clay, determine a
considerable number of elements encompassing chemical diversity, handle large numbers of clay sources
pertaining to particular archaeological problem (Perlman and Asaro 1969). The accuracy required needed
to be no better than the homogeneity of each clay source. At the time the work started it was assumed that
a large number of specimens have to be analyzed for a single problem (Perlman et al. 1972). The hope
was that with time, as more analytical information was gathered, the corpus of data would then be used
for subsequent studies.
By the 1970s adequate equipment was available in the NAA laboratory at Berkeley to measure
Gamma-rays, a standard for the analysis of ceramics was developed and a highly qualied staff centreed
around Isadore Perlman, was gathered which included Frank Asaro, Helen Michel, Duane Mosier and
Harry Bowman. Problems of ceramic provenance were approached and solved in a satisfactory manner
(Asaro et al. 1971; Karageorghis et al. 1972). It was at that time that the rst large study, that of the so
called Palestinian Bichrome Ware, was undertaken (Artzy et al. 1973). Luckily for that study, a rather
large data base, for that time, was available for comparative analysis. It was based on wares which were
sent to the Lawrence Berkeley Laboratory by Einer Gjerstad, and the Mediterranean Museum who had
carried out extensive excavations in Cyprus. Another source for the data bank came from the collection
of ceramics from the coastal area in Israel, especially the excavations of Ashdod by Moshe Dothan. This
13
was part of a study which attempted to localize the production of the Philistine wares and compare them
to the ware named Mycenaean IIIc1. In the last part of the 1960s Perlman was determined to collect
samples and with the co-operation of the Department of Antiquities of Israel and Vassos Karageorghis,
the head of the Department of Antiquities of Cyprus, had a substantial number of analyses, which were
being added to the bank.
In 1973 it was decided to add clay tablets to the provenance studies. It was assumed that many clay
tablets actually divulged their physical origin in the text. A grant of the United States National Science
Foundation made it possible to travel to museums and collect samples of local clays in Cyprus and Israel
as well as wares from different museums in Europe. It was then that the rst samples of the Alashiya
tablets in the British Museum were collected. It was the period in which the excitement of the new
application was at its height and Richard Barnett had the foresight to allow the sampling of two of the
Amarna Letters, EA33 and EA35, a third tablet, which looked to the naked eye different from the other
two, was withheld since it was smaller and it was feared that the sampling might harm it.
The location of Alashiya was being debated by scholars at the time and Enkomi was assigned the
honour to such an extent that on road signs on the east coast Enkomis name appeared as Alashiya.
Because of archaeological disparity it was thought that an answer could be gained by NAA. The data
base available was enough, although the collection of the comparative samples continued, just in case
the answer was negative and other sites would have to be considered. Indeed Enkomi was not Alashiya,
at least the tablets analyzed did not originate in there.
The importance of the Amarna letters for localizing unknown sites continued. Following a rare
concession by the director of the Pergamon Museum in Berlin, Prof. Dr. Gerhard Meier, two visits were
undertaken, in 1975 and 1976. The samples consisted of Amarna letters from Byblos, Jerusalem, Akko,
Megiddo, two from Alashiya and one from Arzawa. At this time changes Perlman left Berkeley to
establish an archaeometry laboratory at the Hebrew University in Jerusalem and I left for the University
of Copenhagen and eventually University of Haifa. The results of the analyses of the tablets gathered at
the Pergamon were never published although the results were collated and evaluated.
DATA MANAGEMENT
SAMPLE EVALUATION
The density of the inclusions varies from tablet to tablet. Therefore, samples of similar size taken from
two different items may vary in their ability to reect the full range of components within each of them.
For example, a 5x10 mm slice taken from a tablet where the inclusions are small and dense, and a similar
sample from a tablet with large and scattered inclusions, do not equally represent the range of materials
present. Differences in the inclusion assemblage can also affect the quality of a sample. For instance, two
samples of the same dimensions, one taken from a monotonous/uniform tablet (having only one inclusion
type) and the other from a heterogeneous item (including various types of inclusions), will differ in their
representation of the whole.
For these reasons, reliability of the results from each sample was estimated based on its nature and
size. These are graded as follows:
High: The sample size is large enough to concur with standard petrographic procedures, i.e., it supplies
the whole range of features expected to be found in the sampled specimen. This is judged by viewing
at least four complete elds through the microscope at X100 magnication (eld diameter: 2.25 mm)
and seeing in each two combined elds a complete range of the features.
14
Satisfactory: The sample size is large enough to supply three complete elds through the microscope at
X100 magnication, where the complete range of features recurs in at least two.
Moderate: The sample size supplies two complete elds through the microscope at X100 magnication,
but apparently not the complete assembly of features.
Fair: The sample size is extremely small but supplies some useful petrographic information.
Unreliable samples were neither described nor discussed.
TERMINOLOGY AND DESCRIPTION OF THE PETROGRAPHIC DATA

PETROGRAPHIC GROUPS
The samples taken from the tablets were divided into petrographic groups. A petrographic group
encompasses items which share similar petrofabrics. In pottery analysis this classication is determined
according to the qualities of raw materials alone, regardless of variables such as typology, chronology
and geographic location of the site. Therefore, it can serve as an independent technical criterion for a
comparative assortment of ceramic assemblages (Porat 1989b; Goren 1992, 1995, 1996a).
The petrographic groups which characterize the pottery of the Levant and correlate them with
specic geological environments have been dened in previous studies (Gilead and Goren 1989; Goren
1987, 1991a, 1992, 1995, 1996a; Goren and Fischer 1999; Goren and Zuckerman 2000; Greenberg and
Porat 1996; Porat 1984, 1987, 1989a, 1989b). These groups bear typical attributes which enable the
reasonable assessment of their geographic origin, and thus may be used for provenance studies.
FIRING TEMPERATURE
Examinations as to whether the tablets had been red or not, as well as estimates of the ring temperatures, were
done by the procedure commonly used in petrographic studies of pottery (cf. Porat 1989b:30). Temperature
determinations are based on the fact that different minerals tend to change their physical structure at varying
temperatures, which affect their optical properties as observed through the petrographic microscope. If the
mineral suite of a given ceramic material is sufciently varied these changes would occur in more than one
mineral. The ring temperature can be estimated as a range between the temperature that affects minerals that
have altered, and that which should have affected minerals that have not.
The main criteria for ring temperatures applied in this study were:
1. Gypsum and anhydrite crystals decompose at around 1500C. However, in arid zones (such as
el-Amarna) these minerals re-crystallize in voids within the clay of the tablet as part of the postdepositional processes. Therefore this criterion should be used with caution.
2. Vegetal material (like plant tissues, straw, etc) chars at around 5000C. After being charred,
the matter remains unaffected until much higher temperatures are reached (London 1981).
Uncharred vegetal material indicates that no signicant ring process has taken place.
3. Dolomite alters into calcite at 5000C. The discrimination between dolomite and calcite by
petrographic means is difcult, but can be done when a sample is being stained by Alizarin RedS (Friedman 1959).
4. Mica minerals become misty in thin section at 5000C and above.
5. Glauconite changes its pleochroism from greenish to yellow at around 5000C. With the increase
of temperature, it changes through orange to deep red, depending on its iron contents.
6. Calcite undergoes structural changes from about 7000C and above, depending on its purity. At
about 8000C and above the calcium carbonate begins a process of decomposing into calcium
15
hydroxide, completely destroying the original crystal structure of the former calcite. The
temperature in which this process takes place depends on the presence of impurities in the
carbonate (Shoval et al. 1992).
7. Hornblende alters into oxyhornblende at 8000C, changing its pleochroism from greenish to
reddish and raising its birefringence value.
8. Most clay types are sintered at around 8000C. At higher ring temperatures they undergo a
vitrication process that turns them into isotropic matter. Most earthenware clay types become
completely isotropic at about 10000C.
9. The mineral gehlenite is created at ca. 11000C. In many instances this mineral can be observed
within the otherwise vitried matrix under high magnications.
For example, if calcite alters at about 7000C and hornblende at about 8000C, then a tablet in which
the calcite has changed but the hornblende has not would have been red between 7000C and 8000C. The
accuracy of this method depends on the mineralogical variability of the item. In ceramics, where many
mineral types appear, these denitions can be quite accurate. In other cases the estimates can be made
only in general (as highly red, moderately red, etc).
In the course of the present study colleagues often cautioned that museum curators might have red
cuneiform tablets in the past as part of their preservation procedures. Firing tablets in an electric kiln at
150 0 C for 48 hours, then at 400 0 C for 5 hours and 700 0 C for 3 hours, appears in the literature as part of
the standard conservation process in the British Museum (Bateman 1966). However, the museum records
showed no evidence that any treatment of this kind was performed on the Amarna tablets, and many
of the tablets showed no evidence whatsoever of any signicant ring process. Since this topic seemed
to be important for some of our interpretations (in theory, un red tablets could have been recycled, as
opposed to red tablets where the irreversible process of sintering occurred), this question was raised
with the museum curators. At least in the case of the largest collections of the Amarna tablets, namely
in the British Museum and the Vorderasiatisches Museum in Berlin, no evidence for such treatment has
ever been recorded.
CHARACTERIZATION OF THE PALAEONTOLOGICAL FEATURES IN THE CLAY1
Lydia Grossowicz
Micropalaeontological studies are applied here for the rst time as an annex discipline for archeological
subjects. The main fossil index used for this purpose is the foraminifera.
Foraminifera comprise one of the main groups of the unicellular organisms Protists (Protozoa).
They are characterized by a single or multi-chambered test built mainly of carbonates although some
genera may include agglutinant material. The bulk live in marine environments, while there are also
some inland and fresh water species. Marine foraminifers are divided into two main groups, those living
in the water mass (planctonic) and those living on the sea oor (benthonic). They are recorded from
the Paleozoic until recent times and they constitute a well-known tool for stratigraphic correlation,
palaeoenvironmental research and geological age determination.
In hard rocks (limestone, sandstone) foraminifers are studied by means of thin-slides (rock
peeling). The assemblage of the different fossil elements presents a palaeoecological picture, known as
a microfacies, which allows the student to interpret ancient marine environments as well as to determine
the age of the rock.
1. In the petrographic catalogues, palaeontological identications made by Lydia Grossowicz were marked: (LG). The
foraminifers were indicated by class: (p) for planctonic and (b) for benthonic.
16
The study of soft rocks (clays, marls, etc.) is usually done by washing and sieving the material, reducing
it to the specic grain size containing foraminifers (average diameter 0.5 mm). This method permits a threedimensional study of the fossils under a binocular microscope ensuring their best recognition.
The samples from the Amarna tablets consist of indurated thin sections of the clays and marls used
in their manufacture, which may include sporadic foraminifers and other fossils (molluscs, corallinean
algae) from the original rocks. Some difculties arose during the study due to the scarcity of fossils, their
poor state of preservation. Many of the most common planctonic foraminifers have basic homologous
shapes that appear repeatedly through the geological record, and the only way to distinguish among
them in thin-sections is by means of their test structure and texture. This leads, for example, to some
uncertainties about the identication of Paleogene or Neogene foraminifers. In some cases the original
test was so poorly preserved or badly damaged that foraminifer determination was almost impossible.
However, in most samples the attempt to dene them was carried out with some success.
Identication was done mostly at a generic level, in some cases at a species level. Geological age
was mainly determined at a System level (Upper Cretaceous, Paleogene, Neogene). An Olympus BX50
optical microscope was used and some samples were microphotographed.
The 1:200,000 geological maps of Israel and the maps of Syria and Lebanon by Dubertret (1945)
were consulted as well as Sartorio and Venturini (1988), Foraminiferi Padani (1982), Buchbinder (1975)
and Luterbacher (1986).
CHARACTERIZATION OF THE BOTANICAL INCLUSIONS IN THE CLAY2
Simcha Lev-Yadun
Small, fragile plant remains may be preserved within sediments for dozens of millennia but totally
disintegrate under the regular procedures of sampling. When petrographic thin sections are prepared by
embedding geological sediments, pottery or similar materials, tissue fragments the size of several m
are still found intact and in situ (Goldberg et al. 1994). This is a great advantage for studies of materials
in which the plant remains are smaller than sieve mesh size or are fragile. Usually, if the plant remains
are very small or in a bad state of preservation, it is not possible to identify the species or even genus to
which they belong. However, in many cases it is possible to characterize the type of material used by the
craftsman for clay preparation: herbaceous or woody material, leaf or stem, seed or bark, monocotyledon,
dicotyledon or conifer. Several rules of thumb can be used to distinguish between these types of plant
material. Young and thin herbaceous material and many leaf fragments or most monocotyledons will
have only primary vascular elements if xylem is included in the specimens. The vascular system of
monocotyledons (usually cereals or reeds) is formed in parallel sectors, a characteristic easily seen if the
sample is not too small. In addition, typical patterns of epidermal cell shape, phytoliths (silicon cells)
and trichomes (hair) are also indicative structures for this group. Wood can be identied in many cases
as conifer or dicotyledon even from fragments the size of several broken cells. Wood made of tracheids
characterizes conifers while dicotyledons have vessel members. Similarly, tree bark has its typical cork
layers and other bark tissues. Many seeds have typical seed coats and dicotyledon leaves have their
branched vein system. Thus, it is possible to positively characterize the type of plant remains in many
instances or else by elimination. Thus, for a well-trained plant anatomist with considerable experience in
identication of ancient plant remains, the examination of plant fragments in pottery or similar materials
gives many indications of their type.
2. In the petrographic catalogues, botanical observations made by Simcha Lev-Yadun were marked: (SLY).
17
QUANTITATIVE ANALYSIS OF THE ELEMENTAL DATA

THE USE OF STANDARDS
Prior to the application of elemental methods, it became mandatory to investigate whether tablets were
made of clays similar to those used for pottery. Letters whose writers location is not disputed, such
as those written by the Kings of Egypt to foreign rulers (i.e., those deposited in the archive as copies,
drafts, etc.), were selected as a case study for comparison with Egyptian ceramics.
The examinations of EA 1 and EA 14 (written to Babylonia), EA 162-163 (written to Amurru) and
EA 190, 367 and 370 (written to rulers in Canaan) disclosed that in most cases (except for EA 163 and
382) the tablets were not made from standard clays used in ancient Egypt for pottery production. Further
research disclosed that this phenomenon is almost unique to the Egyptian correspondence (with a few
more exceptional cases such as the Megiddo letters). Still, these results counsel that standard elemental
analyses should be used with caution and tablets should be compared with pottery after petrographic
examinations have con rmed that they were made of similar clay types. Hence we used the ICP data
mostly for examining the intergrouping of tablets by their element composition and made correlations
with pottery reference data only in cases where petrographic similarity was disclosed.
DATABASE OF WORKSHOP WASTERS FOR SOUTHERN CANAANITE CERAMICS
One of the most crucial points that were raised by the results of the current research was the discovery
of a group of tablets that were presumably sent from the southern coast of Canaan. In addition to the
Egyptian centre of Gaza, there were two city-states in the northwestern Negev and the southern coast in
the Amarna period, namely Yurza and Ashkelon. Thus, a detailed analytical tool was required in order to
differentiate between the fabrics of these rather adjacent and geologically similar areas. The opportunity
to perform such a study was made possible by recent advances in the research of the Byzantine pottery
production systems in the general area between Gaza and Ashdod.
As opposed to most regions in the southern Levant treated in this study, the southern coastal plain
and the western Negev are particularly monotonous in their geology, consisting mostly of aeolian silt
(loess) and coastal sediments that are dominated by sands and their derived calcitic sandstones (kurkar).
This presents a considerable problem when precise provenance determinations are required.
Only a few Middle or Late Bronze Age ceramic workshop sites are known in Israel (Wood 1990;
Killebrew 1996). Yet the use of pottery wasters from other periods within this area seemed to be relevant
for our study. As noted above, sites in the northwestern Negev are typied by CTF = 1. Therefore,
it is probable that similar raw materials were exploited continuously over time. This hypothesis was
con rmed by many petrographic investigations conducted on ceramic assemblages from the northern
Negev (e.g. Goren 1995; 1996a, with references). In order to resolve the methodological problem
presented above and correlate the northwestern Negev tablet to a specic site, a reference collection
of wasters and un red sherds from 20-30 kiln sites in this area was established. In recent years several
attempts have been made to identify the types of amphorae used to store and transport wine from the
production centres of Gaza and Ashkelon reported by ancient sources (Mayerson 1994; 1996). The rst
survey of centres for the production of wine amphorae from the Roman and Byzantine periods along the
southern coastal region was carried out by Israel (1993). In this survey evidence of numerous productions
centres for this type of jar were found in the Western Negev and the southern coast. In excavations of the
3rd Mile Farm next to Ashkelon, Israel (1995) discovered a large workshop for the production of Gazan
amphorae located close to several large, sophisticated winepresses.
18
In the present research ICP AES/MS analyses were carried-out on over a hundred wasters from
positively identied (preferably excavated) Byzantine workshops from the area that lies between BeerSheva, Gaza and Ashdod. The samples included unred pottery or sherds twisted by overring. Special
attention was paid to the examination of pottery from workshops scattered within a range of 5 km.
around the important Late Bronze sites (e.g. Ashdod, Ashkelon, Tel Sera, Tel Jemmeh, Tel Haror).
This petrographic study indicated that the inclusion types within the loess groups varied geographically
and could be isolated and dened. For the reasons explained above the results of this study seem to be
applicable for the Late Bronze Age. Therefore, this data was used for the interpretation of the ICP data
of the southern Canaanite letters.
STATISTICS
The geochemical data retrieved by ICP were manipulated by multivariate statistical analyses, using
StatSoft STATISTICA for Windows (Release 6) programme. One of the shortcomings of ICP for
ceramic characterization studies is the deciency of a database of standards, such as in the case of NAA
studies of pottery. Chemical compositions of clay sources collected by other methods often proved to be
insufcient, since they included only major and sometimes also minor elements but not traces. However,
some data that was revealed during previous studies of Goren (1996a), Porat (1989a) and Porat et al.
1991) by ICP-AES, was useful for this study too. In a more recent study (Mallory-Greenough et al. 1998)
this method was used successfully on ancient Egyptian pottery.
The analytical methods employed were those summarized by Glascock (1992), with modications
for ICP data recently suggested by Mallory-Greenough et al. (1998). Since the characterization of
ceramic materials by multi-element geochemical methods involves a vast amount of data on a relatively
large number of specimens, the associations between elements and specimens are rather complex. The
objective of compositional characterization of the tablets is of course to identify groups that can easily be
discriminated from others in an attempt to reveal a meaningful archaeological interpretation. However, in
the mass of data that forms each group, some elements are poorly related whereas others are interrelated
and affect each other. For example, as is well known from the geochemical literature, Ca is often
associated with elements such as Sr and Ba. Iron, Sc and other transition metals usually exhibit highly
correlated relationships. In pottery, negative correlations often occur between Ca and Si, or Al and Si, as
a result of the dilution of clay elements with those of the non-plastic components. It is also known that
rare earth elements (REE) are commonly correlated. In order to examine this inter-element correlation,
matrixes were calculated between the various element values that were measured by ICP. The results
were used for the plotting of bivariate plots (scattergrams).
The combined ICP-AES and ICP-MS analyses supplied a list of 31 elements, including: SiO2, Al2O3,
CaO, MgO, Fe2O3, TiO2, P2O5, SO3, Zn, Co, Ni, Ba, Mn, Cr, V, Be, Cu, Sr, La, Y, Ce, Pr, Nd, Sm, Eu, Tb,
Dy, Ho, Tm, Yb and Lu. In order to differentiate the signicance of each element measured, the factor
loadings of the elements were plotted in a bivariate plot. However, in the statistical analyses several
elements were omitted due to the risk of bias. Several elements, especially those with high ionic charge
or ionic radius, are more sensitive to post-depositional processes that may occur in buried ceramics due
to their solubility in groundwater. Other elements may lead to biases due to anthropogenic or sedimentary
processes. These include P2O5, SO3, Co and Ba. Phosphorus was omitted because many of the tablets
contained biogenetic matter, either as plant tissues, coprolites or bone fragments, and high values of
P could have been the result of these rather than of the clay composition. Sulphur was left out as it
was found to reect post-depositional processes, such as the crystallization of gypsum that commonly
19
occurs in soils of arid zones such as Amarna. The same applied to barium. Cobalt was eliminated since
it was affected in some cases by the small sample sizes that were subjected to analysis, thus providing
inaccurate values (Appendix:Tables 1, 2).
Previous analyses have demonstrated that treatment of compositional data as lognormal distribution
rather than a normal distribution effects a quasi-standardization of the values retrieved by elemental
analysis (Glascock 1992). It was noted that when treated as logarithms of the measured concentrations the
data appears to be more normally distributed. Moreover, this procedure compensates for the differences
in the magnitudes between the major and the trace elements. Therefore, the values were converted into
logarithms before the application of the multivariate statistical methods.
As a rst stage of interpretation, bivariate plots of the elements were employed in order to observe
partitions in the data set. The element concentrations were plotted as one variable against the other
to create correlation matrices. Examination of these plots reveals possible bias of natural correlations
between, for example, REE. Bivariate plots can be also used to graph linear combinations of the principal
components or discriminant functions.
Cluster analysis was used to examine the complex relations between the analyzed specimens.
This procedure is based on a dissimilarity matrix in which the distances between all pairs of cases are
calculated using a distance measure, usually as squared-mean Euclidean distance (Sayre 1975). The
resulting dendrograms are used to allocate individual specimens to groups and show the degree of
linkage between the members of each group. Cluster analysis, which was favored in the early stages of
the elemental analyses of ceramic pastes (e.g. Sayre 1975), has somewhat lost its popularity in the more
current literature. This is due to its tendency to force data into hyperspherical groups and the known
tendency for pottery compositional groups to be elongated due to inter-element correlation. However, the
initial groups from cluster analysis can provide the starting point for other techniques for group renement
(Glascock 1992). In applying this method, we usually used Wards method but in several cases we found
the complete linkage method to supply more suitable results. The diagrams were constructed using a
Euclidean distance metric, or squared Euclidean distances, as commonly presented in the archaeometric
literature. It was recently suggested that changing the distance metric to Pearson correlation coefcients
yielded more complete separations of closely related sample (Mallory-Greenough et al. 1998).
Further grouping was made by factor analysis and principal component analysis (PCA). The
advantages of this method for multi-elemental analysis of pottery have been described elsewhere (e.g.
Glascock 1992: 17-18).
SOURCES FOR PROVENANCE INTERPRETATION OF THE TABLETS

The interpretation of the geological contexts and the geographic origins of the raw materials (matrix and
inclusions) that were used for the production of each tablet relied on several parameters. The following
criteria were used:
GEOLOGICAL MAPPING
After the identication of the minerals and rock fragments that appear within the sample, an attempt was
made to locate the geological environment reected by its composition. Much attention has been drawn
to details that might reect specic lithology, such as igneous or metamorphic rocks, coastal sands,
travertine, tuff, etc. The lithological data did not rely only on the non-plastic components, but also on the
matrix features.
20
In several cases, when the inclusions were composed of wadi sand, we applied the drainage system
method for the identication of their provenance. The use of this method for the identication of
production centres along water systems has been suggested by Lombard (1987), Goren and Gilead (1987)
and Gilead and Goren (1989), and elaborated by Miksa and Heidke (1995). This method refers to the
inclusions in pottery as sand that can be studied both qualitatively and quantitatively, compared with
naturally occurring sands near a set of sites, and nally used as indication for the exact provenance of
each vessel. Although the samples taken by peeling were extremely small, they were sufcient, in several
cases such as Yurza and Gaza, to supply enough inclusions for comparison between close fabrics.
After determining the geological environment from which the raw materials were derived, we
attempted to correlate them with the origin of the tablet as suggested by its textual evidence. In doing
this, we operated a hierarchical set of considerations, according to the following order:
1. Petrography corresponds with the nearby geology of the city-states capital (yes/no): If the origin
of the tablet was mentioned in the text (e.g. Hazor, Lachish, etc.), we compared the lithology
indicated by the materials to the geological mapping of the area. For this a set of geological maps,
preferably on a detailed scale (1:50,000 or 1:100,000) was commonly used. If the lithology of the
tablet corresponded with the geology of the city-state from which it was said to have come, we
considered it local. If not, we moved further to step 2.
2. Geographic-historical data referring to the boundaries of each city-state was applied. The
lithology of the object was compared to the geology of the entire assumed territory of the citystate. In several cases we could suggest that the clay was not collected from the capitals area,
but from a province within the boundaries of a city-state, or even from a neighboring city-state.
3. In the case of a complete disagreement between the mineralogy and lithology of the tablet to its
origin as determined by the text, the closest possible locality where the lithology may suit the
tablets components was suggested.
4. In the case of letters that do not specify their origin or where the name of the author is missing,
the lithology of the tablet was compared with the geology of their assumed origin as suggested
by scholars. In several instances an elimination of several options could be made, and more
specic locations could be suggested.
SPATIAL DISTRIBUTION OF PETROGRAPHIC GROUPS
In several cases it was possible to demonstrate the core area in the distribution of several key petrographic
groups by using the analytical data that collected in the past from numerous south Levantine pottery
assemblages. Such attempts were undertaken, for example, in the studies of Chalcolithic (Goren 1995)
and EBIV ceramic assemblages (Goren 1996a). The relative frequency of each petrographic group in
every assemblage was recorded in a spreadsheet le which was used as data for a mapping programme.
The maps were then plotted by a Golden Surfer mapping programme, using the distribution of each
given petrographic group as topographic intervals (Goren 1995). Needless to say, the sample size of
each assemblage is not equal nor is the accuracy of the percentage of the petrographic group within
it. Therefore contour intervals of 10% were chosen for demonstrating the spatial traits of each group.
Since the available raw materials in each area were quite limited, and maintained constantly along the
whole archaeological sequence, such maps could be used even if they did not rely on Late Bronze sites.
Preferences of potters could, and in fact did, vary between periods, but if the natural sources of clay were
limited and their quality for ceramic production was reasonable, continuity of their use over the sequence
may be expected.
21
COMPARATIVE EXAMINATION OF RESOURCES

In previous studies, samples of several natural clay formations from the southern Levant were collected
for comparison with key petrographic groups (Goren 1996a). The samples were wetted and formed
into small briquettes, then red at temperatures of 600 0C and 800 0C. The thin-sections prepared from
these briquettes were compared with the thin-sections of the archaeological materials. A comparative
collection of such samples is stored in the petrography laboratory of the Institute of Archaeology, TelAviv University.
CROSS-RELIANCE ON OTHER TABLETS
All the previous criteria enabled a type-list with the characteristics of raw materials that were used in
the main city-states for the production of cuneiform tablets to be established. Through the Egyptian
reference group, the materials used in Egypt for tablets could also be dened. Furthermore, after it was
understood that some Canaanite letters were written in Egyptian administrative centres (Gaza, Bethshean, umur), they could be separated and grouped together. Securely identied original tablets (e.g.
EA 227-228 from Hazor) also formed a reference for the typical materials that were used in each location
for the production of tablets.
22
CHAPTER 3
LETTERS OF THE GREAT POWERS
I. EGYPT
Most of the Amarna archive consists of letters received from foreign kings that dealt with Egypt more
or less on the basis of equality, and of letters sent by Egyptian vassals in Canaan. However, a number
of letters were written by the King of Egypt. This small group includes 11 tablets: three letters and
one inventory directed to Great Kings of independent states (EA 1, 5, 14, 31), and seven letters sent to
vassal kings (EA 99, 162-163, 190, 367, 369-370). Of this group seven letters were analyzed: EA 1, 14,
162, 163, 190, 367, 370. In the course of examination, an additional letter (EA 382) was identied as
belonging to the Egyptian correspondence.
Fig. 3.1: The Near East during the Amarna period.
23
It is not clear why these letters were included in the archive. Moran suggested that letters which
were sent from Egypt were perhaps rst written in Egyptian and then translated into Akkadian.
The Akkadian letters were dispatched abroad whereas the Egyptian drafts were kept for future
correspondence. Only a few copies of the translated texts were occasionally led, possibly due to
oversight (Moran 1992: xix-xx). However, there is no evidence for Egyptian drafts and for the assumed
systematic ling of important letters. Thus it may be tentatively suggested that some tablets are copies
of letters kept as models for future correspondence (e.g. EA 14, 31, 99, 367, 369-370), and the others were
broken, damaged or disqualied for some unknown reason (Naaman 2002:80-81).
CERAMIC ECOLOGY AND PETROGRAPHIC REFERENCES

The information about the clay types used for producing Egyptian ceramics throughout the periods
is well established (Bourriau and Nicholson 1992; Arnold and Bourriau 1993; Bourriau et al. 2000).
Pottery from Amarna and other New Kingdom sites relevant to our research have been investigated
technologically by several scholars (Hope et al. 1981; Bourriau and Nickolson 1992; Arnold and
Bourriau 1993:148-182; Bourriau et al. 2000). These analyses include a thorough study of the ceramic
workshop unearthed at Amarna and a comparative study of recent potters workshops at the nearby
village of Deir el-Mawwas. These studies supply all the necessary information about the potential of
clay sources in this part of Egypt.
CATALOGUE
EA 1 (BM 29784), from the King of Egypt to the King of Babylonia
Sampling method: Peeling.
Reliability: High.
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, almost devoid of foraminifers, with
common small (around 60-70m) greenish and light yellow glauconite concentrations and silt
(about 5%). The matrix is optically active and exhibits weak optical orientation. Opaques (~2%)
appear in bimodal size groups, between a few micrometers and nearly 50m, (very common), and
between 100m and 200m (rare). The silt (2%) contains essentially quartz, but also heavy minerals
in which hornblende, rutile, zircon, biotite, feldspars and opaques were identied.
Inclusions: The inclusions are loosely spread (f:c ratio{0.062mm}=95:5) and contain moderately sorted angular
ne sand particles that maintain a gradual continuum with the silt in terms of grain size. No remains
of vegetal material were traced. The sand fraction contains predominantly subangular to rounded
particles of quartz (up to 400m), a few subangular grains (up to 300m) of micritic limestone and a
very few subangular particles (up to 150m) of microcline and plagioclase.
Firing temperature: Probably very lightly red (below 5000C) judging by the beginning of colour change
in the glauconite.
Geological interpretation: Based on its mineralogical and palaeontological afnities (for the latter see EA
357), this clay is identied as belonging to the shales of the Esna Formation in Egypt, which is equivalent
to the Levantine Paleocene marl (such as the Taqiye Formation of Israel, see Bentor 1966:72-73). The
latter is exposed from central Sinai through Israel and Jordan northwards to Syria and Lebanon.
This formation is almost constant in its stratigraphic position and even in details of its composition.
Equivalent beds appear even in Morocco and Turkey (Bentor 1966:73). The Esna Formation outcrops
in several localities in Upper Egypt, particularly in the Esna-Edfu region (El-Naggar 1966).
24
Reference: The use of Esna shales for pottery production has not been recorded in Egyptian assemblages
except for one unique case where Canaanizing pottery was produced in a Naqada IIIa context
(Porat and Goren 2002). In the southern Levant, the use of the equivalent Taqiye marl for pottery
production is very common.
Conclusions: EA 1 is a royal letter from the King of Egypt to the King of Babylon (Moran 1992:1-5;
Cochavi-Rainey 1993). It was produced of Esna marl with the addition of some sand and straw
temper, a combination that is as yet unknown from New Kingdom Egyptian ceramics.
EA 14 (VAT 1651 + 2711 + Ash 1891.1-41), inventory of gifts for the Babylonian king
Reliability: High. Sampled both in the Ashmolean Museum (Ash 1891.1-41) and in the Vorderasiatisches
Museum (VAT 1651).
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, almost devoid of foraminifers, with small
(~60-70m) yellow to brownish-yellow glauconite concentrations and silt (about 5%). The matrix is
optically active (speckled b-matrix) and exhibits very weak optical orientation. Opaques (~1%-2%)
appear in bimodal size groups, between a few micrometers and nearly 50m, (very common), and
between 100m and 200m (rare). The silt (2%) contains essentially quartz, but also heavy minerals in
which hornblende, olivine, zircon, biotite, feldspars rutile, and opaques were identied.
Inclusions: The inclusions are more densely spread than in EA 1 (f:c ratio{0.062mm}=90:10) and contain
moderately-sorted angular, ne sand particles that maintain a gradual continuum with the silt in
terms of grain size. No remains of vegetal material were traced. Subangular to rounded grains
(up to 1200m but commonly below 40m) of quartz, sometimes displaying undulose extinction,
rarely polycrystalline, predominate in the sand fraction. Subangular micritic limestone grains (up to
1200m) are common and the few accessory heavy minerals: (up to 150m) appearing in the ne
sand fraction include microcline, plagioclase feldspar, hornblende, and zircon.
Palaeontology (LG): Very badly preserved non-diagnostic foraminifers, unidentied. Two samples were
studied, one in thin section and the other by washing. Both supplied the same results. However see
EA 357 for more successful results of the same fabric.
Firing temperature: Probably over 5000C (glauconite changed to yellow) but not exceeding 7000C
(calcite unchanged).
Geological interpretation and Reference: Similar to EA 1.
Conclusions: Another example of a royal Egyptian letter written to the King of Babylonia on Esna marl.
EA 162 (VAT 347), from the King of Egypt to Aziru of Amurru
Reliability: High.
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, almost devoid of foraminifers, with
some light yellow glauconite concentrations (up to 160m) and silt (about 2%). The matrix is
optically active with speckled b-fabric and very weak optical orientation. Opaques (1%) appear in
bimodal size groups, between a few micrometers and nearly 20m, and between 50m and 100m.
The silt contains essentially quartz, but also heavy minerals in which hornblende, zircon, biotite,
feldspars and apatite were identied.
Inclusions: As in EA 14, the inclusions are more densely spread than in EA 1 (f:c ratio{0.062mm}=90:10),
but are less sorted than these of EA 14. They contain angular to rounded sand particles including
frequent subangular to rounded particles (up to 800m but commonly below 500m) of quartz
which sometimes displays undulose extinction but is rarely polycrystalline, and subangular micritic
25
limestone (up to 800m). A few accessory heavy minerals (up to 150m) appear in the ne sand
fraction. These include hornblende, microcline, plagioclase, orthoclase, and zircon.
Vegetal material (SLY): Very few (up to 1mm) with calcium oxalate crystals. A fragment of a stem shows
typical secondary wood with both axial tissues and radial (rye) tissues. This is a dichotyledon plant
which has many bres in its wood. This is not a conifer and not a monocotyledon. No further taxonomic
remarks can be made.
Firing temperature: Probably lightly heated (glauconite changed to yellow) but not exceeding 5000C
since preserved vegetal material is uncharred.
Geological interpretation: EA 162 is similar in its petrographic afnities to EA 1 and EA 14.
Conclusions: An example of a royal Egyptian letter written to a vassal ruler on Esna marl.
EA 163 (VAT 1885), from the King of Egypt to a Canaanite ruler (?)
Reliability: High / adequate.
Matrix: Clayey, brownish-grey in PPL, optically active speckled b-fabric with very abundant (~10%) angular
opaque minerals sizing up to 50m. Also rather common are mica minerals (~5%), sizing up to 50m,
consisting mostly of biotite. Quartz silt is sparse (2%). The silt to ne sand fraction includes lesser amounts
of other heavy minerals: feldspar, hornblende, zircon, pyroxene, and epidote.
Inclusions: This tablet seems to contain no intentionally mixed inclusions apart from the vegetal material.
The other coarse fraction is probably detrital within the clay sediment. Occasional sand-sized grains
of quartz displaying a continuum from silt-sized are sparsely spread. Hornblende, pyroxene, and
feldspar appear in the sand fraction.
Vegetal material (SLY): Remains of vegetal material, up to 200 m, not straw but probably grass or leaf
fragments. Though these are the dominant non-plastics, they occupy altogether less than 1% of the
groundmass. A piece of charcoal (unidentied) and tissue fragments which include primary xylem cells
with their typical dense spiral secondary cell wall thickening are present.
Firing temperature: Probably lightly red judging by the effect of ring on the birefringence of the
biotite, but not exceeding 500oC since preserved vegetal material is uncharred.
Geological interpretation: This tablet is readily identied by its petrographic afnities as being produced
of the so-called Nile silt (or preferably Nile mud). This term refers to pottery manufactured in Egypt
from local Quaternary Nile sediments. It is easily distinguished petrographically. The main features
of Nile silt, when examined under the petrographic microscope are as follows:
A. Contents of poorly sorted sand to silt sized quartz, in varying quantities and size ranges.
B. High proportion of accessory and heavy minerals, including mainly opaques, minerals of the mica,
amphibole, pyroxene and the feldspar groups. Various other minerals can appear too, amounting in
all to 45 mineral species (Hassan 1976:431).
C. Vegetal tissue fragments and related material (phytoliths) are often visible in the clay body.
D. A non-calcareous matrix with abundant vegetal material or its remains as phytoliths or voids.
Reference: There are numerous references for the petrography of Nile silt produced pottery (Bourriau et
al. 2000 with references).
Conclusions: EA 163 is one of two examples (together with EA 382) of an Egyptian letter written on
a tablet made of Nile silt. The text differs from EA 162 so that it is not a copy of the latter. The
identity of its addressee remains unclear. Knudtzon attributed it to the correspondence of Egypt with
Aziru of Amurru on the basis of some textual similarities to EA 162, but not enough text remained
to conrm his suggestion.
26
Based on the examination of the scholarly texts from Amarna (Chapter 4), and especially EA 368 that
is also made of Nile silt, it may be speculated that EA 163 should be included in the category of school
texts. Letters must have been written as exercises by students in the Amarna school for scribes, and
EA 163 might possibly have been an exercise in writing a letter to a Canaanite ruler. We may never
know the circumstances that led to the composition of this letter on a Nile silt tablet, but the fact that
the only obvious equivalent to it is clearly a school text (EA 368) brings this option to mind.
EA 190 (Ash 1893.1-41:411), from Egypt to Etakkama of Qidshu
Reliability: High.
Matrix: Carbonatic, light yellowish-grey in PPL, devoid of any foraminifers, with small (around 60-70m)
greenish and light yellow glauconite concentrations and silt (about 2%). The matrix is optically active
with speckled b-matrix and very weak optical orientation. Opaques (0.5%) appear in bimodal size
groups, between a few micrometers and nearly 10m, (very common), and around 50m (rare). The
silt (1%) contains essentially quartz, but also heavy minerals in which hornblende, rutile, zircon,
biotite, feldspars, and opaques were identied.
Inclusions: Generally, the inclusion suite is comparable with these of EA 1 and EA 14. However, it is
more sparsely spread (f:c ratio{0.062mm}=95:5) and better sorted. Subangular to rounded grains (up to
550m but usually below 400m) of quartz, sometimes displaying undulose extinction but rarely
polycrystalline, predominate. Rounded micritic limestone (up to 400m) is common. No remains of
vegetal material were traced.
Palaeontology (LG): In the thin section only non-diagnostic, badly preserved foraminifers were observed.
in the glauconite.
Geological interpretation: EA 190 is similar in its petrographic afnities to EA 1 and EA 14.
Conclusions: This fragmentary letter was found during Petries excavations at Building 19 (Records
Ofce). Sign forms and formulaic expressions indicate an Egyptian origin (Knudtzon 1915:1719; Campbell 1964:126). The petrographic examination indeed suggests that this is another royal
Egyptian letter written on Esna marl.
EA 367 (AO 7095), from the King of Egypt to Endaruta of Akshapa
Sampling method: SPA.
Reliability: Fair.
Matrix: Carbonatic, light yellowish-grey in PPL, with some small (around 60m) light yellow glauconite
concentrations and quartz silt. The matrix is optically active with speckled b-matrix and very weak
optical orientation. Opaques appear in the matrix. The silt (1%) contains essentially quartz, but also
some heavy minerals (biotite and feldspar were identied in this small sample).
Inclusions: Generally, the inclusion suite is comparable with these of EA 1 and EA 14. No quantitative
estimates could be made due to the small sample size. It comprises subrounded quartz grains (up to
300m) and rounded micritic limestone (up to 250m).
Firing temperature: Perhaps only lightly red to around 5000C (glauconite colour change).
Geological interpretation and conclusions: The petrographic afnities of EA 367 are similar to EA 1 and EA 14.
EA 370 (BM 134870), from the King of Egypt to Yidia of Ashkelon
Reliability: High.
27
Matrix: Carbonatic, yellowish-grey in PPL, devoid of any foraminifers, with small (around 60-70m)
orange to brownish glauconite concentrations and rare opaques (about 0.5%). The matrix is optically
active with strong preferred optical orientation (length fast). The silt (0.5%) contains quartz with
very rare biotite and zircon. The matrix is markedly microlaminated and tends to exfoliate parallel
to the outer surface. The laminae appear either as foils or wavy foils, around 10m in thickness (a
similar phenomenon is observed in EA 355).
Inclusions: As opposed to most other Egyptian tablets, the inclusions contain very few rounded to
subangular grains of quartz, usually up to 300m but one polycrystalline grain is 640m, sometimes
displaying undulose extinction. Very few feldspar grains also appear. The f:c ratio{0.062mm} is low,
about 98:2. There is an unusual combination of organic matter including fragments of charred
organic matter (up to150m), fragments probably of bark, uncharred straw (up to 400m), and
coprolites (remains of manure) that appear as dark reddish-brown irregular phosphatic bodies.
The latter engulf oral remains of arboreal plant tissues with typical carbonate (calcium oxalate)
crystals from the plant tissues. One coprolite fragment contains a fractured but complete long bone
of microfauna, 750m long, with birefringence increased into rst order yellow due to ring. The
bone is perhaps a rib or spine of a sh (R. Rabinovitch, pers. comm.). Hence in this tablet manure
rather than the usual chopped straw was mixed with the clay.
Firing temperature: Up to 5000C for a short time, judging by the colour change in the glauconite, the
increase of birefringence in the bone but the complete preservation of the organic matter.
Geological interpretation: The clay is generally similar to that of EA 1 but far less silty (below 1%)
and exhibits pronounced microlamination. To the naked eye this feature is visible in breaks on the
tablets surface where the clay tends to exfoliate to paper-thin lms. It is possible that Esna marl
was used as clay for this tablet too. However in this case the material is apparently of lesser quality.
The temper that was mixed with it is essentially organic, and contains some straw and animal
(apparently cat/dog) manure.
Conclusions: As EA 1.
EA 382 (BM 58364), small fragment of a letter from an Egyptian ofcial (?)
Sampling method: SPA. The sample that was taken from this tablet included tiny crumbs that were found
in the box where it was stored in the British Museum. Examination under the stereomicroscope
conrmed that the crumbs undoubtedly belonged to this tablet which is distinguished by its unique
very dark and lustrous, nearly ebony like, appearance.
Reliability: Fair. Due to the ne material of this tablet it was sufcient for de ning the clay and silt
characteristics.
Matrix: Clayey, dark tan in PPL (probably due to reducing ring atmosphere), optically active striated bfabric with very abundant (~10%) angular opaque minerals sizing up to 20m. Also rather common
are mica minerals (~5%), sizing up to 50m, consisting mostly of biotite. Quartz silt is sparse (2%).
The silt in this sample includes also hornblende and zircon.
Inclusions: Stereomicroscopic examination has indicated that there were no inclusions.
Firing temperature: There is not enough evidence in this small sample for estimating the ring temperature.
Geological interpretation: By its petrographic afnities the clay of this tablet is identied as Nile silt.
Conclusions: On the assignment of this fragment to the Amarna archive see Walker 1979:249. According
to Moran (1992:369-370, n. 1), the text is a letter and the script, especially the forms of la and ni,
points to an Hittite or Egyptian provenance. The same holds true for the form of address.
28
EA 339 (VAT 1887), Canaanite letter written in Egypt

Reliability: Moderate.
Matrix: Clayey, brownish-ochre in PPL, optically active striated b-fabric with very abundant (~10%) angular
opaque minerals sizing up to ~30m. The matrix contains silt (~10%) of quartz accompanied by mica
minerals, and lesser amounts of other minerals: feldspar, zircon, pyroxene, and epidote.
Inclusions: This tablet seems to contain no intentionally mixed inclusions apart from the vegetal material.
The other coarse fraction is probably detrital within the clay sediment. Occasional sand-sized grains
of quartz displaying a continuum from silt-sized are sparsely spread.
Vegetal material (SLY): Remains of vegetal material, up to 150m, not straw but probably grass or leaf
fragments.
Firing temperature: There is not enough evidence in this small sample for estimating the ring temperature.
Geological interpretation: By its petrographic afnities this tablet most likely represents Nile silt (as EA 163).
Conclusions: EA 339 is a small badly damaged upper part of a letter. The name of its writer is missing, but
the introductory formulae indicate that it was sent to the Pharaoh. The script is typically Canaanite,
and since it is of Egyptian provenance, it must have been written by a Canaanite ruler who arrived
in Egypt and submitted his request in written form.
CHEMICAL ANALYSES OF SELECTED EGYPTIAN LETTERS

Table A1 in the Appendix presents the elemental composition of several Egyptian letters, as disclosed by
ICP AES/MS. It is evident that EA 163 differs from the other three tablets in its composition. Petrography
indicates that this letter was made of a clay type different from that of other letters examined by ICP.
Chemical analysis by other analytical methods (major elements only) of the Esna shales is presented
by Basta et al. (1979:52, and see Ahmed et al. 1992:127-129 for the clay mineralogy. A more detailed
chemical composition is provided by Schreier 1988:104-124).
When compared with the recently available ICP data obtained from Egyptian pottery from Karnak
and Mendes (Mallory-Greenough et al. 1998), including both Nile silt and marly clay types, the
composition of the tablets (excluding EA 163) differs signicantly from these categories. Whilst the
Karnak marly group clusters well and differs from the Karnak and Mendes Nile silt, the latter vary but
still cluster well as a distinct group. The tablets are not related to any of these clay types. Consequently,
the Egyptian letters (except for EA 163) represent different clay sources. Under these circumstances
together with the micropalaeontological evidence from EA 357-358 (that petrographically belong to this
group), marl of the Esna Formation becomes the only possible option.
CONCLUSION: THE EGYPTIAN TEXTS

Our study has shown that royal Egyptian tablets were produced from different clay types than those used
for Egyptian pottery vessels. Since in the case of the Egyptian letters the exact location of the sender
is well known, the letters may be easily compared with the fabrics of local ceramics from Amarna and
Thebes. Moreover, unlike Syrian and Mesopotamian ceramics, Egyptian pottery has long been analyzed
petrographically, hence our comparisons can rely on a solid body of data.
This outcome has far-reaching implications on the methodology that should be adopted in the entire
study. In the two previous analytical researches of the Amarna archive so far undertaken, a group of scholars
from Berkley attempted to investigate the origin of some of the Alashiya tablets and the source of the
Mitannian letters by applying the NAA technique (Artzy et al. 1976; Dobel et al. 1977). From the outset, the
29
possibility of the tablets being copies made in Egypt had been raised (Artzy et al. 1976). The composition of
the Alashiya letters was therefore compared to that of New Kingdom pottery collected at Amarna and related
sites. Though the authors had considered the contingency of dissimilarity between Egyptian ceramics and
cuneiform tablets, this possibility was rejected on the basis of the chemical similarity that was found between
Middle Bronze pottery and clay tablets from Ugarit (Ras Shamra). Indeed, our study of the Ugaritic tablets
from Amarna discloses their similarity to Ugaritic ceramics (Chapter 5.I). However, our results prove that
this is not the case for most Egyptian royal letters (excluding EA 163 and 382). This situation, which is far
more complex than previously expected, warns against sweeping conclusions based on scant samples of a
very composite assemblage such as the Amarna archive.
From the examination of the Egyptian letters it may be concluded that in most cases formal
Egyptian documents were written on Esna marl. The reasons for that became clear when other Egyptian
documents within the Amarna archive (such as EA 163 and the school text of EA 368 [Chapter 4])
were examined. Egyptian Nile mud (=Nile silt), the so-called marl clays and their mixtures, commonly
used for pottery production, are very silty and frequently contain sand and other coarse grits (Bourriau
and Nickolson 1992; Arnold and Bourriau 1993:148-182; Bourriau et al. 2000). They also contain
high proportions of ferrous minerals and organic matter that provide dark reddish-brown or even black
products under reducing ring atmosphere. These properties may blur the signs and produce anesthetic
tablets, unsuitable for the high standard of the royal court. Our study of the Canaanite correspondence
reveals that provincial letters in the Amarna archive were ordinarily written on low quality materials (for
example: loess soil, which is rather similar in texture to Nile mud). The examination of letters sent from
other great powers (Babylonia, Ugarit, Alashiya and probably also Mitanni) has disclosed that they too
were produced of clay types that served in these regions for pottery production. Therefore, the case of
the Egyptian letters is exceptional and needs specic interpretation. We suggest that in Egypt a remote
source of clay was used for the production of cuneiform tablets due to the lack of brightly coloured ne
textured clay near the capitals of Amarna, Thebes and Memphis.
In terms of relative chronology, the Egyptian letters can be arranged on the basis of their textual
context as follows:
1.
EA 1 was apparently sent from the former capital, Thebes, and brought to Amarna when the
royal court moved there.
2.
EA 14, 162, 163, 190, 367 and 370 were all sent from Amarna.
This grouping may explain the slight differences between EA 1, 14, and 370, although chemically
they all cluster into one group. It may indicate the use of several clay sources that are somewhat
associated. It is possible that the variability within the group of letters sent from Amarna resulted from the
exploitation of different beds within the Esna Formation. Elemental analyses of this formation (Schreier
1988:104-124) indicate that these beds differ somewhat in their chemical composition. It may therefore
be assumed that the making of clay cuneiform tablets was not as standardized as pottery production.
In terms of availability of Esna shales, there is a considerable difference between Luxor and
Amarna. While at Luxor the Esna Formation is exposed in numerous locations within a relatively small
distance of 20 km from the site (Geological map of Egypt 1:500,000, 1987, Luxor sheet NG 36 SW), at
Amarna this formation does not exist at all. It outcrops only in the area of Qena, nearly 180 km up the
Nile stream. Therefore, the inconsistency in the composition of the tablets written in Amarna may have
resulted from the collection of raw materials in different places along the Nile. Sailing along the Nile
could have easily bridged the distance between Amarna and the raw material for the tablets. For the time
being we have no explanation for the use of Nile silt for the production of the clay of EA 163.
30
II. ATTI
Two or three of the Amarna tablets were sent from atti. EA 42, probably sent by Shuppiluliuma I and
EA 44 from his brother Zida. EA 43 is fragmentary but may be safely assigned to atti by its dialect,
script and contents.

The environs of Bo=azky is characterized by a Mesozoic ophiolitic suite containing limestone, spilite,
dolerite, basalt, marl, radiolarite and serpentine (Ketin 1962:74). Nearby there are exposures of Neogene
continental deposits of various lithologies and Eocene Flysch containing mainly sandstones and sandy
schists (Ketin 1963:48). We were able to examine thin sections of some of the pithoi from the Great
Shrine at Bo=azky, which supplied only general information as to the possible ceramic raw materials
around the site. Our best reference, however, relies on the preliminary results of a new research project
on the composition and provenance of the Hittite archives from Bo-azky and Kltepe presently being
carried out by Goren, Singer, Wilhelm and Mommsen. The examination of the Bo-azky tablets from the
Vorderasiatisches Museum in Berlin, in particular KUB XIX 20 (VAT 7476) which is a Hittite draft of a
letter sent to Egypt by Shuppiluliuma, provided a denitive reference for the clay used for the production
of Hittite tablets.
CATALOGUE
EA 42 (VAT 1655), from atti (?)
Reliability: High.
Matrix: Clayey, birefringent groundmass with striated b-fabric and chamber structured voids (~2%).
The silt fraction contains abundant opaque minerals and mica (essentially biotite) akes. The mica
particles constitute nearly 5%-10% of the groundmass. The silt-sized fraction also contains quartz,
oxyhornblende, epidote and pyroxene.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=~97:3), sand-sized minerals and rock fragments, most
probably detrital in the matrix and not intentionally mixed by the artisan. No remains of vegetal
material were traced. The sand includes rounded grains between 100m and 500m of quartzbiotite schist (dominant); limestone (frequent) up to 600m, micritic; quartzite (common) up
to 400m, with biotite and muscovite as accessories; quartz (common) up to 200m, angular,
sometimes polycrystalline and commonly with undulose extinction.
Firing temperature: 8000C as evident from the partial alteration of hornblende into oxyhornblende and
the beginning of decalcination of the calcite.
Geological interpretation: The mineral suite indicates an area dominated by metamorphic facies.
EA 44 (VAT 1656), from a Hittite prince to the King of Egypt
Reliability: High.
Matrix: Clayey, birefringent groundmass with striated b-fabric and chamber structured voids (~1%).
The silt fraction contains abundant opaque minerals and mica (essentially biotite) akes. The mica
particles constitute nearly 5%-10% of the groundmass. The silt-sized fraction also contains quartz,
oxyhornblende, epidote, plagioclase, serpentine and pyroxene.
31
Inclusions: More densely spread than EA 42 (f:c ratio{0.062mm}=~93:7), sand-sized minerals and rock
fragments, most likely detrital in the matrix and not intentionally mixed by the artisan. No remains
of vegetal material were traced. The sand includes angular to subangular quartz, between 100m
and 500m with undulose extinction (frequent); micritic limestone up to 400m (frequent); quartzite
(common) up to 700m, with biotite and orthoclase as accessories; rounded phyllite crystals, up to
600m (few); serpentine up to 350m, rounded (few).
Firing temperature: 8000C and above as evident from the alteration of hornblende into oxyhornblende
and the partial decalcination of the calcite.
Geological interpretation: Similar to EA 42.
KUB XIX 20 (VAT 7476), a draft letter from Shuppiluliuma of atti to the king of Egypt
Reliability: High.
Matrix: Clayey, birefringent groundmass with striated b-fabric and chamber-structured voids (~2%).
The silt fraction contains abundant opaque minerals, mica (mostly biotite and rarely muscovite)
flakes together with quartz, calcite, hornblende, zircon and epidote.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=-97.3) sand-sized minerals and rock fragments, mostly
detrital in the matrix and not mixed in intentionally by the artisan. The sand contains predominantly
subrounded grains (up to 350m) of quartz-mica schist and frequent grains of micritic limestone
(up to 600m). Angular quartz (up to 370m), sometimes polycrystalline and commonly with
undulose extinction is common, as is quartzite (up to 450m) with biotite and muscovite as
accessories. There are a few angular polycrystalline grains of biotite (up to 300m).
Firing temperature: Not exceeding 800oC as evidenced by the state of the hornblende, but probably ca.
700oC judging by the beginning of decalcilation of the calcite.
CHEMICAL ANALYSES OF THE HITTITE TABLETS

Table A1 in the Appendix presents the elemental composition of the Hittite letters, as disclosed by ICP AES/
MS. The results con rm the general similarity between the two tablets as suggested by the petrographic
results. Differences occur in the rates of Ni and Cu that are higher in EA 44, whereas EA 42 is higher in
strontium. The high rates of copper and nickel in EA 44 may indicate a contamination by the sampling tool
due to the hardness of the tablet and therefore should be ignored. In any event, Fig. 3.2 (below) indicates that
the Hittite letters cluster closely and separately from letters sent from other great powers.
CONCLUSION: THE HITTITE CORRESPONDENCE

Petrographically, EA 42 and EA 44 indicate a similar origin that is dominated by low to medium grade
metamorphic rocks. Chemically too these tablets are similar. Moreover, their petrographic afnities are
idenical to KUB XIX 20 (VAT 7476) and the main group of about 60 tablets from Bo=azky examined
in Berlin. Being unparalleled in the Amarna archive, the origin of these two tablets can condently be
assigned to the Hittite court.
32
III. BABYLONIA
Ten letters (EA 2-4, 6-12) plus one inventory (EA 13) were sent from Babylonia to Egypt. EA 2-4 were
dispatched from Kadashman-Enlil I, King of Babylonia (ca. 1369-1355), to Amenophis III and are the
earliest letters in this group. EA 6 was sent by Burnaburiash II (ca. 1354-1328), Kadashman-Enlils heir,
to Amenophis III, while most of his letters (EA 7-8, 10-11) were sent to Amenophis IV (Akhenaten). His
last letter (EA 9) was sent to Tutankhamun and is the latest dated letter known from the Amarna archive.
EA 12 was sent by a Babylonian princess, apparently the daughter of Burnaburiash, to a King of Egypt
whose name is not mentioned. It is likely that she was destined for the King of Egypts harem, and the
inventory (EA 13) is possibly a list of her dowry.
The Babylonian letters are easily recognized by their shape, script and language. Hence, even in
letters whose authors and addressees names are broken (e.g. EA 4, 13) it is easy to detect their origin.
Therefore, the corpus of Babylonian letters is well dened and the attribution of letters to this corpus is
almost unanimously agreed among scholars.

Very little research has been done on the technology of pottery from Babylonia. So far no petrographic
studies of Lower Euphrates pottery have been brought to our knowledge. A comparison of the
petrographic data from the tablets with Babylonian pottery is therefore impossible. Yet some information
does exist. First and foremost, we extracted comparative data regarding the petrography of sites along
the Euphrates from the thin section collection of the Gerald Avery Wainwright Archive of Mesopotamian
Ceramics (Minors 1986), in the Department of Scientic Research of the British Museum. Some data
on the petrography of Euphrates sediments in pottery can be extracted from articles relating to ceramics
from sites along the Euphrates drainage system. These include Tell Hadidi, Tell al-Sweyhat, Tell Banat,
and Tell Kebir on the upper Euphrates. Downstream one nds Raqqa (Mason 1994; Mason and Cooper
1999), Terqa and Dilbat (Flint 1980), Dura Europos and Mari (Mason and Cooper 1999). Data on the
Lower Euphrates come from Tell ed-Der, Nippur and Isin (Franken and Van As 1994). However these
studies do not supply any comparable petrographic or chemical data.
The Euphrates drains an area characterized by ophiolitic, basaltic, felsic volcanic and plutonic
complexes that expose around its headwaters (Ponikarov 1966, Sheets I-37-II, I-37-III, I-37-IV). From
the point where the two tributaries, the Balikh and the Khabur, join the Euphrates and downstream,
the petrofabrics display a gradual decrease in the appearance of soft minerals that are derived from
this lithology, namely amphiboles and serpentines, and gradual increase in hard minerals (quartz,
clinopyroxene, felsic volcanics). The latter composition is typical to the Mari area, although there is little
petrographic data from Mari itself. Terqa and Dilbat are located in the same geological environment,
about 50 km north of Mari. The Terqa ceramics include quartz, orthoclase and plagioclase as inclusions,
many opaque minerals and, in a few cases, olivine and augite. The latter, however, are probably from a
different source. The matrix is light olive green to deep red brown in colour. It is essentially vitried. In
the Dilbat assemblage the temper is quartz, orthoclase and plagioclase, and less commonly augite, olivine
and calcite, many opaques including magnetite and haematite. The paste is ne grained, from reddish or
olive green to nearly black. The common feature of all these components is that they are rounded and
sorted by the effect of the river transportation and that the clay is usually the rivers ne deposit. Further
downstream these trends are expected to increase since there is no further source of supply fresh igneous
33
rocks or their derived minerals. Therefore, when one reaches Babylon it is expected that the unstable
mac minerals and their alteration products would diminish considerably, and the relative proportion
of stable minerals (rst and foremost quartz) would increase respectively. These changes, which occur
gradually along the drainage system, can be distinguished only over long distances (as we observed
while examining the Gerald Avery Wainwright Archive of Mesopotamian Ceramics). Therefore, when
examining the Babylonian tablets it is difcult to identify the exact production site, but the overall
picture of lower Mesopotamian origin is quite clear.
CATALOGUE
EA 2 (VAT 148+2706), from the King of Babylonia to the King of Egypt
Sampling method: Peeling
Reliability: Satisfactory.
Matrix: Carbonatic, light tan in PPL, birefringent with speckled b-fabric and preferred optical orientation.
Abundant ne calcite crystals, starting at few micrometers and ranging to 70m make about 30%
of the groundmass. Silt (7%) contains calcite, quartz, opaques, oxyhornblende (sometimes with
hornblende core) and feldspar.
Inclusions: No remains of vegetal material were traced. No intentionally added inclusions appear. The
ne sand that appears as clasts within the matrix is made of rare (f:c ratio{0.062mm}=~98:2) grains of
hornblende party altered into oxyhornblende, serpentine, calcite and quartz.
Firing temperature: Approximately 8000C judging by hornblende alteration.
Geological interpretation: The very ne fabric of this tablet, which is almost entirely devoid of inclusions,
gives the impression that well-rened alluvial sediment was used. The few inclusions that do appear
are clearly clasts that occurred naturally within the sediment and escaped the renement process.
The nature of the clay, as well as the composition of the inclusions, can be linked with the
Euphrates sediments. This conclusion is supported by the information retrieved from the other
Babylonian tablets, some of which are less rened, thus supplying more useful petrographic
information. Chemical analyses (below) indicate that all these letters share a common elemental
composition and therefore are most likely of the same origin.
Reference: No references of pottery from Babylon are found, but amphibole, serpentine, quartz and
calcite appear as temper in Euphrates pottery elsewhere (Mason and Cooper 1999).
Conclusions: Like most other Babylonian letters, this tablet is made of carefully rened Euphrates clay.
The less rened specimens (like EA 12 and EA 13) have the same clay properties and may be related
directly by their temper with the Euphrates alluvium, as indicated by comparisons in the Gerald
Avery Wainwright Archive of Mesopotamian Ceramics.
Reliability: High.
Matrix: Carbonatic, light tan in PPL, birefringent with speckled b-fabric and weakly preferred optical
orientation. Abundant ne calcite crystals, beginning at few micrometers and ranging up to 90m
make about 30% of the groundmass. Silt (5%) contains calcite, quartz, opaques, mica laths of biotite
and muscovite, oxyhornblende and feldspar.
Inclusions: No intentionally added inclusions appear. The ne sand that appears as clasts within the
matrix is made of rare (f:c ratio{0.062mm}=~99:1) sparsely spread grains of oxyhornblende, up to
34
200m, sometimes preserving the original hornblende as core; biotite, up to 120m; feldspar, up to
150m, quartz, up to 150m, angular.
Geological interpretation and conclusions: As EA 2.
Reliability: High.
Matrix: Carbonatic, light tan in PPL, birefringent with speckled b-fabric. Abundant ne calcite crystals,
beginning at few micrometers and ranging up to 90m make about 30% of the groundmass. The
opaques appear in two grain sizes: up to 20m and between 50m and 90m. The silt (7%) contains
also calcite, quartz, biotite laths, oxyhornblende, rutile, glaucophane and feldspar.
ne sand that appears as clasts within the matrix is made of rare (f:c ratio{0.062mm}=~98:2) sparsely
spread grains of quartz, up to 280m, angular; limestone, up to 300m, rounded and spherical;
feldspar, up to 150m, oxyhornblende, up to 150m, sometimes preserving the original hornblende
as core; muscovite, up to 120m.
Reliability: High.
Matrix: Carbonatic, light yellowish tan in PPL, birefringent with speckled b-fabric. Abundant ne calcite
crystals, beginning at few micrometers and ranging up to 90m make about 30% of the groundmass.
The opaques appear in two grain sizes: up to 20m and between 50m and 90m. The silt (7%) contains
also calcite, quartz, biotite and muscovite laths, hornblende, oxyhornblende, rutile, and feldspar.
Inclusions: Sand (f:c ratio{0.062mm}=~95:5) made up of quartz, up to 330m, angular to rounded, sometimes
polycrystalline or with undulose extinction; serpentinized olivine(?), up to 160m, subrounded;
feldspars, usually subrounded, including twinned orthoclase (up to 150m), and microcline (70m);
calcite, up to 250m, rounded and spherical; hornblende, up to 70m.
Vegetal material (SLY): Unidentied uncharred tissue fragment, badly preserved, 280m in size.
Firing temperature: Unred or very lightly red judging by the preservation of uncharred vegetal material.
Geological interpretation and conclusions: As EA 2 but unred and with coarser inclusions.
EA 11 (VAT 151 + 1878), from the King of Babylonia to the King of Egypt
Reliability: High.
Matrix: Carbonatic, light reddish tan in PPL, birefringent with speckled b-fabric. Abundant ne calcite
crystals, beginning at few micrometers and ranging up to 40m make about 20% of the groundmass.
Biotite laths are abundant as part of the matrix (nearly 5%). The silt also contains calcite, quartz,
opaques and feldspar.
ne sand that appears as clasts within the matrix is made of rare (f:c ratio{0.062mm}=~99:1) sparsely
spread grains of quartz, up to 200m, angular; limestone, up to 200m, rounded micrite; chert, up
35
to 280m, rounded grains of radiolarian chert (radiolarite) stained with ochre to dark reddish-brown
limonite around the radiolaria spheres.
Firing temperature: undetermined.
Geological interpretation and conclusions: As EA 2. The occurrence of radiolarian chert may be linked
with the north Syrian or Tauric ophiolites where these rocks are common (Chapter 5.I).
EA 12 (VAT 1605), from a Babylonian princess (?)
Reliability: High.
Matrix: Carbonatic, light greyish tan in PPL, birefringent with speckled b-fabric and weak preferred
optical orientation. Abundant ne calcite crystals, beginning at few micrometers and ranging up to
40m make about 20% of the groundmass. Biotite laths are abundant as part of the matrix (4%). The
silt contains also calcite, quartz, opaques, feldspar and zircon.
Inclusions: No remains of vegetal material nor any intentionally added inclusions were traced. The ne
sand that appears as clasts within the matrix is made of rare (f:c ratio{0.062mm}=~98:2) sparsely spread
grains of diorite, 1.6 mm in size; microcline and orthoclase, up to 220m,; quartz, up to 300m,
angular; limestone, up to 320m, rounded and spherical; hornblende, up to 130m, subangular;
granitoid rock fragments, 320m in size, rounded, with orthoclase, hornblende and opaques;
epidote, rounded, 180m, clear; mollusc shell fragment, 1 mm.
Firing temperature: Below 8000C (no hornblende alteration), undetermined.
Geological interpretation: As EA 2.
Conclusions: In his commentary on the Amarna tablets (EAT II) Weber (1907) suggested that this is
either an Egyptian copy, or a draft written in Egypt. Knudtzon (1914:483-486) disagreed and
suggested that the letter was sent from Babylonia. His main arguments were: (a) the style of the
script is Babylonian; (b) the name Kidin-Addu, which appears in the tablet, is certainly Babylonian;
(c) the title daughter of the king does not t a princess who lives in Egypt. He thus concluded that
the princess must have been the daughter of Burnaburiash who lived in Babylonia. Most scholars
supported Knudtzons interpretation (for literature, see Moran 1992:24, note 1).
Both the petrographic and chemical examinations clearly link this letter to the rest of the
Babylonian correspondence and thus support Knudtzons conclusions.
EA 13 (VAT 1717), an inventory from Babylonia
Reliability: High.
Matrix: Carbonatic, greyish tan in PPL, birefringent with speckled b-fabric and weak preferred optical
orientation. The silt is sparsely spread (about 1%) and contains quartz, calcite, opaques, feldspar, mica
laths, hornblende and oxyhornblende, pyroxene and zircon.
Inclusions: The inclusion assemblage in this tablet is somewhat richer than in the other Babylonian letters.
The sand that appears as clasts within the matrix is made of spread (f:c ratio{0.062mm}=93:7) grains
of minerals and rock fragments including feldspar, up to 220m, clear, rarely zoned; quartz, up to
400m, angular to subangular, commonly with undulose extinction; limestone, up to 320m, rounded
and spherical; metamorphic rock fragments, up to 420m, rounded, including fabrics comprising
quartzite and chert intergrowth textures (one rounded phyllite fragment of 350m); serpentine, up
to 300m, rounded, rarely preserving olivine core; hornblende, up to 130m, subangular; epidote,
rounded, up to 100m; glaucophane, rounded, up to 120m; rutile, rounded, up to 120m.
36
Vegetal material (SLY): A fragment of a dicotyledonous stem. The vascular rays are obvious. The rays
are two to three cells wide and about ten cells high. Further identication is impossible.
Firing temperature: Below 8000C (no hornblende alteration), probably not above 5000C judging by the
preservation of uncharred vegetal material.
Geological interpretation and conclusions: As EA 2 but unred or very lightly red.
CHEMICAL ANALYSIS OF SELECTED BABYLONIAN TABLETS

Table A1 in the Appendix presents the elemental composition of selected Babylonian letters as disclosed
by ICP AES/MS.
The results show that EA 13 has very high contents of SO3 as compared with the other tablets. As
explained above, this may be the result of secondary (post-depositional) gypsum crystallization; gypsum
(CaSO4) is likely to crystallize in soils of arid zones such as the Egyptian desert. When the SO3 is
eliminated and the dilution factor corrected, EA 13 becomes similar to the other Babylonian tablets.
Principal component analysis was performed on the Egyptian, Babylonian, Mitannian, Hittite,
and Ugaritic letters in order to examine their intergrouping by their elemental composition (Fig. 3.2).
The results prove that the letters sent by other great powers group in agreement with their textual
provenance, providing another indication that they are originals sent from Western Asiatic kingdoms to
Egypt, not copies made in Egypt.
CONCLUSION: THE BABYLONIAN CORRESPONDENCE

The Babylonian letters, though lacking a signicant body of comparative material, are rather easy to
dene. The clay used for them was most likely extracted from the Euphrates sediments. In most cases it
was rened by soaking in order to eliminate the coarse fraction. The tablets were usually well red, often
at around 8000C, as opposed to the Egyptian tablets that were very lightly red if at all. Chemically the
Babylonian letters indicate a more or less solid group (Fig. 3.2).
Fig. 3.2: Principal component

analysis of ICP data of
selected international
letters and scholarly
texts (27 elements).
37
IV. MITANNI
Mitanni was one of the major powers in Asia in the Amarna period. Its centre was in northern
Mesopotamia and its sphere of inuence reached middle Syria and bordered on the Egyptian province of
Canaan. In the early Amarna period Egypt and Mitanni were close allies whose relations were tightened
by marriage conducted between the two royal houses. Eleven letters (EA 17-21, 23-24, 26-29) plus two
inventories (EA 22, 25) were sent from Mitanni to Egypt. Tushratta, King of Mitanni, dispatched letters
EA 17-21 and 23-24 to Amenophis III, and letters EA 27-29 to Akhenaten. Letter EA 26 was sent by
Tushratta to Teye, the widow of Amenophis III, after the death of her husband. The correspondence
between Egypt and Mitanni came to an abrupt end in the early years of Akhenaten, when Shuppiluliuma,
King of atti, defeated Mitanni. It lost its military power and vassal states as well as its international
status as a member of the great powers club.
EA 24, sent by Tushratta to Amenophis III, is the only letter in the Amarna archive that is written
in Hurrian, the native language of the kingdom of Mitanni. The reason for writing the letter in Hurrian
is not clear. Other exceptional letters are the two Arzawa letters (EA 31-32) written in Hittite, and one
Assyrian letter (EA 15) written in Assyrian. Either the messengers who brought these letters translated
them to the Egyptian scribe, or some Egyptian scholars who studied foreign languages were able to
translate them into Egyptian.
The Mitannian letters are easily recognized by their shape, script and language, so that even in letters
whose authors and addressees names are broken (e.g. EA 18, 25) it was easy to detect their origin.
Therefore, the corpus of Mitannian letters is well dened and the attribution of letters to the corpus is
unanimously agreed among scholars. EA 30 is of special interest. It is a kind of passport addressed to the
kings of Canaan, servants of my brother by the King. The brother is the King of Egypt and all scholars
agree that the letter was written by the King of Mitanni. This is indicated by the messengers name (Akiya,
a typical Hurrian name), by the formulae (e.g. to speed posthaste), by certain terms (e.g. alzulu) and by
the seal impression on the tablet (see Moran 1992:100).
The location of Waukanni (Assyrian Uukani), the capital of Mitanni and the site where the
tablets were possibly written, is debated among scholars. The bilingual Aramaic-Akkadian inscription on
a statue from Tell Fakhariyeh (Abou-Assaf, Bordreuil and Millard 1982) was dedicated to the god Hadad
who dwells in Sikan. It thus discloses the location of the Neo-Assyrian city of Sikan at Tell Fakhariyeh.
This Neo-Assyrian toponym is most probably derived from the Middle Assyrian Uukani, hence the
identication of Waukanni, Mitannis capital, with Tell Fakhariyeh (Abou-Assaf et al. 1982:84-85;
Zadok 1982:123-124). Tell Fakhariyeh is located on the Khabur River, near Ras al->Ain. We set out to
establish whether this site was indeed the place of origin of the Mitanni tablets.

Archaeometric research of the Mitannian letters from the Amarna archive began with the Laurence
Berkeley laboratory teams attempt to disclose the provenance of selected letters through NAA (Dobel
et al. 1977). The goal was to identify the location of Waukanni, the capital city of Mitanni. Six of the
thirteen Mitannian letters were sampled and analyzed. Of these, four (EA 22, 24, 25 and 29) were found
to be remarkably homogeneous, and hence suitable for establishing the chemical prole of Waukanni.
The publication does not specify the details of the other two tablets which were perhaps the two different
fragments of EA 18 (Moran 1992:43, n. 1). However, while one publication refers to nine examined tablets
38
of which the fragments of EA 18 were chemically different (Dobel et al. 1977), the other refers to six
tablets of which two unnamed tablets stood out (Dobel et al. 1977). The publications do not supply any
details about the outliners: their chemical compositions, how they differed from the other four tablets and
what was the possible explanation for their different composition. Moreover, the method of comparison
between the chemical concentrations of the elements in the examined samples is statistically invalid. The
possibility that the area of Waukanni could have had more than one chemical prole resulting from the
existence of different geological formations that include clay deposits, has not been considered.
The Berkeley researchers attempted to establish the chemical prole of the four Mitanni letters
with a chemical database of standard ceramics from several sites that had been suggested as the site of
Waukanni. The comparative material included data obtained by Davidson and McKerrel (1976) in their
study of Halaan pottery from the Khabur headwaters and the clay samples that the latter collected in the
vicinity, some Halaan sherds from Tell Halaf, as well as specially selected pottery from Tell Fakhariyeh
and other neighbouring sites. The possibility that Halaan pottery (early fth millennium BCE) was made
of clay types different from the Mitanni letters was not considered. Indeed, other studies indicate that
Halaan ceramics were made of specically selected clays that would highlight their special decorations
(Noll 1976; Tite et al. 1982). As for the raw clay samples, Davidson and McKerrels study includes many
such samples from Wadi Dara and Wadi Jaghjagh, but only two samples from the Khabur region, collected
15 and 30 miles away from Tell Fakhariyeh (Davidson and McKerrel 1976:48). Both were taken from an
area that is geologically dissimilar to the surrounding of Tell Fakhariyeh (as can be seen in Ponikarov
1964, sheet J-37-V). The selection of fteen items from Tell Fakhariyeh as standards for the local prole
(including two Middle Assyrian economic tablets, not necessarily local or representing clay selection for
Mitannian royal letters) is also problematic. The publication gives no answer to the question as to how the
remaining 13 sherds were selected and why they were considered to be of local production. The fact that
only nine (60%) of these samples proved to be homogeneous makes the method used in this study highly
questionable. Finally, no typological details are given for these sherds, to support the assumption that they
are indeed local to Tell Fakhariyeh. As a result of these shortcomings, and regardless of the importance of
this research in establishing a new tool of analysis for cuneiform tablets, we tend to disregard the results
although they were accepted uncritically by some (e.g. Wilhelm 1989:27).
The entire assemblage of Mitannian tablets in the Amarna archive was examined. For comparison we
used several reference collections of ceramics from the Euphrates drainage system. These include the Gerald
Avery Wainwright Archive of Mesopotamian Ceramics, deposited in the British Museum, Department of
Scientic Research, by S. Mynors (1986). Petrographic data relating to Tell Hadidi, Tell al-Sweyhat, Tell
Banat, Tell Kebir and Raqqa (Mason 1994; Mason and Cooper 1999) were also used. Geological mapping
is supplied by Ponikarov (1964, sheet J-37-V).
CATALOGUE
Reliability: High.
Matrix: Dense, light tan to ochre in PPL, birefringent with speckled b-fabric. Fine silt (2%) containing
quartz, biotite, calcite, epidote (?) and opaques makes up most of the aplastic component.
Inclusions: Very sparsely spread (less than 1%) inclusions, including rounded micritic limestone (up to
250m) and angular to subangular quartz, up to 100m.
39
Firing temperature: There are no indicators in this tablet for determining the ring temperature.
Geological interpretation: Petrographically, the materials of this tablet supply little information about its
provenance. However they indicate a purely sedimentary environment with no nearby igneous rocks.
Reference: No valid references for this petrofabric and others reected by the Mitannian letters were found.
Conclusions: See below.
EA 18 (VAT 1880 + 1879), from Mitanni (?)
Reliability: High.
Matrix: Dense, ochre in PPL, birefringent with striated b-fabric and very strong optical orientation. Fine silt
(4%) containing quartz, biotite, calcite, opaques and feldspar makes up most of the aplastic component.
Inclusions: No inclusions appear. No remains of vegetal material were traced. Very few accidental grains
reach the ne sand fraction including micritic limestone, and chert.
Firing temperature: Probably around 7000C8000C as the calcite exhibits some evidence of decalcination.
Geological interpretation: Petrographically, the materials of this tablet supply little information about its
provenance apart from the pronounced sedimentary nature of its materials.
Conclusions: Apparently this tablet can be related to the Mitanni correspondence by its chemical
composition and its petrographic similarity to other secured Mitannian letters, such as EA 19 (contra
Dobel et al. 1977; Moran 1992:43, n.1).
Reliability: High.
Matrix: Dense, ochre with grey areas in PPL, birefringent with striated b-fabric and strong optical
orientation. Fine silt (4%) containing quartz, biotite, calcite, opaques and feldspar makes up most
of the aplastic component.
Inclusions: Sparsely spread sand (~1%) of angular calcite, up to 400m, sometimes as aggregates
(sparite) and rounded quartz, up to 100m, sometimes with undulose extinction.
Firing temperature: Probably around 7000C, as the calcite does not exhibit any evidence of decalcination
but the matrix is sintered in places.
Geological interpretation and conclusions: This tablet is almost identical in its materials to EA 18,
indicating that the latter too should be included in the Mitanni correspondence.
Reliability: High.
Matrix: Orange-tan in PPL, birefringent with speckled b-fabric and very strong striated optical
orientation. The silt (10%) is rich in mineral types including biotite, quartz, calcite, oxyhornblende,
opaques, pyroxene, zircon, and feldspar.
Inclusions: No inclusions appear. No remains of vegetal material were traced. Very few accidental
grains reach the ne sand fraction including angular quartz with undulose extinction, calcite, and
hornblende in the process of alteration into oxyhornblende.
Firing temperature: 8000C judging by the alteration of hornblende.
40

Matrix: Dense, greyish-tan in PPL, birefringent with striated b-fabric and very strong optical orientation.
Fine silt (4%) containing quartz, biotite, calcite, opaques and feldspar makes most of the aplastic
component.
Inclusions: Sparsely spread sand (~1%) of subrounded micritic limestone, up to 2 mm, subangular
calcite, up to 20m, and rounded quartz, up to 100m, sometimes with undulose extinction.
Firing temperature: Probably around 7000C8000C, as the calcite exhibits some evidence of decalcination.
Matrix: Dense, tan in PPL, birefringent with speckled b-fabric and zoned optical orientation. Fine silt
(2%) containing quartz, calcite, opaques, hornblende and feldspar makes up most of the aplastic
component.
Inclusions: No inclusions nor remains of vegetal material were traced. Very few accidental calcite grains
reach the ne sand fraction.
Firing temperature: No indicators were found to estimate the ring temperature, although the calcite and
hornblende show that 8000C was not reached.
quartz, mica, calcite, and opaques makes up most of the aplastic component.
Inclusions: Very sparsely spread (less than 1%), including rounded micritic limestone, up to 1200m,
and angular to subangular quartz, up to 100m.
Firing temperature: There are no indicators in this tablet for determining the ring temperature.
Geological interpretation: This tablet is identical to EA 17.
Conclusions: As EA 17. See conclusions of EA 27 regarding the Egyptian ink inscriptions.
Reliability: High.
Matrix: Dense, tan in PPL, birefringent with speckled b-fabric and zoned optical orientation. Silt (2%)
containing quartz, calcite, opaques, and zircon makes most of the aplastic component.
Inclusions: No remains of vegetal material were traced. The inclusions appear as accidental clasts within
the matrix (f:c ratio{0.062mm}=93:7). They include rounded micritic limestone, up to 800m, and
subangular quartz:, up to 100m.
Firing temperature: Probably 8000C judging by the partial decalcination of calcite.
Geological interpretation and conclusions: See EA 17 above. Chemical analysis links this tablet to EA
18 (Fig. 3.1) and places it out of the Egyptian and Babylonian clusters.
41
EA 25 (VAT 340 + fragments 2191 a-c, no. 2), inventory of gifts from the King of Mitanni
Reliability: High.
Matrix: Dense, tan in PPL, birefringent with speckled b-fabric and zoned optical orientation. Silt (3%4%) containing quartz, calcite, opaques, serpentine, and hornblende makes up most of the aplastic
component.
Inclusions: No remains of vegetal material were traced. The inclusions appear as accidental clasts within
the matrix (f:c ratio{0.062mm}=98:2) and include rounded micritic limestone, up to 800m, subangular
quartz, up to 100m, and rounded clear twinned calcite, up to 350m.
Firing temperature: Undetermined but lower than 8000C judging by the unaffected calcite and hornblende.
Geological interpretation and conclusions: See EA 17.
EA 26 (BM 29794), from the King of Mitanni to the Queen of Egypt
Reliability: High.
quartz, mica, calcite, and opaques makes up most of the aplastic component.
Inclusions: Very sparsely spread (less than 1%) inclusions, including rounded micritic limestone, up to
350m, and angular to subangular quartz, up to 100m.
Vegetal material (SLY): Unidentied bre-rich uncharred tissue fragment, 400m in size.
Firing temperature: Probably unred or very lightly red judging by the preservation of uncharred
vegetal material.
Geological interpretation and conclusions: This tablet is identical to EA 17.
EA 27 (VAT 233 + 2197, no. 1; 2193), from the King of Mitanni to the King of Egypt
Matrix: Dense, tan in PPL, birefringent with striated b-fabric and very strong optical orientation. Fine
silt (4%) containing quartz, biotite, calcite, opaques and feldspar makes up most of the aplastic
component.
Inclusions: No inclusions appear. No remains of vegetal material were traced. Very few accidental grains
reach the ne sand fraction including calcite and some angular quartz.
Firing temperature: Probably around 7000C 8000C, as the calcite exhibits some evidence of decalcination.
Geological interpretation: See EA 18.
Conclusions: Two tablets in the Mitanni correspondence (EA 23 and 27) bear hieratic dockets. The
inscription on EA 27 states that it is a copy of a letter from Naharin (Moran 1992:90, n. 20).
Ostensibly it seems logical to suggest that EA 27 is a copy that was made in Egypt of an original
Mitannian letter. Yet the petrographic analysis of these two tablets leaves it beyond doubt that
they are similar in all respects to the other Mitannian letters and differ from all Egyptian tablets
included in the Amarna archive. This is apparent not only in the mineralogical traits of the tablets,
but also in the high ring temperature that is typical of the Mitannian and Babylonian letters but
not of the Egyptian letters.
Based on the similarity to other Mitannian letters examined here, this tablet is undoubtedly a
letter from Mitanni and not an Egyptian back-up copy.
42

Reliability: High.
quartz, mica, calcite, epidote, and opaques makes up most of the aplastic component.
Inclusions: Very sparsely spread (less than 1%) rounded micritic limestone, up to 600m, and angular to
subangular quartz, up to 100m.
Vegetal material (SLY): Uncharred bre fragments, up to 500m.
Firing temperature: Probably un red or very lightly red since the vegetal material is uncharred.
Reliability: High.
Matrix: Dense, reddish tan to tan in PPL, birefringent with speckled b-fabric. Fine silt (1%) containing
quartz, mica, calcite, hornblende, and opaques makes most of the aplastic component.
Inclusions: No inclusions were found in this tablet apart from an accidental calcite crystal, 120m in size.
Firing temperature: Undetermined but lower than 800 0C judging by the unaffected calcite and
hornblende.
EA 30 (BM 29841), from the King of Mitanni to the mayors of Canaan
Matrix: Dense, light yellowish tan PPL, birefringent with speckled b-fabric. Fine silt (2%) containing quartz,
mica, calcite, feldspar, epidote, and opaques makes most of the aplastic component.
Inclusions: Sparsely spread (2%) inclusions, including rounded micritic limestone, up to 200m, and
subangular calcite, up to 200m.
Firing temperature: Undetermined but lower than 8000C judging by the unaffected calcite and hornblende.
Geological interpretation and conclusions: This tablet is similar to EA 17.
CHEMICAL ANALYSES OF TWO MITANNIAN TABLETS

Table A1 in the Appendix presents the elemental composition of EA 18 and EA 24, as disclosed by ICP
AES/MS. Fig. 3.1 presents their relationship with other international letters, as indicated by principal
component analysis of the logged data.
Of the two tablets, we take EA 24 (the Hurrian letter) as the more obvious representation of
Mitannian clay, since EA 18 is very fragmentary and its origin has been doubted. Table A1 indeed shows
that the elemental compositions of the two tablets differ in several respects. EA 18 is richer in Si and
Al and higher by far in Zn than EA 24. However, when plotted on a multivariate basis (by principal
component analysis of 27 elements) against the other international letters (Fig. 3.1), the two tablets are
still found to be closer to one another than in relation to any other tablet. This means that although EA
18 and EA 24 differ somewhat, they are made of two related clay types that are markedly different from
those of the Babylonian and Egyptian letters. Therefore, we consider them as made of two different types
of clay that were used simultaneously in Mitanni for the production of tablets. As stated before, this does
43
not necessarily mean that the Mitannian tablets were produced in two different locations if the area of
Waukanni had a CTF value that was greater than 1, which is indeed the case for Tell Fakhariyeh and
most of the Jezireh area in general. Consequently, both clay types can be used in future studies in order
to disclose the possible location of Waukanni.
CONCLUSION: THE MITANNIAN TABLETS AND THE LOCATION OF WAUKANNI

All the Mitannian letters in the Amarna archive were analyzed in the present study. They may be
summarized as follows:
1. Two distinctive clay types were used by the Mitannian scribes: a marly type (including EA 17,
20, 22, 23, 24, 25, 26, 28, 29, and 30) and a clayey type (including EA 18, 19, 21, and 27). These
are probably the two chemical proles referred to by Dobel et al. (1977). However, as explained
above, the idea that each site is represented by only one chemical prole is over simplistic.
Unfortunately, such approaches still appear sometimes in the current literature regarding NAA
data (McGovern 2000 and the methodology of reference materials selection presented; Goren
2003 for a review of this issue). The environs of Waukanni apparently included two types of
workable clay that were occasionally alternated by the scribes of the Tushratta court.
2. Signicantly, the two fabrics include similar types of clastic components (silt and sand),
indicating a very similar depositional environment.
3. The composition of the clastic components indicates a sedimentary environment with no supply
of minerals or rock fragments derived from igneous rocks (apart from some wind-blown mica
and feldspars that occur in the ne silt fraction).
4. This indicator eliminates the entire upper Euphrates basin. Volcanic, metamorphic and intrusive
clasts and their derived minerals are reported from ceramics made in sites located in that area.
5. Sedimentary provinces that are dominated by calcareous rocks are found in the upper Khabur
and Jaghjagh basins. Wide-ranging basalt exposures appear along the middle Khabur River at
Ard esh-Sheikh. Therefore, taking the Khabur basin as the most likely area where Waukanni
should be sought, it should be located northwest of the Ard esh-Sheikh plateau.
6. In this area, Tell Fakhariyeh is apparently the most likely candidate. When these petrographic
results are combined with the textual and archaeological evidence, Tell Fakhariyeh seems to
become the only possible site for the location of Waukanni, the capital city of Mitanni.
44
V. ARZAWA
Only one letter sent by Tarundaradu, King of Arzawa, appears in the Amarna archive (EA 32). The
kingdom of Arzawa is mentioned once in a letter sent to him by Amenophis III (EA 31:2). EA 32 does
not carry the name of the sender and is identied as an Arzawa letter thanks to EA 31. At the request of
the Arzawan scribe, the correspondence is carried out in Hittite (EA 32:24-25).
Arzawa is believed to be located in western Anatolia, between the Hermos and Meander Rivers
(Hawkins 1998).
EA 32 (VAT 342), from the King of Arzawa to the King of Egypt
PETROGRAPHIC ANALYSIS
Matrix: Homogenous, orange to reddish tan in PPL, carbonatic with some silt (2%) and akes of mica
minerals including muscovite and biotite. Other heavy minerals that appear in the silt fraction contain
prismatic zircon, garnet, feldspars, oxyhornblende and epidote.
Inclusions: There are no intentionally mixed inclusions in this tablet, but the quartz and feldspar grains
occasionally reach the ne sand grain size. Calcite occurs in the same grain sizes too. No remains
of vegetal material were traced.
Firing temperature: About 8000C judging by the alteration of hornblende into oxyhornblende and the
partial decomposition of the calcite.
Geological interpretation: The combination of mica, quartz and garnet, the latter being rather sensitive
to chemical and physical weathering, indicates a schistose environment.
Conclusions: The petrographic evidence that EA 32 supplies is scanty. Since garnet-mica schists are
quite common in western Turkey, the conclusion that they originate in a schistose environment
cannot serve for any detailed provenance determination.
NEUTRON ACTIVATION ANALYSIS OF EA 32

Michal Artzy, Hans Mommsen and Frank Asaro
This tablet, among others, was already analysed in the 1970s by Neutron Activation (NAA) in Berkeley
and never published. Since the supposed location of the kingdom of Arzawa is western Anatolia, the
chemical composition of this tablet should be compared with compositional data of pottery from
there. Since trace elements are involved, care should be taken to use precise absolutely measured data
(preferably by NAA) for the study.
The results obtained at the NAA laboratories in Berkeley and Jerusalem are directly comparable
with those from Bonn since the Bonn pottery standards are calibrated with those of the other two
laboratories (Perlman and Asaro 1969; Mommsen et al. 1991). This strongly suggested that the Bonn
data be used for comparison with the data of the Arzawa tablet given in the Appendix: Table 3. At the
moment the Bonn NAA data bank holds more than 350 samples from different sites in western Anatolia,
mainly from Ionia (Miletus, Ephesos, Smyrna, Phokaia) and the Aeolis (Kyme). They cover the time45
span from the Late Bronze Age till the Archaic period with a few samples from Hellenistic and Roman
times. The statistical evaluation of these data to form groups of samples of similar composition resulted
in a number of different characteristic chemical patterns. Each of these patterns can be assigned to a
clay paste prepared for pottery production in one or several workshops exploiting the same clay beds.
For some of the groups the production site or at least the production area could be established by the
occurrence of kiln material or wasters in these groups. For others the provenance could only be de ned
due to archaeological considerations of the group members. A rst set of about 100 of these samples was
already published (Akurgal et al. 2002).
In a statistical evaluation of compositional data each sample is commonly represented by a point
in the m-dimensional space of elemental concentrations where m is the number of elements under
consideration (in this case 17 elements). Al and Mn are not measured in Bonn, and Ca and Na are not
considered since they are known to vary strongly in groups made from the same clay paste (Mommsen
2001). Samples with a composition similar to that of the tablet will be located in the 17-dimensional
space in the neighbourhood of the point given by the tablet, therefore the similarity is measured by a
distance. The lter procedure developed in Bonn (Beier and Mommsen 1994) will sort out of the data
bank all such points which have distances below a given cut-off value. The distances are calculated
considering two important facts. One is the consideration of a possibly varying dilution agent like
different amounts of sand in the paste raising or lowering all concentration values by a constant factor.
Secondly, the measurement errors have to be taken into acount, since for large errors larger differences
in composition (large distances) are statistically acceptable (e.g. [10.0 1.0] g/g and [10.5 1.0] g/g
is considered to be similar, since the difference normalized to the error is only 0.5/1.0 = 0.5, but [10.0
0.1] g/g and [10.5 0.1] g/g is dissimilar, since in this case the same absolute difference 0.5 g/g
gives, normalized to the error 0.5/0.1 = 5).
For the data of the tablet EA 32 no errors are reported. To use our lter, we have introduced constant
errors of 5% for all concentration values of the tablet except for Ni, Rb, and U, where 10% are taken.
These values have been estimated from a publication quoting Berkeley NAA data including errors
(Karageorghis et al. 1972). Before the lter search, the Cr and Yb values of the tablet have been corrected
(Cr 11% down, Yb 5.7% up) according to a recent recalibration of the Berkeley pottery standard values
(Strange et al. 1995).
Not only the western Anatolian data bank, but the whole Bonn data bank with more than 5000
additional samples from different regions of Greece, Macedonia, the Troad and a few samples from the
Levant and Egypt can be searched by the lter procedure for samples similar to EA 32. Furthermore, a
data bank holding more than 800 samples of Greek pottery measured in Berkeley was added (Mommsen
et al. 2002). According to our criteria the result of the search was that no sample with an exactly
similar composition was found and therefore the paste of the tablet was not used to produce any of
the pottery samples measured. However, many samples of a western Anatolian group with not very
different elemental abundances were ltered out by the search (and only these). It turned out that the
tablet has a composition which is closely associated to a group of samples which was published as group
G in Akurgal et al. (2002). This small and at that time still not well-dened group of seven samples
has increased meanwhile to 40 members (one from the site Klazomenai, ve from Smyrna, eight from
Phocaia, 10 from Larisa and 16 from Kyme). Therefore, its average elemental compositions m and its
spread values (root mean square deviations) changed slightly compared to the published pattern. It is
compared in Table 3.1 to the values of EA 32 which are multiplied by the best relative t factor of 1:
31 correcting for dilutions. One of the dilutents of the tablet is Ca, having a much higher concentration
46
in the tablet. For all the other elements the differences are less or not much larger than twice either the
errors of the tablet or the spreads of group G. It must be also taken into account that there might be
small laboratory differences considering the early age of the measurement of the tablet in Berkeley and
that the assumed errors might be too small. Yet even in our standard statistical evaluation the sample
EA 32 would be classied as being associated with the group G. The workshop producing the vessels of
this group has not yet been located exactly but, according to the distribution of members of this group, a
provenance for EA 32 in northern Ionia or even the Aeolis seems very probable. There is no agreement
in composition with several groups in our data bank which can be assigned with high probability to
workshops in Ephesos (Kerschner, forthcoming).
TABLE 3.1: ELEMENT CONCENTRATIONS (C) OF EA 32 (ELAM 23) COMPARED TO AVERAGE
CONCENTRATIONS (M) OF GROUP G.
EA 32 (Elam 23)
1 sample
factor 1.31
C +/- %
Al%
Ca%
Co
Cr
Cs
Eu
Fe%
Hf
La
Lu
Mn
Na%
Ni
Rb
Sc
Sm
Ta
Th
Ti\%
U
Yb
11.5
11.6
28.0
154.
18.6
1.82
6.72
5.21
49.6
0.50
11.6
0.69
145.
187.
21.5
8.42
1.19
19.0
0.49
5.10
3.61
0.6
0.6
1.4
8.
0.9
0.09
0.34
0.26
2.5
0.03
0.6
0.03
15.
19.
1.1
0.42
0.06
0.9
0.02
0.51
0.18
Group G (northern Ionia)

40 samples
factor 1.00
M +/- %
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
10.
10.
5.0
5.0
5.0
5.0
5.0
10.
5.0
4.94
27.7
197.
24.8
1.92
5.95
5.74
54.3
0.59
0.96
188.
177.
21.2
9.49
1.18
20.6
0.47
3.79
4.12
0.89
1.3
21.
3.4
0.10
0.13
0.40
2.0
0.03
18.
4.8
11.
14.
5.0
2.2
7.0
3.7
5.8
0.12
54.
6.
0.5
0.65
0.04
0.6
0.09
0.27
0.20
12.
29.
3.1
2.2
6.9
3.7
2.7
19.
7.2
4.9
47
VI. ALASHIYA
Yuval Goren, Shlomo Bunimovitz, Israel Finkelstein and Nadav Naaman
For more than a century the location of the Bronze Age kingdom of Alashiya, that old conundrum
as Muhly (1996:49) so aptly dubbed it, has continuously been debated with diminishing returns (for
comprehensive and critical discussions of the data and previous literature see Merrillees 1987; Knapp
1996a). The textual evidence from both Amarna and Ugarit indicates that during the 14th and 13th
centuries BCE Alashiya maintained economic and political contacts with Egypt and north Syria. The
documents suggest that it produced and exported large amounts of copper and that it was an independent
state, since its king was referred to as brother in his correspondence with the King of Egypt a
designation that was used between rulers of equal rank.
Most scholars associate the copper producing-land of Alashiya with part or all of Cyprus. This
conclusion is supported by an overall interpretation of the relevant documents within the historical,
geopolitical and archaeological background of the eastern Mediterranean in general and Cyprus in
particular (e.g. Holms 1971; Muhly 1972; 1989; 1996; Knapp 1985; 1996a). The place of origin of
the Alashiya tablets was generally identied at Enkomi near the eastern coast of the island. A minority
view considers the data circumstantial and inconclusive (e.g. Hellbing 1979; Merrillees 1987). Scholars
advocating the latter view tend to identify Alashiya in either part or all of Cilicia, or part of northwestern
Syria (see Merrillees 1987 for a review of literature).
Provenance studies could have resolved the problem by establishing the origin of copper oxhide
ingots and the Alashiya clay tablets found in Amarna and Ugarit. Unfortunately, the studies conducted
so far have produced ambiguous results (for the difculties involved in copper analyses see Muhly 1996:
47-49; Gale 1989; Budd et al. 1995 with responses; 1996; for the tablets see below).
Eight tablets dispatched from Alashiya to the Egyptian court were discovered at Amarna (EA 3340). Seven of these letters (EA 33-39) were sent by an unnamed King of Alashiya to an unnamed King
of Egypt, while the eighth letter (EA 40) was dispatched by the governor of Alashiya to his equal in
the Egyptian court. The omission of names of authors and addressees makes it difcult to establish the
chronology of the Alashiya correspondence within the Amarna period.
Two of the Alashiya tablets (EA 34 and EA 35) now in the British Museum were examined by NAA
in order to determine their origin (Artzy et al. 1976). They were compared with Late Bronze Age pottery
from different sites in Cyprus and the Levant. The report states that the two tablets, which have similar
chemical composition patterns, were not made of eastern Cypriote clay and hence are most probably
not from the vicinity of Enkomi and Kalopsidha. They are also chemically different from the clays of
Toumba tou Skourou near Morphou Bay, Lapithos and other places in the north of the island, Nitovikla
and other sites in the Karpas, Kition and Hala Sultan Tekke in the south on the Larnaca Bay, and Amathus
on the Akrotiri Bay. Certain chemical similarities were found between the two tablets and Mycenaean
IIIC1 sherds from Kouklia (Palaeopaphos), although a real match between the tablets and the Kouklia
clay composition could not be proved. Artzy and her colleagues also concluded that the two tablets are
not copies made in Egypt, since they differ from clays used for the production of New Kingdom pottery
in Amarna and elsewhere in the Nile Valley. A later analysis by the same team of two other Alashiya
tablets (EA 33 and EA 38) now in the Vorderasiatisches Museum in Berlin showed similar results (M.
Artzy pers. comm.; Hellbing 1979:71, n. 103).
48
Four Alashiya letters from Amarna (EA 33, 34, 37 and 38) and one Alashiya letter from Ugarit were
examined in the course of the present research project. We attempted to enlarge the analytical basis that
has been established by Artzy and her colleagues by using the following research strategies:
1. Examination of tablets that were not included in the previous research, especially EA 37 that, as
already noted by Knudtzon (1915:1272, n. 2; 1275, n. 1; 1276, n. 1; 1294, n. 2; 1298, n. 1), was
different in its general fabric and inclusion type from all other Alashiyan tablets.
2. Analysis of the Alashiya letters from Ugarit.
3. Examination of the previously analyzed tablets by other methods, namely petrography and ICP.
4. Examination of clay documents from Late Cypriote sites in Cyprus, especially Enkomi, for
comparison.
As a working hypothesis, the widely accepted identication of Alashiya in Cyprus was adopted but
without ignoring other possibilities. The mineralogical and chemical data of the tablets were therefore
compared with the geology of Cyprus, Cilicia, and northwestern Syria. The catalogue below presents the
tablets according to an order of priorities, from the most informative to the less important.
CATALOGUE: THE ALASHIYA LETTERS FROM AMARNA

Description: To the naked eye this tablet differs in colour and appearance from all other Alashiya tablets
examined [EA 33, EA 34, and EA 38 (and the other Alashiya tables, as explained earlier)]. The fabric
is generally reddish-brown rather than cream/whitish-grey. There are large inclusions (reaching 3
mm in size) that do not appear in other Alashiya tablets, and the texture is coarser than that of the
other Alashiya letters from Amarna.
Sampling method: Peeling supplemented by SPA (collection of inclusions from other parts of the tablet).
Reliability: High.
Matrix: Petrographically, this tablet is totally different from the other Alashiya tablets from Amarna so
far examined. The matrix is carbonatic, slightly silty (about 1%), yellowish-tan in PPL and contains
abundant reddish spherical bodies that sometimes preserve skeletal structures of columnar plagioclase.
These may accord with the ferric illite described by Vaughan (1989). They vary in size from about
50m up to 500m. The origin of these bodies is made clear by the examination of the inclusions in
which weathered spilitic basalt is apparent. These weathered rocks were probably the origin for their
residues in the matrix in which remnants of skeletal feldspar crystals can still be seen. Therefore, the
matrix of this tablet may be de ned as clay containing highly weathered basaltic clasts.
Inclusions: No remains of vegetal material were traced. Apart from the highly weathered basalt remnants
that were described above, the inclusions contain rounded fragments of highly weathered spilitic
rock including skeletal plagioclase laths (sometimes branching) ophitically enclosed by melanocrats
partly or completely altered into red clay. These inclusions, exceeding 3 mm in size, also include
some clasts of better preserved spilite. A fragment of volcanic glass also appears in the aplastic
component. All this hints at a pillow-lava complex. Since the method of sampling by SPA is not
quantitative, it is impossible to determine the relative proportions of these components in the
overall assemblage of the inclusions. However, surface examination under the stereomicroscope
suggests that this is the dominant inclusion category and this is corroborated by the abundance
of weathered basaltic clasts in the matrix. Also present are sparitic limestone fragments, up to 3
mm in size, and low-grade metamorphic greywacke represented by rounded spherical fragments
of quartzitic microgranular sandstone sizing about 1 mm. The sandstone falls into the greywacke
49
classication, though it is microgranular with rounded to subangular grains of quartz that commonly
exhibit undulose extinction showing that the quartz has been subjected to some straining. Weathered
feldspars constitute a secondary component together with the quartz. These minerals are intertwined
with chloritic biotite and fewer opaques (presumably iron oxides). Radiolarian mudstone appears as
subrounded tan grains up to 3 mm in size with 3%-5% radiolaria microfossils. Single clinopyroxene
crystals, about 250m in size, as well as a large aggregate of similarly sized clinopyroxene crystals
are visible. Epidote occurs as rounded, single crystals, about 200m in size, as do subangular
to angular sand-sized (around 100m-200m) grains of quartz. Angular grains up to 1 mm of
replacement chert are also among the inclusions.
Palaeontology (LG): Small planctonic foraminifers: Hedbergella (p), Heterohelix (p). Age: Upper
Cretaceous (Senonian).
Firing temperature: Undetermined but below 700 0 C since there are no signs of anomalous colour in
the calcite.
Geological interpretation: Clay derived from the Moni Mlange Formation of the southeast Troodos
foothills.
Reference: The fabric of this tablet is similar in all details to one of the petrofabrics of the Base Ring
Ware from Kalavasos Ayios Dhimitrios (Vaughan 1989) and the inscribed cylinder K-AD 82, 405.
Conclusions: See below.
EA 33 (VAT 1654), from the King of Alashiya to the King of Egypt
Reliability: High.
Matrix: Pale yellow in PPL, optically active speckled b-fabric, with high carbonatic component. Mica
akes (both muscovite and biotite) are abundant within the matrix, as well as small (up to 100m)
bodies of yellow to orange clay. Fine silt (less than 1%) includes calcite, quartz and feldspar.
Calcitic foraminifers appear.
Inclusions: This tablet is almost devoid of non-plastics. The sample was therefore enlarged in order to
obtain more inclusions. This was possible due to the presence of two large breaks on both edges of
the tablet from which larger fragments could be chipped off without causing any damage whatsoever
to the written surfaces. The few coarse fraction components that appear in it seem to be detrital
within the clay matrix and not intentionally added temper. These include a single angular 200m
grain of dolerite containing twinned plagioclase and serpentinized clinopyroxene which, though
small, permits the rock type to be determined as well as very few subangular grains of micritic
limestone, (up to 200m). No remains of vegetal material were traced.
Palaeontology (LG): Globigerina (p), Globorotalia (p). Age: Neogene.
Firing temperature: Undetermined, seemingly unred.
Geological interpretation and conclusions: Marl of the Pakhna Formation of Cyprus.
Matrix: The matrix is a light carbonatic marl containing ne brous carbonate crystals exhibiting weak
optical orientation. The matrix contains very few foraminifera and yellowish clay concentrations.
There is almost no silt.
50
Inclusions: There are almost no inclusions. No remains of vegetal material were traced. The only nonplastic components in this sample include a quartz grain, a foraminiferous chalk grain and a wellpreserved goat hair.
Palaeontology (LG): The matrix contains ghosts of unclear foraminifers. One is probably of Paleocene
to Eocene age.
Firing temperature: Probably unred judging by the preservation of hair.
EA 38 (VAT153), from the King of Alashiya to the King of Egypt
Reliability: Fair.
Matrix: Pale yellow in PPL, optically active speckled b-fabric, with high calcitic component. Mica akes
(both muscovite and biotite) are abundant within the matrix, as well as small (up to 100m) bodies
of yellow to orange clay. Fine silt (less than 1%) includes calcite, quartz and feldspar. Calcitic
foraminifers appear.
Inclusions: No inclusions appear in this sample. No remains of vegetal material were traced.
Firing temperature: Undetermined, seemingly unred.
COMPARATIVE STUDY OF CYPRO-MINOAN TEXTS FROM ENKOMI

E-1687, a Cypro-Minoan tablet
Description: A Cypro-Minoan inscribed clay tablet, published by Dikaios (1971:881-891, Pl. 317-18),
found in Level III A (late 13th-early 12th century BCE).
Reliability: High.
Matrix: Light yellowish-tan in PPL, carbonatic and lightly optically oriented in thin section. The matrix is
slightly silty (about 1%) and stained by orange-brown clay concentrations and iron-rich (limonitic)
bodies, sizing up to 100m. Foraminifers are extremely rare.
Inclusions: A well-sorted set of various minerals and rock fragments occurs, all spherical to subangular
and usually rounded grains, including cloudy to slightly cloudy quartz with uid inclusions,
commonly with undulose extinction, up to 250m (frequent); sparitic or micritic limestone:
usually slightly decarbonated due to ring, sizing up to 350m (frequent); spilitic basalt: usually
weathered, including fabrics comprising plagioclase laths ophitically enclosed by augite usually
altered to clay, up to 300m, (common); chert, usually clear but sometimes ferrous replacement
chert, up to 250m (common); plagioclase feldspar, twinned with sharp boundaries, up to 200m
(few); serpentine with relict olivine, up to 150m (very few); pyroxene, generally clinopyroxene
with light greenish pleochroism, up to 100m (very few); epidote, up to 100m (trace).
Firing temperature: Probably around 7000C judging by anomalous interference colours in the calcite.
Geological interpretation: The unimodal nature of the inclusions, in terms of their shape and sorting, may
be considered to represent alluviated detrital sediments. These include volcanic and sedimentary
rocks of varying types. The matrix can be generally dened as marl. The inclusion deposits may
have been derived from an area containing several lithological environments, including igneous and
marine sedimentary rocks.
51
The effect of stream transportation may sort material by mineralogy (weight), resistance to
mechanical and chemical weathering, abundance along the drainage basin and the ow regime of
the conduit. The general sphericity of all the inclusion components, especially the resistant chert and
quartz, may indicate a considerable distance between the bedrock and the depositional area where
the inclusions were collected.
Enkomi is situated in eastern Cyprus, 7 km north of Famagusta. It is located in an area of
Quaternary deposits containing gravel, sand, silt and clay. Sediments now obstruct the stream of the
Pedhieos River, passing near the southern edge of the site. However, in the Late Cypriote period it
was probably still viable as a sailing channel for boats between the coast and Enkomi, and perhaps
even further inland (Catling 1964:17; Dikaios 1969:10). The Pedhieos river drainage system included
the Mesaoria plain, the southeastern part of the Kyrenia terrain and the northeastern slope of the
Troodos massif. Since sedimentary rocks cover all the Mesaoria terrain, the Pedhieos River becomes
the only possible supplier of volcanic rocks and their derived minerals to the Enkomi area.
Conclusions: Petrographic considerations may link this tablet to the general area of eastern Cyprus,
although Enkomi cannot be considered the only option. However, from the broader view of the
archaeological evidence and the petrographically homogeneous assemblage of the Enkomi texts, it
is reasonable to assume that it was produced there.
H-1885, a Cypro-Minoan tablet
Description: A Cypro-Minoan inscribed clay tablet, published by Dikaios (1971:881-891, Pl. 314), found
in Level I B (mid-late 15th century BCE).
Reliability: Moderately to adequate.
Matrix: Yellowish-tan in PPL, argillaceous, optically speckled with clear optical orientation in thin
section. In contrast to the matrix of E-1687, it is not carbonatic. The matrix is slightly silty
(about 1%) and extremely rich in akes of mica minerals, essentially biotite with some additional
muscovite, ranging from single micrometers up to 100m. The matrix is partly isotropic in places
due to the high ring temperature.
Inclusions: A badly-sorted set of various minerals and rock fragments occurs, most of which are
subrounded, including sparitic or micritic limestone, usually slightly decarbonated due to ring, up
to 350m (frequent); spilitic basalt, usually weathered, up to 250m (common); cloudy to slightly
cloudy quartz with uid inclusions, commonly polycrystalline with undulose extinction, up to
500m but usually below 200m (common); replacement chert, up to 200m (common); plagioclase
feldspar, twinned with sharp boundaries, up to 200m (few); serpentine with relict olivine, up to
150m (very few).
Firing temperature: Probably around 8000C judging by anomalous interference colours in the calcite, the
beginning of decalcination of limestone and the partial isotropism of the matrix.
Geological interpretation: Although the matrix is somewhat different from that of E-1687, the inclusion
assemblage is generally similar and the same geological conclusions can be drawn.
Conclusions: As E-1687.
E-1610, a Cypro-Minoan inscribed cylinder
Description: A Cypro-Minoan inscribed clay cylinder found by the French Mission headed by Schaeffer
and published by E. Masson (1971).
52
Reliability: Fair.
Matrix: Similar to E-1687, but almost isotropic due to the high ring temperature, estimated at 9000-10000C.
Inclusions: Only very small pieces were sampled by SPA because this inscribed cylinder was completely
preserved. Therefore frequencies could not be estimated. Obviously, not all the components
are represented in the sample. The inclusions in the sample consist of limestone, up to 350m,
decarbonated due to ring, usually just remnants of deformed carbonate crystals within ghosts
preserving their shapes, with reaction facades around the void where the clay matrix reacted with the
carbonate; serpentine with relict olivine, up to 400m (very few); quartz, sometimes polycrystalline,
up to 200m; replacement chert, up to 350m; feldspar with sharp boundaries, up to 200m.
Firing temperature: Probably around 9000 -10000C judging by anomalous interference colours in the
calcite, decalcination of the limestone that already reacted with the clay around the edges of the voids
and the isotropism of the matrix.
Geological interpretation and conclusions: As E-1687.
E-1193 + Enk. F.E20.01, a Cypro-Minoan tablet
Description: Two joined fragments of a Cypro-Minoan inscribed clay tablet. One fragment (E-1193) was
found in Level IIIB (12th century BCE) and published by Dikaios (1971:881-891, Pl. 319, 320). The
second fragment (Enk. F.E. 20.01) was found by Schaeffer (Karageorghis 1970:249, Fig. 99; for the
join see Michaelidou-Nicolaou 1980).
Reliability: High.
Matrix: Light yellowish-tan in PPL, carbonatic and lightly optically oriented in thin section. The matrix is
slightly silty (about 1%) and stained by orange-brown clay concentrations and iron-rich (limonitic)
bodies, sizing up to 150m. Large foraminifers are common.
Inclusions: A well-sorted set of various minerals and rock fragments occurs, all spherical to subangular
and usually rounded grains, including sparitic or micritic limestone, usually slightly decarbonated
due to ring, up to 400m (frequent); cloudy to slightly cloudy quartz with uid inclusions,
commonly with undulose extinction, up to 250m (common); spilitic basalt, usually weathered,
including fabrics comprising plagioclase laths ophitically enclosed by augite usually altered to red
clay, up to 400m (common); serpentine, up to 200m, one inclusion 3.5 mm (common).
Conclusions: As E-1687.
AM 2336, a Cypro-Minoan tablet
Description: A Cypro-Minoan inscribed clay tablet now in the Muse du Louvre published by Dikaios.
Reliability: Fair.
Matrix: Similar to E-1610.
Inclusions: Only very small fragments were sampled by SPA. Frequencies could not be estimated since
obviously not all the components are represented in the sample. The inclusions present consist
of limestone, decarbonated due to ring, up to 270m; serpentine, 250m;.quartz, sometimes
polycrystalline, up to 200m.
Firing temperature: Probably around 9000C judging by decalcination of the limestone and the isotropism
of the matrix.
Geological interpretation and conclusions: As E-1687.
53
COMPARATIVE STUDY OF CYPRO-MINOAN TEXTS FROM KALAVASOS AYIOS DHIMITRIOS

Description: A Cypro-Minoan inscribed clay cylinder, published by E. Masson (1983:131).
Matrix: Clayey, reddish-tan in PPL, sparkled birefringent and optically oriented. Silt (~10%) includes
quartz, very abundant mica laths, epidote, and plagioclase. Opaques appear in the silt fraction (~5%)
as angular bodies, presumably magnetite.
Inclusions: The rich and dense inclusion assemblage (f:c ratio{0.062mm}=85:15) includes subangular quartz,
up to 240m; subangular plagiogranite, a 2.5 mm grain is included in the sample; quartz-mica
phyllite sometimes showing foliation of the micas, up to 600m; rounded epidote, up to 120m;
rounded replacement chert, up to 120m; mica laths, up to 100m; rounded apatite, up to 120m;
rounded to angular opaque minerals, up to 130m.
Geological interpretation: Autochthonous Troodos clays of the Kannaviou Formation (similar to those
discussed in Vaughan 1989).
Reference: A main fabric of the local Base-Ring Ware at Kalavasos Ayios Dhimitrios (Vaughan 1989).
Matrix: Carbonatic, tan in PPL and silty (~5%). The silt contains quartz, mica laths, pyroxenes, and calcite.
Opaques (~2%) appear in two grain sizes: ne (up to about 10m - 15 m) and coarser (50m - 100m).
Inclusions: These include rounded fragments of weathered spilitic basalt and its constituent minerals,
up to 330m; foraminiferous chalk fragments, rarely glauconitic, up to 300m; disintegrated
clinopyroxenes, amphiboles and feldspars derived from the above rocks also appear; rounded
fragments of welded microgranular quartzitic sandstone, 1.35 mm in size. The sandstone falls into
the greywacke classication, though it is microgranular subangular grains of quartz that commonly
exhibit undulose extinction. Therefore, the quartz has been subjected to some straining. Chlorite,
chloritic biotite and opaques (presumably iron oxides) form the secondary components. Single
crystals and an aggregate of clinopyroxene, about 250m in size, are present. Other inclusions are
subangular to angular sand-sized grains of quartz, up to 300m and rounded to subrounded grains
of replacement chert, up 270m in size.
Firing temperature: Undetermined but probably low as there are no signs of isotropism of the matrix.
Geological interpretation and reference: As EA 37.
Reliability: Fair.
Matrix: Clayey, reddish-tan in PPL sparkled birefringent and optically oriented. Silt includes quartz, mica
laths, amphibole, and plagioclase. Opaques appear in the silt fraction as angular bodies, presumably
magnetite. No frequency estimates were done due to the small sample taken.
54
Inclusions: These comprise grains of subrounded quartz with mineral inclusions, up to 380m; rounded
epidote, up to 120m; chloritic mica as aggregates up to 200m; angular clinopyroxene and
amphibole crystals, up to 300m; rounded to angular opaque minerals, up to 130m; rounded chert,
up to 150m.
Firing temperature: Undetermined.
Geological interpretation and reference: Although the sample is small, it is probably similar to K-AD 82, 388.
Description: A Cypro-Minoan inscribed clay cylinder, published by E. Masson (1989:38-39).
Reliability: High.
Matrix: Clayey, reddish-tan in PPL, speckled birefringent and optically oriented. Silt (~5%) includes
quartz, mica laths and plagioclase. Opaques appear in the silt fraction (3%) as angular bodies,
presumably magnetite.
Inclusions: The inclusion assemblage (f:c ratio{0.062mm}=93:7) contains rounded fragments of weathered
spilitic basalt (up to 700m) and single grains of clinopyroxene and feldspar; subangular quartz (up
to 300m), sometimes polycrystalline; rounded fragment of welded quartzitic siltstone (250m)
with subangular strained grains of quartz exhibiting undulose extinction and chloritic biotite and
opaques as secondary components; rounded grain (530 m) of zebraic chalcedony nodule; angular
to subrounded quartz (up to 400m) sometimes polycrystalline; rounded epidote, up to 150m;
rounded replacement chert, up to 230m.
Firing temperature: Undetermined but probably low, as there are no signs of isotropism of the matrix.
Geological interpretation and reference: As EA 37.
A LETTER OF THE KING OF ALASHIYA FROM UGARIT

RS L.1 (AO 21087), from the King of Alashiya to the King of Ugarit
Sampling method: Peeling supplemented by SPA. Under the stereomicroscope the following fabric had
been observed: light reddish matrix with badly sorted dense inclusions, reaching millimeter size.
The great majority of the inclusions are dolerite fragments and their derived minerals, perhaps
with some additional gabbroic inclusions. Serpentine is common in lesser grain sizes. Limestone
and chert form a minority of the inclusions. Since most of the tablets surface was covered with
Polyvinyl Acetate (PVA) in the process of restoration, it was sometimes difcult to dene some of
the inclusions. However, a break on one of the corners was cleared out of the PVA in order to enable
the sampling, and most of these observations were made on this exposed surface. No samples were
taken for elemental analysis due to the risk that the immersion of the tablet in glue would provide
biased results.
Reliability: High.
Matrix: Dense, tan in PPL and containing about 5% opaque minerals ranging between subangular to
subspherical. The silt fraction (~4%) contains quartz, twinned plagioclase, biotite, epidote, calcite
and serpentine.
Inclusions: Badly sorted, usually angular (f:c ratio{0.062mm}=~90:10) assemblage of igneous and
sedimentary rock fragments and their derived minerals. These include dolerite/microgabbro and
its derived minerals which are frequent, up to 1.2 mm, commonly angular or subangular. Rarer
are subrounded grains of holocrystalline dolerite consisting of clinopyroxene, plagioclase, ore and
55
occasionally some quartz. The clinopyroxene is commonly serpentinized. Single detrital crystals of
derived minerals include clinopyroxene (reaching 250m) and plagioclase (reaching up to 400m).
Fossiliferous limestone is common, rounded micrite (up to 700m), single cleavage calcite crystals
(up to 380m), sparite (up to 450m) and a fossiliferous corallinean alga fragment (1.3 mm).
Rounded grains of brous serpentine (up to 570m) usually with iron concentrations or horizons are
common. There are a few grains (up to 500m) of subangular to subrounded replacement chert with
iron mineral stains and a very few grains of subrounded quartz (up to 270m).
Firing temperature: Estimated at 500 0 C-600 0 C judging by the increase in pleochroism of the ironrich serpentine from greenish-yellow to yellowish-red, but not exceeding 800 0 C since the calcite
is not affected.
Geological interpretation: Dolerite-derived reworked clay from the Troodos contact zone between the
volcanic series and the sedimentary formations.
Reference: The fabric of this tablet is similar to that of Cypriote dolerite-based clays. However, this clay
is reworked and mixed with components of the supra-ophiolitic sediments.
Excursus: Letter RS L.1 and Carchemish
Most scholars follow the hypothesis raised by Nougayrol et al. (1968:86, n.1, 695-697), that RS L.1 was
sent from Alashiya and replied to by letter 20.238 of Ammurapi, King of Ugarit. A minority of scholars
opposes this view due to the lack of any toponym of the sender. Singer (1983:217, 1999:720, n. 394) and
Yamada (1992:438-450) suggested that RS L.1 must have been a letter from the King of Carchemish, the
overlord of Ugarit at the time of Ammurapi. A similar view has recently been adopted by Malbran-Labat
(1999:122) and Yon (1999:118). Therefore, before using the petrography of RS L.1 as evidence for the
location of Alashiya, it is mandatory to examine the possibility that this letter originated at Carchemish.
The petrographic data of this tablet can be correlated with the geology of Carchemish in particular
and the Upper Euphrates area in general. Additionally, a letter from the King of Carchemish to the King
of Ugarit (RS 8.333) which we also examined may serve as reference to the materials that were in use at
this site for the production of tablets at the same period.
The general as well as detailed geological maps of the area between Gaziantep and Carchemish
(Tolun and Pamir 1975; Ulu 1996a; 1996b) indicate that the sediments around the site and upstream
along the Euphrates are very homogeneous. The site is located on recent Euphrates uvial sediments.
Immediately near it outcrops the Gaziantep Formation of the Upper Eocene, composed of silty, clayey
or chert-including limestone, or chalk with glauconite concentrations. North of Carchemish one nds
the Quaternary Old alluvium with partly consolidated clay, sand and gravel. Igneous rocks expose only
45 km to the north as the crow ies, near Hamut Dai. They include tholeiitic and alkali olivine basalts
but not dolerite or gabbro. Ophiolitic exposures of the Koali Complex appear only some 34 km further
north, near Yaylaek, but like the Tauric ophiolites they contain only serpentinized ultrabasics, silicied
shale, radiolarites and very limited pillow-lavas. Therefore, the lithology indicated by RS L.1 does not
exist around Carchemish.
Fabrics of ceramics that are made of Upper Euphrates sediments are expected to reect some of the
above-described components. Although no petrographic reference to the Carchemish ceramics has been
recorded so far, there is some relevant data on the Bronze Age pottery from Tell Hadidi, somewhat south
of Carchemish but still within the same geological environment (Mason and Cooper 1999). As one may
expect, there is no reference to any fabric resembling of RS L.1 in this data.
The petrography of the clay used in Carchemish for the production of cuneiform tablets at the time of
RS L.1 may be represented by the following letter:
56
RS 8.333 (AO 19.955, PRU III, 7-8) from the King of Carchemish to the King of Ugarit
Matrix: Ochre to orange-tan in PPL, birefringent with speckled b-fabric and very strong striated optical
orientation. The silt (5%) is rich in mineral types including muscovite, quartz, calcite, serpentine,
hornblende, opaques, rutile, zircon, and feldspar.
Inclusions: Very few, accidental grains of micritic limestone are the only representatives of the coarse
fraction.
Firing temperature: Below 8000C judging by the lack of any signs of alteration in the hornblende, but
probably above 7000C according to the deformation of calcite.
Discussion: The petrographic traits of RS 8.333 resemble those of the Mesopotamian (Babylonian
and Mitannian) tablets from the Amarna archive. They indicate that the Euphrates clay-silt was used
without any intentional addition of sand inclusions, due to the silty nature of these sediments. Similar
characteristics are known from the pottery that was produced in this region (Franken and Van As 1994).
This, again, stands in contrast with the characteristics of RS L.1.
Consequently, the hypothesis that RS L.1 was sent from Carchemish can be readily dismissed.
Taking into consideration the content of the letter, there can be no doubt that this is a letter from
Alashiya. Therefore, it can (and should) be used for the investigation of the location of the political
centre of Alashiya.
INTERPRETATION OF THE PETROGRAPHIC DATA

THE GENERAL PROVENANCE OF THE ALASHIYA LETTERS
A macroscopic comparison of the fabric of the eight Amarna tablets from Alashiya indicates that all except
one (EA 37) are homogeneous in terms of colour of clay, quality, type of grits, polish and texture (Hellbing
1979:71). EA 37 differs in its fabric, grits and general texture. Preliminary petrographic observation of EA
33, 34 and 38 supplies inconclusive evidence for the nature of the geological environment from which their
clay was derived. Marl and limestone (the latter seen in EA 34) are too widely distributed to pinpoint a
specic geographic location within the Near East. Even if the Enkomi area is excluded on the basis of the
NAA results, the petrographic data still leave open many other possibilities within the various regions that
have been suggested for the location of Alashiya.
The two remaining tablets, EA 37 and RS L.1, are coarser and gritty and therefore enable better
petrographic determination. Both tablets share the same afnity: the main inclusion constituents were
derived from an area consisting of basic igneous rocks (spilitic basalt, dolerite, micro-gabbro), with a
set of sedimentary rocks (limestone, chalk, sandstone, mudstone and chert) accompanying the igneous
assemblage. These compositional, textural and mineralogical characteristics suggest that the source area
should be sought on the margin of an ophiolite complex where pillow-lavas and dolerite are exposed.
In other words, the lithology reected by the raw materials of these two tablets points to a contact zone
between an ophiolitic complex and a sedimentary area containing calcareous, argillaceous, and siliceous
sedimentary rocks.
Ophiolite complexes are presumed to represent oceanic crust which was thrust onto continental crust.
When complete, an ophiolite consists of a thin uppermost veneer of oceanic sediment (which may include
oceanic clay and radiolarian cherts) overlying quenched pillow basalts and more mature lavas, which in
turn overlie a sheeted dolerite complex. Beneath the dolerites are texturally isotropic gabbros, which overlie
57
layered gabbros, peridotites and pyroxenites. These largely basic and ultrabasic components are cut by latestage intrusions of co-eval plagiogranite and overlay older oceanic sediments including radiolarites and
limestone. As a consequence of its formation at spreading ridges, oceanic crust (and therefore ophiolites)
experience ocean-oor metamorphism which characteristically produces assemblages of greenschist and
amphibolite facies. These metamorphites are often undeformed.
Of all the above, a very limited range of rock types is represented in EA 37. The magmatic
constituent includes only spilites, or rather their weathering products (with a few other mac minerals),
with the addition of a set of sedimentary rocks. This hints at a clay type that is directly derived from
the weathering of pillow lavas in a zone that is proximate to exposures of marine sedimentary rocks
that include limestone and replacement chert. The localized nature of the clay and the selection of the
inclusions indicate that the area from which they were derived was very limited and bounded within
the margin of the ophiolite complex. The data supplied by RS L.1 supplements this picture. The
homogeneous assembly of inclusions, dominated by basic intrusive rocks (dolerite), indicates an area
where these rocks form the dominant constituent of the lithology. Apart from their degree of weathering
and the relative proportions of the accompanying sedimentary rocks, the two tablets seem to have been
derived principally from closely related geological environments.
In the Eastern Mediterranean region ophiolite complexes are found in Cilicia, northwest Syria and
Cyprus. Thus, it is an ironic coincidence that ophiolitic occurrences can be found in each and every area
suggested for the location of Alashiya. Ophiolites form the Troodos massif in southwestern Cyprus, the
Mersin and Pozanti-Karsanti massifs in Cilicia, the Kizilda massif in Hatay Province, Turkey, and the
Bar-Bassit massif of northwest Syria (Whitechurch et al. 1984). Nevertheless, a closer examination of
the geological data enables the elimination of most of these locations and a rather explicit identication
of the source area of EA 37 and RS L.1.
In this context, the rst hypothesis that should be evaluated is that Alashiya was located in Cilicia.
This possibility was rst raised by Hall (1913) and later by Merrillees (1972; 1987). Merrillees (1972:118)
suggested identifying it with the Gulf of Iskenderun, possibly with a site such as Kinet Hyk, but admitted
that this identication must await further archaeological inquiry. To examine this general location an area
that extends between Mersin on the western Cilician coast and the Gulf of Iskenderun must be examined.
The Mersin and PozantiKarsanti massifs are part of the median or Tauric ophiolitic belt of Turkey
(Thierry 1980:205). They form the most southeasterly front of the Taurus ridge, between Faraa and the
plain north of Silifke (ibid.: 215-220). The two ophiolites are in structural continuity with each other but
are separated by the strike-slip fault of Ecemi. The Mersin ophiolite exposes northwest of the city of
Mersin, under the Miocene and Oligocene conglomerate reworking of the ophiolite. To the north, it lies
on the Tauric limestone series. The Mersin massif continues the corresponding lithology of the Pozanti
Karsanti ophiolite north of it. It is made of a pile of tectonic slices, mostly of serpentinized harzburgite
intersected by isolated tholeiitic dolerite dikes. In the western part gabbroic and pyroxenic cumulates
appear, interbedded with pillow-lavas and radiolarites. In this complex the basaltic component is
therefore minor and interlayered with the far more dominant peridotitic constituent (Thierry 1980:216).
Consequently, it is very unlikely that this assembly would supply sediment where basalt clasts would be
dominant. This is especially true for the post-tectonic Miocenic sedimentary plain of the Cilician coast,
where large ancient settlements (such as Mersin and Tarsus) are located. Therefore, the Cilician coast
option must be eliminated.
The same applies to the more northeasterly area of Pozanti-Karsanti. The ophiolite outcrops over
more than 100 km in length, covering more than 1500 km2. Tectonically it lies over a series of limestones
58
and radiolarites. Here too, thick outcrops of harzburgite form the main lithology, overlaying a crushed
series of tectonic slices made of pillow-lavas, radiolarites, pelagic limestones and metamorphic rocks.
Apart from these infra-peridotitic slices, the rest of the massif is formed entirely of coarse-grained
ultramac rocks (Bingl 1978; Thierry 1980). Therefore, the same argument raised against the Mersin
ophiolite namely, basalt clasts are not dominant enough compared to what we see in the tablets are
equally true for this region.
In summary, the petrography of EA 37 and RS L.1 does not match the Cilician ophiolitic complex,
or the lithology of neighbouring areas. A coastal site in the Gulf of Iskenderun is unlikely to be the source
for these tablets, especially in the case of RS L.1, which portrays a localized assembly of materials
adjacent to the massif, rather than materials that were washed off it into a sedimentary plain.
According to a second theory, rst raised by Wainwright (1914-1915) and later supported by
Merrillees (1987), Alashiya was located in northwest Syria or the Hatay. Two ophiolitic complexes
expose in the area of northwest Syria and the related province of Hatay in Turkey: the Kizilda massif
in the southwestern part of the Amanos range in the Hatay, and the Bar-Bassit massif west and south of
the lower Orontes plain.
The Kizilda massif forms the westernmost extension of the Amanos range and stretches onto the
Gulf of Iskenderun. A rich body of geological data (Dubertret 1955, Erendil 1984, Robertson 1986, Tekeli
and Erendil 1986) enables a detailed investigation of this area. The constituent rock types of the ophiolite
complex crop out over an area of approximately 1100 km 2 and although they resemble those of the Troodos
massif in Cyprus, the effects of ocean-oor metamorphism are signicantly more developed. It is unlikely
that the source of the Alashiya tablets was in the plains around, or in the Kizilda massif itself, because the
pillow basalts crop out over a very limited area which comprises under 0.5% of the region (Erendil 1984).
It is difcult to see how an alluvial clay could be formed, which would contain the required percentage of
pillow lava clasts. Moreover, the same argument raised against Cilician coastal sites, that basalt clasts are
not dominant enough, can also be applied in the case of northwestern Syrian coastal sites.
The Bar-Bassit massif of northwest Syria lies south of the Kizilda massif being separated from it
by the lower valley of the Orontes River and the Plain of Antioch. Chenevoy (1959), Parrot (1974; 1977;
1980) and Whitechurch and Parrot (1974) have described the geology of this massif, which covers an
area of about 112 km2. It differs from both the Troodos massif in Cyprus, and the Kizilda massif. Firstly,
the constituent rock types are more strongly deformed, and secondly, they are more varied in composition
because the ophiolite was tectonically interleaved with a variety of older sedimentary, igneous and
medium to high-grade metamorphic rocks. None of these rock types have been recorded in the paste of
the two Alashiya tablets under review.
Matson (in Braidwood and Braidwood 1960) presented detailed petrographic data on the pottery
assemblages of all stages in the Amuq Valley. Additional information was retrieved from our thin-section
collection of about 100 representative Amuq ware fabrics (made from a systematic study collection prepared
by Braidwood in the 1960s for the Israel Department of Antiquities). Both sources of information reduce
the possibility that EA 37 and RS L.1 originated in this region. Entirely different clay and temper types,
in which serpentine plays the dominant role in the inclusions, characterize the ceramics of the Amuq sites.
A wide range of metamorphic facies is also represented in the inclusion assemblage of the pottery. These
components are not represented in the paste of EA 37 and RS L.1. Conversely, basalt and dolerite clasts,
when appearing in the inclusions in the Amuq ceramics, are marginal. Additionally, in the Amuq pottery
the chert fragments are always of the radiolarian class, an occurrence that conforms with the radiolarian
oceanic chert formations that are preserved near the Bar-Bassit and Kizilda ophiolites (Dubertret 1955:
59
91-94; Parrot 1977) (Chapter 5.I) This is contrary to the situation in the Troodos area of Cyprus, where
these sediments were removed and most of the chert found is a much younger Eocene replacement chert (of
the Middle Lefkara Formation). The Mamonia Complex in Cyprus and the Mlange that is formed by the
ophiolite reworking on the southeastern slopes of the Troodos contain radiolarian cherts to some extent,
but these are accompanied by entirely different rock types.
In conclusion, the areas of the plain of Antioch and the Hatay are not likely to be the source of EA
37 and RS L.1 despite their nearby ophiolitic complexes. Hence the only option left for the provenance
of EA 37 and RS L.1, and consequently all the Alashiya tablets, is on the island of Cyprus.
THE PROVENANCE OF THE ALASHIYA LETTERS WITHIN CYPRUS
A general look at the geology of Cyprus (geological map of Cyprus 1:250,000, 1995 edition) indicates
that the ophiolitic complex of the Troodos massif occurs in the central-western part of the island (Fig.
3.3). The massif forms much of the Troodos Mountains, the outcrop covering an area of 3,000 km2. Two
small outliers occur in the Akamas peninsula and at Troulli. Erosion has exposed all the constituent rock
types of the ophiolite complex, which is exceptional for its relative lack of deformation.
The concentric arrangement of the members of the ophiolite succession in the Troodos massif
provides the most rened control on the mineralogical content of the alluvial deposits that develop both
within and on the periphery of the mountains. The pillow lava series form the outer, incomplete ring
that completes the structural units of the massif. Two units have been recognized: in the oldest series,
designated the Lower Pillow Lava (LPL), the main rock types include andesitic basalt, quartz andesitic
Fig. 3.3: Generalized geological map of Cyprus showing the formations referred to in the text.
60
basalt, quartz microdolerite and quartz-microgabbro. Of these the most common are the andesitic basalts.
These lower division rocks are silica rich and contrast with the younger Upper Pillow Lava (UPL) series,
which are basic in composition, with olivine as one of the most common constituent minerals. The rock
suite includes olivine-basalts and mugearites (Gass 1980). An inner ring of the Troodos succession is
formed mainly of sheeted dolerite dykes.
A signicant feature of the pillow lava outcrops of the Troodos range is that the rocks alter to form
ne plastic clays, some of which have in the past, and continue today, to be used for pottery, tile and brick
production (Wilson and Ingham 1959, Gass et al. 1994). One of the most important clay deposits is near
Kornos, where alluvial clays and silts from the river valley support one of the best known pottery and brick
production centres in Cyprus (Bagnall 1960:103; Hemsley 1991; London 1991; Ionas 1998:23, 109-110).
Amongst their other uses, these clays were known to serve until the 1960s for the production of pithoi
(locally termed pitharia). The main production centres existed in Kornos, Phini and Ayios Dhimitrios in
the Troodos area (Hampe and Winter 1962; London 1986, 1989; Keswani 1989:17; Ionas 1998:125-131).
For the sake of our discussion we have adopted the results of the comprehensive petrographic research
conducted by Vaughan (1987, 1989, 1991), primarily on Cypriote Base Ring Ware. We have also used
our own reference collection of clays from the relevant formations (Table 3.2). The petrographic data
concerning pottery production around the Troodos in all periods clearly indicate similar clays and clastic
assemblies to the ones observed in EA 37 and RS L.1. The inclusion composition of both tablets suggests
a contact area between the volcanic and marine sedimentary terrains. In the case of EA 37, another
signicant trait is the presence of the low-grade metamorphic greywacke. Contact regions between marine
sedimentary and volcanic lithologies appear around the Troodos Mountains. The combined lithology of
EA 37 and RS L.1 indicates that their origin should be looked for inland, not on the coast. The immediate
contact zone between the dolerite and the basalt outcrops on the one hand, and the limestone exposures on
the other, can be found in the northern, southern, and eastern edges of the Troodos area.
Before attempting to pinpoint the origin of the tablets, it should be borne in mind that EA 37 and
RS L.1 represent two related, but still different, fabrics. The fresh detrital igneous inclusions that
characterize RS L.1 indicate localized colluvial clay, which overlies the dolerite series of the Troodos.
However, the presence of carbonates and chert is incompatible with the use of locally available doleritederived clays that were, and still are utilized in the Troodos area for pottery production (Vaughan 1991:
353). Therefore, the origin of this tablet should be sought in an area where the dolerite-derived clays
are reworked and mixed with components of the supra-ophiolitic sediments that surround the Troodos
complex. This area should contain limestone, reefal limestone and chert. These requirements suit
the region of the southern Cypriote coast from Petra tou Romiou to the Tremithes River, following
through the eastern Troodos foothills to Petrofani in the north. In this region, the Lef kara and Pakhna
Formations contain a set of limestone, biogenic and corallinean limestone, chert and marls that may
contribute these components (Gass et al. 1994:115-123, with the relevant references). The dominance
and fresh state of the dolerite clasts suggests an area in the immediate contact zone between the
igneous and sedimentary outcrops. This area should be looked for inland, where the Lef kara and
Pakhna Formations overlay the dolerite series.
The petrography of EA 37 indicates an assembly of volcanoclastic, calcareous, siliceous and
argillaceous rock fragments. These include weathered basalt clasts, radiolarian mudstone, chert, limestone,
ne-grained low-grade metamorphosed sandstones, and detrital minerals. This set corresponds with the
data supplied by Vaughan (1989:80, 1991:358-359) for the typical clays of the Moni Mlange in the
southeastern anks of the Troodos Massif. This formation contains detrital igneous rock fragments and
61
minerals that are frequently interbedded with silty lenses, mudstones, and umberiferous or radiolarian
shales. Olistoliths that are made of lithologies that typify the Mamonia Complex (considered to be part
of an Upper Triassic to mid-Cretaceous continental margin) are embedded in the bentonic clay matrix,
especially along the northern margin of the Mlange. These include quartztose sandstone, clay-rich
siltstone with other minerals that suggest a metamorphic or granitic provenance. These allochthonous
materials are overlain in places by the Lefkara chalks of the Eocene age, which are rich in replacement
chert, and by the Pakhna chalks (Miocene) which contain arenite lenses. As a result of weathering
processes, the upper layers of the clays often contain random particles of the overlaying carbonatic rocks
(Gass et al. 1994:111-114). This unique combination, all of which is represented in EA 37, is directly
lined with the Moni Formation.
The Moni Formation is exposed over restricted areas along the southern and southeastern foot
of the Troodos (Pantazis 1967:38-46; Gass et al. 1994:111-115). Although the nomenclature and
subdivision of this formation is somewhat complex (see Gass et al. 1994:111 for details), it is obvious
that the sediments reworked by the Mlange form the origin for this petrofabric. Exactly the same
petrofabric was found to characterize Late Cypriote IIC vessels at Kalavasos Ayios Dhimitrios; a
natural exposure is found just southwest of the site (Vaughan 1989:79-80). On a broader scale, the
distribution of pottery that is made of the Kannaviou clays of the Moni Mlange characterizes the
general area of the south coast (Vaughan 1991), but the outcrops of the formation are restricted to the
Troodos foothills between Yerasa and Pentacorno. We may add that one of the inscribed cylinders from
Kalavasos Ayios Dhimitrios that we examined (K-AD 82, 405) contains exactly the same inclusion
suite as EA 37.
Returning to the Amarna letters from Alashiya that were made of pure marl (EA 33, 34 and 38),
since EA 37 and RS L.1 direct us to the southern slopes of the Troodos, we should start our search for
the source material of the other tablets there. When the NAA study of the Amarna tablets was carried
out (Artzy et al. 1976), no major site in this area had been excavated and hence ceramic material
from the region could not be compared with the clay of the tablets. Still, the correlation between
the Alashiya letters and some of the pottery from Kouklia Paleopaphos hints that the clay type that
was used for some of the vessels at that site should be associated with that used for the tablets. The
petrographic examination of EA 33, 34 and 38 indicates a type of cream-coloured marl with occasional
volcanic clasts. The palaeontological data from EA 33 (the only case where the foraminifers were
in an identifiable state of preservation) indicates Neogene marl. In the general area of the southern
Troodos foothills, there is only one possible location for such a clay source: the marl member of the
Pakhna Formation (Fig. 3.3), dating to the Lower Miocene (Gass et al. 1994:123). This formation is
distributed both in Kouklia and along the southern Troodos slopes, but it does not extend to Enkomi or
other sites that were surveyed in the NAA study. The apparent, but still rather vague relation between
the Alashiya tablets and the Kouklia pottery suggests that they were both made of marl of the same
formation, but of different exposures or localities.
Marl deposits form part of the lowest chalk-marl member of the Pakhna Formation (Pantazis 1967:
64-66). The marl is remarkable for its cream to buff-colour, its plasticity and the relatively hard fabric that
it forms after hardening (Pantazis 1967:65; Gass et al. 1994:119), similarly to the tablets. The plasticity
of the Pakhna marls is one of their most prized features, containing as they do signicant percentages
of the Troodos-derived montmorillonite. The shrinkage of this mineral seems to be naturally tempered
by its mixture with illitic components from the carbonate sediments, making it an ideal potting material
(S. Vaughan, pers. comm.).
62
CHEMICAL ANALYSES OF THE ALASHIYA LETTERS AND SELECTED CYPRIOTE CLAYS

It was decided to supplement the results of the petrographic study by a chemical analysis. Since there
seemed to be no reason to carry out a repeat NAA study (Artzy et al. 1976), this was aimed at disclosing
the elemental compositions of the tablets and comparing them with the relevant Cypriote clay types
suggested by the petrographic results.
The samples are listed in Table 3.2. By measuring the elemental composition of the Pakhna marl
from several outcrops, preferably near the relevant Late Cypriote sites, it was hoped to identify certain
differences between the various locations and accordingly to suggest a better-de ned origin for EA 33
and EA 34.
Table 3.3 presents the elemental composition of several Alashiya letters, as revealed by ICP-AES/
MS. Only limited samples could be extracted from two of the tablets 60 mg in the case of EA 34
and 30 mg in the case of EA 37. This must have had some effect on the accuracy of the measurements,
especially for the latter. Although the results indicate that the concentrations of most elements are
presented in true numbers (except for P and Co) this deciency did not affect the results severely. It
was not possible to examine RS L.1 because it was immersed in PVA and perhaps other types of glue.
The NAA study carried out by Artzy et al. (1976) revealed that EA 33, 34, 35 and 38 were chemically
very similar. Therefore there seemed to be no point in examining EA 35 in the British Museum and EA
38 in the Vorderasiatisches Museum in Berlin since they were obviously produced from the same clay.
Alhough our ICP analyses disclosed some differences in the chemical composition of EA 33 and EA 34
as can be seen by the standard deviation row in Table 3.5, the results are closely related.
The above conclusions led us to perform a comparative study of clays from the Pakhna and the Moni
Formations as well as the lava-derived clays from Kornos and Phini. Samples of the Pakhna marl were taken
from selected locations between Kouklia and Larnaca, especially near relevant Late Cypriote sites (Table
3.2). The Moni Formation was sampled in several locations around Moni and Pentacomo. A satellite Global
Positioning System (GPS) was used to take their grid references. The clay samples were formed into small
briquettes and red at 500 0C. The red briquettes were thin-sectioned and examined petrographically for
comparison with the tablets. While doing this, we attempted to estimate the workability of each sample
by assessing its plasticity, behaviour upon drying and in ring. The Pakhna marl and Moni Mlange clay
samples were analyzed by ICP-AES/MS. The results are presented in Table 3.4. Samples that proved to
be absolutely unworkable were not included in the ICP analysis. The composition of the Alashiya letters
was added for comparison.
Petrographic examination con rmed that RS L.1 is similar to the clay from the ceramic workshop at
Kornos (sample CY-24), although the latter lacked the sedimentary rock fragments. EA 37 is similar to
a sample of reworked exposure of the Moni Mlange that was partly covered by the Lefkara Formation
chalks (sample CY-23). EA 33, 34 and 38 are petrographically similar to many of the samples taken from
various sections of the Pakhna marls.
A close look at the ICP results reveals that there is considerable chemical variability between the
samples, even within the same geological formation (e.g. the Pakhna marl). Such variability was observed
in fact, also in the petrographic analyses of the red briquettes that were made from the Cypriote clay
samples. Moreover, even the Alashiya letters assigned to that formation (EA 33, 34 and 38) differ in
some respects as can be seen by their detailed descriptions. This may be due to the fact that the marly
member of the formation includes sets of beds that are intercalated with other rock types (especially
chalk). While some of the samples could have been reworked, others were taken directly from a clean
section of the outcrop. Therefore, the next step was to examine whether the Alashiya letters fall within
63
the compositional limits of the clays to which they are attributed. In Table 3.5 such pro les are presented
through the mean and standard deviation of the Pakhna and the Moni clay samples, compared with the
mean values of EA 33 and EA 34 (to represent tablets presumably made of Pakhna marl), and EA 37 (of
the Moni Mlange clay)
It is obvious that the samples of the Pakhna marl give large standard deviation values, meaning that
there is a high compositional variability within this formation. Furthermore, the SO3 concentrations in
the Alashiya tablets are higher than in the geological samples. All the other elements in the tablets have
lower levels than the corresponding ones in the Pakhna marl samples. This may point to dilution due
to post-depositional accumulation of gypsum (CaSO 4) in the tablets that did not appear in the original
clay. Secondary crystallization of gypsum is common in arid zones such as the area around el-Amarna.
Consequently, the concentration of elements in the tablets was depressed by the addition of a major
component that did not appear to this level in the original clay. If this is true, there should be a constant
proportion between all of the other elements in the Pakhna marl and the tablets. If this were not the case,
one would expect to nd irregular uctuations between the two sets of numbers and not a constant factor.
Indeed, such a constant exists, and its value is approximately 0.7. The signicance of this relationship is
that the Alashiya tablets and the Pakhna marls indeed share a common origin but the tablets are diluted
by the addition of gypsum. A similar phenomenon has been observed in the case of the Babylonian tablet
EA 13, which was offset from the Babylonian cluster due to its gypsum contents, but fell well within it
after the SO3 value was omitted. In such cases it is a common procedure to compensate for the dilution
factor by considering this constant. Table 3.6 presents the composition of the tablets and the Pakhna clay
samples after the former were adjusted for dilution.
Fig. 3.4: Cypriote clay samples and the Alashiya letters, tree diagram, Wards method, squared Euclidean distances
64
Using logarithms of the measured concentrations, the relations between the analyzed specimens
may be examined by cluster analysis. Fig. 3.3 shows that the Alashiya tablets cluster with the Cypriote
clays, EA 33 and EA 34 cluster with samples of the Pakhna marl, while EA 37 clusters with samples of
the Moni Mlange.
It becomes evident again that the Pakhna Formation includes two differing marl categories. One
(represented in our collection by samples CY-8, 9, 11, 14) has excellent plasticity and bright (even
whitish) colour. For some reason this type was not utilized for the production of tablets. Table 3.5 reveals
that it has signicantly lower silica and higher calcium rates. The letters belong to a second marl category
within the same formation, that is typied by good plasticity (though not as fat as the former type) and
a creamier colour. Within this subgroup, both EA 33 and EA 34 are most closely related to sample CY-10
from near the Cardboard Factory, right above the site of Alassa Paliotaverna. The other samples (from
Kouklia, Kalavasos and Konou) are more remote, although they are closely related to each other.
These results do not necessarily mean that EA 33 and EA 34 were made in Alassa Paliotaverna.
The close relationship of samples within this clay member of the Pakhna Formation from Kouklia
Palaeopaphos and Kalavasos Ayios Dhimitrios means that it is extremely homogeneous in its composition
over broad areas. It can only be established that EA 33 and EA 34 (and the like) were made of a member
of the Pakhna Formation which exposes in various locations between Larnaca and Kouklia. Thus,
although the ICP analysis indicated the relation of EA 33 and EA 34 with a certain member of the marl
unit within the Pakhna Formation, it failed to relate the tablets to a more specic provenance due to the
homogeneity of this member over southern Cyprus.
In order to investigate the reason for the preference of the less plastic creamy marl over the brighter and
more pliable variety we simulated the production of a tablet-like shape and its incision with a sharpened stick.
This experiment made it obvious that incising cuneiform signs in wet, very plastic clay may be frustrating
since such clay tends to cling to the stick and produce blurred signs. Moderate clay plasticity was therefore a
desired trait of which scribes must have been aware when preparing the tablets for writing.
EA 37 clearly falls within the cluster of the Moni Mlange samples that come from several outcrops
of this formation. Sample CY-23, which is reworked under the overlying Lefkara Formation, is better
related to the pure sample of the Lefkara Formation (CY-2). Therefore, although clearly too small in size,
the sample taken from EA 37 conrms the petrographic interpretation that associated it with this clay
type. Of the Moni clay samples only CY-19 showed good plasticity. We therefore suggest that the clay
selected for EA 37 was taken from a reworked rather than from an in situ exposure of the clay.
TABLE 3.2: CYPRIOTE CLAYS: SAMPLE LIST

GPS extracted locations are in Universal Transverse Mercator Grid (UTM), Zone 36, International Spheroid
European Datum.
Sample
Formation
Location
Grid (UTM)
Notes
Ceramic behaviour
CY-1
CY-2
Pakhna
Lefkara
Khirokitiya
Khirokitiya
053152.385061
053038.385104
Near the site

In the village
CY-3
Pakhna
Khirokitiya
053149.384938
In the village
CY-4
Pakhna
Near Konou
053283.385189
Insoluble in water.
Moderate plasticity, dries without
cracking, slightly sinters at 5000C.
Gritty, plastic but crumbles in
drying.
Good plasticity, slightly cracks in
drying, sinters at 5000C.
65
CY-5
CY-6
Pakhna
Pakhna
CY-7
CY-8
Pakhna
Pakhna
CY-9
Pakhna
CY-10
Pakhna
CY-11
Pakhna
CY-12
Pakhna
CY-13
Pakhna
CY-14
Pakhna
Kouklia
046059.384086
CY-15
Pakhna
Kouklia
046050.384088
CY-16
CY-17
Pakhna
Doleritic
reworked
clay
Dolerite
derived clay
Kouklia
Kato Platres
046073.384098
048668.385999
CY-18
Amathus
053156.384780
051261.384102
West of Kourion
Alassa-Pano
Mandilaris
AlassaPaliotaverna
048817.383664
049250.384700
Alassa
Cardboard
factory
Alassa
Cardboard
factory
Kalavasos
049240.384700
049295.384630
049240.384700
052648.384453
Section of the
northern wall
foundation trench
North of
Paliotaverna, from a
section.
North of
Paliotaverna, from a
section.
Near the old road
crossing the site
(east of Ayios
Dhimitrios).
053149.384863
Phini
CY-19
Moni
SE of
Pentacomo
052250.384300
CY-20
Moni
SE of Moni
051900.384350
CY-21
Moni
SE of Moni
051900.384325
CY-22
Moni
Pentacomo
052218.384453
CY-23
Moni
West of Moni
051785.384229
CY-24
Dolerite/
pillow lava
derived clay
Kornos
66
Carpark near the

Agora
Road section
Upper (humic) layer

from a road section.
Potters workshop
clay, said to be
locally collected
From a mining area,
section
Reworked under the

Lefkara Fm.
Potters workshop
clay (said to be from
near Stravovouni).
Insoluble in water.
Insoluble in water.
Insoluble in water.
Excellent plasticity, dries without
cracking, sinters at 5000C.
Good plasticity, dries without
cracking.
Rather gritty, good plasticity, dries

without cracking, cracks at 5000C.
Insoluble in water.
Gritty but plastic, dries without
cracking,
Excellent plasticity, perfect
sintering at 5000C.
Good plasticity, shrinks in drying,
sinters at 5000C but slightly
cracks.
Moderate plasticity, cracks while
drying, crumbles in ring to
5000C.
5000C.
5000C.
crumbling, perfectly sinters at
5000C.
TABLE 3.3: ELEMENTAL ANALYSIS OF THE ALASHIYA LETTERS

TABLE 3.3A: MAJOR ELEMENTS (IN WEIGHT/PERCENT)
EA
SiO2
Al2O3
CaO
MgO
Fe2O3
TiO2
P2O5
SO3
33
34
37
0.55
0.50
0.63
0.21
<0.2
<0.3
0.64
0.60
0.64
41.86
38.25
41.51
9.66
8.42
9.87
18.79
17.51
10.86
2.39
1.88
1.78
4.28
3.99
5.43
TABLE 3.3B. MINOR AND TRACE ELEMENTS (PPM)

EA Zn Co Ni
33 99.24 16.70 80.21
34 55.46 <10 91.67
37 57.17 <15 69.08
Ba
Mn
Cr
Be Cu
Sr
Ce
Pr
Nd
Sm Eu Tb Dy Ho Tm Yb Lu
142.59
523.15
87.94
98.94
1.72 111.35
417.34 23.45
La
18.10
50.33
6.82
25.45
4.76
0.91
0.65
3.66
0.87
0.37
2.10
0.28
175.00
454.17
128.33 77.92
1.38 38.96
410.00 26.54
19.50
45.15
6.06
22.16
4.14
0.85
0.62
3.67
0.73
0.30
2.02
0.28
315.79
480.26
123.03 101.32 1.71 49.93
221.05 35.07
27.37
46.24
7.01
26.97
5.33
1.10
0.86
4.65
0.94
0.37
2.46
0.34
TABLE 3.4: ELEMENTAL ANALYSIS OF THE CYPRIOTE CLAY SAMPLES

Sample
SiO2
Al2O3
CaO
MgO
Fe2O3
TiO2
P2O5
SO3
CY-2
CY-4
CY-8
CY-9
CY-10
CY-11
CY-12
CY-14
CY-15
CY-19
CY-20
CY-21
CY-23
EA33
34.0
27.1
8.5
19.3
31.4
14.6
19.3
13.7
24.3
67.0
60.7
57.1
42.2
41.86
9.7
7.5
1.4
3.4
5.1
3.0
5.9
3.2
5.8
11.1
12.0
13.0
9.6
9.66
19.6
29.7
51.2
40.4
31.4
44.7
37.2
45.7
29.9
<0.5
1.7
0.4
15.6
18.79
4.1
2.3
0.6
1.6
1.8
1.5
1.7
0.8
4.6
1.7
1.9
2.4
5.1
2.39
6.18
3.28
0.53
1.81
1.96
1.24
3.18
1.50
2.42
5.08
5.45
6.30
6.79
4.28
0.55
0.49
0.10
0.23
0.35
0.20
0.32
0.20
0.38
0.54
0.64
0.62
0.41
0.55
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.2
<0.1
0.2
0.4
0.4
0.6
<0.1
0.21
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.7
0.5
<0.1
0.64
EA34
38.25
8.42
17.51
1.88
3.99
0.50
<0.2
0.60
EA37
41.51
9.87
10.86
1.78
5.43
0.63
<0.3
0.64
TABLE 3.5: PROFILES OF CYPRIOTE CLAYS AND THE ALASHIYA LETTERS

Sample
SiO2
Al2O3
CaO
MgO
Fe2O3
TiO2
P2O5
SO3
Pakhna Avg
Pakhna SD
Moni Avg
Moni SD
EA33/4Avg
EA33/4SD
21.34
8.51
56.77
10.55
14.93
2.55
5.01
2.56
11.43
1.43
3.78
0.88
36.65
9.92
5.91
8.45
23.28
0.90
2.12
1.38
2.77
1.56
1.75
0.37
2.46
1.65
5.91
0.78
2.05
0.21
0.31
0.15
0.56
0.11
0.23
0.03
0.24
0.02
0.44
0.11
0.13
N.D.
N.D.
N.D.
0.57
0.15
N.D.
0.03
67
68
97
61
16
36
46
29
63
42
42
88
104
113
68
992
55.5
57.1
CY-2
CY-4
CY-8
CY-9
CY-10
CY-11
CY-12
CY-14
CY-15
CY-19
CY-20
CY-21
CY-23
EA33
EA34
EA37
24
16
<10
14
19
<10
14
10
23
25
37
38
37
16.7
<10
<15
Co
Ba
Mn
Cr
48
61 536
71
146
68 115 547
96
77
41
58 247
26
23
56 111 289
74
51
69
73 459 103
55
41
75 420
45
40
68 116 953
68
81
41 397 462
49
45
44 796 1531
54
116
75 112 997 121
128
99 142 1977 118
129
107 108 747 160
145
197 245 985 434
144
80.2 142.6 523.1 87.9 98.94
91.7 175.0 454.2 128.3 77.92
69.1 315.8 480.3 123.0 101.32
Ni
Cu
0.42
75
1.15
45
0.32
24
0.59
33
0.79
28
0.59
18
1.04
73
0.48
64
1.04
46
1.73 105
1.91
92
1.99 122
0.50
85
1.7 111.3
1.4 39.0
1.7 49.9
Be
13
15
6
14
16
9
14
8
16
19
22
23
13
18.1
19.5
27.4
Y
958
1499
603
858
726
1144
905
474
859
98
107
134
272
417.3
410.0
221.0
Sr
5
18
5
11
14
9
15
7
18
21
27
24
10
23.4
26.5
35.1
La
9
38
8
20
29
16
28
13
39
41
57
52
14
50.3
45.1
46.2
Ce
Nd
1.5
7
4.8
19
1.2
5
2.9
12
3.9
16
2.2
9
4.0
15
1.9
8
4.9
20
5.4
21
7.1
28
6.4
26
2.1
9
6.8 25.45
6.06 22.2
7.01 27.0
Pr
1.8
3.7
1.0
2.4
3.2
1.8
3.2
1.6
3.9
4.2
5.7
5.2
2.0
4.8
4.1
5.3
Sm
0.50
0.70
0.19
0.50
0.63
0.33
0.64
0.35
0.62
0.80
1.02
1.06
0.49
0.91
0.85
1.10
Eu
11.0
35.7
662.4 N.D.
6.2
19.8
34.2 119.7 151.9 1176.4 208.1 136.5
93.3
8.1
24.5
22.9
48.8
28.6
44.8
64.1 546.0 151.7
32.8 223.6 503.4
53.5
12.6 246.8 401.6
14.9
54.8
8.9
39.2
70.4
5.0
65.2
Cr
23.5
52.9 200.3 605.1
Mn
17.0
Ba
48.0
Ni
Pakhna
Avg
Pakhna
SD
Moni
Avg
Moni
SD
EA33/
4Avg
EA33/
4SD
Co
Zn
Sample
21.4
45.3
Cu
0.2
0.5
0.7
51.2
33.3
16.3
1.5 101.0
0.3
0.7
Be
Sr
80.8
1.0
5.2
8.0 596.4
4.6
19.0 152.7
3.8 301.0
12.3 891.7
2.2
8.2
7.4
20.5
5.2
11.2
La
3.7
17.0
19.2
41.0
11.8
22.2
Ce
0.5
1.1
2.2
5.3
1.4
3.0
Pr
2.3
4.4
8.5
20.8
5.5
12.2
Nd
0.4
0.9
1.6
4.3
1.0
2.5
Sm
0.0
0.2
0.3
0.8
0.2
0.5
Eu
Tb
0.0
0.1
0.2
0.6
0.1
0.4
Tb
0.37
0.52
0.13
0.38
0.48
0.24
0.46
0.23
0.55
0.58
0.74
0.73
0.36
0.65
0.62
0.86
TABLE 3.5B: MINOR AND TRACE ELEMENTS (PPM) OF CYPRIOTE CLAYS AND THE ALASHIYA LETTERS
Zn
Sam.
TABLE 3.4B: MINOR AND TRACE ELEMENTS (PPM) OF CYPRIOTE CLAY SAMPLES
0.0
0.7
0.9
3.6
0.8
2.2
Dy
2.3
2.9
0.9
2.1
2.8
1.4
2.6
1.3
3.1
3.4
4.2
4.2
2.4
3.66
3.67
4.65
Dy
0.1
0.2
0.2
0.7
0.2
0.5
Ho
0.52
0.59
0.19
0.47
0.58
0.30
0.52
0.28
0.63
0.66
0.84
0.85
0.50
0.87
0.73
0.94
Ho
0.0
0.1
0.1
0.4
0.1
0.2
Tm
0.28
0.28
0.10
0.21
0.28
0.14
0.26
0.13
0.30
0.34
0.39
0.42
0.27
0.37
0.30
0.37
Tm
0.1
0.4
0.3
1.9
0.4
1.2
Yb
1.47
1.48
0.50
1.18
1.57
0.78
1.35
0.72
1.56
1.83
2.15
2.19
1.49
2.10
2.02
2.46
Yb
0.0
0.1
0.1
0.3
0.1
0.2
Lu
0.23
0.22
0.08
0.19
0.24
0.11
0.21
0.11
0.25
0.29
0.32
0.35
0.21
0.28
0.28
0.34
Lu
69
34.0
27.1
8.5
19.3
31.4
14.6
19.3
13.7
24.3
29.7
27.2
22.6
8.2
CY-2
CY-4
CY-8
CY-9
CY-10
CY-11
CY-12
CY-14
CY-15
EA33
EA34
Average
SD
6.9
6.0
5.3
2.4
9.7
7.5
1.4
3.4
5.1
3.0
5.9
3.2
5.8
Al2O3
13.3
12.4
32.3
13.1
19.6
29.7
51.2
40.4
31.4
44.7
37.2
45.7
29.9
CaO
1.7
1.3
2.0
1.3
4.1
2.3
0.6
1.6
1.8
1.5
1.7
0.8
4.6
MgO
23.7
96.9
60.7
15.6
36.0
46.3
29.2
63.2
41.5
42.2
704.5 11.9
39.4
106.9 N.D.
199.3 N.D.
CY-2
CY-4
CY-8
CY-9
CY-10
CY-11
CY-12
CY-14
CY-15
EA33
EA34
Avg.
SD
N.D.
23.0
10.0
14.3
<10
18.5
14.0
<10
15.8
Co
Sample Zn
54.4
11.9
65.1
57.0
44.0
40.7
67.9
41.1
69.4
55.5
40.7
68.4
48.4
Ni
Cr
536.1 71.2
Mn
247.0 26.4
420.2 45.4
184.4 558.1 67.3

223.6 374.3 23.3
124.3 322.5 91.1
101.2 371.4 62.4
796.0 1531.1 53.6
397.0 462.3 49.0
Be Cu
Sr
3.0
2.8
2.5
1.5
6.2
3.3
0.5
1.8
2.0
1.2
3.2
1.5
2.4
Fe2O3
La
51.3
23.5
77.2
1144.5 9.0
0.5 63.6 8.2
473.5 7.0
1.0 73.5 14.1 905.0 15.0
0.6 18.4 8.6
0.8 27.6 16.2 726.0 14.0
69.1
35.4
55.3
70.2
0.8 46.8 12.5 783.0 12.4

0.3 22.6 3.4 362.0 5.4
1.0 27.7 13.8 291.1 18.8
1.2 79.1 12.9 296.3 16.7
115.9 1.0 46.4 16.2 858.9 18.0
45.1
80.6
39.7
602.9 4.7
0.6 33.5 13.6 858.4 10.7
0.3 24.2 5.9
1.2 45.1 14.5 1498.9 17.7
1.5
Pr
1.2
24.3 3.3
11.6 1.4
32.1 4.3
35.7 4.8
39.0 4.9
13.0 1.9
28.0 4.0
16.0 2.2
29.0 3.9
20.0 2.9
8.0
SO3
1.8 0.5 0.4 2.3 0.5 0.3 1.5 0.2
Sm Eu Tb Dy Ho Tm Yb Lu
0.5
0.4
0.4
0.0
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
1.0 0.2 0.1 0.9 0.2 0.1 0.5 0.1
1.8 0.3 0.2 1.4 0.3 0.1 0.8 0.1

1.6 0.4 0.2 1.3 0.3 0.1 0.7 0.1
13.1 2.6 0.5 0.4 2.2 0.5 0.2 1.2 0.2

5.3 1.0 0.2 0.1 0.7 0.1 0.1 0.4 0.1
15.7 2.9 0.6 0.4 2.6 0.5 0.2 1.4 0.2
18.1 3.4 0.6 0.5 2.6 0.6 0.3 1.5 0.2
20.0 3.9 0.6 0.5 3.1 0.6 0.3 1.6 0.3
7.7
15.3 3.2 0.6 0.5 2.6 0.5 0.3 1.4 0.2
9.1
15.8 3.2 0.6 0.5 2.8 0.6 0.3 1.6 0.2
11.6 2.4 0.5 0.4 2.1 0.5 0.2 1.2 0.2
4.7
19.1 3.7 0.7 0.5 2.9 0.6 0.3 1.5 0.2
7.0
Nd
0.2
N.D.
N.D.
N.D.
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.2
<0.1
0.2
P2O5
38.0 4.8
9.0
Ce
0.4
0.4
0.3
0.1
0.6
0.5
0.1
0.2
0.3
0.2
0.3
0.2
0.4
TiO2
145.9 0.4 75.4 13.4 957.6 4.8
459.2 103.0 54.9
116.3 953.2 68.3
75.3
72.6
110.6 289.5 73.6
58.4
115.5 547.0 96.2
60.9
Ba
TABLE 3.6B: MINOR AND TRACE ELEMENTS (PPM)
SiO2
Sample
TABLE 3.6A:MAJOR ELEMENTS (IN WEIGHT/PERCENT)
TABLE 3.6: PAKHNA MARL AND EA 33-34 AFTER CALCULATING THE DILUTION FACTOR
CONCLUSION: THE ALASHIYA TABLETS AND LATE BRONZE AGE CYPRUS

THE GOVERNMENT OF CYPRUS IN THE LATE BRONZE AGE: AN OVERVIEW OF CURRENT
OPINIONS
It is widely accepted that the political system of Late Bronze Cyprus should be reconstructed on its
own terms without a priori reference to the palatial polities of the ancient Near East or the Aegean
(e.g. Knapp 1996a:4; Peltenburg 1996:27; but see strm and Herscher 1996:10, 12). In addition, since
the archaeological record changes considerably toward the Late Cypriote II (13th century BCE), both
Peltenburg (1996:28) and Knapp (1996b:67-69) warned against extrapolation backward, and envisaged
diachronic political changes during the Late Cypriote period.
The key in this discussion has always been copper production the economic backbone of Cyprus in
antiquity. The earliest commercial exploitation of Cyprus copper ores took place along the northern rim
of the Troodos mountains at the beginning of the second millennium BCE. Although the main settlements
along the northern coast of the island are still unexcavated, the rich cemeteries of Lapithos Vrysi tou
Barba, Bellapais Vounous, Vasilia Kaphkalla and Karmi Palealona testify to their economic importance.
Indeed, all known imports from the Aegean, the Levant and Egypt into Cyprus during this period were
found at these four sites, considered to be primary nodes for external trade (Knapp 1994:418-419).
During the rst half of the second millennium BCE, however, the centre of gravity of both settlement
and economy shifted from the northwest to the central and eastern parts of the island. This is clear not only
from the clusters of settlements in the north-eastern foothills of the Troodos mountains, south of Nicosia,
and in the eastern Mesaoria, but also from the heavy concentration of Near Eastern imports in these
regions. It is certainly not a coincidence that the early references to copper from Alashiya in Syrian and
Babylonian cuneiform documents appear at the same temporal horizon (late 19th-17th centuries BCE
Courtois 1986; Knapp 1996a:5; Keswani 1996:219). During the Middle Cypriote period Kalopsidha might
have been the main gateway for Cypriote copper exported to the Levant. However, the archaeological
record suggests that in the early part of the Late Bronze Age (MC III-LC I, ca. 1700-1400 BCE) it was
replaced by Enkomi Ayios Iakovos, soon to become the most important site in eastern Cyprus.
About the same time that Enkomi was established as a centre in the east coast of the island (ca. 1600
BCE), Toumba tou Skourou was founded as the main centre of the Morphou Bay (a possible explanation
for the mode of urbanization of these two sites is suggested by Keswani 1996). However, only scant
remains of metallurgical activity were discovered at this site, compared to the proliferation of such
remains at Enkomi even in its early phases (though it should be remembered that large parts of Toumba
tou Skourou were bulldozed away prior to excavation) (Keswani 1996:220-221). Thus, it appeared to
many that at the beginning of the Late Bronze Age Enkomi controlled the copper industry of the entire
island and served as its main seat of power (Muhly 1989:299; Peltenburg 1996:29-35; Knapp 1993a:99;
1996b:67; 1997:65). The arguments raisesd by Merrillees (1992) and especially Keswani (1993; 1996:
234) against the existence of a unied, island-wide polity, founded and dominated by a paramount centre
at Enkomi, were refuted by Peltenburg (1996:28; see also Knapp 1997:48-49) on the ground that they
pertain to the situation in the later part of the Late Cypriote period.
The emergence of Enkomi as a new and dominant centre may have been accompanied by
the establishment of a network of hinterland forts aimed at securing control of the mines and the
communication lines connecting them to the coastal gateway (Peltenburg 1996:30-33; but see Keswani
1996:219). Most scholars have argued that the political situation in Cyprus did not change during the
15th-14th centuries BCE and is still reected in the Amarna tablets. They have suggested that a king,
70
most probably seated at Enkomi, controlled the copper production of the island and participated in the
international trade conducted in the eastern Mediterranean (e.g. Knapp 1997:65-66). The political and
economic situation seems to have considerably changed in the Late Cypriote IIC period (13th century
BCE). Fragmentation of central authority appears to be supported by the new data acquired by recent
eldwork. Many urban sites, some of them much larger than Enkomi, were established, mainly along
the southern coast (Negbi 1986; strm and Herscher 1996). Their monumental ashlar architecture,
large storage facilities, industrial installations for olive oil and copper processing and administrative
technology (e.g. seal impressions and Cypro-Minoan inscriptions) as well as rich burials, seem to
indicate that they must have been centres of regional polities and the seat of local elite. It has therefore
been suggested that power and the islands resources were shared in the post-Amarna period by a group
of peer polities (see e.g. Muhly 1989:301-303; 1993; Peltenburg 1996:28, 36; Knapp 1996b:68; 1997:6668). Since the above political reconstruction suggests that the focus of power on Cyprus had continuously
shifted during the second millennium, scholars noted that Alashiya may have been a general designation
for the entire island rather then a name of a city (cf. Masson 1973; Knapp 1996a:8).
In a few texts Alashiya appears with the URU determinative. However, in certain cases scribes who
worked in the peripheral kingdoms did not distinguish between determinatives, and sometimes used
URU, or KUR.URU, for KUR. An examination of all the references where Alashiya appears with the
URU determinative reveals that not a single text refers unequivocally to a city. A town named Alashiya
is not borne out by the textual evidence. We may conclude that Alashiya was a name for the island of
Cyprus, or sometimes for part of it, and that the claim that a certain city was called Alashiya is yet to be
conrmed (Knapp 1985:237; 1990:799; 1996a:4; Naaman 1997:611).
THE POLITICAL-ADMINISTRATIVE CENTRE OF ALASHIYA DURING THE 14TH-13TH CENTURIES
BCE ACCORDING TO THE ANALYTICAL RESULTS
In the light of the analytical results, all the Alashiya letters examined can be attributed to the southeastern
margins of the Troodos mountains in Cyprus. Before attempting to locate a possible candidate site
for their origin, the old identication of the capital of Alashiya at Enkomi should be re-examined.
Our ndings support the results of the Berkeley team, which tested and dismissed this hypothesis a
generation ago (Artzy et al. 1976).
A general look at the large-scale geological map of Cyprus (1:250,000, 1995 edition) reveals that
the whole area of eastern Cyprus in general, and Enkomi in particular, is characterized by Pleistocene
to Holocene sands, silts, and sedimentary calcareous rocks including mainly biocalcarenites and marl.
This means that the source materials of the tablets are far beyond the exploitable ranges of Enkomi,
Kalopsidha, or in fact any other site in the eastern part of the island. Yet is it possible that carefully
selected clay sources were imported to Enkomi from other parts of the island in order to produce
clay tablets? This possibility, raised by Vassos Karageorghis in a discussion of our results, cannot be
dismissed on petrographic considerations alone. Nonetheless we are condent that this is not the case
for several reasons:
1. The local marl at Enkomi is itself suitable for tablet production under the restrictions suggested
above, i.e., high plasticity, low shrinkage, light colour and no grits. Creamy colour and ne
texture distinguish the pottery produced at Varosi, the Greek quarter of Famagusta. Potters who
ed from Varosi during the civil war in Cyprus testify that one of their main sources for the high
quality white earth was in Enkomi (Ionas 1998:133). Therefore, there seems to be no reason for
a scribe to prefer remote sources of clay to this high quality and easily available source.
71
2. All the Alashiya tablets discussed here differ drastically in their materials from the Cypro-Minoan
tablets found at Enkomi. The Enkomi tablets that were examined cover a time-span of ca. 300
years, which overlaps the periods of the Amarna and Ugarit letters, and display an extremely
homogenous composition. It is clear that the raw material selected for tablet production at
Enkomi remained basically the same.
3. While the locally produced pottery and Cypro-Minoan tablets at Enkomi excel in their ne fabric
and light shades, EA 37 and RS L.1 are reddish-brown in colour and contain coarse inclusions.
The scribes who produced these two tablets no doubt selected any clay that was handy. This clay
typies the margin of the Troodos.
Therefore, the Alashiya tablets from Amarna and Ugarit were produced far from the Enkomi region
and the search for their point of dispatch can return to the southern slopes of the Troodos mountains.
The two different clay types that were used for the production of the Alashiya tablets, namely the
predominantly calcareous fabric derived from the Miocene Pakhna marls and the volcanic-derived clays
from the ophiolite margins, can be interpreted in two different ways. They can be understood either as
representing two different sources for the tablets, or else a single location where both clay types were
used simultaneously. Since the volcanic-derived tablets were also made in the contact zone with the
sedimentary area where both clay types are exposed in near proximity, the second possibility should
be favoured. Indeed, in sites located on the margins of the Troodos this twofold use of clay types was
customary for pottery production from the Early Bronze Age through the Late Bronze Age (Barlow and
Idziak 1989; Vaughan 1991:353-354; Weisman 1996; Barlow and Vaughan 1999). Therefore, it is very
likely that all the Alashiya letters originated from a single site.
The combination of Pakhna marl and the Moni Mlange clay is rather limited in distribution. In
the immediate area where the Moni Formation is exposed there are only a few large Late Cypriote II
sites that may have served as administrative centres. These include rst and foremost Kalavasos Ayios
Dhimitrios and Amathus. Within an exploitable range of 10 km from the clay sources lie also Maroni
Vournes (on the coast) and Alassa Paliotaverna/Pano Mandilaris (in the piedmont), where Moni clays
and Pakhna marls are bounded in the CTF. Adding the doleritic-derived clay of RS L.1 as another
possible resource does not change this picture signicantly, although it makes the coastal sites (Maroni
and Amathus) less probable. The petrography of the letters indicates a location in the Troodos margins
in both the 14th and the 13th centuries BCE. Thus it can be assumed that the administrative-political
centre of Alashiya was the same during this time.
Other considerations may help in narrowing the possibilities further. Historical sources indicate that
copper played a major role in the relations between Alashiya and its eastern neighbours. A site that is
directly related to copper production may therefore be favoured. In the area limited by the distribution of
the above clay types, the sites of Kalavasos and Alassa best answer this description.
The site of Kalavasos Ayios Dhimitrios is situated in the Vasilikos Valley, northeast of Akrotiri
Bay. It is located in an area of the Lefkara Formation of Paleogene age, dominated by chalks and marls
(Gass et al. 1994:115-119). It is adjacent to the Kalavasos copper mines at the southeastern corner of the
Troodos massif. Imposing ashlar masonry, as well as extensive olive oil production and storage facilities
distinguish this site as a major administrative centre (South 1996, 1997). It ourished in the Late Cypriote
IIC, though excavations beneath and around ashlar Building X indicate that it was occupied already in
the Late Cypriote IIA. The evidence from the tombs, which exhibit the wealth and international contacts
enjoyed by the local elite, corroborates this observation. Although located not far from the coast,
Kalavasos Ayios Dhimitrios is situated close to the volcanic complex of the Troodos massif. Therefore it
is suitable for the identication of the origin of EA 37.
72
The site of Alassa Paliotaverna/Pano Mandilaris was discovered in 1983 in the course of a
salvage survey prior to the construction of the Kouris dam. The excavations exposed a considerable
Late Cypriote IIC-IIIA site, the only one in the hilly zone of the Troodos piedmont excavated thus far
(Hadjisavvas 1989; 1996). The site is situated close to an area rich in sulde ores and other copper alloys,
and the excavations revealed rich evidence for metallurgical activities (Hadjisavvas 1989). Impressive
architecture characterized by ashlar masonry, dating mainly to the 13th century BCE but apparently
showing indications of 14th century BCE activity as well, was exposed at Paliotaverna (Hadjisavvas
1996). The pottery from the foundation trenches of Building I is attributed to the Late Cypriote II
(Hadjisavvas 1996:28, 32). New 14C measurements seem to support a 14th century BCE date for the
foundation of Building II (Hadjisavvas, pers. comm.). Moreover, the long sequence of the burials from
Alassa Pano Mandilaris starts at Late Cypriote IB (Hadjisavvas 1991). The imposing architecture and the
impressive storage facilities for olive oil (i.e., pithoi, some bearing seal impressions) indicate that the site
was a major administrative centre (Hadjisavvas 1996:34-36).
Geologically, the site of Alassa is located in an area dominated by the Pakhna Formation. The rst
exposures of the pillow-lava series appear only several kilometers to its north, and the westernmost
margins of the Moni Formation can be found at Yerasa 10 km to the northeast. In terms of CTF these
clays are within the exploitable territory of the site.
THE GOVERNMENT OF CYPRUS IN THE LATE BRONZE AGE: ALTERNATIVE MODELS
Our analysis seems to resolve the debate about the identication of Alashiya. Firstly, it shows that
Alashiya, acknowledged as a member of the club of great powers in the Amarna correspondence (see
Zaccagnini 1987:61-65 for the diplomatic rhetoric of the day), must be located in Cyprus. Secondly, it
indicates that the political and administrative centre of Alashiya in the 14th-13th centuries BCE was
situated in the southern Troodos piedmont region.
The four authors could not reach a consensus on the location of the centre from which the Alashiya
tablets originated. With no more information available, Bunimovitz tends to leave the identication open
between Kalavasos and Alassa. Finkelstein, Goren and Naaman opt for the latter. It is an imposing site,
with evidence of administrative activity, located close to the copper mines of the southern Troodos and
ideally situated to control both the hilly, inland areas of the Troodos and the centres and harbours of the
coast. Whoever rules in Alassa controls the ow of commodities between the Troodos and the coast.
Courtois (quoted by Hadjisavvas 1996:23) noted the similarity between the toponyms Alashiya
and Alassa. However, this is not a decisive argument for the identication of Alashiya which was a
designation for a vast area whose name might theoretically have been preserved in any part of the island.
Even if it were preserved in Alassa, it may point to no more than a general area in which the political
centre of Cyprus might have been located.
Finally, there is textual evidence which seems to support the identication of the capital of Alashiya in
the hilly area, away from the coast. RS 20.18 is a letter sent by Eshuwara, the chief prefect of Alashiya, to the
King of Ugarit. It was probably written at the same time as RS L.1, in the last days of Ugarit. The relevant
passage (lines 16-28) runs as follows (Nougayrol et al. 1968, no. 22): But now, (the) twenty enemy ships
- even before they would reach the mountain - have not stayed around but have quickly moved on, and where
they have pitched camp (?) we do not know. I am writing you to inform and protect you. Be aware!.
Having Enkomi in mind, scholars have always translated the mountain (shore). Nevertheless, one
should follow the verbal translation mountain (there is no shore in the text). It seems that Eshuwara
reports a threat to his place, located in a mountainous area, which was avoided because the twenty ships
73
of the enemy moved forward, and now may threaten Ugarit. This textual evidence nicely ts the location
of the political-administrative centre of Alashiya at Alassa.
Other textual references to Alashiya indicate that, in addition to the desire to keep a close eye
on the copper production, ongoing security problems at the coastal regions of the island may have
dictated an inland location for its capital. In the famous Hittite Indictment of Madduwatta text, which
is dated to the end of the 15th century BCE, one reads of pirates raids on Alashiya by people from
western Anatolia (Bryce 1989:298-299; Knapp 1996a:31). A few generations later, the King of Alashiya
complains that Lukkians (a people located in southwestern Anatolia) continuously raided his villages
(EA 38; Moran 1992:111-112). Apparently, hostages were the main booty of these sea-borne raids
(Wachsmann 2000:103-104). Christodoulou (1959:62) has noted that, in some later periods as well, fear
of piracy or invasion inhibited village settlement on the coasts of Cyprus.
With the origin of the Alashiya tablets established at Alassa or Kalavasos, we can go back to the
general socio-political picture of Late Bronze Age Cyprus. Building on Catlings (1962:144-145; 1975:
189-193) seminal tripartite division of the Late Cypriote settlement system and economy into: coastal
industrial, administrative and trading urban centres; inland rural settlements; and copper production sites
in the Troodos foothills, Keswani (1993) and Knapp (1996b; 1997) have presented updated versions of
settlement cum society hierarchy in Late Bronze Age Cyprus. The essence of their models, which added
the important category of ceremonial or sanctuary sites to the settlement structure proposed by Catling,
is a tentative explanation of how subsistence goods, raw materials and luxury items were produced,
distributed, transported and administered within the Late Cypriote settlement system. Interwoven within
these models are suggestions about the sociopolitical and economic mechanisms (e.g. staple/wealth
nance) that integrated a variety of functionally different sites on a regional scale. Both scholars, however,
have emphasized the dynamic nature of their constructs, whose conguration and elements must change
as new data emerge (Keswani 1993:79-80; Knapp 1996b:66-67). Knapp (1997:61) noticed that Kalavasos
and Alassa, de ned in the above models as primary centres, do not conform to the ideal type of such
centres and are unique in their multiplicity of functions. Closer to the copper mines than primary coastal
centres, these two sites must have controlled the mining, production and transport of copper and served as
centres of commercial administration. Knapp further suggested that intermediary coastal sites for these
inland centres could have been Maroni Tsaroukkas (for Kalavasos) and Kourion Bamboula (for Alassa).
Reformulating these models about the Late Bronze Age Government of Cyprus in the light of our
ndings, three alternative models seem viable. According to the rst, there was a single, centralized
authority for the island polity during the 16th-13th centuries BCE. The seat of power was located in the
mountainous area throughout this period, but the pre-14th century centre has not been located yet.
The second model acknowledges a single centralized authority for Alashiya during the 16th-13th
centuries BCE, but would argue that the location of the centre of power within this polity and its internal
organization changed in the course of time. Possibly emerging about 1600 BCE, together with the
establishment of Enkomi, the rulers of Alashiya may have established their government at this early stage
over the eastern part of the island if not already on its entirety. As Dikaios (1969:11) claimed long ago, in
analogy to the Prodromos villagers carrying Troodos ice to Nicosia in the summer nights, copper ore could
have been transported to Enkomi from the Skouriotissa and other mines which lie on the northern foothills
of the Troodos. Later on, other coastal urban centres emerged, and one may wonder if their foundation was
initiated by the central government or by the local elite. Clearly, however, lacking storage facilities and
relying on transport from the hinterland, the coastal cities must have been integrated within the overall
settlement and economic system for their livelihood. In the late 15th - beginning of 14th centuries BCE the
74
main seat of power moved inland to Alassa (Finkelstein, Goren and Naaman), or to either Kalavasos or
Alassa (Bunimovitz). The royal letter RS L.1 as well as the new Alashiyan letters from Ugarit mentioning
a king of Alashiya named Kushmeshusha (Yon 1999:118; Malbran-Labat 1999:122) were written at the end
of the Late Cypriote IIC (13th century BCE). It seems, therefore, that the picture of political fragmentation
of Cyprus depicted for that period is exaggerated if not completely wrong. The unprecedented urban
ourishing during the Late Cypriote IIC reects the expansion of the commercial system in the eastern
Mediterranean region and the success of the Alashiyan ruler(s) to integrate within it in order to exploit it
fully. Enkomi continued to serve as a gateway community for the Cypriote copper trade with the Levant.
According to the third model, Alashiya=Cyprus may have been comprised of a number of competing
regional factions, or of a federation of such independent polities as envisioned by Merrillees and
Keswani. However, since Near Eastern rulers of the 14th-13th centuries BCE acknowledged a King of
Alashiya, which according to our research was situated at Alassa or Kalavasos, he must have been either
the overlord of the other Cypriote polities or a primus inter pares.
It is difcult to decide which of the above scenarios is closer to the Late Cypriote reality. The
identication of Alashiya with Cyprus and the location of its inland capital are essential steps in the
process of unravelling the complicated picture of the Government of Cyprus in the Late Bronze Age.
Many questions are still to be answered before the picture becomes clear. In the meantime Cadogans
(1996:15) optimistic view is comforting: ...it is a good sign that Late Cypriote society and history become
ever more complicated to explain. It means that we are coming a little closer to the realities of those days.
VII. INTERNATIONAL LETTER OF UNKNOWN PROVENANCE

EA 172 (VAT 1877)
Reliability: High.
Matrix: Greyish tan in PPL, with speckled b-fabric and high calcitic component. Dark reddish-brown and
opaque iron oxide minerals are relatively common (~3%). Fine silt that gradually continues to ne sand
in size (around 2%) includes mostly quartz but with the addition of abundant accessory heavy minerals
including biotite and muscovite, hornblende, plagioclase, epidote, zircon, sphene, augite, titanaugite,
serpentine and chlorite. Many of the latter reach also ne sand size.
Inclusions: No inclusions are found in this tablet except for the naturally occurring ne sand fraction.
Very few quartz grains and a rounded grain of quartzite size up to 300m.
Firing temperature: There are not enough indicators to determine the ring temperature of this fragment
but it must have been lower than 800oC since there is no colour change in the hornblende.
Geological interpretation, reference and conclusions: EA 172 is a small fragment. According to its
petrography it may be assigned to an area neighbouring a schistose province judging by the richness
of mica and other unstable heavy minerals in its silt to nd sand fraction. If so, it is denitely a nonCanaanite tablet and should be related to the international section of the archive. The fragmented text
(four badly broken lines) looks like a list of some kind, and hypothetically it might have been a list of
gifts sent by an unknown king to Egypt.
75
CHAPTER 4
SCHOLARLY TEXTS
Thirty-two documents in the Amarna archive are classied as scholarly texts. This assemblage was in
all likelihood part of the local school for scribes where students were trained in producing tablets and
inscribing them in cuneiform script. This task, alien to the Egyptian scribes who were accustomed to
writing with ink on papyri, required special skill and long training. The scholarly texts discovered at
Amarna are either library documents that were used as textbooks, or students exercises. A few texts are
myths and epics, and among them is the so-called Triad (EA 356-358) a group of three tablets containing
Mesopotamian mythic and narrative compositions. Others belong to different genres such as syllabaries,
lexical texts, a god-list, a tale and an amulet. In antiquity these texts were used for training young scribes
and for obtaining certain knowledge of the literary tradition of the Syro-Mesopotamian world.
In order to achieve a better understanding of the work of the Amarna school of scribes, we shall
rst try to establish whether a given tablet is of Western Asiatic origin, or was written in Egypt. In the
former case, we shall try to establish the provenance of the tablet, rst and foremost by comparing it
with other Amarna tablets (such as the letters from Babylonia, Mitanni, atti and Ugarit). In the latter
case we shall examine whether local clay was used or, as in the case of the royal letters, a special type of
clay was brought from distant places. Needless to say, in the case of a mixed assemblage we shall check
which texts were inscribed on local clay and which were written on foreign materials.
CATALOGUE
EA 340 (VAT 1583), an historical tale or a letter fragment
Reliability: Satisfactory/moderate.
Matrix: Carbonatic, light grey to greyish-brown in PPL, devoid of foraminifers, with few small (around
60-70m) greenish to light greenish-yellow glauconite concentrations and silt (about 5%). The
matrix is optically active with speckled b-fabric and no preferred optical orientation. Opaques (~2%)
are between a few micrometers and nearly 50m. The silt contains mainly quartz, but hornblende,
zircon and feldspar were also identied.
Inclusions: The inclusions (f:c ratio{0.062mm}=92:8) contain moderately sorted angular, ne sand particles that
maintain a gradual continuum with the silt in terms of grain size. The sand fraction contains subangular
to rounded particles of quartz, up to 250m (dominant); subangular calcite crystals, up to 180m, (few);
subrounded hornblende, up to 150m (very few).
Firing temperature: Probably unred judging by the colour of the glauconite.
Geological interpretation: This tablet is identical in all components to the Egyptian letters. From a
technical point of view, it is unred (or very lightly red), a trait that is characteristic of the Egyptian
documents (unlike the Mesopotamian tablets which are usually well red).
Reference: See EA 1 (Chapter 3.I).
Conclusions: Scholars debated whether this tablet is a letter or some kind of literary text (Izreel 1997:15-16).
Petrographically it is of Egyptian provenance. The text differs from all the Amarna letters and establishing
its origin in Egypt fully corroborates the latter assumption, rst suggested by Knudtzon (1915:17-19, 23).
76
EA 341 (Ash. 1893 1-41: 415), the story of Kei

Matrix: Carbonatic, light yellowish-tan to brown in PPL, almost devoid of foraminifers, with few small
(around 60-120m) yellowish to light greenish-yellow glauconite concentrations and silt (about
5%). The matrix is optically active with speckled b-fabric and no preferred optical orientation.
Opaques (~1%) are between a few micrometers and nearly 50m. The silt contains mainly quartz,
but also hornblende, zircon and feldspar.
maintain a gradual continuum with the silt in terms of grain size. The sand fraction contains subangular
to rounded particles of quartz, up to 350m (dominant); subangular micritic limestone, up to 300m
(few); rounded hornblende, up to 180m (very few).
Vegetal material (SLY): A single fragment. Badly preserved non-characteristic tissue fragment.
Firing temperature: Very low (up to 5000C) judging by the beginning of colour change in the glauconite.
Conclusions: EA 341 is a fragment of an Akkadian version of the Hurrian tale of Kei (Izreel 1997:17-19).
Petrographically it is of Egyptian provenance. The tablet must have been copied in Egypt from an
original brought there from an unknown northern kingdom.
EA 342 (Ash. 1893 1-41: 414), an exercise in letter writing?
Reliability: High.
Matrix: Carbonatic, light reddish tan to reddish tan in PPL, birefringent with speckled b-fabric and
preferred optical orientation. Abundant ne calcite crystals, beginning at few micrometers and
ranging up to 40m make about 20% of the groundmass. Biotite laths are abundant as part of the
matrix (nearly 5%). The silt contains also quartz, opaques, serpentine, oxyhornblende and feldspar.
Inclusions: No remains of vegetal material were traced. No intentionally added inclusions seem to appear.
The sand that appears as clasts within the matrix (f:c ratio{0.062mm}=~99:1) is made of grains of rounded
to subangular quartz, up to 400m (dominant); rounded calcite, up to 200m (common); rounded
serpentine, up to 180m (few), rounded feldspar including twinned plagioclase and microcline, up to
200m (few).
Firing temperature: Probably around 8000C judging by the alteration of hornblende into oxyhornblende.
Geological interpretation: This tablet was probably made on re ned Euphrates sediment. It is similar in
most respects to the Babylonian letters, especially to EA 11 (Chapter 3.III).
Conclusions: The ductus of the tablet is non-Egyptian (Knudtzon 1915:24), and Izreel (1997:2021) noticed its similarity in colour and fabric to EA 344. Petrographically it is different from
the Egyptian tablets. It is made neither of Esna shales as are most of the Egyptian letters and
scholarly texts nor of Egyptian Nile Silt. Technically it is well red like most Mesopotamian
tablets and unlike most Egyptian documents. It is similar in many respects (especially in matrix) to
the Babylonian letter EA 11. However, it contains a higher proportion of naturally occurring clasts
and therefore the clay is not as well re ned as most of the Mesopotamian letters. The inclusion
assemblage is typical of sites that are located in the lower Euphrates region and the tablet may be
attributed to Babylonia, although this is not entirely certain.
77

Reliability: High.
Matrix: Carbonatic, light yellowish grey to greyish-brown in PPL, devoid of foraminifers, with few small
(up to 120m) greenish to deep greenish-yellow glauconite concentrations and silt (about 2%).
The matrix is optically active with speckled b-fabric and no preferred optical orientation. Opaques
(~1%) are between a few micrometers and nearly 70m. The silt contains mainly quartz, but also
hornblende, zircon and feldspar.
Inclusions: The inclusions (f:c ratio{0.062mm}=98:2) contain moderately sorted, rounded to subangular sand
particles including subangular to rounded particles of quartz, up to 500m (dominant); rounded
micritic limestone, up to 230m (few to common); subrounded aquatic mollusc shell fragments, up
to 350m (very few).
Firing temperature: If red at all, probably very lightly (below 5000C) judging by the beginning of
colour change in the glauconite.
Conclusions: EA 343 is a practice tablet (Izreel 1997:22). Petrographically it is of Egyptian provenance.
EA 344 (Ash. 1893 1-41: 417), an exercise?
Matrix: Carbonatic, reddish tan in PPL, birefringent with speckled b-fabric and preferred optical
orientation, but tends to isotropism in places. Abundant ne calcite crystals, beginning at few
micrometers and ranging up to 40m, make about 10% of the groundmass. Biotite laths are common
as part of the matrix. The silt also contains quartz, opaques, serpentine, oxyhornblende, zircon and
feldspar.
Inclusions: No remains of vegetal material were traced. No intentionally added inclusions seem to
appear. The sand that appears as clasts within the matrix (f:c ratio{0.062mm}=~99:1) is made up of
grains of rounded to subangular quartz, up to 650m (dominant); rounded serpentine, up to 120m
(common); rounded micritic limestone or clear calcite, partially decomposed by ring, up to 200m
(few); subangular oxyhornblende, up to 130m (very few).
Firing temperature: Probably around 8000C judging by the alteration of hornblende into oxyhornblende.
Geological interpretation: This tablet was probably made on rened Euphrates sediment. It resembles in
most aspects the Babylonian letters, especially EA 11.
Conclusions: As EA 342. The few remaining signs on this fragment make it uncertain whether it was indeed
a school text or a letter. In the latter case it may be regarded as a fragment of a Babylonian letter.
Matrix: Carbonatic, greyish-brown in PPL, devoid of foraminifers, with silt including opaques, quartz
and hornblende.
Inclusions: The sand fraction contains subangular to rounded particles of quartz, up to 300m; rounded
micritic limestone, up to 200m.
78

Conclusions: EA 345 is perhaps an exercise in writing lexical sequence (Izreel 1997:24). Petrographically it
is of Egyptian origin.
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, devoid of foraminifers, with small
(up to 120m) greenish-tan to orange glauconite concentrations and silt (about 3%-4%). The
matrix is optically active and has no optical orientation. Opaques (~1%) appear between a few
micrometers and nearly 50m. The silt (2%) contains mainly quartz, but also hornblende and
rutile could be identied.
Inclusions: The inclusions (f:c ratio{0.062mm}=95:5) contain moderately sorted angular, ne sand particles
that maintain a gradual continuum with the silt in terms of grain size. The sand fraction contains
angular to rounded particles of quartz, up to 400m (predominant); rounded micritic limestone, up to
300m (few); rounded microcline up to 150m (very few).
Firing temperature: Probably red at around 5000C or slightly higher judging by the colour change in
the glauconite.
Conclusions: EA 346 is an exercise, possibly of a lexical list (Izreel 1997:25-26). Petrographically it is
of Egyptian provenance.
EA 347 (Ash. 1893 1-41: 422), a lexical list?
Matrix: Carbonatic, grey to greyish-brown in PPL, devoid of foraminifers, with few (up to 100m) deep
orange tan to orange glauconite concentrations and silt (about 2%). The matrix is birefringent with
speckled b-fabric and no optical orientation. Opaques (~1%) appear between a few micrometers and
nearly 50m. The silt (2%) contains mainly quartz, but also heavy minerals (hornblende and rutile
could be identied).
(predominantly subangular to rounded particles of quartz up to 500m in size and a few up to 300m
rounded micritic limestone grains) and added vegetal material.
Vegetal material (SLY): A few (up to 700m lengthwise), uncharred plant tissues. Probably a piece of
wood with a large ray and lignied axial bres.
Firing temperature: Probably red at around 5000C or below, since there is a slight colour change in the
glauconite but the vegetal material remained uncharred.
Conclusions: EA 347 may be a fragment of a lexical list (Izreel 1997:27). Petrographically it is of
Egyptian origin.
79
EA 348 (Ash. 1893 1-41: 419) a fragment of a Sa signlist

Reliability: Moderate/Fair.
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, devoid of foraminifers, with few small
(up to 80m) greenish-tan to yellowish green glauconite concentrations and silt (about 3%-4%).
The matrix is birefringent with speckled b-fabric and no optical orientation. Opaques (~1%) appear
between a few micrometers and nearly 50m. The silt (2%) contains mainly quartz.
Inclusions: The inclusions contain subangular to rounded quartz, up to 200m; sub-rounded micritic
limestone, up to 300m; a single grain of rounded microcline, up to 320m.
Firing temperature: If red at all, probably very lightly (below 5000C) judging by the beginning of
colour change in the glauconite.
Conclusions: EA 348 is a fragment of a palaeographic syllabary (Izreel 1997:28-29). Petrographically
it is of Egyptian provenance.
EA 349 (Ash. 1893 1-41: 428), a fragment of a syllabary?
Reliability: High.
Matrix: Carbonatic, light brownish-grey to greyish-brown in PPL, devoid of foraminifers. The matrix
is birefringent with speckled b-fabric and no optical orientation. Opaques (~1%) appear between a
few micrometers and nearly 60m. The silt (2%) contains mainly quartz, but also heavy minerals
(hornblende and feldspar could be identied).
subrounded to rounded particles of quartz, up to 400m (predominant); rounded micritic limestone,
up to 120m, (few); rounded feldspar of unidentied type, up to 400m (very few); rounded
tourmaline, 220m (rare).
Firing temperature: Slight anomalies in the calcite suggest ring at about 7000C, but this sample contains
no indicators to estimate the ring temperature.
Conclusions: EA 349 is probably a syllabary (Izreel 1997:30). Petrographically it is of Egyptian provenance.
EA 350 (Ash. 1893 1-41: 425), a fragment of a tu-ta-ti exercise
Matrix: Carbonatic, light yellowish-grey to greyish-orange in PPL, devoid of foraminifers, with a few
small (up to 70m) yellow to orange glauconite concentrations and silt (about 3%). The matrix is
birefringent with speckled b-fabric and no optical orientation. Opaques (~0.5%) appear between a
few micrometers and nearly 50m. The silt (2%) contains essentially quartz, but also some feldspar.
Secondary gypsum re-crystallization is visible in part of the sample. This is the result of postdepositional processes in the arid zone of el-Amarna.
Inclusions: The inclusions (f:c ratio{0.062mm}=95-6:4-5) contain moderately sorted angular, ne sand
particles that maintain a gradual continuum with the silt in terms of grain size. The sand fraction
80
contains subangular to rounded particles of quartz, up to 370m (predominant); rounded micritic

limestone, up to 300m (few); rounded microcline, up to 130m (very few); rounded augite, up to
110m (very few).
Firing temperature: Probably lightly red (around 5000C-6000C) judging by the colour change in the
glauconite but no effect on the calcite.
Conclusions: EA 350 is a fragment of a syllabary (Izreel 1997:31). Petrographically it is of Egyptian
provenance.
EA 352 + 353 (Ash. 1893 1-41: 413+421, two pieces joined), a fragment of diri
Reliability: High.
Matrix: Carbonatic, yellowish-grey to greyish-brown in PPL, almost devoid of foraminifers, with small
(up to 60m) yellowish glauconite concentrations and silt (about 3%-4%). The matrix is birefringent
with speckled b-fabric and no optical orientation. Opaques (~1%) appear between a few micrometers
and nearly 50m. The silt (2%) contains mainly quartz, but also heavy minerals, such as hornblende,
muscovite, epidote and feldspar .
subangular to rounded particles of quartz, up to 600m (dominant); rounded micritic limestone, up to
280m (common); rounded microcline: up to 150m (very few); angular to subangular hornblende;
up to 120m, (very few); rounded aquatic mollusc shell fragments, up to 500m (very few).
Firing temperature: Probably very lightly red (around 5000C) judging by the colour change in the
glauconite.
Conclusions: The two combined fragments of EA 352 and EA 353, most likely together with EA 351
(which was destroyed and therefore not sampled) and EA 354 below, are parts of a diri lexical list
(Izreel 1997:37-38). Petrographically it is of Egyptian provenance.
EA 354 (Ash. 1893 1-41: 418), a fragment of diri
Reliability: High.
Matrix: Carbonatic, yellowish-grey to greyish-brown in PPL, almost devoid of foraminifers, with small
(up to 80m) yellowish-green glauconite concentrations and silt (about 3%-4%). The matrix is
birefringent with speckled b-fabric and no optical orientation. Opaques (~1%) appear between a
few micrometers and nearly 50m. The silt (2%) contains mainly quartz, but also heavy minerals
(hornblende, muscovite and feldspar could be identied).
subangular to rounded particles of quartz, up to 940m (dominant). The largest particle is spherical
and made of polycrystalline quartz. Also rounded micritic limestone, up to 1.5mm (common); rounded
microcline, up to 180m (very few); angular to subangular hornblende, up to 100m (very few).
Vegetal material (SLY): A badly preserved unidentied tissue fragment.
81
Firing temperature: Probably very lightly red (around 5000C) judging by the colour change in the glauconite.
Conclusions: Izreel (1997:34, 38-39) suggested joining EA 354 with EA 352 and 353. The petrographic
similarity in many details between these tablets supports his suggestion.
EA 355 (Ash 1893.1-41: 416), inscribed cylinder
Matrix: Carbonatic, yellowish-grey to greyish-brown in PPL, devoid of foraminifers, with small (up
to 70m) orange glauconite concentrations. The matrix is birefringent with speckled b-fabric and
no optical orientation. In cross section the matrix is microlaminated and tends to crumb along the
lamination planes. Opaques (~0.5%) appear between a few micrometers and nearly 50m. The silt
(2%) contains mainly quartz, but also hornblende, zircon and feldspar.
Inclusions: The inclusions (f:c ratio{0.062mm}=95:5) contain moderately-sorted angular ne sand particles
subangular to rounded particles of quartz; up to 500m (predominant); rounded micritic limestone,
up to 150m (few); rounded feldspar unidentied to type, up to 320m (very few).
Firing temperature: Probably lightly red (around 5000C or slightly above) judging by the colour change
in the glauconite.
Geological interpretation: The microlamination may be the result of the use of shales of lower quality
than in the other tablets.
Conclusions: The clay cylinder was found during Petries excavations in Building 19 (the Records
Ofce) at el-Amarna (Petrie 1894:36 and Pl. XXXII no. IX; Izreel 1997:41-42). The only parallel
to this unique artefact is a letter written on a clay cylinder that has recently been found at Beth-shean
(Horowitz 1996; see Chapter 12.IX). On the basis of its materials EA 355 can be related to the group
of Egyptian letters written on Esna marl.
EA 356 (VAT 348), the myth of Adapa and the South Wind
Reliability: High.
Matrix: Carbonatic, light reddish tan to reddish tan in PPL, birefringent with speckled b-fabric and weak
optical orientation. In cross section the matrix is microlaminated and tends to crumb along the
lamination fronts. Abundant ne calcite crystals, beginning at a few micrometers and ranging up to
40m make about 20% of the groundmass. Finer, tan argillaceous bodies are embedded within the
matrix. Biotite laths are abundant as part of the matrix (nearly 5%). The silt contains also quartz,
opaques, serpentine, glaucophane, and hornblende altering into oxyhornblende, epidote and feldspar.
Inclusions: No remains of vegetal material were traced. The inclusions (f:c ratio{0.062mm}=~93:7) are made
predominantly of rounded to subangular grains of quartz (up to 430m), sometimes polycrystalline,
occasionally with undulose extinction. Subrounded calcite grains (up to 200m) are common, as are
rounded grains of feldspar (up to 450m), including twinned plagioclase and microcline. There are
very few subrounded grains of olivine (up to 230m).
Firing temperature: probably around 800 0C judging by the alteration of hornblende into oxyhornblende.
Geological interpretation: This tablet was probably made of Euphrates sediment.
82
Reference: See Chapter 3.III.

Conclusions: EA 356 is a mythic tale written in Babylonian ductus (Izreel 1997:43-50; 2001:47-54).
Petrographically, it can be attributed to the lower Euphrates clays. It is obviously not Mitannian, since
as we demonstrated above, Mitannian letters were not produced of Euphrates sediments. The presence
of glaucophane is notable; it is typical to the Babylonian tablets but absent from the Mitannian
specimens. Therefore we suggest that EA 356 is a literary text that originated from Babylonia. As
opposed to the Babylonian letters, the clay is not re ned and contains naturally occurring clasts. The
chemical analyses (NAA and ICP) somewhat complicate the picture (see below).
EA 357 (VAT 1611 + 1613 +1614 + 2710; BM 29865), the myth of Nergal and Ereshkigal
Reliability: High.
Matrix: Carbonatic, light yellowish-grey PPL, devoid of foraminifers, with common small (up to 120m)
yellowish-green and light yellow glauconite concentrations. The matrix is birefringent with speckled
b-fabric and weak optical orientation. Opaques (~1%) appear in a bimodal size groups, between a
few micrometers and nearly 50m, (very common), and between 100m and 200m (rare). The silt
(2%) contains mainly quartz, but also heavy minerals (hornblende, zircon, biotite, feldspars, and
opaques could be identied).
subangular to rounded particles of quartz, up to 700m (dominant); rounded micritic limestone,
up to 680m (common); rounded microcline, up to 150m (very few); fragment of an insect leg,
170m in size.
Paleontology (LG): A sample was examined by the washing method. A few badly preserved foraminifers
were found. Two species were clearly identied: Brizalina (b) sp., and Morozovella (p). This
indicates a Paleogene age.
in the glauconite.
Conclusions: EA 357 is a mythic tale written in Babylonian ductus (Izreel 1997:51-61). Petrographically
it is of Egyptian provenance. This is also supported by the ICP analysis (Fig. 3.1). The palaeontology
supports the identication of the matrix as Esna shales.
EA 358 (VAT 1612 + 2708), a narrative of still undetermined genre
Matrix: Carbonatic, light yellowish-grey to greyish-brown in PPL, devoid of foraminifers. The matrix is
birefringent with speckled b-fabric and no optical orientation. Opaques (~2%) appear in a bimodal
size groups, between a few micrometers and nearly 50m, (very common), and between 100m and
200m (rare). The silt (2%) contains mainly quartz, but also heavy minerals (hornblende, zircon,
feldspar, and opaques could be identied).
maintain a gradual continuum with the silt in terms of grain size. The sand fraction contains subangular to
rounded particles of quartz, up to 1.5mm (dominant) and micritic limestone, up to 450m (common).
83
Palaeontology (LG): A sample was examined by the washing method. Identied: Barren, non-diagnostic.
Firing temperature: Probably very lightly red (below 5000C) but there are not enough indicators to
estimate the ring temperature in this sample.
Conclusions: EA 358 is a mythic tale written in Babylonian ductus (Izreel 1997:62-65). Petrographically it
is of Egyptian provenance. This is also supported by the ICP analysis (Fig. 3.1).
EA 368 (Ash. 1921-1154), Egyptian-Akkadian vocabulary
Reliability: High.
Matrix: Argillaceous, non-carbonatic, light tan in PPL, with striated b-fabric and strong optical orientation.
Opaque minerals are extremely common in the groundmass (over 10%). The non-plastics are badly
sorted and maintain a gradual continuum from the ne silt to the sand fraction. Therefore, they will be
all treated as inclusions. The non-plastics are abundant (f:c ratio{0.010mm}=70:30) and the above-silt
sized inclusions are common too (f:c ratio{0.062mm}=80:20).
Inclusions: The silt to sand fraction is extremely rich in mineral types, including (in their abundance
order from common to rare) quartz, subangular in the ner fraction, becoming rounded in the coarser
fraction, reaching 400m in size. Sometimes exhibiting undulose extinction; biotite and muscovite,
reaching 300m lengthwise; hornblende, usually angular (subrounded in the ner fraction), reaching
300m; pyroxenes: augite (the more common), usually angular or subangular, reaching 200m and
aegirine-augite (rare) reaching 200; micas: microcline reaching 320m and twinned plagioclase in
both silt and ne sand grain sizes, rarely polycrystalline; chalcedony, very few grains up to 150m;
epidote, very few rounded grains, up to 100m; a single aggregate of radial brous zeolite, 150m;
sphene, a single rounded grain, 250m.
Geological interpretation: The very rich mineral suit that appears within the clay of this tablet and its
overall petrographic characteristics are typical of the siliceous category of the Egyptian Nile silt
(Hope et al. 1981, Bourriau and Nicholson 1992; Bourriau et al. 2000). The main attributes of this
category are the non-carbonatic clay, the bad sorting of the inclusions and their gradual transition
from silt to sand, and the contents of quartz, mica minerals, feldspars, amphiboles and pyroxenes.
Reference: Similar clay is recorded in 18th Dynasty pottery from el-Amarna (Hope et al. 1981, Bourriau
and Nicholson 1992).
Conclusions: EA 368 is part of a dictionary or a lexicon that was used in the Amarna school of scribes
(Izreel 1997:77-81). Its ductus indicates that it may have been written by an Assyrian or Mitannian
scribe. However, petrographically it is of Egyptian provenance. Its clay reects an opportunistic
selection of the nearest raw material available. Indeed, the silty character of the clay and its dark
shade make this document far less aesthetic and readable than the bright, relatively smooth Esna
marl used for the formal royal letters and the other scholarly texts.
A blank tablet (Ash. 1893 1-41: 429)
Reliability: Fair.
Matrix: Carbonatic, yellowish-grey to greyish-brown in PPL, birefringent with speckled b-fabric and
no optical orientation. Opaques appear up to nearly 50m. The silt contains quartz.
Inclusions: The inclusions contain angular, fine sand particles that maintain a gradual continuum
84
with the silt in terms of grain size. The sand fraction contains subrounded quartz particles and
angular microcline.
Firing temperature: Probably very lightly red but there are no indicators in this sample for its estimate.
Conclusions: This is a blank tablet that was probably produced for an exercise but was never inscribed.
Petrographically it is of Egyptian provenance.
NEUTRON ACTIVATION ANALYSIS OF TABLETS EA 356-358 AND ITS IMPLICATIONS

Joseph Yellin
NAA has been applied to EA 356 (the myth of Adapa), the Berlin fragment of EA 357 (the myth of Nergal
and Erekigal), and EA 358.1 The results are shown in Table 4.2. For the sake of brevity only a few of the
abundant measured elements are shown. These are calcium (Ca), cobalt (Co), europium (Eu), iron (Fe),
lanthanum (La) and scandium (Sc). Likewise, to save space, errors of measurement are not shown but are
much smaller than the differences in composition between the three tablets. The errors of measurement
expressed as a percentage of the measured values for Ca, Co, Eu, Fe, La, and Sc are respectively 4%, 0.5%,
1.5%, 0.5%, 0.1% and 0.1%. Thus, for example, the value of Co for EA 356 in Table 4.2 is 20.4 ppm +/0.5% or 20.4 ppm +/-0.1 ppm. Note that the lines from Co to Sc for EA 356 and 357 are nearly parallel.
The Ca levels in EA 356 are much smaller than the Ca levels in EA 357. However, all of the other
elements in EA 356 have higher levels than the corresponding ones in EA 357. This may be evidence of
dilution, meaning that the concentration of elements in EA 357 was depressed by the presence of a major
component, such as calcium. If this is true, then there should be a constant proportion between all of the
other elements in EA 356 and 357, which indeed there is. Its value is approximately 0.9. The signicance
of this relationship is that EA 356 and 357 are made of the same clay and therefore share a common origin.
EA 357 is diluted by calcium. Perhaps calcite was added to EA 357 to temper it for structural reasons
such as to make it more resilient to heat. Table 4.3 presents the composition of EA 356 and 357 after the
former was adjusted for dilution. The mean value and root-mean-square deviation (RMSD) (expressed
as a percentage) of the pair is also given. The smallness of the RMSD for all elements is convincing
evidence that their compositions are identical. If they were not, wild uctuations in the RMSD would be
seen. Note that after adjustment for dilution, EA 356 and 357 are nearly coincident.
The composition of EA 358 is different from that of EA 356 and 357 and cannot be related to them
by a dilution factor. Therefore, EA 358 has a different origin. We have not found a compositional match
for any of the three tablets and therefore can say nothing about their origins.
ICP ANALYSES OF THE TRIAD (EA 356-358)

Table A1 in the Appendix presents the results of the ICP analysis of the Triad (EA 356-358). The
data indicate that while EA 357 and 358 are very similar, EA 356 is different in many respects. Even
disregarding the extremely high values of Zn and Cu, all major and most trace elements have higher
1. See also Izreel 2001:56-58. It should be noted that the NAA results reported here were carried out on samples as small as
20 mg, while typical NAA samples are 100-200 mg. This was the rst study in which non-standard sample sizes were used
and the results were most satisfying. Since this study was undertaken (1994), other projects involving precious pottery
requiring delicate sampling procedures were successfully carried out with micro sampling techniques and small sample
size (Yellin et al. 1999).
85
86
37.37
4.21
39.34
5.61
38.30
10.58
0.70
10.80
2.25
10.20
Al2O3
Zn
Co
Ni
Ba
Mn
Cr
125.03 30.59 214.19 195.24 737.93 239.83

52.99 1.79 15.14 32.94 76.28 33.67
342.92 30.93 351.54 137.78 557.50 300.45
332.18 8.07 24.46 11.31 56.87 16.28
513.00 14.00 58.00 276.00 519.00 121.00
120.49
19.70
127.26
7.83
156.00
13.57
1.44
14.93
4.81
21.50
CaO
4.8
20.4
1.138
4.04
26.8
14.2
Ca%
Co
Eu
Fe%
La
Sc
Cu
Sr
La
Ce
5.97
0.34
6.91
0.37
5.50
Fe2O3
14.2
18.8
0.990
3.70
24.9
12.3
EA 357
Nd
4.76 18.10
0.55 2.00
4.37 16.71
0.81 2.97
6.25 23.68
Pr
0.59
0.06
0.52
0.09
0.82
Sm
3.75
0.37
3.32
0.65
4.69
TiO2
Eu
Tb
0.61
0.08
0.53
0.10
0.79
7.2
9.93
0.677
2.16
18.8
7.1
EA 358
0.84
0.09
0.75
0.11
1.11
0.17
0.14
0.20
0.01
0.30
P2O5
Dy
3.55
0.46
3.07
0.53
4.27
Ho
0.84
0.10
0.72
0.13
0.84
0.35
0.04
0.33
0.05
0.42
Tm
2.72
3.16
0.68
0.00
1.50
SO3
Yb
2.21
0.24
1.86
0.31
2.27
Lu
0.28
0.04
0.24
0.03
0.30
18.2
1.02
3.61
24.0
12.7
18.8
0.99
3.70
24.9
12.3
EA 357
* Values are in parts per million except for iron, which is a percentage.
Co
Eu
Fe%
La
Sc
EA 356 (adjusted)
18.5
1.00
3.66
24.4
12.5
MEAN
2.3
2.1
1.7
2.6
2.3
RMSD (%)
TABLE 4.3: COMPARISON OF THE COMPOSITION OF EA 356 AFTER ADJUSTMENT FOR DILUTION WITH THE COMPOSITION OF EA 357*
*Values are in parts per million except calcium and iron, which are percentages.
EA 356
Tablet / Element
Be
5.05
0.35
6.86
0.47
5.20
MgO
1.58 43.05 557.48 17.48 17.23 37.31

0.24 2.70 318.44 1.57 1.56 4.51
1.45 54.63 246.83 15.31 14.99 33.89
0.32 6.80 29.78 3.56 3.00 7.00
1.60 321.00 288.00 30.00 23.00 49.90
TABLE 4.2: PARTIAL COMPOSITION OF EA 356-358*
Bab Avg
Bab SD
Mit Avg
Mit SD
356
MINOR AND TRACE ELEMENTS (PPM)
Babylon Avg.
Babylon S.D.
Mitanni Avg.
Mitanni S.D.
EA 356
SiO2
MAJOR ELEMENTS (IN WEIGHT/PERCENT)
TABLE 4.1: PROFILES OF THE BABYLONIAN AND MITANNIAN LETTERS AND EA 356
concentrations in EA 356 than in the other two tablets. This means that the differences between them are
not the result of dilution created by the presence of a major component that was added as temper. When
plotted on a principal component analysis graph together with the Egyptian and Mesopotamian letters
(Fig. 3.1), EA 357 and 358 cluster closely with the Egyptian letters, while EA 356 is isolated. Though the
ICP analysis cannot match EA 356 with any other cluster, it con rms its distinctiveness.
Table 4.1 indicates that EA 356 does not match either the Babylonian or the Mitannian pro les.
Since petrographically it does t the composition of Euphrates sediments, it may have been brought to
Amarna from another Mesopotamian site.
Finally, the fact that the NAA undertaken by Yellin did not disclose any compositional match for
the scholarly tablets (below) needs some clarication. The lack of equivalent for EA 356-358 suggests
that in certain cases in Mesopotamia, as in Egypt, clays other than the types employed for pottery were
used for preparing tablets. It raises again the problem of the chemical dissimilarity between the Mitanni
tablets and northern Mesopotamian ceramics that was part of our criticism on the study conducted by
Dobel et al. (1977). We now see that this trend was customary also for literary texts in some courts of
ancient Near Eastern rulers. This technical trait was typical only for the Mesopotamian and Egyptian
courts, but not Canaan and Ugarit.
CONCLUSION: THE AMARNA SCHOLARLY TEXTS

Most of the scholarly texts discovered at Amarna were written in Egypt, including exercises of different
kinds, literary and lexical texts, syllabaries and an amulet. Noteworthy are the myth of Nergal and
Ereshkigal, the story of Kei (EA 341) and some other narratives (EA 341, 358). Presumably texts of
northern origin were brought to Egypt at the request of the royal court and copied there by local scribes.
When the royal court left Amarna, the original texts were transferred to the new capital whereas copies
of inferior quality were left in the palace. Preference of the master copies by the Egyptian scribes may
explain why only a few texts of northern origin were discovered at Amarna.
Only three texts of Mesopotamian origin were discovered at Amarna: the myth of Adapa and the
South Wind (EA 356) and two fragmentary texts whose genre is not clear (EA 342, 344). We have
not discovered any text of Anatolian or north Syrian origin. Unfortunately, the texts of the epic of ar
tamari (EA 359, 375), which exhibit some linguistic peculiarities that can be attributed directly to
the Akkadian of Bogazky, are stored in the Cairo Museum and thus have not been investigated in the
course of this study. Izreel (1997:71) noted that the ductus (of EA 359) is distinguishable from the
contemporaneous Hittite tablets, and can be de ned on the basis of many signs to be Egyptian. Also,
the clay looks similar to that of the Egyptian tablets. He therefore suggested that the tablet might be a
copy made in Egypt of a Hittite-Akkadian recension of the ar tamari epic. This suggestion ts well
with the results of our study. It is possible that in this case too, the original tablet was transferred to the
new capital when the royal court moved there.
87
CHAPTER 5
NORTH SYRIAN KINGDOMS
I. UGARIT
Four of the ve Amarna letters which had been identied as originating from Ugarit (EA 45-49) were
examined. The names of the authors of EA 45 and 49 are mentioned on the tablets, whereas the names of
the authors of the other three letters (EA 46-48) are broken. Knudtzon attributed them to Ugarit on the basis
of their clay, script and certain formulae. Of special interest is EA 48, a letter from a certain lady [-]eba,
who sent her mistress a pot of balsam. If this letter was indeed sent from Ugarit, it would add an otherwise
unknown name to the list of Ugaritic queens, and quite signicantly, a Hurrian one.
The site of Ugarit - Ras Shamra on the Syrian coast was excavated over many years and the
pottery and tablets discovered there can be used as comparative material for the clay of the Ugaritic
tablets from Amarna.

The immediate surroundings of Ras Shamra are characterized by lower to middle Quaternary marine
sediments including shelly limestones, sandstones, conglomerates and calcareous tuffs. Further inland, at
a distance of about 1-2 km south and east of the site, one nds younger (upper Quaternary) clays, loams,
limestones and sands (Ponikarov 1964: Sheets I-36-XXIV; I-37-XIX, Geological map of Lataqia 1:50,000).
Several kilometers north of the site lies the ophiolitic complex of Bar-Bassit (Chapter 3.VI).
Artzy et al. (1976) report the results of a chemical comparison between Middle Bronze pottery
and clay tablets from Ugarit, which proved that they had been made of similar clays. If these results are
acceptable, the examination of Ugaritic ceramics and tablets can supply an immense body of comparative
data. Our ndings were rst compared to a small reference collection of thin-sections of Ugaritic pottery
which was prepared by L. Smith and J. Bourriau of the McDonald Institute of Cambridge University.
At a later stage we were able to conrm Artzys observation by the inspection of two texts from Ugarit,
presumably locally made, at the Louvre. In addition we produced a reference collection of thin sections
from about fty well-contexted typologically identiable Middle Bronze Late Bronze Age ceramic
vessels from Ras Shamra. These data, together with the detailed geological mapping of the area, supplied
all the necessary comparative information.
CATALOGUE
EA 45 (VAT 1692), from Ammishtamru to the King of Egypt
Reliability: High.
Matrix: Yellowish to bright tan in PPL, optically active groundmass with high calcitic component. Fine
silt (around 2%) includes predominant quartz with few zircon, mica and hornblende (?). Calcitic
foraminifers appear.
Inclusions: These include sand with minerals and rock fragments. No remains of vegetal material were traced.
The sand contains predominantly rounded grains (up to 500m) of radiolarian chert (radiolarite)
stained with ochre to dark reddish-brown limonite around the radiolaria spheres. Under the
88
stereomicroscope these inclusions appear as pink to deep jasper red bodies. Rounded to subangular
grains (up to 400m) of micritic limestone with the addition of rare clear calcite crystals are common.
In addition there are few idiomorphic clinopyroxene crystals (up to 400m) and a few subangular
quartz grains (up to 400m).
Palaeontology (LG): Badly preserved (recrystallized) small planctonic foraminifers, non-diagnostic.
Firing temperature: Probably around 700 0 C judging by partial colour change in the hornblende.
Geological interpretation: The petrography of this tablet indicates a source environment with a
combination of minerals of igneous origin and sedimentary rocks. However, the inclusion
assemblage that is generally rounded, thus clastic in nature, may have been dragged some distance.
Therefore, it may have originated from a drainage system that collected minerals and rock fragments
from a nearby environment.
The most characteristic component in the materials of this tablet is the radiolarian chert.
Radiolarites, which characterize the ophiolitic regions of north Syria and the Hatay province,
are almost absent from the Troodos ophiolitic complex in Cyprus (see EA 37 in Chapter 3.VI).
Radiolarites are not found in the Levant south of Lataqia either. In the eastern Mediterranean
zone radiolarites may therefore be related directly with the ophiolitic complexes of the Kizilda=
massif in the Hatay province, Turkey, and the Bar-Bassit massif of northwest Syria, near Ugarit
(Whitechurch et al. 1984:306-307). To the north of Lataqia, on the northern edge of Sahel elLataqia and towards the upper streams of Nahr el-Qandil, a layer of radiolarites is deposited on
top of the ophiolitic complex (Dubertret 1955:91-92). In the eld the radiolarites appear as jasper
red, rose or brighter pink due to their staining by ferrous minerals which can be observed also in
the thin sections (ibid.: Pl. XVI: Fig. 2). Alteration of radiolarites, mudstones, lavas and tuffs of
basic or intermediate composition, limestones and ints characterize this unit, which is dated to
the Upper Triassic Lower Jurassic ages (Ponikarov 1964: Sheets I-36-XXIV; I-37-XIX). This unit,
which is part of the Bar-Bassit massif, appears in numerous locations north of Lataqia through the
Bar-Bassit massif to the Hatay province. The largest exposures near Ugarit are found around Nahr
el-Qandil (about 15 km northeast of the site), and near Qasmin (about 8 km to its east). The origin
of this tablet should therefore be sought in the area stretching from the Syrian coast north of Lataqia
to the Iskenderun Bay. Consequently, both the petrographic and textual data point to the obvious
provenance of this tablet at Ugarit.
Reference: Similar radiolarites form the most common and distinctive feature of the inclusion assemblage
in the petrographic reference collection of the ceramics from Ugarit. In fact, most of the ceramics in
this collection are petrographically identical with the Amarna tablets assigned to Ugarit.
Conclusions: Combination of the textual and petrographic evidence clearly indicates that EA 45 was sent
from Ugarit.
Matrix: Identical in all details to EA 45.
Inclusions: Identical in all details to EA 45 containing radiolarite: (up to 300m), as in EA 45. Rounded
grains of micritic to sparitic limestone (up to 250m) are common. A few idiomorphic clinopyroxene
crystals (up to 250m) are present as are a few subangular quartz grains (up to 220m).
Firing temperature: As EA 45.
89
Conclusions: The text of this tablet is very fragmentary. Knudtzon attributed it to Ugarit on the basis of
its fabric and some verbs that are similar to EA 47. Petrographically and chemically this tablet was
made of exactly the same raw materials as EA 45. This letter was undoubtedly sent from Ugarit.
Reliability: High.
Matrix: Similar in all details to EA 45.
Inclusions: Very much like EA 45, but with some additions. It contains radiolarite (up to 200m), rounded
grains of micritic limestone (up to 350m), quartz (up to 120m), large grains (1.5mm) of rounded,
entirely serpentinized dolerite (?), where ghosts of the original crystals can still be observed
as pseudomorphs of serpentine after pyroxene (?) and a very large (3 mm) grain of beachrock
containing grains cemented by sparitic calcite. The grains include radiolarite, olivine, corallinean
alga, clinopyroxene, and serpentinized pyroxene. No remains of vegetal material were traced.
Firing temperature: Probably as EA 45.
Geological interpretation and reference: Though this tablet has not been analyzed chemically, the
petrographic data connect it with Ugarit. The matrix and inclusions are similar to those of EA 45,
but there are clearer indications for both a neighbouring ophiolitic environment and present-day
coastal vicinity.
EA 48 (VAT 1690), from the Queen of Ugarit to the Queen of Egypt
Reliability: High.
Matrix: Identical to EA 45-47.
Inclusions: Very much like EA 45, but with some additions. It contains radiolarite (up to 250m),
including free spheres of silicied radiolaria that appear as single inclusions; rounded grains of
micritic limestone (up to 150m); spherical quartz grains, including one polycrystalline grain,
apparently a fragment of an acid igneous rock (up to 350m); angular chalcedony (up to 250m);
idiomorphic clinopyroxene crystals (up to 250m) and a large (1.5 mm) grain of beachrock which
contains a large marine mollusc shell fragment cemented by sparitic calcite. No remains of vegetal
material were traced.
Palaeontology (LG): Globigerina (p), Globigerinoides (p), Globorotalia (p), Mollusca, Orbulina (p).
Age: Neogene.
CHEMICAL ANALYSES OF TWO UGARITIC TABLETS

Sampling the Ugaritic tablets both for petrographic and chemical analyses was very difcult due to their
delicate condition and extreme hardness. Therefore, only two tablets are represented in our elemental
analyses: EA 45, certainly a Ugaritic letter, and EA 46. Since the petrographic evidence indicates that
EA 45-49 form a homogeneous group, it may be assumed that the similarity between the two tablets
examined chemically may apply for the other two as well.
Table A1 in the Appendix presents the elemental composition of the two Ugaritic letters, as disclosed
by ICP AES/MS. Fig. 3.1 shows that EA 45 and EA 46 are similar and cluster very closely. Therefore
90
their chemical composition too points to a common source.
CONCLUSION: THE UGARITIC CORRESPONDENCE

Our study indicates that letters EA 45-48 are similar in their mineralogical composition, and that they
differ from other letters within the Amarna archive. Petrographically they suit the rather unique lithology
of the Ugarit area and are identical to the reference material from Ras Shamra.
II. NUASHE
Nuashe was a territory (rather than a city) in the middle Orontes Valley, between ama and Aleppo
(Klengel 1969:18-57; 1992:151-156; for a detailed discussion and map, see Astour 1969 and Pl. LI).
Some scholars suggested the city of Ukulzat (del Monte and Tischler 1978:451-452) as the capital of
Nuashe (Astour 1969:409; Helck 1971:176, 285). Astour (1977:57 n. 56) proposed identifying it at Tell
alawa, a large mound situated on the road leading from the north to Qatna and Qidshu, but Klengel
(1969:48) doubted this identification.
CATALOGUE
EA 51 (VAT 559), from Addu-nirari to the King of Egypt
Reliability: High.
Matrix: Carbonatic, light yellowish-tan in PPL, birefringent with speckled b-fabric and slight optical
orientation, containing varying amount of foraminifera and ostracoda. The matrix is speckled at
higher magnications (>X200). It is optically active and displays weak optical orientation. The
foraminifers are often lled with calcite or iron oxide. In many cases the microfauna is broken
into small fragments. Silt to ne sand size iron minerals that appear at a range of sizes from few
micrometers to about 80-90m are relatively common (~2%-3% of the matrix), angular in the ne
fraction and rounded in the coarser ones. These include opaque, through deep red and nearly opaque,
to reddish-brown minerals (presumably magnetite and haematite), the latter occasionally staining
the clay. Well-sorted, sparsely distributed (about 1%) ne silt quartz grains appear together with
lesser grains of iddingsite, plagioclase, rutile, biotite, pyroxene and hornblende.
Inclusions: Badly sorted sand, (f:c ratio{0.062mm}=~95:5) including minerals and rock fragments. The ner
fraction tends to be more angular and the coarser more rounded. Some prismatic crystals occur. Some
grains are coated by calcitic encrustation. Frequently occurring are rounded micritic, sparitic, and
sometime biogenic limestone (up to 750m) and subangular to rounded quartz (up to 400m). Angular
to sub-rounded replacement chert (up to 500m) is common with occasional haematite inclusions and
some grains coated by calcitic incrustation. There are a few up to 250m foraminiferous chalk grains
and a single well-rounded globule of glauconite which is yellow to orange in PPL. No remains of
vegetal material were traced.
Palaeontology (LG): Most fossils occur in the matrix but some were also observed within inclusions:
Acarinina (p), Cibicides (b), Echinoids, Morozovella (p), Planorotalites (p). Age: Paleogene.
Firing temperature: Probably 7000C judging by colour change in the glauconite and the beginning of
deformation in the clear calcite crystals.
Geological interpretation and reference: No analytical work on ceramic materials from the Nuashe
91
territories has ever been reported. Therefore our petrographic interpretations rely only on the
geological mapping of the area.
The clay type used in this tablet is extremely common amongst the Amarna letters from the
Levant and is also recorded in many ceramic assemblages from this area. It is commonly associated
with the Paleocene marls and shales, equivalent to the Taqiye formation in Israel, the marl member
of the Muwaqqar formation in Jordan (Sneh et al. 1998) and part of the recently suggested Chekka
formation of southwest Lebanon and the Beqa> (Walley 1997:103). Similar units of the same
geological age are lithologically consistent around the eastern Mediterranean, from Turkey through
the Levant to Egypt (the Esna Shales) and as far west as Morocco (Bentor 1966:73). In EA 51 this
marl is accompanied by silt containing some minerals of basaltic association, namely plagioclase,
iddingsite and pyroxene. The last two are relatively unstable and are easily subjected to physical
and chemical weathering. This means that where the clay was collected, it was enriched by aeolian
silt from an area neighbouring basaltic ows, but not from an area that directly supplies their clasts
into the local sediments through alluviation. Therefore, the basaltic terrain was not close to the
source area of this letter. The clay and the main set of inclusions reect Campanian or Lower Eocene
lithology (marl, chalk, chert, and glauconite), perhaps with some Cenomanian Turonian terrains
(limestone, geode quartz). The rounded nature of most of the inclusions and the coating that occurs
on some of them indicate that this sand originated in a body of water.
In the broad area which constituted the territory of Nuashe such combinations are quite common
(Ponikarov 1964: Sheets I-37-XIX; I-37-XX).
Conclusions: The results of the petrographic and chemical analyses accord with the general area
suggested for Nuashe between Aleppo and ama. However, the petrographic data is too general
to suggest a specic location for the capital city of Addu-nirari.
III. NII (?)

Nii was a city between Aleppo and ama presumably located east of the Ghab, in the area of the Roman
city of Apamea (Klengel 1969:58-74). One letter in the Amarna archive was identied as originating in
this area on the basis of its petrographic and textual evidence.
CATALOGUE
EA 67 (VAT 1591), from an unknown ruler north of Canaan
Reliability: High.
Matrix: Carbonatic, yellowish-tan in PPL, birefringent groundmass with speckled b-fabric containing
sparsely spread (~3%), well sorted calcitic bodies and less common calcite crystals between 10m
and 50m in size. The calcite crystals are commonly idiomorphic or subidiomorphic in the ner
fraction but tend to become rounded sparite in the coarser. Foraminifers appear (about 2%). Quartz
silt comprises about 2%. Occasional heavy minerals appear in the silt fraction and include augite,
zircon and twinned plagioclase. Opaque (angular) to reddish-tan (spherical and rounded) iron
mineral grains (magnetite and haematite ~2%) appear in a range of sizes from a few micrometers to
about 30-40m.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) made up of rounded micritic (and more rarely sparitic)
92
limestone (up to 500m) which is frequent. Rounded to subrounded quartz (up to 400m) is
common. Common, rounded opaque or nearly opaque spheres (as in EA 100 but lacking the skeletal
plagioclase), up to 250m. Subrounded to rounded, clear to partially serpentinized titanaugite,
sometimes containing mineral inclusions or twinned (up to 400m) is also common. In addition
a few grains of rounded foraminiferous chalk (up to 760m), subangular twinned plagioclase (up
to 300m) and very few subrounded hornblende grains (up to 300m) can be seen. A few rounded
coprolites measuring up to 350m, with no spherulites or druzes.
Firing temperature: Most likely unred judging by the preservation of the coprolites.
Geological interpretation: The matrix of this tablet is similar to that of EA 100 (Chapter 7), the
palaeontological examination of which most likely indicates Neogene marl. The presence of fresh,
physically unstable igneous minerals in large grain sizes (augite, hornblende) points to a nearby
outcrop of volcanic or basic intrusive rocks.
Reference: As EA 169.
Conclusions: EA 67 was written after the conquest of umur, possibly by Aziru of Amurru (cf. Liverani
1998:185). Its rst part is broken and the name of the writer and his city are missing. On the one
hand the letter lacks the typical horizontal lines that separate the text into passages, which are
characteristic of letters of northern origin. On the other hand, Moran (1992:137 n. 2) examined
the language and writing of the tablet and suggested that they have a northern cast. Moreover,
Liverani (1998:185 n. 75) suggested that the Hurrian term alzulu (line 15) is a synonym for
Akkadian azannu (mayor), the latter being used by all Canaanite rulers (see the literature cited in
Moran 1992:100 n. 3). Finally, we suggest rendering lines 10-12: All the inhabitants of [my?] land
(KURK[I-ia?]) who had resided in umur came out and are residing in my land. In the light of
these considerations the seat of the writer must be sought in the former Mitannian territories, north
or northeast of umur.
The petrographic data point to the use of Neogene marl and inclusions containing both
sedimentary rock fragments and fresh minerals derived from volcanic rocks (augite, plagioclase).
The best parallels in the Amarna archive are EA 100 and the Amurru letters which, in our opinion,
were sent from Irqata (Chapter 7). However, the historical evidence makes this interpretation
impossible. After the conquest of umur by Aziru, Irqata was under Amurrus control and could
not have been the seat of the writer of EA 67. As noted above, his city must be sought in the former
Mitannian territories, north or northeast of umur. Therefore another locality, outside the borders of
the Egyptian domination in Canaan, must be sought. The geological map of Syria reveals that the
only province in the relevant areas where Neogene marls and basalt ows appear together is the area
east of the Ghab, identied as the land of Nii (Klengel 1969:58-74). We therefore suggest that EA
67 was a letter of the king of Nii.
Locating the writer in the city of Nii (Qal>at el-Mudiq near the Roman city of Apamea) ts the
text of EA 67. umur was probably an important commercial centre at the terminus of the main road
that led from the Middle Orontes to the Mediterranean. The king of Nii reported to the Pharaoh that
inhabitants of his kingdom, who formerly were engaged in commerce in the city, left it after it was
conquered by Aziru of Amurru (EA 67:10-12).
93
CHAPTER 6
SYRIAN KINGDOMS IN THE MIDDLE ORONTES AREA
This chapter deals with Syrian kingdoms which were located outside the connes of the Egyptian Empire in
Asia. When Egypt and Mitanni concluded a peace treaty in the late 15th century BCE, the border between
their vassal states passed between the Beqa> of Lebanon (>Amqi) and the kingdom of Qidshu (Tahshi). With
the collapse of Mitanni and the conquest of its territories by Shuppiluliuma I of atti, confrontations along
the border zones of the two empires broke out once again. The former vassals of Mitanni sought military aid
from Egypt, which remained the sole empire that could withstand the Hittite military offensive. This is the
reason for the presence of letters from Nuashe, Qatna, Tunip and Qidshu in the Amarna archive.
Klengel (1965, 1969, 1970) has systematically collected and analyzed all the documentary evidence
on north and central Syria in the second millennium BCE (see also Helck 1971). In discussing the
identication of Middle Orontes Syrian kingdoms mentioned in the Amarna letters we refer to his
geographical-historical conclusions (Klengel 1970; 1992). For the identication of toponyms mentioned
in the Hittite documents we used the works of del Monte and Tischler (1978, 1992; an updated
literature and short discussions of some toponyms have recently been published by Myers 1997). While
archaeological research in this area is in an initial phase, two of the sites mentioned in the Amarna letters
Qidshu (Tell Nebi Mend) and Qatna (Tell el-Mishrife) have been excavated. No petrographic study
of the pottery of these sites has been published to date.
I. QATNA
Qatna is identied at Tell el-Mishrife, between oms and ama, north of present-day Qatana (Klengel
1969:96-138; 1992:156-157; for recent excavations see Novak and Pflzner 2000).

Tell el-Mishrife is situated in an area combining Upper Eocene to Miocene chalk and chert, marls, and
conglomerate. To its west, north of Buheiret Qattineh, lie the middle series of the Pleistocene basalt
sequence. These include olivine basalts. Closer to the site extends the alluvial Quaternary ood plain of
the Orontes basin (Ponikarov 1964: Sheets I-36-XVIII; I-37-XIII).
CATALOGUE1
EA 55 (BM 29819), from Akizzi to an Egyptian ofcial
Reliability: High.
Matrix: Bright tan in PPL, optically active with speckled b-fabric and very weak optical orientation,
with high calcitic component. Foraminifers are spread throughout the matrix, usually inlled with
calcite but occasionally silicied or stained by iron oxides. Sparsely spread silt (~2%) includes
1. The tablets are presented here in order of petrographic importance and not according to their EA numbers.
94
predominantly quartz, with the addition of few twinned feldspars, iddingsite and hornblende. Iron
minerals are common in various grain sizes within the silt fraction, either as angular opaques or deep
reddish-brown rounded bodies.
Inclusions: Rather dense (f:c ratio{0.062mm}=90:10) sand made up of frequent rounded micritic limestone (up
to 280m). Rounded to subangular (commonly angular in the smaller fraction) quartz (up to 450m)
is common as is replacement chert (up to 200m) usually with iron mineral inclusions. Fragments of
fresh (unfossilized) aquatic mollusc shell fragments (up to 400m) are also common.
Vegetal material (SLY): Common, up to 450m in length. Fibre and tissue fragments with no specic
character, uncharred.
Firing temperature: Probably unred or very lightly red judging by the uncharred state of preservation
of the vegetal material.
Geological interpretation and reference: The lack of reference material from Tell el-Mishrife limits the
comparison to the geological mapping of the surrounding area. The petrography depicts a source
area where a sedimentary rock suite is exposed, with more distant volcanic occurrences. The
sedimentary sequence includes limestone, chert, sandstone, marl and deposits from a recent body of
water. These data accord with the geology around Tell el-Mishrife.
Conclusions: EA 55 is used here as reference for the Akizzi letters, both on the basis of its clear reading
and the quality and size of the sample taken.
EA 53 (BM 29820), from Akizzi to the King of Egypt
Sampling method: SPA/Peeling.
Matrix: Bright tan in PPL, optically active with speckled b-fabric and weak optical orientation, with high
calcitic component. Foraminifers are uncommon. Sparsely spread silt (~2%) includes predominantly
quartz, with the addition of few twinned feldspars, iddingsite, and hornblende. Iron minerals are
common in various grain sizes within the silt fraction, either as angular opaques or deep reddishbrown rounded bodies.
Inclusions: Rather dense (f:c ratio{0.062mm}=90:10) sand containing frequent grains of rounded micritic
limestone (up to 430m). Rounded to subangular (up to 250m) quartz (commonly angular in
the smaller fraction) is common as are replacement chert (up to 280m), often with iron mineral
inclusions, and fragments of fresh (unfossilized) aquatic mollusc shells (up to 370m).
Vegetal material (SLY): A few (up to 250m long) tissue fragments with no specic character, uncharred.
Geological interpretation and conclusions: Identical to EA 55.
EA 54 (VAT 1868 + 1869 + 1721), from Akizzi to the King of Egypt
Matrix: Bright tan in PPL, optically active with speckled b-fabric and very weak optical orientation, with
high calcitic component. Frequent foraminifers are spread throughout the matrix, usually inlled
with calcite but occasionally silicied or stained by iron oxides. Sparsely spread silt (~2%) includes
predominantly quartz, with the addition of few twinned feldspars, iddingsite, and hornblende. Iron
minerals are common in various grain sizes within the silt fraction, either as angular opaques or deep
reddish-brown rounded bodies.
95
Inclusions: Rather dense (f:c ratio{0.062mm}=95:5) sand comprising frequent grains of rounded micritic
limestone (up to 900m). Rounded to subangular (commonly angular in the smaller fraction) quartz
(up to 350m) are common. There are a few grains of replacement or radiolarian chert (commonly
with iron mineral inclusions) (up to 150m) and a few rounded fragments (up to 450m) of volcanic
glass or tuff, sometimes almost entirely weathered to clay. There are also very few fragments of
aquatic mollusc shells (up to 350m).
Vegetal material (SLY): Few, up to 500m in length. A tissue fragment with no specic characters.
Geological interpretation and reference: Identical to EA 55 (with the addition of highly weathered
volcanic rock fragments).
Conclusions: The tablet is badly broken, but on the basis of the introductory lines and the comparison of
the remaining text to other letters it was safely assigned to the correspondence of Akizzi of Qatna.
Our petrographic data conrms this conclusion, since EA 54 is identical to the securely identied
letters from Akizzi (EA 53, 55).
EA 57 (VAT 1738), from an unknown king (in the area of Qatna?) to the King of Egypt
Reliability: High.
Matrix: Yellowish-tan in PPL, optically active with speckled b-fabric, brous in higher magnications
with high calcitic component. The marl matrix contains no foraminifers. Silt to ne sand size iron
minerals that appear in a range of sizes from few micrometers to about 80-90m are relatively
common (~2% of the matrix), angular at the ne fraction and rounded at the coarser. Silt of other
minerals (~3%) includes predominantly quartz with the addition of accessory plagioclase, pyroxene,
hornblende, olivine, iddingsite, epidote and zircon.
Inclusions: Sand (f:c ratio{0.062mm}=97:3) made up of frequent rounded micritic limestone grains (up to
400m), Rounded to subangular (commonly angular in the smaller fraction) quartz (up to 350m) is
common, as is replacement or radiolarian chert (up to 450m) commonly with mineral inclusions.
Also seen are a few rounded fragments of very weathered nely crystalline basalt, volcanic glass,
and serpentine (up to 450m) and a few fragments of aquatic mollusc shells (up to 400m).
Vegetal material (SLY): Few, up to 500m in length, including a fragment of a plant tissue including
several primary vessel members. No further identication is possible.
Geological interpretation and reference: Petrographically, EA 57 is similar to EA 53-55 and should be
assigned to Qatna.
Conclusions: Since Akizzi, King of Qatna is mentioned in this letter in the 3rd person (line 2), the letter
may have been sent from one of Qatnas neighbouring kingdoms (for discussion, see Klengel 1963:45
n. 3, 52). Petrographically it may be readily assigned to the Qatna assemblage of EA 53-55. Therefore,
it must have been written in a neighbouring area to the north or south of Qatna. In the light of the
petrographic results, there is also some possibility that EA 57 was sent from Qatna under circumstances
which are not clear to us, e.g. Akizzi is referred to by a later king of the city.
96
II. QIDSHU
Qidshu on the Orontes is unanimously identied with Tell Nebi Mend (Klengel 1969:139-177, 1992:157160; for a short summary of the nds at the site see Parr 1997).

Tell Nebi Mend is located in an area that borders on two lithological units. The rst, extending from the
immediate surrounding of the site southwards, is characterized by Pliocene continental conglomerates,
sandstones, limestones, clays, marls, marine clays and tuff-breccia. Sediment deposits of the Orontes
oodplain typify the immediate surrounding of the site. The second, extending to the north of Buheiret
Qattineh, is characterized by the middle series of the Pleistocene basalt sequence, which includes alkaliolivine basalts (Ponikarov 1964: Sheets I-36-XVIII; I-37-XIII).
For lack of any published report on ceramic resources and ceramic technology in the area in question,
we turned to a collection of selected Late Bronze vessels from Tell Nebi Mend published by Bourke
(1992). Thin sections of some of these vessels were made by D. Grifths of the Institute of Archaeology,
University College, London. Of these, we examined especially simple wares, namely undecorated plain
vessels, such as bowls and kraters, which are more likely to represent local raw materials.
CATALOGUE
EA 189 (VAT 336), from Etakkama to the King of Egypt
Reliability: Satisfactory to High.
Matrix: Yellowish to bright tan in PPL, optically active with speckled b-fabric and weak optical
orientation, with high calcitic component. Infrequent foraminifers are sparsely spread throughout
the matrix, including fragments of ostracoda. Silt-sized iron minerals are spread in the matrix (~1%).
These include opaque, through deep red and nearly opaque, to reddish-brown minerals (presumably
magnetite and haematite), the latter occasionally staining the clay. Well-sorted, sparsely distributed
(about 0.5%) ne silt quartz grains appear together with fewer grains of iddingsite and rare epidote,
hornblende, twinned plagioclase.
Inclusions: Voids left after organic matter and sparsely spread particles. The common inclusions include
usually rounded micritic limestone and chalk (up to 250m) and a few clear calcite crystals as well
as rounded and spherical grains of quartz (up to 550m). There are also a very few fresh aquatic
mollusc shell (aragonitic) fragments ( up to 150m).
Vegetal material (SLY): Frequent, elongated voids with sharp boundaries (ratio within matrix: 3%),
sizing up to 900m in length, indicate vanished vegetal material. No remains of actual vegetal
material were traced.
Firing temperature: Perhaps lightly red at 500-6000C due the vanishing of the vegetal materials, but
not reaching 7000C.
Geological interpretation and reference: From a petrographic viewpoint this tablet supplies rather
poor indications as to its origin since it contains mostly artefactual inclusions (chopped vegetal
materials). Using the scant comparative data that could be obtained from the thin sections of Late
Bronze pottery from Tell Nebi Mend, we can cautiously state that EA 189 accords with the main
97
group of samples in this assemblage. The marly matrix (probably of Paleocene age) in the silt, which
includes occasional minerals of basaltic origin (plagioclase, iddingsite), is common in the pottery
too. Spherical quartz plays a major role in the inclusions, with the addition of chert. Geologically,
all the above features reect an area with basically sedimentary features (marl, sandstone, limestone
and chalk) with remote basaltic terrain that supplies some of its derived minerals as silt.
Conclusions: The petrographic data from EA 189 accords with the geology in the vicinity of Tell
Nebi Mend.
III. LABANA
Several scholars identied Labana with the site of Lebwe (Lebo-Hamath) in the Beqa> Valley (e.g.
Weber in Knudtzon 1915:1111; Aharoni 1967:66, 137, 147; Moran 1992: 390; Liverani 1998:287).
However, Labana and Labu/Lebwe are two different places since they are mentioned side by side in
the topographical list of Thutmose III (Nos. 10 and 82; Edel 1953:153-154; Kuschke 1954:103; Helck
1971:130). Late Bronze Lebwe was a small village (for the survey of the site see Marfoe 1995) and
could not have been the centre of a city-state. Labana, on the other hand, was a city-state (EA 193) and
is mentioned in the topographical list of Thutmose III (No. 10) and in two Amarna letters (EA 53:35, 57;
54:27, 32; Klengel 1970:60-61).
Tiwate/Teuwatti, Labanas ruler, was an ally of Arzawiya of Ruizzi and they cooperated with the
ruler of Qidshu in attacking the land of Upi (EA 53:36, 56; 54:26, 31; 191:2; 192:4; 197:26, 33; Klengel
1970:78-79). Ruizzi is mentioned in the topographical lists of Thutmose III (No. 79) and Amenophis III
(BN rev. 2) and in several Amarna letters (EA 53:36, 56; 191:2). These topographical lists do not include
toponyms north of Qidshu. The list of Thutmose III includes the three cities of Ruizzi (79), Hermel
(81=Hermel in the northern Beqa>) and Labu (82=Lebwe, also in the Beqa>). Hence Ruizzi might have
been located near the southern border of the kingdom of Qidshu. The city of Ruizzi (written Raium)
is also mentioned in a Mari letter and must be sought between Qatna and Canaan (Rainey 1979).
The exact location of Labana and Ruizzi cannot be established with certainty. However, in view
of the textual evidence, it is reasonable to assume that it was situated on the northern border of Upi
(Damascus area), east of the Beqa> and south of Tashi (the Qidshu area).
CATALOGUE
EA 193 (VAT 1608), from Teuwatti to the King of Egypt
Matrix: Tan in PPL, optically active with speckled b-fabric, very weak optical orientation, with high
calcitic component. Frequent foraminifers are spread throughout the matrix, usually inlled with
calcite but occasionally siliceous. Sparsely spread silt (~2%) includes quartz, with rather common
minerals of basaltic origin similar to those that appear in the inclusions (twinned feldspars, clear
orange-brown to red iddingsite, hornblende). Iron minerals are common in various grain sizes within
the silt fraction, either as angular opaques or deep reddish-brown, nearly opaque bodies.
Inclusions: Rather dense (f:c ratio{0.062mm}=85:15) sand comprising basalt and its derived minerals.
These grains are frequent, up to 1 mm, rounded fragments of nearly doleritic olivine basalt. The
98
olivine is almost always entirely altered into iddingsite ranging between deep orange to ruby
red. The augite is sometimes serpentinized. Small polycrystalline grains (e.g. iddingsite and
augite, etc) appear in various sizes. More common are single crystals of derived minerals, of
which iddingsite is the most common. In these grains, it tends to split along the {010} cleavage
plain. This feature is quite unique to this sample within the Amarna tablets which we examined
and may have some significance for distinguishing the type of basalt concerned. Rounded chalk
grains (up to 800m) are common and there are also a few rounded quartz grains (up to 200m).
Hornblende is present (up to 80m) but very rare.
Vegetal material (SLY): Few, up to 700m lengthwise. Several fragments with the typical vessel
members of the primary xylem with spiral secondary cell wall thickening. No further interpretations
can be made.
Palaeontology (LG): A few small non-diagnostic planctonic foraminifers. Identied: Heterohelix (p).
Age undetermined, perhaps Upper Cretaceous.
Firing temperature: Probably unred or very lightly red judging by the complete preservation of the
organic matter and the fact that there is no inuence of heating on any of the minerals.
Geological interpretation and discussion: The pronounced basaltic component within this tablet indicates
an area immediately adjacent to basaltic ows. The alkali-olivine composition with the typical
alteration into iddingsite and the holocrystalline structure point to young (Pliocene to Pleistocene)
rather than older (Cretaceous) basalt. The coarsely crystalline basalt that is nearly doleritic is typical
of the Lower Basalts of Pliocene age. This should be located near outcrops of sedimentary rocks
including limestone and marl (as evident by the matrix and the other inclusions).
In the light of these results, the equation of Labana = Lebwe should be rejected. In the southern
and central Beqa> Valley in general, and near Lebwe in particular, there are no basaltic exposures.
The nearest volcanic occurrences in the neighbouring area of the Beqa> are the Lower Cretaceous
volcanics that include tuffs and alkali basalts. These are exposed on the topmost areas of the
Lebanon Mountains in locations too remote from Lebwe to make them a possible source. Moreover,
these basalts differ drastically in their mineralogy and structure from the younger types, such as
the ones that typify the inclusions of EA 193 having a typical hypocrystalline structure with typical
alteration of the olivine into bowlingite and chlorite.
The closest exposures of younger basalts appear further north, in the northern Beqa> and
more signicantly in the Middle Orontes basin north of the Qidshu lake (Buheiret Qattineh, see
Ponikarov 1964: Sheets I-36-XXIII; I-37-XIII; Bartov 1994). This is the southernmost area where
large exposures of young basalts of Upper Pliocene age are exposed. Two small outcrops of such
basalt appear near Hermel and at Fadil (west of Quseir) in the northern Beqa> Valley. Hermel may
be eliminated since it is mentioned under the name Hermel and Harnam/Arnam in several Egyptian
topographical lists, in a Ramesses II inscription and in Papyrus Anastasi I (Helck 1971:132, 200203, 317; Kitchen 1982:52-56). Tentatively we suggest locating Labana (and Ruizzi) east of the
Orontes, possibly in the area of Quseir or south/southeast of it (for Late Bronze sites in that area see
Philip et al. 2002).1
Naaman (1999:421-425) proposed locating Iron Age Zobah, the capital city of Aram-Zobah, at Quseir. Either Labana or
Ruizzi of the Late Bronze Age could have been located at the same site. Hence the northern part of the land of Upi, where
Ruizzi and Labana may possibly be sought, covers part of the rst millennium BCE territory of Aram-Zobah.
99
IV. RUIZZI
For the possible location of Ruizzi see the identication of Labana above.

Considering the proposed proximity between Labana and Ruizzi, the general geology of the Labana area is
surveyed above. No petrographic references have ever been published from this area.
CATALOGUE
EA 192 (VAT 1674), from Arsawuya to the King of Egypt
Reliability: Moderate. Sampled again to supply a highly reliable sample.
Matrix: Tan to dark greyish-tan in PPL, optically active with speckled b-fabric, very weak optical
orientation, with high calcitic component. Foraminifers occasionally appear in the matrix. Sparsely
spread silt (~2%) includes quartz, with rather common minerals of basaltic origin similar to those
that appear in the inclusions (twinned feldspars, iddingsite). Iron minerals are common in various
grain sizes within the silt fraction, either as angular opaques or deep reddish-brown, nearly opaque
bodies.
Inclusions: Sand of subangular nely-crystalline olivine basalt and its derived minerals (up to 900m).
The olivine is entirely altered into iddingsite ranging between deep orange to ruby red. The augite
is sometimes serpentinized. Iddingsite and augite appear as single grains of various sizes. The
former is clear and deep orange-red in colour. The augite (appearing as one grain in this scant
sample) is 200m in size .There is rounded chalk (up to 200m) and subrounded quartz (up to
100m) in this sample.
Vegetal material (SLY): Up to 400m lengthwise in this sample. Several fragments with unclear
characteristics.
Firing temperature: Probably unred or very lightly red judging by the complete preservation of the
organic matter and no inuence of heating on any of the minerals.
Geological interpretation and conclusions: The neighbourhood of Labana and Ruizzi as suggested by
scholars on the basis of the documentary evidence is conrmed by the petrographic analyses of
EA 192 and 193. EA 192 does not supply additional clues as to the exact location of Ruizzi but it
should nevertheless be sought within the same boundaries suggested for Labana.
100
CHAPTER 7
THE KINGDOM OF AMURRU AND NEIGHBOURING POLITIES
This chapter discusses a group of northern polities connected with the Kingdom of Amurru: the letters of
the kings of Amurru, letters from Irqata which were not sent by Aziru, a letter from the citizens of Tunip
and a letter from an unspecied kingdom in the neighbourhood of Amurru.
The early history of the Kingdom of Amurru has been examined by many scholars ever since the
discovery of the Amarna letters (for detailed summaries see Klengel 1969:178-299; Izreel and Singer
1990; Singer 1991:135-195). Amurru was initially a small highland kingdom situated in the mountainous
regions on the western slopes of Mount Lebanon and along Nahr el-Kebir (Eleutheros of the Greek
sources), that gradually expanded during the Amarna period. In its heyday it covered the territory between
Tripoli on the Lebanese coast and the Middle Orontes area of western Syria. Amurru rst emerged under
Abdi-Ashirta, who was able to expand his territory and conquer cities in his neighbourhood. After his
death it was lead by his son and successor Aziru who continued his fathers offensive and expanded the
territory of his realm to the Orontes basin.
Several problems related to the Amurru le in the Amarna archive, such as the order of events
depicted by the Amarna letters, have been resolved in scholarly research. Other issues, such as the location
of the capital of Amurrus rulers at various stages of their career and the sequence of events in the time
of Aziru are still debated. A major question is the identication of the city of Tunip mentioned in four of
Azirus letters (EA 161, 165-167) and in a letter sent by its citizens to the Pharaoh (EA 59). Research on
these problems has reached a deadlock, which may be broken by the petrographic analysis of the tablets.

The arena where the Amurru correspondence takes place may be enclosed by a schematic rectangle
that lies between the TripoliTartous line in the west, and the Qidshu (Tell Nebi Mend)Aleppo line
in the east. This area may be divided into four broad geographic and geological units, each with a few
subdivisions:
1. The coast: from Tripoli to Nahr el->Arqa; the >Akkar plain from Nahr el->Arqa to Simerian; and
from the northern limits of the >Akkar up to Tartous.
2. The western mountainous range: the northern part of the Lebanon Mountains; the Tell Kalakh
volcanic province; Jebel Anariyeh and the Shin volcanic plateau.
3. The inner Valleys: the northern Beqa> and the oms Lake up to oms; the Middle Orontes Valley
and the Ghab.
4. The eastern mountainous range: the Salamiyeh Plateau and Jebel Zawiye. Different geological
environments that are expected to yield different petrofabrics characterize each one of these
geographical units.
Due to the present formative state of archaeological research in these areas, our sources of
information were rather more fragmentary than usual, especially where comparative petrographic data
are concerned. Contrary to the case for sites located along the Euphrates, no petrographic work has
been published so far on sites such as Tell Kazel, ama, Tell >Arqa, and Qatna. Preliminary work has
101
Fig. 7.1: The Land of Amurru showing sites mentioned in the text.
102
been done on Tell >Asharneh and we were able to examine petrographic collections of ceramics from
other sites in the region. When combined with the geological data, these collections supplied reasonable
documentation that enabled us to draw important conclusions.
The rst source of data that can be used is the geological map of the area, drawn rst by Dubertret
(1949a, 1951a, 1951b) and later supplemented by others (Ponikarov 1964: Sheets I-36-XXIV; I-37-XIX;
Ponikarov et al. 1966; Kozlov et al. 1966; Shatsky et al. 1966; Sanlaville 1977; Sanlaville et al. 1993).
Broadly speaking it enables distinguishing between various geographical zones within this wide area:
the Orontes basin and the northern Beqa>, Mount Lebanon, the Jebel Ansariyeh ridge, and the Coastal
Plain. Within each area the lithological landscape is varied enough to enable a higher resolution of
differentiation between smaller units. Thus, the geological mapping supplied the basic information that
can be correlated with the petrographic data obtained from the tablets.
Furthermore, in certain cases the geological literature supplies detailed information that may be used
for even more precise identication of places mentioned in the Amurru correspondence. This applies,
rst and foremost, to the coastal plain, and especially to the >Akkar plain and the mouth of Nahr el-Kebir
(Sanlaville 1977; Kozlov et al. 1966). The micropalaeontological identications, combined with very
detailed geological literature concerning the Paleogene (Luterbacher 1986; Krasheninnikov et al. 1996)
and the Neogene (Dubertret 1945, Buchbinder 1975), supplied enough information to set the clay types
used by the scribes in their chronological-stratigraphic context.
I. AMURRU
CATALOGUE
EA 60 (VAT 343), from Abdi-Ashirta to the King of Egypt
Matrix: The matrix is clayey, yellow to yellowish-orange in PPL and optically active with speckled bfabric and strong optical orientation. It is mottled with common bodies of clay in various colours
sizing up to 250m, seemingly the alteration products of various minerals, constituting of nearly 5%
of the matrix. These include yellow through orange to dark red clay, frequently optically active, tuff
and iddingsite. Opaque minerals, usually angular at the ner fraction and subrounded to spherical at
the coarser, are also widespread forming about 3%-4% of the matrix, sizing up to 60-70m. Some of
the larger opaque or nearly opaque particles are oolitic. Silt, essentially of quartz but with accessory
plagioclase (sometimes twinned), forms ~2%-3% of the matrix. Dark reddish-tan, ferruginous
shales that are frequently microlaminated and silty (~2%-3%) appear as massive bodies reaching
millimeter size. Other shale fragments are of yellowish clay, with speckled b-fabric.
Inclusions: Basaltic minerals are dominant and include rounded globules (up to 100m) of glassy phases,
yellow to orange in PPL, brous with undulose extinction. These are most likely serpentinized
minerals. There are a few clinopyroxene crystals and occasional iddingsite particles. A few fragments
of nely crystalline alkali basalt (up to 350m) of trachytic texture with elongated and oriented
plagioclase laths also occur. The pyroxene is partly or entirely serpentinized.
Vegetal material (SLY): A fragment of plant tissue with elongated cells resembling a piece of conifer
wood or a band of bres.
Firing temperature: There are no indicators to determine the ring temperature of this sample.
103
Geological interpretation: This tablet can be linked by its characteristic matrix and inclusions to a wellrecorded petrographic group that is known from many vessels in the Levantine ceramic repertoire. It
is typied by diverse ferruginous to argillaceous shale fragments and by ferruginous ooliths. In many
instances quartz sand may be present in this group, usually as sub-spherical grains. Other attributes
are aggregates of spherical quartz grains cemented by a carbonate and/or iron oxide matrix. There
are also siltstone, oolitic limestone, spheroids of iron oxide (sometimes with an internal concentric
structure) and aggregates of such spheroids embedded in micritic limestone. In the case of EA 60
some of these are probably not represented due to the size of the sample. Other indicators of this
group are tuff and weathered basalt fragments, and typical rhombohedral limonitic pseudomorphs
after dolomite. The clay matrix is typically dark red to nearly opaque, or as in this case, yellowish
and remarkably optically oriented under crossed polarized light. The last feature usually indicates
kaolinite-rich non-carbonatic clay.
All these features are characteristic of Lower Cretaceous clay and shales (Porat 1989a:64,
71-72; Greenberg and Porat 1996). The presence of basalt and/or tuff in the inclusions suggests
a nearby exposure of the basal Lower Cretaceous volcanics, whereas the presence of diversied
shales, siltstones, quartzitic sands and ferruginous ooliths points to the use of shales from the
Lower Cretaceous sandstone units. Trachytic textures and an alteration of the olivine into chlorite
characterize the basalts of the Lower Cretaceous section (Mimran 1972; Amiran and Porat 1984) as
opposed to the Miocene-Pleistocene basalts (cf. EA 169).
The lower formations of the Levantine Lower Cretaceous lithological section outcrop widely in
Mount Lebanon, along the slopes of Mount Hermon and less frequently in the Anti-Lebanon. Many of
the attributes of this group are unique to these formations. Geologically, they are included in the Hatira
formation in Israel, or the Kurnub Group in Israel and Jordan. In Lebanon the terms Grs de Base or
C1 have been used since the fundamental mapping by Dubertret (1949a), but recently the term Chouf
Sandstone formation was formalized (Walley 1997) to describe the entire Lower Cretaceous basal
unit including the sandstone series. The ferruginous oolites are characteristic of the Aptian deposits
of Israel, Lebanon and Syria. They contain about 80% goethite and a shaly, limy or clayey matrix
(Rohrlich and Metzer 1980).
The presence of basalt and pyroclastic material among the inclusions may be related to the
proximity of the clay source to an exposure of the Lower Cretaceous volcanic complex (termed
the Tayasir volcanics in Israel or Basalte Crtac in Lebanon). These layers are widely exposed
in Mount Lebanon. The distribution of the Aptian formations in Lebanon is limited to a belt
that extends between Mount Hermon in the south and the >Akkar Plain in the north, covering the
upper parts of Mount Lebanon and their slopes (Dubertret 1962). A narrow band of this formation
outcrops along this ridge, from Merj >Ayyun northeastwards. The largest exposure appears in the
area between Zale in the Beqa> and Aaley on the western slopes of Mount Lebanon. In the AntiLebanon, a strip exposes between Rashiya el-Fukhar and the Zebedani. However, the volcanics
appear in signicant exposures only in Mount Lebanon, north of the Beirut-Zale line.
Consequently, this tablet is more likely to have originated in the mountainous area rather than in
the lower plains to the east and west. More specically, its source may be sought in the area south
of the >Akkar Plain, but not in or around it. Lower Cretaceous exposures, including outcrops of the
Basalte Crtac, occur along the slopes of Jebel Neghas, about 10 km southeast of Arde (Dubertret
1949a). Thus, the mountainous area east of Tripoli from which the clay of this tablet came must have
been the core area of the Kingdom of Amurru (see Singer 1991:141-148).
104
Reference: This petrographic group has been described in detail by Greenberg and Porat (1996) and
Goren (1992, 1995, 1996a) with reference to pottery from Israeli sites. There is no mention of
it in reports dealing with the petrography of Syro-Lebanese pottery, although we observed it in
abundance in several thin section collections from this area. This clay is usually considered to be of
superior quality for pottery production since its high iron content made it possible to achieve a high
degree of sintering at lower ring temperatures due to the uxing properties of the iron. Therefore,
Lower Cretaceous ferruginous shales were often used to produce vessels that required extra strength,
especially liquid containers such as pithoi, jugs and storage jars. At the Chalcolithic site of Teleilat
Ghassul, for example, most of the locally-made pottery is formed of this iron-rich clay (Goren 1987:
48-53; 1991a: Appendix 2), the typical pithoi being sintered to a surprisingly high quality (Edwards
and Segnit 1984). Similar raw materials were used during the Early Bronze Age II-III in order
to produce the high quality Metallic Ware as well as other superior vessels (Porat 1989a:71-74;
Greenberg and Porat 1996). Recent examination of Iron Age I collared-rim and Galilean pithoi
from several sites in the Samaria and Galilee regions (Glass et al. 1993; Cohen-Weinberger and
Goren 1996) demonstrated that in several cases such clays were preferred by the potters in order to
produce high quality vessels. However, there is no obvious advantage in producing tablets from this
material. Hence EA 60 must reect the clay type that existed close to the senders location.
Conclusions: EA 60 was sent from the mountainous areas east of Tripoli.
EA 61 (Ash. 1893.1-41: 410), from Abdi-Ashirta
Reliability: High.
Matrix: Carbonatic, light yellowish-tan in PPL, containing infrequent badly preserved foraminifers and,
more commonly, their fragments. It is rather silty (about 2%) and very rich in opaque iron minerals
that appear in a range of sizes from a few micrometers to about 30-40m. Under higher magnications
(>X200) the matrix is brous, optically active and displays weak optical orientation.
Inclusions: Apart from the vegetal material, the inclusions appear to be naturally detrital within the
reworked clay of the matrix and not intentionally mixed in by the artisan. Rounded grains of micritic
and sparitic limestone (up to 250m) and single grains of pure calcite are frequent to dominant. Also
present are a very few subangular grains of quartz (up to 100m).
Vegetal material (SLY): Up to 750m, frequent. Fragments of plant tissue with large and small
parenchyma (soft tissue) cells.
Palaeontology (LG): Abundant fossils in matrix: Bryozoa, Bulimina (b), Catapsydrax (p), Globigerina
(p), Globigerinoides (p), Globorotalia (p), Orbulina (p). Age: Neogene.
Firing temperature: This tablet was red at very low temperature, if at all, as evidenced by the uncharred
state of the vegetal material.
Geological interpretation: The matrix of this tablet indicates Neogene marl by virtue of its petrofabric and
palaeontology. Such marls, dating to the Miocene or the Pliocene, do not appear in the Levant south
of the Lebanese coast. In Lebanon their outcrops are restricted mainly to exposures east and south of
Tripoli (Dubertret 1951b). Notable among them is the outcrop of Nahr el-Awdeh, including the site
of Tell Arde (ancient Ardata), which is the only signicant mound near any of these exposures that is
found in an entirely sedimentary area (as opposed to Irqata, below). Tell Arde is a large site of ca. 35
dunam (Salam-Sarkis 1972; 1973; Izreel and Singer 1990:119-120) located 8 km from the coast and
commanding the area east of Tripoli.
105
Reference: No reference materials from Arde were available to us. Our interpretation relies only on the
geological data.
Conclusions: The petrographic data combined with the geology of the arena of the Amurru early
correspondence indicate that EA 61 must have been sent from the area east of Tripoli, probably
from the city of Ardata (Tell Arde). Tell Arde is situated over 12 km downhill from the nearest
Lower Cretaceous exposure on Mount Lebanon. It is therefore suggested that EA 60 and EA 61 were
prepared and sent from two different sites.
EA 62 (VAT 1680), from Abdi-Ashirta to Pahanata, the Commissioner of umur
Reliability: High/satisfactory.
Matrix: The matrix is carbonatic, light yellowish-tan in PPL, with a few badly preserved foraminifers and
their fragments. It is rather silty (~ 2%-3%) and very rich in opaque iron minerals that appear in a
range of sizes from a few micrometers to about 30-40m, the ner fraction of which is angular and
the coarser tends to be more rounded. The silt includes essentially quartz, with additional hornblende,
mica and feldspar. Under higher magnications (>X200) the matrix is brous, optically active and
displays weak optical orientation.
Inclusions: Rather dense (f:c ratio{0.062mm}=~95:5) badly sorted sand. Quartz is dominant appearing as
sometimes subrounded but commonly subangular grains up to 300m in size. Frequent rounded
grains of micritic limestone (up to 400m) are present and subrounded to angular clear replacement
chert (up to 1mm) is common. A few fragments (up to 350m) of aquatic mollusc shells appear. No
remains of vegetal material were traced.
Firing temperature: There are no petrographic indicators to determine the ring temperature of this sample.
The lack of any effect on the carbonates indicates that in any case it did not exceed 7000C.
Reliability: Satisfactory/high.
Matrix: The matrix is carbonatic, light yellowish-tan in PPL, with uncommon badly preserved foraminifers
and their fragments. It is slightly silty (less than 1%) and extremely rich in opaque to reddish-tan iron
minerals (magnetite and haematite respectively) that appear in a range of sizes from a few micrometers
to about 30-40m, the opaques of which are angular and the translucent tending to be spherical and
rounded. Haematite also appears as inllings within foraminifers. Another component that is abundant
within the matrix is vegetal material, most likely related to the grassy matter that appears in the
inclusions. It is represented by either charred plant bres, charred amorphous bodies, uncharred but
lignied plant tissues and phytoliths. Under higher magnications (>X200) the matrix is brous,
optically active and displays weak optical orientation.
Inclusions: Different from EA 61 and EA 62 consisting of a few rounded grains (up to 350m) of micritic
limestone and very few angular grains of quartz (up to 100m).
Vegetal material (SLY): Predominant, reaching up to 400m in length. Either charred or uncharred but
shrunken plant tissues, all non-indicative.
Palaeontology (LG): Brizalina (b) aeraniensis, Catapsydrax (p), Globigerinoides (p). Age: Upper
Miocene to Upper Pleistocene.
106
Firing temperature: Probably un red or lightly red judging by the state of preservation of the vegetal
material.
Geological interpretation: Although the matrix is undoubtedly Neogene marl as in EA 61 and 62, the
inclusion assemblage is composed predominantly of vegetal material. This is probably the result
of the scribes preference. The conclusions reached in the analysis of EA 61 and 62 may also be
applied to EA 156.
Conclusions: Most likely as EA 61 and EA 62.
Reliability: High.
Matrix: The matrix is argillaceous, ferruginous and reddish-tan in PPL. It is optically speckled and has
a pronounced optical orientation, rich (~5%-6%) in silt to ne sand sized ferruginous opaque to
dark reddish-tan bodies, with a gradual continuum between the grain-sizes. The ner grains are
angular whereas the coarser bodies develop into rounded shapes. The matrix also contains a very
few ferruginous shale fragments, irregular in shape.
Inclusions: These are sparsely spread (f:c ratio{0.062mm}=99:1). Predominant are angular grains very clear
volcanic glass (up to 370m) with pale green pleochroism, commonly with liquid inclusions and
less commonly mineral inclusions, isotropic or slightly birefringent to very low rst order grey. In
many cases it appears as laths or short bres. The coarser particles are also angular. A few rounded
calcite bodies of varying sizes (up to 120m), usually single crystals and rarely as aggregates are
present as are very rare rounded and spherical grains of quartz (100m).
Firing temperature: There are no petrographic indicators to determine the ring temperature of this sample.
Geological interpretation and conclusions: This letter belongs to the same petrofabric as EA 60, namely to the
Lower Cretaceous clay or shales group. The presence of volcanoclastics in this sample is noteworthy.
Reliability: High.
Matrix: The matrix is carbonatic, light yellowish-tan to light yellowish grey in PPL, with very few badly
preserved foraminifers and their fragments. It is slightly silty (less than 1%) and extremely rich in
opaque and dark reddish-tan iron minerals appearing in a range of sizes from a few micrometers
to about 30-40m, the ne fraction of which is angular and the coarser fraction tending to become
rounded. The matrix tends to be isotropic in places due to high ring temperatures. Under higher
magnications (>X200) the matrix is brous, speckled optically active and displays pronounced
optical orientation.
Inclusions: Sparsely spread sand (f:c ratio{0.062mm}=~99:1). Dominant are grains of sparitic limestone (up to
250m) and single grains of pure calcite, sparsely spread. Quartz is common being sparsely spread as
subangular grains (up to 100m). A small fragment (250m) of basalt is present. As in EA 61, apart from
the vegetal material the inclusions appear to be natural and not intentionally mixed in by the artisan.
Vegetal material (SLY): Frequent, sparsely spread uncharred plant tissues reaching 350m in length. Non
indicative small tissue fragments.
Firing temperature: Probably un red or lightly red judging by the preservation of uncharred vegetal
material.
107
EA 161 (BM 29818), from Aziru to the King of Egypt

Reliability: Sampled twice in order to increase the amount of inclusions, thus moderately reliable.
Matrix: Carbonatic, yellowish-tan in PPL and rich in cloudy, badly sorted carbonate micrite. It is
extremely rich in opaque to reddish-tan iron minerals (magnetite and haematite respectively) that
appear in a range of sizes from a few micrometers to about 30-40m, the opaques being angular
and the translucent tending to be spherical and rounded. Haematite also appears as inllings within
foraminifers. Under higher magnications (>X200) the matrix is brous, optically active and
displays weak optical orientation.
Inclusions: The SPA sample includes a large inclusion (ca. 1 mm) of coarsely crystalline alkali-olivine
basalt. The augite is partly serpentinized and the olivine altered to iddingsite. Additional sand size
basaltic minerals (iddingsite, pyroxene) are found in the matrix as monocrystalline inclusions. There
is some micritic limestone.
Firing temperature: There is not enough evidence in this sample for determining the ring temperature.
Geological interpretation conclusions: Although this sample is dened as being moderately reliable by
its size, it supplies enough evidence to suggest that it may be similar to EA 169.
EA 164 (VAT 249), from Aziru to the Egyptian ofcial Tutu
Reliability: Fair.
Matrix: Carbonatic, yellowish-tan in PPL, and rich in cloudy, badly sorted carbonate micrite (20%).
It is extremely rich in opaque to reddish-tan iron minerals (magnetite and haematite respectively)
that appear in a range of sizes from a few micrometers to about 30-40m, the opaques of which are
angular and the translucent minerals tend to be spherical and rounded. Haematite also appears as
inllings within foraminifers.
Inclusions: The SPA sample includes micritic limestone (up to 120m) and an iddingsite inclusion
(100m) most likely from olivine basalt as in EA 169. Stereoscopic examination of the surface of
this tablet reveals nely crystalline basalt inclusions as well.
Vegetal material (SLY): Up to 500m, a non-indicative tissue fragment.
Firing temperature: Most likely unred or very lightly red judging by the preservation of uncharred
vegetal material.
Geological interpretation: Although this sample is minimal in size and contents, the petrographic data
suggests that it may be similar to EA 161 (in itself a meagre sample but still more informative), and
to the highly reliable sample of EA 169. It undoubtedly belongs to the group represented by EA 169-171.
Reliability: High.
Matrix: Clayey, carbonatic, orange-tan to tan in PPL with scarce foraminifers. Opaque minerals are
rather common (~2%) sizing up to 100m, angular at the ne fraction and subrounded at the coarser
grain sizes. Quartz silt occurs (~1%) together with smaller amount of plagioclase. The carbonate
crystals within the matrix are dense (15%), usually sizing around 10m but occasionally reaching
20m-30m.
108
Inclusions: Sand (f:c ratio{0.062mm}=~95:5). Rounded fragments of fossiliferous coastal limestone

(beachrock) and more commonly separate fossils (up to 650m) are frequent to dominant. The fossils
consist mainly of articulated fragments of the calcareous corallinean algae Amphiroa, together with
some mollusc shell fragments. Rounded grains (up to 1.3mm) of smoky to brown-stained replacement
chert with local intergrowth of chalcedony are frequent. Subrounded to subangular fragments (up to
1mm) of micritic limestone usually with localized brownish staining are common. There are quite a
few subangular to subrounded quartz grains (up to 120m) and isolated polycrystalline fragments of
geode quartz with common liquid and mineral inclusions. Also present are a very few angular grains
(up to 380m) of serpentinized mineral crystals. These are most likely alteration products of mac
minerals (pyroxene or olivine).
Palaeontology (LG): Corallinean algae, mollusca.
Firing temperature: There is no petrographic evidence to determine the ring temperature of this sample.
Geological interpretation: EA 165, 167 (and probably 166) identical in their matrix and inclusions
differ signicantly from all other Amurru letters. The inclusion assemblage contains rounded
fragments of bioclastic coastal limestone (beachrock) and more commonly separate fossils of
coastal fauna and ora. The fossils consist predominantly of articulated fragments of the calcareous
corallinean alga Amphiroa, together with scarcer mollusc shell fragments. Chert, usually smoky
to brown with local intergrowth of chalcedony, is also common, together with subrounded to
subangular micritic limestone with common localized brownish staining, and subangular fragments
of geode quartz with common inclusions. These are accompanied by serpentinized minerals, most
likely alteration products of pyroxene, and more rarely volcanic glass. Geode quartz typied by its
coarse grain size, vast inclusions and the accompanying chalcedony forms a secondary constituent.
Along the Levantine coast corallinean algae of the genus Amphiroa occur in Quaternary bioclastic
sediments of the Pleshet, Hefer and Kurdane formations of Israel (Buchbinder 1975; Almagor and
Hall 1980; Sivan 1996). So far no equivalent geological terminology has been formalized for the
coast of Lebanon, but similar traits are recorded from the contemporary and analogous beachrocks
and sands (Sanlaville 1977:161-177; Almagor and Hall 1980; Walley 1997). While in other localities
this alga appears in older sediments, in the eastern Mediterranean it is absent even from the Miocene
reefal formations and appears only from the Pleistocene onward (Buchbinder 1975). On the basis of
the dominance of this component within the inclusions of EA 165-167 this group should be related
with Quaternary beach deposits.
Evidently, in EA 165 the alga fragments have no relation with the foraminifera that appear in the
clay (identied in EA 96 below, which belongs to the same petrographic group). The presence of
bentonic and deeper planctonic foraminifera and the lack of specic shallow benthic foraminifera in
the clay indicate that it is related to an older source rock other than the beachrock and sand. While
the coarser material is apparently beach sand, the planctonic foraminifers, which live in deep water,
probably come from the clay. Therefore, the source of the materials should be sought in an area
with exposures of marine clays of relatively young ages (Pliocene to Pleistocene, as attested by EA
96), together with Pleistocene-to-recent beach deposits of mainly calcareous character. The other
components represent different units within the Levantine lithostratigraphic section. Chert is almost
always related with formations of Santonian-Campanian or Eocene age, and geode quartz is typical
of the CenomanianTuronian transition. The igneous mac minerals (serpentine, olivine, pyroxene)
and the volcanic rock fragments that appear as detrital but rather large grains can represent PliocenePleistocene basalt ows, or earlier Lower Cretaceous basalts that are generally scarcer in extent.
109
Therefore, the source of the materials should be sought in an area where exposures of chalk
appear together with Pleistocene to Recent beach deposits of mainly calcareous character, chert
and occasional basalt exposures. While in the southern Levant the coastal sediments are dominated
by quartzitic sand which originally comes from the Nile, from Acco northwards this type of sand
diminishes and the sediment becomes increasingly calcareous. A systematic examination of thin
sections made from Holocene coastal sand from various localities along the coast indicates that
quartz is the dominant component as far north as Haifa Bay. At Bat-Galim (a neighbourhood
of Haifa) the sand is still dominated by quartz, but in Acco the beach sand is composed almost
exclusively of carbonates (see also Nir 1989:12-15; Sandler and Herut 2000). Even the sands that
exist near the eolianites of the coast of Galilee (resulting from the weathering of kurkar) are reported
to contain below 10% quartz (Sivan 1996:155). This implication is signicant, because it indicates
that EA 165 and petrographically similar tablets should be related a priori to the coastal area north
of Acco.
While in sand samples from south of the Acco area, where quartz is predominant, alga fragments
are rare and it is unlikely to have even one of them in a standard thin section, from Acco northwards
they form nearly 70% of the sand components. In Lebanon quartz may still appear as a minor
component in the beach sand dunes, but near Tyre and at Shoueifat (slightly to its north) the sand is
made essentially of carbonates, mostly from bioclasts (Sanlaville 1977:162-164). Further north the
beach sand dwindles quickly and at Sidon it is virtually absent. Still further north, beach sands appear
again in the >Akkar Plain (ibid.:161). Therefore, coastal sediments dominated by calcareous bioclastic
deposits are a clear attribute of the northern Levantine coast at some specic localities. This picture
does not change until one reaches Lataqia, where a different lithology appears (Chapter 5.I).
Other components within the inclusion assemblage may further limit the possibilities. In the
Levantine lithostratigraphy, chert is connected with either Senonian or Eocene exposures. Northward,
the thick Senonian deposits of the Mishash formation of Israel and the equivalent Amman formation
of Jordan lessen and the greater majority of cherts can be linked directly with Eocene exposures.
Such exposures are found predominantly between Tyre and Sidon, and again north of Tripoli. There
are also small outcrops east of Acco. When serpentine and weathered volcanic rock fragments
appear, they can be linked to an inland area where volcanic rock types expose. The only area where
Quaternary beach deposits, Senonian or Eocene chert, and mac minerals of volcanic origin may
appear together is the coastal area of the >Akkar Plain between Tripoli and Tartous. The mac
minerals were most likely dragged there from the basaltic ows of Nahr el-Kebir, where volcanic
elements are embedded in the local sediments. Therefore, the origin of tablets containing basaltic
minerals should be looked for in the area of the >Akkar Plain, not far from the seashore.
Considering the archaeological evidence, the only signicant site in this area where Late Bronze
remains are reported is Tell Kazel. The area to the north is characterized by Pliocene brown alluvial
soils, limestone, chert and basalt pebbles and Quaternary marine deposits (Kozlov et al. 1966:
43-44; Sanlaville 1977:270-284, Map 1). Basalts appear on the ridges east of the plain, and their
derived minerals and alteration products are drained by Nahr el-Abrash that passes near the site.
Thus, the combination of all the petrographic elements points to Tell Kazel as the most likely origin
of this group of tablets. Tel Kazel (Badre et al. 1990; 1994; Badre and Gubel 1999-2000) is widely
accepted as the location of the ancient city of umur (Klengel 1995).
Reference: No petrographic report has been published so far on pottery from Tell Kazel. This petrofabric
is common in the Canaanite jars of the Middle Kingdom and Second Intermediate Periods at Tell el
110
-Dab>a (Goren and Cohen-Weinberger, forthcoming), as well as in the New Kingdom Canaanite jars
from Amarna and Memphis (Serpicao et al. 2003). It was also found to form the Phoenician amphoras
from two Iron Age shipwrecks from deep water off Ashkelon (Ballard et al. 2002:160).
Conclusions: Considering the textual, the historical and the archaeological evidences, the only possible
city from where Aziru could have sent these letters is umur.
EA 166 (VAT 250), from Aziru to the Egyptian ofcial aay
Reliability: Fair. Only very small fragments were taken from this tablet. However they supply some valid
petrographic information.
Matrix: Similar in most details to EA 165: clayey, carbonatic, orange-tan to tan in PPL with scarce
foraminifers. Opaque minerals are rather common (~2%) sizing up to 50m, angular at the ne
fraction and subrounded at the coarser grain sizes. Quartz silt occurs (2%) together with smaller
amounts of plagioclase and pyroxene. The carbonate crystals within the matrix are dense (15%),
usually around 10m but occasionally reaching 20m-30m.
Inclusions: Sand, represented in this scanty sample by rounded brown stained replacement chert (300m),
subrounded to subangular fragments of micritic limestone (up to 250m) and a rounded 400m
mollusc shell fragment.
Geological interpretation: Petrographically it is similar to EA 165 and 167, although the small sample
size does not leave this beyond doubt.
Conclusions: See EA 165.
EA 167 (VAT 326), from Aziru (?) to the Egyptian ofcial Tutu (?)
Reliability: High.
Matrix: Identical to EA 165. Clayey, carbonatic, orange-tan to tan in PPL with scarce foraminifers.
Opaque minerals are rather common (~2%) sizing up to 50m, angular at the ne fraction and
subrounded at the coarser grain sizes. Quartz silt (~2%) occurs together with smaller amount of
heavy minerals including iddingsite and plagioclase. The carbonate crystals within the matrix are
dense (15%), usually sizing around 10m but occasionally reaching 20m-30m.
Inclusions: Sand (f:c ratio{0.062mm}=~95:5) of frequent to dominant rounded fragments of fossiliferous
marine limestone (beachrock) and more commonly separate fossils derived from it (up to 650m).
The fossils consist predominantly of the corallinean algae Amphiroa, together with mollusc shell
fragments. Rounded grains (up to 900m) of smoky to brown stained replacement chert with local
intergrowth of chalcedony, occasionally containing pseudomorphs after dolomite are fequent.
Subrounded to subangular fragments (up to 1mm) of micritic limestone usually with localized
brownish staining are common. Subangular to subrounded quartz grains (up to 120m) are fairly
common. There are a few polycrystalline fragments of geode quartz, commonly with liquid and
mineral inclusions, and a very few angular grains (up to 400m) of serpentinized crystals.
Palaeontology (LG): Corallinean algae, mollusca.
Geological interpretation: EA 167 is identical to EA 165 in all respects. The perfect match may suggest
that the two tablets were made of the same body of clay and inclusions.
111

Reliability: High. In light of the surprising results and due to its fragmentary state of preservation, this
tablet was sampled twice to produce a larger sample. This conrmed the rst examination.
Matrix: Silty (10%), carbonatic, yellowish-tan in PPL, optically active with speckled b-fabric. The silt
is essentially of quartz, but also contains recognizable quantities of accessory heavy minerals,
including hornblende, augite, zircon, plagioclase, microcline, biotite, muscovite, epidote and rarely
tourmaline, rutile, and garnet(?). Opaques are relatively common (2%-3%), sizing between few
micrometers and about 60m.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal material (f:c ratio {0.062mm}=95:5).
Dominant is quartz sand with accessory minerals: (up to 380m). The quartz is rounded to subrounded,
occasionally with undulose extinction, sometimes polycrystalline, rarely with mineral and liquid
inclusions. Grains of other minerals rarely appear at the same sizes, all rounded, including: zircon,
hornblende, augite (?). There are a few rounded grains (up to 320m) of micritic limestone. Fresh
(unfossilized) aquatic mollusc shell fragments are common (up to 400m lengthwise).
Vegetal material (SLY): Common (up to 400m), uncharred plant tissues (straw), unidentiable.
Firing temperature: No evidence for signicant ring is available. The vegetal material shows no trace
of charring. The tablet may have been lightly heated to preserve its shape, but not red.
Geological interpretation: This document is one of a group of tablets that to the naked eye are characterized
by a coarse and rather crumbly light tan fabric (around Munsell colour 10YR5/3). The coarse fraction
contains quartzitic sand that can be readily observed under a magnifying glass, coarse straw chaff,
and often millimeter-size fragments of marine mollusc shells. In thin section the matrix is silty and
rather carbonatic. The silt is well sorted and contains mainly quartz but also a recognizable quantity
of other minerals, including hornblende, zircon, mica minerals, feldspars, tourmaline, augite and
more rarely garnet, epidote and rutile. Ore minerals are abundant too in this fraction. The non-plastic
assemblage includes dense, well sorted, rounded sand-sized quartz grains with the occasional addition
of other minerals (feldspars, hornblende, zircon, augite), straw and marine mollusc shell fragments in
recognizable sizes (up to 3 mm and sometimes even more). In several cases they are accompanied by
calcarenite or calcite-cemented quartzitic sandstone (kurkar).
Based on a wealth of published data (Porat 1987:112-115; 1989a:50-52; Goren 1987; 1988; 1991a:
101-104; 1991c; 1996a; Goren and Gilead 1987; Gilead and Goren 1989:7; Goldberg et al. 1986;
Rognon et al. 1987) the matrix is readily identied as loess soil. In the Levant loess soil occurs
mainly in the northern Negev and the southern Shephelah of Israel.1 It does not appear in Lebanon
and therefore this tablet is undoubtedly not made of materials from Amurru. Our collection of
materials from Levantine sites, including ceramic workshops, indicates that loess-made pottery was
produced in the region between Lachish (or slightly to its north), Ashkelon, Gaza and Beersheba.
This is true for all periods between the Pottery Neolithic and Mediaeval times.
In pottery assemblages that belong to this group, the inclusions accompanying the loess matrix
are variable, and indicate different geological environments within the area where loess soil is
distributed. The use of loess with inclusions in which limestone is the dominant component is
prevalent mainly at sites north-east of the Beersheba valley, whereas in the southern Shephelah chalk
1
Note that the term loess refers to a set of aeolian and alluvial silty-clay sediments (Dan et al. 1976) which cannot be
distinguished clearly in ceramics by petrographic means.
112
sand is commonly the dominant or even sole non-plastic component. Quartz is the major constituent
at northwestern Negev sites (Gilead and Goren 1989:Fig. 2). When quartz sand dominates the
inclusions, sand-sized grains of accessory minerals, mainly hornblende, zircon, feldspar and augite,
commonly accompany it. These indicate a littoral origin, since the sandstone-derived quartzitic
sands of the Lower Cretaceous formations (such as those that appear in Lebanon) are devoid of
all of these. To this evidence we should add the presence of mechanically and chemically unstable
minerals (like hornblende, pyroxene, rutile, garnet, etc.) that appear with the quartz as rounded
sand-sized grains. These also hint at a littoral origin, since in sands of coastal origin that are swept
inland by aeolian activity they are subjected very rapidly to mechanical and chemical weathering
(Slatkine and Pomerancblum 1958; Pomerancblum 1966; Nahmias 1969). This point is important
because sands of coastal origin, transpoted by aeolian activity, appear as far inland as the southerncentral Beersheba valley. To this we may add the fresh fragments of marine mollusc shells, and
the kurkar rocks. The latter are derived from fossilized sand dunes, locally termed kurkar, that are
generally grouped into the Pleshet formation (Issar 1961; see also Chapter 14.VI).
So far, we have not recorded the combination of loess soil with coastal sand much further
north of Ashkelon. In our study of ceramic workshop wasters, pottery depicting this combination
was restricted to the area between Raphia in the south, Ashkelon in the north and the Mefalsim Kissum area in the east. Pottery produced at Ashdod already exhibits the use of hamra soil rather
than loess as clay (see also Edlestein and Glass 1978 for the Iron Age I).
In conclusion, the materials of this tablet originated in the area of Gaza or Ashkelon.
Reference: The combination of loess, straw and coastal sand is well known from the Late Bronze Age
ceramic assemblages from Tell Abu Salima, Tell Riddan and Deir el-Bala, all located in the coastal
plain of Gaza and northern Sinai (Dothan 1978, 1987). In the last two, evidence for pottery production
was encountered (G. Edelstein, pers. comm.; Cohen-Weinberger 1998; Killebrew 1996, 1998:525-59).
This combination is also characteristic of Egyptian pottery production in sites of the southern coast
(Goldberg et al. 1986; Killebrew 1996; 1998:206-211; Cohen-Weinberger 1998).
Conclusions: In this study the combination of loess soil and coastal sand appears systematically only in
the Ashkelon tablets (Chapter 14.VI). Loess soil or loessial alluvium also appears in letters of other
cities located in the area where it naturally occurs, namely Lachish and Gath. Yet, in the latter the
composition of the inclusions differs and does not show such emphasized coastal character. Two
groups of inclusions accompanying loess matrix may be dened: coastal (Ashkelon), and inland,
southwestern Shephelah (Lachish). The rst group is characterized by purely coastal sediments
(quartz with abundant accessory heavy minerals, kurkar, marine mollusc shell fragments), while the
second is characterized by a dominance of limestone and chalk inclusions.
In EA 168, like other tablets that belong to the loess matrix category, the inclusions clearly
indicate at a coastal environment. The specic raw materials point to the coastal strip between
Raphia and Ashkelon. The only feasible explanation is that EA 168 was sent by Aziru from Gaza,
the main Egyptian administrative centre in southern Canaan, where he may have landed on his
way to Egypt. Indeed, in lines 4-7 he declares that he will arrive safely to see the face of the king,
a declaration that he already made in his former letters (EA 165:31-33; 167:17-19). Then he adds
that his envoy, the Egyptian messenger (atip) and the presents he brought to the king will have
arrived in ships (lines 7-11). It seems that Aziru stopped for a short time in Gaza while sending his
messenger and presents, with tablet EA 168, to placate the Pharaoh and prepare for his arrival at the
royal court.
113
EA 169 (VAT 1660), from DU-Teshup or Bet-ili (?) to the Egyptian ofcial
Reliability: High.
Matrix: Carbonatic, yellowish-tan in PPL, with few foraminifers, and rich in cloudy, badly sorted
carbonate micrite (20%). It is extremely rich (~7%) in opaque (angular) to reddish-tan (spherical
and rounded) iron minerals (magnetite and haematite) that appear in a range of sizes from few
micrometers to about 30-40m. Haematite also appears as inllings within foraminifers.
Inclusions: Moderately sorted sand (f:c ratio {0.062mm}=~95-93:5-7). Frequent to dominant is alkali-olivine
basalt (up to 1.2mm) which is usually subrounded or spherical and coarsely crystalline, sometimes
doleritic. The olivine phenocrysts are partly or entirely altered to iddingsite. Both micritic and sparitic
limestone are frequent as subrounded grains up to 800m. A few angular cleavage crystals of calcite (up
to 300m) are present as are a very few angular quartz grains (up to 120m).
Vegetal material (SLY): Few, up to 380m (lengthwise). Non-indicative tissue fragments.
Firing temperature: Probably unred or very lightly red because uncharred vegetal material is preserved.
Geological interpretation: This tablet is used as the type-sample for the petrographic group that includes
EA 161, 164, 169, 170 and 171. In terms of the clay used, it represents marl with characteristic cloudy
micritic limestone fragments that densely occupy it. Foraminifers are uncommon. The inclusions are
typied by a signicant basic igneous content, usually of basalts and seldom of dolerite, together
with limestone and some quartz. EA 169 and 171, presenting the best and largest samples belonging
to this petrographic group, include foraminifers of Neogene to Holocene age (see EA 171).
In terms of provenance, the lithological combination that is presented here may be limited to a
rectangle stretching between present-day Tripoli and Tartous on the coast and Tell Nebi Mend and
ama inland. This area features three geographic and geological units: the >Akkar Plain in the west,
the Nahr el-Kebir basin in the centre and the Middle Orontes Valley in the east. Due to the state
of archaeological research in these areas, the petrographic research relies mainly on the geological
mapping. Still, it may supply reasonable documentation concerning the possible origin of this
petrofabric. In this area Pliocene marine deposits and volcanics appear together only in the >Akkar
Plain, specically near alba in the south and immediately north of the Nahr el-Kebir channel in
the north. The only signicant site in this area is Tell >Arqa. Indeed, EA 100, which was sent from
Irqata (Tell >Arqa), is identical to the Amurru letters of this petrofabric (see Section IV below).
Tell >Arqa is situated on the southern ank of the >Akkar Plain, near Nahr el->Arqa that drains the
mountainous area to the southeast of the plain. The site is located on a plain of Pliocene argils and
marl, near the lower terrace of quaternary colluvium that collects its materials from the Turonian
and Cenomanian calcareous formations and the volcanic terrain east of alba (Sanlaville 1977:25,
243-280, Map 1). The nearby channel of Nahr el->Arqa collects sediments from the plateau to the
east and the plains north and south of Tell >Arqa. In these areas, Cenomanian-Turonian limestone
series, Senonian chalk and chert, Lower Cretaceous sandstones and marls and Jurassic limestones
are exposed. To the northeast, Pliocene volcanics contribute the basalt component.
The volcanics of the Nahr el-Kebir area are of Upper Miocene age, being represented by basalts and
dolerites and, more rarely, volcanic glass. They are leucocratic with either porphyric, glomeroporphyric,
ophitic, poikiloophitic or tholeiitic texture. Phenocrysts are olivine (Kozlov et al. 1966:33).
Reference: No petrographic report on the pottery from Tell >Arqa has been published. EA 100 is therefore
the only reference for this petrofabric.
Conclusions: Based on its petrographic traits and similarity to EA 100, EA 169 was probably made at Tell
>Arqa. The same applies to EA 161, 164, 170 and 171.
114
EA 170 (VAT 327), from Baluya and Bet-ili to Aziru in Egypt

Matrix: Carbonatic, yellowish-tan in PPL, and rich in cloudy, badly sorted carbonate micrite (20%).
Common (~5%) opaque to reddish-tan iron minerals (magnetite and haematite respectively) appear
in a range of sizes from a few micrometers to about 30-40m, the opaques of which are angular and
the translucent minerals spherical and rounded. Under higher magnications (>X200) the matrix is
brous, optically active and displays very weak optical orientation.
Inclusions: Sand of subrounded and spherical coarsely crystalline alkali-olivine basalt, sometimes doleritic
(up to 500m), the olivine phenocrysts being partly or entirely altered into iddingsite. Subrounded
micritic and sparitic limestone (up to 700m), subrounded and spherical aggregates of chalcedony (up
to 750m) and a very few angular grains of quartz (up to 120m) make up the inclusion suite.
Vegetal material (SLY): Up to 300m (lengthwise), non-indicative tissue fragments.
Firing temperature: As EA 169
Geological interpretation: Similar in most details to EA 169.
Conclusions: Several scholars have suggested that the same scribe wrote both EA 169 and 170 (Moran
1992:257, n. 1, with additional references). The petrographic data may support this hypothesis,
although the sample size of EA 170 does not allow a semi-quantitative analysis of the inclusions.
Reliability: High.
Matrix: Carbonatic, yellowish-tan in PPL, containing densely spread (~20%), badly sorted calcite crystals
and micritic bodies ranging between 10m and 80m. The calcite crystals are commonly idiomorphic,
the micritic bodies are rounded. Foraminifers of Pliocene age are rather common. Quartz silt is scarce
(1%). Occasional heavy minerals appear in the silt fraction and contain twinned plagioclase and augite.
Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~2%) that appear in a range
of sizes from few micrometers to about 30-40m, the opaques of which are angular and the translucent
minerals tend to be spherical and rounded. Haematite also appears as inllings within foraminifers.
Inclusions: Badly sorted sand (f:c ratio{0.062mm}=93:7) in which volcanic rocks, their derived minerals and
alteration products are dominant. An assemblage of minerals and rock fragments, all related to basalts
and tuffs, includes the following: 1) A 900m relict of volcanic rock, entirely weathered into clay
but preserving the magnetite inclusions and pseudomorphs after the olivine phenocrysts. Iddingsite
crystals (most likely after olivine) are common in grain sizes ranging between 120m and 650m.
2) Rounded and spherical fragments of coarsely crystalline olivine basalt, reaching 300m in size,
usually partially weathered. 3) Loose crystals of iddingsite (up to 300m), sometimes preserving
the original olivine as core, augite (380m), and plagioclase (up to 100m). Angular to rounded
quartz, (up to 650m) sometimes with undulose extinction and uid or mineral inclusions but rarely
polycrystalline is frequent. Usually rounded micritic limestone (up to 500m) and rarely sparite
stained by iron minerals, is common. There are a few calcite cleavage crystals (up to 250m).
Palaeontology (LG): Abundant fossils in the matrix and some in the inclusions: Bolivina (b), Bulimina
(b), Globigerina (p), Globigerinoides (p), Globorotalia (p). Age: Neogene to Holocene.
115
II. UMUR
EA 96 (VAT 1238), from an army commander to Rib-Hadda of Gubla
Reliability: Moderate. The sample size adequately represents the tablets fabric and enables a reasonable
petrographic interpretation. Re-sampled (by SPA again) to supply more data, which made the
sample satisfactory.
Matrix: Somewhat carbonatic, rather silty (about 5%), dark range-brown in PPL. The matrix contains
opaque minerals (~3%) ranging between 10m to 50m. These are commonly angular in the smaller
grain sizes and spherical in the larger fraction. Calcite, either as clear crystals or as sparite, is
rather common in the silt fraction. Secondary are quartz (usually angular), feldspar and some heavy
minerals (identied: zircon, rutile, and iddingsite).
Inclusions: Due to the small sample size, only part of the original inclusion assemblage may be
represented in the thin section. Frequencies were not estimated due to the partial sampling process.
Distinguished were rounded micritic and bioclastic limestone (up to 420m in this sample)
containing coralline algae fossils (most likely Amphiroa sp.) and marine mollusc shell fragments;
diversied shale fragments (some more clayey and reddish-brown in PPL, others ferruginous and
opaque or nearly opaque) up to 510m in this sample and quartz, usually subangular, up to 250m
in this sample.
Palaeontology (LG): Brizalina spathulata (b), Globigerina (p), Globorotalia. Age: Pliocene to
Pleistocene, determined according to B. spathulata.
Firing temperature: If red at all, the temperature did not reach 750 0C judging by the lack of changes in
the calcite.
Geological interpretation: Petrographically this tablet is similar in its details to a group of tablets sent
by Aziru of Amurru (EA 165, 166, 167). The palaeontological data indicates Tertiary to Quaternary
marine clay. See discussion of EA 165.
Conclusions: Tablet EA 96 was written by an Egyptian army commander. The contents of the letter
clearly indicate that he was then staying at umur and corresponded with Rib-Hadda on the affairs
of the city. Petrographically EA 96 can be connected with a group of tablets sent by Aziru of Amurru
from umur (EA 165-167). The historical data of the letter strongly supports the conclusions reached
independently by the petrographic analysis, indicating that this group was sent from the Egyptian
centre of umur.
III. TUNIP
The location of the city of Tunip is disputed. Since its identication has been considered a key issue in
the interpretation of Azirus correspondence, we have included EA 59, written by the citizens of Tunip
to the Pharaoh, in this section.
The kingdom of Tunip played an important role in the history of Syria in the 15th-14th centuries
BCE. A rich body of textual data suggests that the city of Tunip, its capital, was located either in the
Orontes basin or west of it (Astour 1969:390-398; 1977; Klengel 1969:75-95; 1995). The textual
evidence for the location of Tunip is as follows:
116
1. The city is mentioned in the Ebla, Kltepe, Mari, Babylon and Alalakh VII tablets. Hence we
must look for a site that was occupied in the Early, Middle and Late Bronze Ages.
2. Troops from Tunip took part in the defense of Ullasa, a coastal city probably located in modern
Tripoli. This seems to indicate that Tunip was Ullasas strong neighbour.
3. In the course of his 42nd year, Thutmoses III destroyed the city of Irqata (Tell >Arqa) and then
proceeded to conquer Tunip. Afterwards he reached the region of Qidshu (Tell Nebi Mend) in the
northern Beqa> (ANET: 241b).
4. In his 7th year Amenophis II conducted a campaign along the Middle Orontes, from Qatna
northward to Nii and back to Qidshu, but does not mention the city of Tunip.
5. The treaty between Ir-Addu of Tunip and Niqmepa> of Alalakh (AT 2) established the relations
between the two neighbouring kingdoms (Dietrich and Loretz 1997). It indicates that Tunip
must have had a common border with Alalakh. The city of Nii, commonly identied at Qal>at
el-Mudiq, was included in the territory of Alalakh (Klengel 1969:58-59, 66-67). Therefore, a
location for Tunip south of the Ghab, in the Middle Orontes Valley, is favoured (Klengel 1992:
89-90; 1995:130).2
6. According to EA 165-167, Tunip was located two marching days from the land of Nuashe a
large kingdom located in northern Syria, south of Aleppo and north of the Middle Orontes.
7. In an inscription of his 8th year, Ramesses II relates that he conquered Dapur, a city located in
the region of the city of Tunip in the land of Naharin. According to another text, Dapur was
situated in the land of Amurru (ANET: 256b; Kitchen 1982:54-56, 68-69). These references
indicate that Tunip was considered to be located in the territory of Amurru, not far from the
borders of Qidshu.
8. A Hittite text (KBo VIII 38) mentions Tunip in connection with the verb to cross. Astour (1977:
56) suggested that this refers to a ford where the Hittite army crossed the Orontes on its way to
the lands of Upi and Amqi. Tunip is probably the last in a list of towns that might have marked
the road leading there, but none of these towns can be identied with certainty.
In the light of these references, several locations have been offered for Tunip:
A. Qal >at el-usn (Krak des Chevaliers of the Crusaders), near Nahr el-Kebir (Weber 1915),
or a site located in northern Phoenicia, not far from Ullasa (Helck 1973). However, the
documentary evidence hardly supports a location near the coast (see Klengel 1969:75-78).
B. The majority of scholars locate Tunip north of Tell Nebi Mend in the middle Orontes valley
(Astour 1969:390-398; 1977; Kuschke 1979:23-25; Klengel 1969:75-78; 1995:127-129). Astour
(1977) suggested placing it at ama, a major site that was the capital of the Kingdom of Hamath
in the rst millennium BCE. This proposal is questionable, as texts from Ebla mention both
ama (Amadu/Amad) and Tunip (Duneb/Dunib) (Klengel 1995:127).
C. Courtois (1973:55 n. 5), who conducted a survey in the Middle Orontes, proposed either Tell
Sikkine Sarute or Tell >Asharneh in the southern Ghab. The identication of Tunip with Tell
>Asharneh was also suggested by Klengel (1995) and accepted by Liverani (1998:298 n. 42).
Tell >Asharneh is located on the Orontes, ca. 40 km northwest of ama, southwest of Apamea. It is 70
ha in size, thus by far the largest site in the Middle Orontes. The site is comprised of two acropoli, separated
2
Another treaty, probably dated to the time of Tudalia I, was conducted between an unnamed Hittite ruler and Labu of the
city of Tunip (CTH 135; Klinger 1995). The historical relations between the two sides are described in the rst part of the
treaty. The text apparently refers to Ilim-ilimma, king of Alalakh, and his relations with Tunip and atti, but is badly broken.
Thus, CTH 135 does not help establishing Tunips border.
117
by a wide opening, perhaps representing an ancient gate, and an extensive lower town surrounded by an
earthen rampart. It is strategically located, dominating an important ford on the Orontes (Courtois 1973:
63-650). A stele of Sargon II was found at the site (Thureau-Dangin 1933). A small-scale excavation has
recently (1998) been conducted at Tell >Asharneh (directed by M. Fortin of Laval University in Quebec). It
indicates that the site was occupied during the Early and Middle Bronze Ages, as well as in the Iron Age
(ca. 900-700 BCE). No Late Bronze Age nds have been discovered as yet..

Sanlaville (1977) surveyed the geology of the >Akkar Plain with an emphasis on the Quaternary deposits.
In a later work he covered also the Nahr el-Kebir area (Sanlaville 1979). A geological map of the Middle
Orontes Valley has recently been published by Sanlaville et al. (1993). It includes the surroundings of
ama, Tell >Asharneh and the Ghab. Geological mapping of the entire area (north of the present Syrian
Lebanese border) in various scales, the most detailed of which is 1:200,000, is supplied by Ponikarov
(1964) and Ponikarov et al. (1966), with explanatory notes by Kozlov et al. (1966) and Shatsky et al.
(1966). The comparative petrographic data include preliminary results from Tell >Asharneh (Cooper
and Fortin 2001) and a collection of thin sections from ama made of pottery housed in the Israel
Antiquities Authority storerooms.
CATALOGUE
EA 59 (BM 29824), from the citizens of Tunip to the King of Egypt
Reliability: High.
Matrix: Carbonatic, yellowish-tan with local changes to light reddish-tan in PPL, very rich in foraminifers
(up to 7% in places). Densely spread (~15%), badly sorted calcite crystals and calcitic bodies ranging
between 10m and 80m appear in the matrix. The calcite crystals are commonly idiomorphic; the
calcitic bodies are micritic and rounded. Quartz silt is scarce (less than 1%). Opaque to reddish-tan
iron minerals (probably magnetite and haematite respectively, over 2%) appear in a range of sizes
from few micrometers to about 30-40m, the opaques of which are angular and the translucent
minerals tend to be spherical and rounded.
Inclusions: Sand (f:c ratio{0.062mm}=~85:15-90:10) of well-rounded, badly sorted grains. Glauconite
globules, the major inclusion of this tablet, is visible to the naked eye as olive-green, spherical
bodies that are embedded in the dark reddish-tan clay. In thin section this inclusion class forms
dominant rounded, usually spherical globules (up to 650m) of glauconite with green to olive-green
pleochroism (thus unred, as glauconite becomes yellowish to reddish upon ring). The globules are
usually massive but in several cases they preserve zoned internal structures with numerous silt-sized
bodies of opaque minerals and local stains of iron minerals (most likely ghoethite). Foraminiferous
chalk is frequent (up to 300m) and there are a very few fossilized mollusc shell fragments (up to
300m lengthwise). No remains of vegetal material were traced.
Palaeontology (LG): Fossils are clustered in the matrix and the inclusions: Acarinina (p), Chiloguembelina
(p), Globorotalites (p) sp., Subbotina (p) sp. Age: Paleogene.
Firing temperature: Unred or only slightly red judging by the green pleochroism of glauconite.
Geological interpretation and reference: The composition of this tablet is unique in the Amarna
correspondence. Its most signicant feature is the dominance of glauconite globules, which are
unlikely to be a substance added by the scribe who produced the tablet. This is due to the softness
118
of this mineral and its scarcity as pure sand. It is more logical to assume that the glauconite was
an integral part of the clay utilized for producing the tablet. Therefore, the source material for this
tablet was a variety of glauconite-bearing marl, where glauconite is expressed in whole percents by
volume, thus acting as a signicant rock-forming element.
Glauconite is known to occur almost exclusively in marine sediments (Deer et al. 1971:35-41).
The narrow range of chemical composition of glauconites is consistent with the restricted set of
geological environments that are related to its deposition. It is commonly accepted that glauconite is
formed by marine diagenesis of materials in shallow water at a time of slow sedimentation. In this
glauconite is closely related to phosphorite and often replaces it. Although glauconite concentrations
may be found as impurities in limestones and marls, when they appear in high proportions they form
greensands, a term coined after the green colour of the mineral.
The composition of EA 59 suggests an area where glauconite-containing marl is exposed.
This composition is quite uncommon in the Levant. The palaeontological evidence links it to
the Paleogene. Since glauconite is a potential substance for age determination of geological
strata through potassium-argon dating, there is a rich body of data regarding its stratigraphic and
geographic distribution in the Paleogene sequence of Syria (Krasheninnikov et al. 1996, with further
references). Given the combination of the textual evidence concerning Tunip and geological data
about glauconite distribution in Syria, it may be possible to identify the general area, and perhaps
the specic site, where the materials of EA 59 were extracted.
Glauconite sediments are distributed in Syria in very restricted areas. According to the conditions
of glauconite formation, its deposits replace organic matter and phosphates in enriched sediments.
In Syria and Lebanon glauconite is represented in various units of the entire Paleogene section
(Krasheninnikov et al. 1996:65). The majority of glauconite horizons are found in the Palmyrids
depression, the southern slope of the alab uplift, the western slope of Jebel Anariyeh, and much
less frequently on the eastern slope of the Anti-Lebanon (Krasheninnikov et al. 1996:67). The
textual evidence indicates that most of these areas are irrelevant for the location of Tunip.
In the relevant area glauconite horizons are reported from merely four localities. These include
Nahr Marqiyeh and Safarqiyeh near the coast, and Jebel Abu Dardeh and Jebel Zawiye inland. These
proles differ in their age and hence their palaeontological and lithological characteristics (Shatsky
et al. 1966:37, Fig. 5; Krasheninnikov et al. 1996:66, Fig. 12). The Nahr Marqiyeh, Safarqiyeh and
Jebel Abu Dardeh horizons date between the Upper Paleocene to the Lower Eocene, whereas the
Jebel Zawiye deposit dates to the Middle Eocene. Re-deposited Senonian and Paleogene glauconite
beds are reported also from the Pliocene continental deposits that ll the tectonic basin of the Ghab
and the margins of the >Asharneh Valley (Shatsky et al. 1966:54-56).
Considering the sites and areas which were suggested for the location of Tunip in the light of the
distribution of these glauconite bearing beds, the following picture evolves:
The Nahr el-Kebir area: This area is characterized by Miocene volcanics. No signicant second
millennium BCE site is recorded in this region. Therefore this option should be eliminated. In the
opening of Nahr el-Kebir to the >Akkar Plain, where Tell Jamous is the most prominent site, no
glauconite bearing strata are found. This area is still dominated by the volcanic effusives and basalt
ows of the Tell Kalakh area and by Pliocene to Quaternary marine and beach deposits (Ponikarov
1964: Sheets I-36-XVIII, I-37-XIII; Kozlov et al. 1966:34-37, 41-45).
Sites in northern Phoenicia: Only two restricted locations along the coastal area may fit the
composition of EA 59. The first is the small Lower Paleogene outcrop near the mouth of Nahr
119
Marqiyeh and the second is a similarly restricted area of the same age near Safarqiyeh. The
Nahr Marqiyeh exposure is located north of Tartous, a few kilometers from the seashore. The
Safarqiyeh deposit, nearly midway between Jebleh and Lataqia, is located further inland, about
10 km from the seashore.
The lithostratigraphic section at Safarqiyeh (Shatsky et al. 1966:35-38, Fig. 5: section II) reveals
that beds rich in glauconite appear at the base (Paleocene) and higher up into the Early Eocene. In
places they change into glauconitic sandstones. Considering the age and density of these beds, they
may very well suit the petrography of the Tunip letter. However, there is no site in these locations
signicant enough to be identied with a city of such importance. Moreover Safarqiyeh is too far
to the north, already in the territory of Ugarit, whereas Nahr Marqiyeh is too close to umur. In
addition, the historical record seems to indicate that Tunip was not a coastal city.
ama: The geological evidence cannot support Astours identication of Tunip in the ama area
(1969:390-398; 1977). Coniacian to Campanian chalks and limestones interbedded with chert typify
the immediate vicinity of ama. The site is located on a terrace of Quaternary sediments (Sanlaville
et al. 1993). These elements are also reected in the petrography of most of the vessels from ama that
we examined. Although there are Middle Eocene exposures east of the city, no glauconite horizons
are recorded from this area. For this reason, and because both ama and Tunip are mentioned in the
Ebla texts (Klengel 1995:127), ama cannot be considered as a candidate site for Tunip.
Tell Hana: In the light of the petrographic data and the presence of glauconite deposits at Jebel
Abu Dardeh (Shatsky et al. 1966:37, Fig. 5; Krasheninnikov et al. 1996:66), Dussauds suggestion
(1927:109) should be re-examined. Tell ana is a medium-sized mound located 3 km east of a
village named Duneybah, which might have preserved the ancient name Tunip.
Jebel Abu Dardeh lies about midway between Qatna (el-Mishrife) and ama. The mountain range
is composed mainly of Middle Eocene sediments. Along the channel cut by the Orontes River, older
(Paleocene to Lower Eocene) sediments are exposed. A full section appears east of the village of
Taqsis in the cliffs located near the bottom of the Orontes channel. The section is built up of grey
marls occurring immediately above the basal glauconitic marl. Other beds appear in several parts of
the Lower Eocene section (Shatsky et al. 1966:38-39, Fig. 6). However, Tell ana is located about 10
km east of these deposits and outside the range of Jebel Abu Dardeh. In other words, the glauconite
deposits are quite remote from the site. From an archaeological point of view the site is too small and
too close to Qatna to be considered as a signicant kingdom of the calibre of Tunip. Historically Tunip
must have had a common border with Alalakh, and this is not possible regarding Tell Hana.
Tell Sikkine Sarute: No glauconite deposits are known from the entire area around this site.
Tell >Asharneh: Middle Eocene sedimentary rocks in the area of Jebel Zawiye are reported to
contain glauconite, sometimes in considerable concentrations and even as glauconitic sandstones.
The range of Jebel Zawiye is characterized by Cenomanian and Turonian limestones and dolomites
in the south, with local exposures of Campanian clayey limestones and chert. Pliocene basalts cover
the central and northwestern areas and a broad Middle Eocene province extends in the northern
and northeast parts of the anticline (Ponikarov 1964: Sheets I-36-XXIV, I-37-XIX). In this area
glauconite-bearing strata are reported from sections near M>arret en-No<man (Shatsky et al. 1966:
42-44) and Aria (Krasheninnikov et al. 1996:66).
The Ghab valley lies to the west of Jebel Zawiye. The western side of the valley, bordering
the Jurassic anticline of Jebel Anariyeh, is irrelevant to our discussion. Geologically it lacks
glauconite-bearing deposits and archaeologically there is no signicant site there. Several sites are
120
located on the eastern edge of the seasonally ooded Ghab valley (identied as the ancient Sea of
Niya). The largest and most important of them is Tell >Asharneh. Detailed geological mapping of
the area (Malinov and Fati 1985) as well as a larger scale map (Shatsky et al. 1966; Ponikarov 1964:
Sheets I-36-XXIV, I-37-XIX) were used for analyzing the petrographic situation in this area.
The margins of the Ghab and the >Asharneh tectonic basin are lled with Pliocene continental
rocks. On the basins margins they come into contact with older rocks of different ages. Sandstones
interbedded with red clay dominate the lower part of the section. The sandstones consist of
debris of carbonate rocks and foraminifera fragments. The detrital material includes grains of redeposited glauconite, surrounded by clay-carbonate cement. Upwards in the section the amount of
clastic materials increases, including clayey limestones, foraminifera and grains of glauconite redeposited from Senonian and Paleogene rocks. In the eastern ank of the Ghab basin the thickness
of this part of the section is 50 m. This means that the area around Tell >Asharneh does include
suitable materials to t the EA 59 petrography, although these were translocated and re-deposited
in a later age, after their formation.
A current study of the pottery assemblage from Tell >Asharneh (Cooper and Fortin 2001)
reveals a petrographic group described as >Asharneh Petrofabric 1 (ibid.: Fig. 1:a-o), which seems
to be similar in many details to EA 59. This petrofabric is characterized by a red clay matrix,
distinguished chiey by the presence of glauconite whose quantity varies in the samples between
3-13%. The glauconitic particles are rounded, averaging about 0.2 mm in size. They are well to very
well sorted and are usually clear bright red or bright orange under plane polarized light, although in
a few samples the grains are brown-yellow or dark brown. The majority of the glauconitic particles
are semi-isotropic, taking on a dark green speckled appearance under cross-polarized light.
In the >Asharneh pottery the inclusion assemblage is slightly more varied than in EA 59. This
may result from the larger sample sizes taken from the vessels (as compared with the tablet),
or from less selective addition of temper by the potters. The inclusions contain quartz, trace
plagioclase feldspars, trace amphiboles, trace clinopyroxenes, trace biotite and muscovite, chert
and phosphorites. Additionally, and as in EA 59, the samples contain poorly-sorted micrite and
sparry calcite. About half of the samples feature large grit-sized particles of sandstone, comprising
particles of glauconite, phosphates and occasionally sparry calcite in micritic cement.
The presence of glauconite in varying proportions, in combination with phosphates, distinguishes
this from other petrofabrics found at Tell >Asharneh. This glauconitic fabric was regarded as a local
product since the presence and proportions of other inclusions (quartz, phosphates, micrite, etc) are
more or less consistent with several other fabrics that characterize the pottery from this site. It is also
noteworthy that glauconitic specimens were limited to pottery forms that date to the Middle Bronze
Age. It did not characterize any of the Early Bronze Age sherds nor any of the later Iron Age material.
The study did not include any safely-dated Late Bronze pottery (Cooper and Fortin 2001).
Conclusions: Based on a combination of geological, archaeological and textual considerations, Tell
>Asharneh is the most likely candidate for the city of Tunip. Needless to say this relies on the
assumption that the site was signicantly inhabited also in the Late Bronze Age. The unique
composition of EA 59 seems to indicate that none of the Amurru letters was written at this site
(contra Izreel and Singer 1990:137-138; Singer 1991:152).
121
IV. IRQATA
Irqata is identied with Tell >Arqa in the >Akkar Plain. The mound, ca. 5 ha in size, commands the
main road leading from alba to the south. It is the most important Bronze Age mound between Tell
Kazel (umur) and Tripoli (probably the location of Ullasa). The site is being excavated by the French
Archaeological Institute in Beirut (Thalmann 1991; 2000).

Tell >Arqa is situated on the southern ank of the >Akkar Plain, near Nahr el->Arqa which drains the heights
to the southeast of the plain. The site is located on a plain of Pliocene argils and marl, near the lower terrace
of quaternary colluvium that collects its materials from the Turonian and Cenomanian calcareous formations
and from the volcanic terrain east of alba (Sanlaville 1977:25, 243-280, Carte No. 1).
CATALOGUE
EA 100 (BM 29825), from the elders of Irqata to the King of Egypt
Reliability: High.
Matrix: Carbonatic (marl), greyish-yellow tan to light tan in PPL, tan in XPL, f:c:v{0.01mm}=70:20:
10, birefringent with speckled b-fabric. Densely spread (~20%), badly sorted calcite crystals and
micritic bodies ranging between 10m and 80m appear in the matrix. Also common are opaques,
subangular to subspherical and angular quartz, and foraminifers.
Inclusions: Sand (up to 900m) of volcanic rocks, their derived minerals and alteration products is the most
frequent component. This assemblage is made upof minerals and rock fragments which are all related to
basalts and tuffs. It includes: 1) Rounded spheres of weathered basalt, altered to dark red clay but still
preserving some skeletal plagioclase laths. 2) Rounded and spherical fragments of coarsely crystalline
olivine basalt, reaching 300m in size, heavily weathered. 3) Loose crystals of iddingsite (up to 400m),
sometimes preserving the original olivine as core. Also present are a few grains of micritic limestone
and foraminiferous chalk (up to 800m), subrounded quartz up to 300m (common - few) and very few
angular (up to 400m) chert grains.
Palaeontology (LG): In the matrix: Bulimina (b), Globigerinoides (p), Globorotalia (p). Age: Neogene
to Holocene.
Firing temperature: Probably very lightly red or unred.
Geological interpretation: This tablet is petrographically similar to EA 161, 164, 169, 170, and 171 sent
by Aziru of Amurru.
V. AN UNSPECIFIED CITY IN THE NEIGHBOURHOOD OF AMURRU

EA 58 (VAT 1716), from Teu-Teshup to the King of Egypt
Reliability: Satisfactory..
Matrix: Carbonatic (marl), ochre to light tan in PPL, highly birefringent with strial b-fabric. Densely
spread (~20%), badly sorted calcite crystals and micritic bodies ranging between 10m and 80m
122
appear in the matrix. Also common are opaques, subangular to subspherical and angular quartz.
Foraminifers are sparsely spread.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) of rounded micritic and sparitic limestone up to 600m,
(dominant) and rounded quartz up to 300m (common).
Vegetal material (SLY): Small fragments of distorted tree tissue fragments, each with less than ten cells.
Palaeontology (LG): Small globigeriniids in the matrix. Only one partial test of spinose texture.
Globigerina (p), age: Neogene (?).
Firing temperature: Probably un red or very lightly red since uncharred organic material is preserved.
Geological interpretation: The clay of this tablet is similar to EA 61 (probably sent from Ardata). See
EA 61 for details.
Conclusions: EA 58 hints at its provenance in different directions. The name of its writer is Hurrian,
which is unexpected for a city located on the Phoenician coast. The introduction (Great King),
showing parallels to EA 160-161 (Amurru), EA 260 (Balu-Mer) and EA 317-318 (Dagan-takala),
also points to the north. On the other hand, Knudtzon (1915:1122) has noted that the script is similar
to that of the letters from Rib-Hadda and that the clay is quite similar to that of the Gubla letters.
In content EA 58 is close to EA 85 (the campaign of the king of Mitanni to umur) and other letters
which refer to Tushrattas campaign to the coast of Lebanon (e.g. EA 101).
At the time of the Mitannian campaign, Irqata was already in the hands of Abdi-Ashirta. Since
this letter is similar in clay (most likely Neogene marl) and inclusions to EA 61-62, and EA 156
and 159, we suggest that it originated from the area of Tripoli or east of it. As Ardata had already
been captured at that time by Aziru, Teu-Teshup could have arrived at the Egyptian centre at
Ullasa and written his letter there.
HISTORICAL CONCLUSIONS: THE AMURRU CORRESPONDENCE

The petrographic data makes it possible to trace the territorial expansion of Amurru during the time of
the Amarna correspondence through the proposed provenance of the tablets (Table 7.1).
The earliest letters, including those of Abdi-Ashirta and the early letters of Aziru, originated from
two closely-related locations: the mountainous area and the city of Ardata in the foothills, both east of
Tripoli. EA 60 and 157 were sent from the former area. We therefore suggest that in the early days of
the Amarna period the seat of Abdi-Ashirta was located in the mountains east of Tripoli, and that this
was the core area of the Kingdom of Amurru. No signicant Late Bronze Age site has been recorded
thus far in this area. Yet the letters could have been sent from a small highland stronghold possibly
the hometown and place of origin of the family. This conrms the perception that Amurru was initially a
small kingdom which was located on the slopes of Mount Lebanon, on both sides of Nahr el-Kebir, and
that it was inhabited by farmers, pastoral groups and uprooted elements (Liverani 1965; Klengel 1969:
245-253; Mendenhall 1973:130-135).
EA 61, 62, 156, 159 were sent from the city of Ardata (Tell Arde) (Salam-Sarkis 1972; 1973) which
was located in the foothills, not too far from the Egyptian centre of Ullasa (probably to be identied with
modern Tripoli). Initially Ardata was governed by its own ruler (EA 139:15; 140:12; see 75:30-31). It was
then captured by Abdi-Ashirta (EA 88:5) and held by his heir, Aziru (EA 104:10). Based on the petrographic
analysis we suggest that Ardata was the capital of Amurru during the later years of Abdi-Ashirta and the early
years of Aziru (Izreel and Singer 1990:119-120). The fact that both Abdi-Ashirta and Aziru dispatched letters
from the mountainous area as well as Ardata indicates that the Amurru correspondence in the Amarna archive
starts after the take-over of Ardata.
123
EA 161, 164, 169, 170 and 171 were sent from Tell >Arqa. It seems that, after consolidating his reign,
Aziru transferred his capital to Irqata. The city was governed by a local ruler named Aduna. He was
murdered on the initiative of Abdi-Ashirta (75:32-33; 139:15; 140:10), who took possession of the city
(EA 62:13, 17, 22; 88:6). After the death of Abdi-Ashirta the elders of Irqata sent a letter to the Pharaoh
in which they explained their deeds and expressed their loyalty to Egypt (EA 100; for interpretation of the
letter, see Moran 1992:173 n. 6). The loyalty of Irqata to Egypt is also mentioned in a letter sent by RibHadda of Gubla (Byblos) at roughly the same time (EA 103:11-13, 34-36). Shortly afterward Irqata was
seized by Aziru. This is indicated by EA 109:9-15, which says: Now the sons of Abdi-Ashirta have
taken the cities of the king and the cities of his mayors, just as they please; they are the ones that [took
Irqa]ta (sic!) for themselves. And you did nothing about their [actions] when you heard that they have taken
Ullasa (Moran 1992:183 and Liverani 1998:212 restored in line 12 [Arda]ta). The capture of Irqata by
Aziru is con rmed by the letters of Ili-rapi, Rib-Haddas successor at Gubla (EA 139:15; 140:10).
EA 171, one of Azirus early letters, was sent from Irqata. Yet, his other early letters (EA 156, 159) were
sent from Ardata. This may indicate that Aziru moved his seat to Irqata only in a later stage in his reign.
Irqata served as Azirus capital until the end of the Amarna period. Letter 170, sent by Azirus
brother and son while he was in Egypt, was dispatched from this city. EA 161 probably Azirus latest
letter written after his return from Egypt and his conquest of Tunip was also sent from Irqata.
According to the famous Generals letter Irqata functioned as the headquarters of the army that
held the territory of the Kingdom of Amurru against an impending attack of Egyptian troops (Izreel and
Singer 1990:117-121). Ardata was probably an advance post of the Generals troops, thus located near
the southern border of Amurru. The letter was written either not long after the end of the Amarna period,
thus reecting the post-Amarna stage in the political development of the Kingdom of Amurru, or after
the battle of Qidshu (ca. 1275 BCE) (Dietrich 2001).
The archaeology of Tell >Arqa no doubt the location of ancient Irqata presents a problem.
The Late Bronze Age II-III, including the Amarna period, is characterized by a signicant decline, with
no major architectural activity attested (Thalmann 2000:70). This problem is not unique to Tell >Arqa.
Many of the south Levantine mounds, including important city-states such as Gezer and Lachish, have
not yielded signicant 14th century BCE remains.
EA 165, 166 and 167 were sent by Aziru from umur (contra Izreel and Singer 1990:138; Singer
1991:152, who proposed that they were sent from Tunip). These letters all represent one event in the
history of Amurru. The paucity of letters sent from umur indicates that although Aziru conquered
the city, he avoided turning it into his permanent seat. He might have arrived there to meet Hatip, the
Egyptian messenger, and on that occasion wrote the three letters under discussion. Soon afterwards he
went to Egypt, accompanied by Hatip, probably on a ship sailing from a port near umur (see EA 168).
Aziru might have visited umur on other occasions (e.g. the situation described in EA 161:11-16), but
his letters were written in another place (i.e., Irqata).
No Amurru letter was sent from Tunip. Although Aziru captured the city, at least in the Amarna period
it did not serve as his capital. The identication of Tunip with Tell >Asharneh may help explain this, as the
city was too remote from the major arena in which Aziru operated. For military and economic reasons the
coast of Lebanon remained his main concern. The Generals letter fully illustrates this.
In summation, the petrographic analysis proves to be an indispensable tool for analyzing certain
aspects of the history of Amurru that cannot be approached by conventional historical research. It enables
us to establish the sequence of political centres of the kingdom of Amurru and its development from a small
highland entity with its centre in the highlands to a large territorial kingdom with a capital near the coast.
124
TABLE 7.1: PETROGRAPHIC GROUPING OF THE AMURRU LETTERS
EA
Matrix
Inclusions
LB
YB
AM
CT
GQ
LS
MS
***
Senders
VG
VM
QZ
60
LCC
157
LCC
58
NGM
***
61
NGM
***
62
NGM
156
NGM
159
NGM
100
NGM
***
**
67
NGM
**
**
161
NGM
164
NGM
@?
169
NGM
***
***
170
NGM
171
NGM
***
165
QCC
***
***
166
QCC
167
QCC
***
***
168
LSP
**
***
SP
Location
Mountains east
of Tripoli
**
Ullasa
**
Ardata
***
***
***
SF
**
*
*
Irqata
@
*
@
**
**
**
umur
@
*
**
*
**
**
***
Gaza
LEGEND
Matrix types: LCC = Lower Cretaceous clay. PLM = Paleogene marl. NGM = Neogene marl. QCC = Quaternary coastal
clay. LSP = Loess soil of southern Palestine.
Inclusion types: LB = Lower basalt (Cretaceous), YB = Younger olivine basalt and dolerite (Miocene Pleistocene) and
its derived minerals, AM = Amphiroa algae fossils (Pleistocene Holocene), CT = chert, GQ = geode quartz,
LS = limestone and calcite, MS = mollusc shell fragments, VG = volcanic glass, VM = vegetal material, QZ
= quartz, SF = shale fragments, SP = serpentinized minerals.
Frequency: *** dominant, ** frequent, * scarce, @ undetermined (SPA samples).
125
CHAPTER 8
THE LEBANESE BEQA >
The Beqa> letters fall into three categories:

1. Letters from specic cities.
2. A group of identical letters, probably written by the same scribe (which includes two tablets from
cities of the rst category).
3. A group of letters of unspecied locations, possibly in the Beqa>.
I. ENISHASI
Enishasi is a city in the Lebanese Beqa> which is mentioned in the topographical lists of Thutmose III
(No. 5) and Amenophis III (CN 11), and in two Amarna letters (EA 187:12; EA 363:4 (see Section IV
below). The following description appears in Papyrus Anastasi I (ANET: 477a): You have not gone
forth to Qidshu and Tubiu. You have not gone to the region of the Shasu (>3 n 3-sw) with the bowmen
of the army. Weippert (1970:265) and Rainey (1975b:14-15) suggested rendering the latter >en-3-sw
and translated You have not gone to >Ain-shasu, identifying the toponym with the Enishasi of the
topographical lists and Amarna letters (cf. Ahituv 1984:57-58). However, the rendering of the text is
uncertain and indeed, Weippert later (1974:273) abandoned it (for Enishasi and its location see Weippert
1970; Rainey 1975b; Naaman 1988c:188-190).
CATALOGUE
EA 187 (BM 29860), from Shatiya to the King of Egypt
Reliability: High.
Matrix: Yellowish grey tan to yellowish tan in PPL, with high calcitic component and speckled b-fabric. Dark
reddish-brown and opaque iron oxide stains are common. The silt (around 5%) consists almost entirely
of quartz with occasional epidote, plagioclase, and muscovite.
Inclusions: Frequent grains of rounded chalk (sometimes foraminiferous) and limestone (commonly
micritic, rarely sparitic) ranging from below 100m up to 1.5 mm. Opaque to nearly opaque
spherical to oval ferruginous ooliths (up to 350m) which rarely preserve any internal structure are
common as is subangular to subrounded quartz (up to 550m).
Vegetal material (SLY): Several non-indicative tissue fragments. Further sampling indicated common
vegetal tissues, up to 750m lengthwise, including a piece of a plant which includes a proto and
meta xylem vessel member with helical cell wall thickening. It is probably a fragment of a very
young stem. Other plant tissues are non-indicative. No further identication is possible.
Firing temperature: Probably unred since the vegetal material is uncharred.
Geological interpretation: While the matrix of this tablet is not different from that of most of the Beqa>
letters, several components of the inclusions suggest marl or rendzinal soil that developed on the
marine calcareous formations of the Lower Cretaceous lithological section. A typical fingerprint
126
of the Lower Cretaceous limestone and marl of the Hammana Formation (Walley 1997:98-99) can
be found in the ferruginous ooliths which are not known as a major component from any other
part of the lithostratigraphic column of the Levant. This formation (marked C2b and C3 on the
geological maps of Lebanon) includes a varied sequence of carbonates, marls and terrigenous
sands, including the three recognized lithostratigraphic units (after Dubertret): the Couches
Orbitolina, Blanc de Zumoffen and Couches Knemiceras. This formation is broadly exposed on
the ridges bordering the Beqa >.
Conclusions: The typical clay type, widely distributed in the Lebanese mountains and in certain areas on
the edges of the Beqa> Valley, makes the exact provenance determination impossible. The location
of the city, perhaps in the hilly forested areas bordering the Beqa>, may explain why the Egyptian
scribes etymologized the Canaanite name Enishasi and rendered it Ain-shasu, after the Shasu, the
common name in the Egyptian sources for the Asiatic nomads who must have wandered in these
hilly areas (Ahituv 1984:58; Naaman 1988d:189).
II. ASI
asi is identied with Tell izzin in the Lebanese Beqa>, 10 km south of Baalbek (Klengel 1970:61-62).
Three letters of asi are known: EA 185-186, described here, and EA 175, discussed in a different subchapter below. For the surveys conducted at the site, see Galling 1953; Kuschke 1958:99-101; Weippert
1970:262 n.10; Marfoe 1995:241.

Tell izzin is situated on the colluvial and alluvial deposits of Nahr el-Li ani, at the lower part of the
Beqa> Valley. The valley is bordered on both sides by Neogene lacustrine deposits, and on Eocene
nummolithic chalks and Senonian marl and chalk (Dubertret 1950b). Geologically, Tell izzin and Tell
ashbe are found in a similar environment. Indeed, the aerial distance between the two sites is about
5 km with the Li ani river running between them.
EA 185 (VAT 1725), from Mayarzana of asi to the King of Egypt
Reliability: High.
Matrix: Yellowish tan to ochre in PPL, brous with high calcitic component and speckled b-fabric.
Badly preserved foraminifers and their fragments are spread in the matrix, as well as dark reddishbrown and opaque iron oxide stains. A few small glauconite globules are sparsely spread in the
matrix. Fine silt (around 1%) includes almost entirely quartz with a few pyroxenes, twinned
plagioclase and biotite. A few bodies of Nile silt (see EA 163, Chapter 3.I) appear together with the
main sample. These are most likely related to post depositional inlling of a void within the tablet
that was included in the sample.
Inclusions: Sparsely spread sand of rounded quartz (up to 400m) and rounded micritic and sparitic
limestone (up to 600m). Also present are a few idiomorphic calcite spars (up to 220m).
Vegetal material: Voids left after some vegetal material that vanished by ring, often leaving phytoliths
in the voids.
Firing temperature: Probably lightly red due to the disappearance of the vegetal material.
Geological interpretation: As EA 175 (below), but red to some extent.
127

Reliability: High.
Matrix: Tan to reddish tan in PPL, silty (7%) with speckled b-fabric. Small glauconite globules (altered
to deep red by ring) are sparsely spread throughout the matrix. Foraminifers and their fragments
are sparsely spread in the matrix, as well as dark reddish-brown and opaque iron oxide stains.
Ferruginous ooliths up to 230m in size are scattered in the matrix. The silt is dominated by quartz
with secondary mica laths, plagioclase, and hornblende.
Inclusions: Sand (f:c ratio{0.062mm}=90:10). The dominant component is badly sorted, usually subrounded
to rounded micritic limestone with calcite spars in the ner fraction, ranging between 500m and to
1.5mm in size. Subrounded to subangular quartz (up to 500m) is frequent. Some grains are angular
and very rich in mineral and liquid inclusions, probably representing quartz from druzes. There are
also a few subrounded to subangular aggregates of chalcedony (up to 600m), smoky yellow in PPL,
most likely from druzes.
Vegetal material (SLY): Common, up to 1mm long. A) Non-indicative tissue fragments. B) Part of a seed
including the coat and endosperm. C) A small tissue fragment with many primary xylem elements
with their typical spiral secondary cell wall thickening.
Firing temperature: Probably unred or very slightly red since the begetal material is uncharred.
Geological interpretation and conclusions: This tablet is different from the other asi letters. Its matrix
represents terra rossa soil, with inclusions made of the parent hard limestone.1 The frequency of
quartz with abundant inclusions and chalcedony aggregates indicates the presence of druzes in this
rock. Based on southern Levantine equivalents, it may be established that the parent rock was hard
limestone of Cenomanian age. Such lithology is widely distributed and therefore cannot serve for
re ned provenance determination.
III. GUDDASHUNA
Guddashuna is mentioned in the topographical list of Thutmose III (No. 4) and in EA 177. It should be
a locality in the Lebanese Beqa> (Klengel 1970:63). Kuschke (1958:92, 101) suggested locating it at the
mound of Jeditheh near Shtorah, but the identication remains uncertain.
CATALOGUE
EA 177 (VAT 1684), from Yamiuta to the King of Egypt
Reliability: High.
Matrix: Carbonatic with rare foraminifers, greyish-tan in PPL, with speckled b-fabric. Opaque minerals
are infrequent (below 1%), sizing up to 40m. Quartz silt is uncommon (below 1%) and is
accompanied by rare zircon and epidote.
Inclusions: Moderately sorted sand (f:c ratio {0.062mm}=90:10) comprising predominantly subangular to
rounded quartz grains (up to 250m), sometimes with undulose extinction and mineral inclusions.
Rounded micritic, and rarely sparitic, limestone (up to 450m) is common. There are also a few
1. See EA 64 below for detailed discussion on terra rossa soils as ceramic materials.
128
angular calcite grains up to 600m in size. A very rare single angular crystal of iddingsite (100m)
tending to split along the {010} cleavage plane is present.
Vegetal material (SLY): Rare non-indicative tissue fragments.
Firing temperature: Probably unred due to the preservation of uncharred vegetal material.
Geological interpretation and conclusions: The matrix is most likely Paleocene marl (see EA 51 for
details). The singular iddingsite crystal, split along the {010} cleavage plain, is similar to those
that are abundant in EA 193 of Labana (above). Since this feature is quite exceptional within the
Amarna tablets examined by us, it may indicate some relation between the two localities, where
Labana is closer to this basalt source. However, this interpretation is highly speculative and the
other petrographic details of this tablet are too general for any specic provenance determination.
IV. A GROUP OF IDENTICAL LETTERS SENT BY FOUR BEQA> RULERS

EA 174-176 and 363 form an assemblage of letters from cities in the Beqa>, all containing an identical
text indicating that they were either written at about the same time in one place, or that a single scribe
traveled between several cities in the Beqa> (cf. Chapter 11). Petrographic analysis is the ideal tool to
decide this matter.2 EA 173 deals with the same subject although it does not follow the same formula.
Petrographically it may be attributed to this group.
CATALOGUE
EA 174 (VAT 1585), from Bieri of ashabu to the King of Egypt
Reliability: High.
Matrix: Light yellowish tan to greyish tan in PPL, with high calcitic component. Foraminifers scarcely
appear in the matrix. Dark reddish-brown and opaque iron oxide minerals are relatively common
(~3%). Fine silt (around 2%) which is almost entirely quartz with a few muscovite, hornblende and
zircon grains.
Inclusions: Rounded grains of micritic limestone (up to 800m), often containing sparsely spread ne
quartz silt is dominant. Subangular to rounded grains of quartz (up to 400m) are frequent and there
are a few rounded sparitic limestone (up to 800m) and subrounded single calcite spars.
Vegetal material (SLY): Few to common, including: A big chunk of tissue with rhombic calcium oxalate
crystals in dense groups, also showing a part of a ray. The tissue resembles a piece of oak stem, indicating
that this is probably a fragment of the bark. Another, non-indicative tissue fragment, is also apparent.
Firing temperature: Probably unred or very slightly red judging by the preservation of uncharred
vegetal material.
Geological interpretation and conclusions: ashabu is identied at Tell ashbe, in the Lebanese
Beqa>, about 16 km southwest of Baalbek (Klengel 1970:62). The site was surveyed by Kuschke
(1954:120-122; 1958:95-96) and Marfoe (1995:240-241). Tell ashbe is situated on the Neogene
lacustrine deposits of Zale, bordered by Eocene nummolithic chalk and Senonian marl and chalk.
A short distance away lies the alluvial valley of the Beqa>, where on both sides of Nahr el-Liani
colluvial and alluvial soils are deposited (Dubertret 1950b).
2. Unfortunately EA 176 was not available for examination during our visits to the British Museum.
129
Petrographically EA 174 reveals very little concerning its provenance. The matrix is probably
made of Paleocene marl containing some rounded quartz grains and limestone fragments. Therefore
it suits the environs of Tell ashbe, although this composition is widespread along the eastern side
of the Beqa> from Marj >Ayyun to the north of Baalbek.
EA 175 (VAT 1588), from Ildayyi of asi to the King of Egypt
Reliability: High.
Matrix: Yellowish grey to bright yellowish tan in PPL, with high calcitic component and speckled bfabric. Badly preserved foraminifers and their fragments are spread in the matrix, as well as dark
reddish-brown and opaque iron oxide stains. Fine silt (around 2%) consists almost entirely of quartz
with a few hornblende and muscovite grains.
Inclusions: Rounded grains of chalk or nari (up to 1.5 mm) often containing sparsely spread ne quartz
silt are dominant. There are a few usually subrounded quartz grains (up to 500m) and subrounded
single calcite spars (up to 220m).
Vegetal material (SLY): A fragment of a piece of cork from the bark of a tree as well as non-indicative
tissue fragments.
Paleontology (LG): Few unidentied small foraminifers in the matrix. Briozoa, non-diagnostic regarding age.
Firing temperature: Seemingly unred or very slightly red judging by the preservation of uncharred
vegetal material.
Geological interpretation: This tablet is similar in most details to EA 174 and likewise reveals very little
concerning its provenance. Therefore it is practically useless for re ned provenance determination.
EA 363 (AO 7097), from Abdi-Risha of Enishasi to the King of Egypt
Matrix: Yellowish grey to greyish tan in PPL, with high calcitic component and speckled b-fabric.
Opaque iron oxide stains are common. The silt (around 2%) is almost entirely quartz with a few
muscovite laths.
Inclusions: Rounded to subrounded, commonly micritic, rarely sparitic limestone, up to 350m in this
small sample. A single ferruginous oolith (as in EA 187), subangular to subrounded quartz (up to
300m), subangular to subrounded partially weathered tuff (up to 480m) and a grain of olivine
basalt make up the assemblage. There is also a 380m long fossil mollusc shell fragment.
Vegetal material (SLY): Several non-indicative tissue fragments.
Firing temperature: Probably unred since uncharred vegetal material is preserved.
Geological interpretation and conclusions: As EA 187 of Enishasi, with some volcanoclastic component.
Reliability: High.
Matrix: Silty (10%), carbonatic, yellowish-tan in PPL, optically active with very weak optical
orientation. The silt is essentially made of quartz, but it also contains recognizable quantities of
accessory heavy minerals, including hornblende, augite, zircon, plagioclase, biotite, muscovite,
epidote and rarely tourmaline and garnet. Opaques are relatively common (3%), measuring
between few micrometers and about 60m.
130
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal material (f:c ratio{0.062mm}=93:
7). Rounded to subrounded quartz (up to 430 m), sometimes with undulose extinction, sometimes
polycrystalline, rarely with mineral and liquid inclusions, is dominant. A grain of plagioclase
feldspar appears in the same grain size.
Vegetal material (SLY): A large piece of a soft plant tissue, showing many thin-walled parenchyma cells and
many primary xylem vessel members with their spirally arranged secondary cell walls. It is probably a
dicotyledon.
Firing temperature: Probably unfired since vegetal material is uncharred.
Geological interpretation and conclusions: EA 173 is petrographically similar to EA 174-175. The text
deals with the same matter as EA 174, 175 and 363, and although it does not follow their formulaic
manner, it may be attributed to this assemblage.
V. LETTERS OF UNSPECIFIED LOCATIONS, POSSIBLY FROM THE BEQA> VALLEY

Reliability: High.
Matrix: Yellowish grey to bright yellowish tan in PPL, with high calcitic component and speckled bfabric. Badly preserved foraminifers and their fragments are sparsely spread in the matrix, as well
as dark reddish-brown and opaque iron oxide stains. Fine silt (around 2%) composed almost entirely
of quartz with a few biotite, pyroxene, zircon, hornblende and iddingsite grains.
Inclusions: Very few (f:c ratio{0.062mm}=99:1) subrounded quartz (up to 190m) and idiomorphic calcite (up
to 120m) grains.
Firing temperature: Probably unred or very lightly red but there are not enough indicators to determine
the temperature.
Geological interpretation and conclusions: Similar to EA 174 and 175, revealing very little concerning
its provenance. Therefore it is practically useless for rened provenance determination.
EA 178 (VAT 1677), from ibiya to an Egyptian ofcial
Reliability: High.
Matrix: Yellowish grey to bright yellowish tan in PPL, with high calcitic component and speckled bfabric. Dark reddish-brown and opaque iron oxide stains are sparsely spread in the matrix. Fine
silt (around 2%), which includes almost entirely quartz with a few plagioclase, hornblende and
muscovite.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=97:3) subspherical to subrounded quartz (up to 250m).
Vegetal material (SLY): Non-indicative very small tissue fragments.
Firing temperature: Probably unred since uncharred organic material is preserved.
Geological interpretation and conclusions: See EA 179.
131
EA 179 (VAT 1703), from the deposed ruler of Tubiu to the King of Egypt
Reliability: High.
Matrix: Yellowish grey to bright yellowish tan in PPL, with high calcitic component and speckled bfabric. Badly preserved foraminifers and their fragments are sparsely spread in the matrix, as well
as dark reddish-brown and opaque iron oxide stains. Fine silt (around 2%) which is almost entirely
quartz with a few plagioclase, hornblende and oxyhornblende and muscovite grains.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=97:3) rounded grains of chalk (up to 1 mm), sometimes
foraminiferous, and subrounded quartz (up to 250m).
Insect: Single (cross-sectioned) part from a small beetle measuring 850m lengthwise.
Firing temperature: Probably unred since the insect is uncharred.
Geological interpretation and conclusions: Tubiu is mentioned in the topographical list of Thutmose
III (No. 6), in EA 179 and in Papyrus Anastasi I (Klengel 1970:62-63). It also appears in several
biblical texts (Gen 22:24; 2 Sam 8:8; 1 Chr 18:8). In Papyrus Anastasi I it is mentioned together with
Qidshu and therefore it was probably located in the northern Beqa>. The proposed identications
with Tell Siran, or Tell Deir Zenun in the central Beqa> (Kuschke 1954:115-117; Weippert 1970:
262 n. 16 and the map on p. 270) are uncertain. In any event, the writer was deposed from his throne
by his brother and wrote his letter from a neighbouring city. Thus, the location of Tubiu cannot be
established from the content of EA 179.
Petrographically, this tablet gives no specic indication as to its provenance. Since EA 178 and
179 are chemically and petrographically identical, it is possible that Tubius deposed ruler ed to
ibiya, probably his neighbour, and sent this letter from his city. This situation has two parallels
in the Amarna correspondence: Yashdata of Ta>anach was deposed from his city, found shelter at
Megiddo and wrote a letter (EA 248) from there. Rib-Hadda of Byblos left his city, found shelter in
Beirut and wrote his latest letters (EA 136-138) from that city.
ibiya was probably the ruler of the city of [Z]ayitanu (or [A]yatanu, or Yatanu) (line 5). The
place was possibly called after the olive trees grown in the area around it. ibiya was ordered by
the Pharaoh to assemble grain (lines 12-14) and reports to the Egyptian ofcial that there is much
grain in the land of Amurru (lines 20-24). This might indicate that ibiyas city was located not far
from Amurru.
CONCLUSION: THE LEBANESE BEQA> CORRESPONDENCE

The tablets of ve city-states located, according to the documentary evidence, in the Beqa> (ashabu, asi,
Enishasi, Guddashuna and possibly Zayitanu) were analyzed petrographically. The analysis has con rmed
their attribution to the Beqa> although it failed to pin-point their exact location within this region.
Examining the natural data, surveying the area and comparing settlement patterns from earlier and
later periods, Marfoe (1979:12-14) reconstructed the number and size of political units in the Beqa> in the
Late Bronze Age. In his opinion the region was fragmented into a number of small, virtually autonomous
petty kingdoms or city-states, resulting for the most part in a fairly evenly spaced distribution of city
territories that averaged about 50 km2 in area. Distances between major centres varied between 5 and 25
km, but the average (particularly in the centre of the valley) was slightly less than 10 km. Consequently,
the radius of a city domain was generally somewhat less than 5 km ... Late Bronze Age Tell ashbe and
Tell izzine, which both can be identied within the same textual contexts (EA 174-175) as aabu and
132
azi, are no more than 5 km apart. The countryside of each urban centre included about half a dozen
villages and hamlets (Marfoe 1979:14).

The number of city-states known from the written sources is about one-half of the number estimated
by Marfoe, for example. Tablets of some centres, e.g. Tushulti mentioned in EA 185-186 and Shazaena
mentioned in a letter unearthed at Kamid el-Loz (Edzard 1970:56-57) are missing in the Amarna
archive.3 Theoretically, the petrographic analysis could indicate the origin of the extant tablets, while
the overall number of city-states could be established only by the combination of the petrographic,
textual and archaeological evidence. According to Naaman, such a situation also exists in the case of the
Galilean correspondence (Chapter 12). However, in terms of their petrographic traits, the Beqa> letters
depict a surprisingly homogeneous pattern.
The Beqa> Valley extends over 150 km from southwest to northeast. It is bordered by varied
lithologies, ranging between the Jurassic and Quaternary, including volcanics, marine and continental
sedimentary rocks, as well as lacustrine and alluvial deposits. In such a broad area, with varied
lithology, at least some variability could be expected in the petrography of products that come from
different locations. Yet most of the Beqa> letters seem to represent closely related materials, a pattern
that usually indicates a common origin. This implies that most if not all of the Beqa > tablets may share
a common provenance. The most logical explanation for this phenomenon is that most, if not all, the
letters were sent from a central site which served the local city-state rulers who could not afford their
own scribes. Given the limited size and resources of the Beqa> city-state, this hypothesis is plausible.
The only option that comes to mind is that this was the Egyptian administrative centre at Kumidi.
Kumidi was unique among the Egyptian administrative centres in the Levant in that it was
governed by both a Canaanite ruler (as indicated by EA 198) and an Egyptian high official. It should
doubtless be identified with the present-day mound of Kamid el-Loz in the southeastern Beqa>, halfway
between Jibb Jenin and Sultan Yaaqub et-Tata. Geologically, the site is located in an area where
Eocene chalk and rendzina soils are exposed. Paleocene marls, bordered by a series of Cenomanian
limestones outcrop a short distance to the east, at Sahel el-Faluj. In the same area and further east, a set
of Lower Cretaceous to Jurassic shales and limestones are exposed. Hence petrographically, Kumidi
could be the source of the lions share of the Beqa> letters.
In our discussion on the letters from Mushi una (Chapter 10: EA 182-184) we refer to their
petrographic similarity to most of the Beqa> letters, a fact already observed by Knudtzon (1915:1278 n. 1).
It is possible that the scribe(s) of Kumidi also served that city, which can therefore be identified in the
area between Damascus and the Lebanese Beqa >.
3. Naaman (1988d:190) raised the possibility that Shazaena is a mistaken writing (metathesis) of Enishazi and that KL 69:279
was sent to the ruler of Enishazi = Enishasi. This, of course, is no more than a hypothesis and cannot be veried.
133
CHAPTER 9
THE LEBANESE LITTORAL
I. BYBLOS (GUBLA)
The city of Gubla is identied with biblical Gebal, Byblos of the classical periods, situated in Jbail on
the northern coast of Lebanon.

A collection of thin sections which we prepared from Early and Middle Bronze Age pottery from Byblos,
now deposited in the Muse du Louvre, is the only reference material at our disposal for comparative
research. The Middle Bronze Age samples (Tell el-Yahudiyeh vessels) appeared to be comparable
with the petrographic data from the main group of tablets, representing Neogene marine clay (below).
However this data cannot be considered to be denitive because, theoretically, the vessels could have
been imported to the site from another region. Hence our petrographic interpretation relies mainly on
data retrieved from 1:50,000 scale geological maps of the Jbail area (Dubertret 1951a; 1956) and other
geological reports (Fevret and Sanlaville 1966; Tejirian 1970; Sanlaville 1977).
The ancient mound of Byblos is located in an area where Neogene to recent marine sediments
were deposited. These include a series of chalks, clays, marls and sand dunes. There are no detailed
mineralogical analyses of the clay and marl deposits, nor any micropalaeontological data that would
be comparable with the foraminifer suite that appears in the thin sections. However, as Walley (1997:
103) states, the entire Neogene marine sequence in Lebanon is equivalent to strata from Israel that were
studied by Ben-Gai et al. (1996).
In terms of CTF, another source of clay does exist within the exploitable range of Byblos. This is
the Lower Cretaceous shales deposit which appears at Nahr Ibrahim, about 7 km southeast of the mound,
but from a practical point of view is too far away for exploitation. Indeed, no evidence for the use of this
clay has been traced in the Rib-Hadda tablets.
CATALOGUE
EA 72 and 81 are described rst, since they served as reference for one of the main fabrics of the RibHadda letters. This is due to the large sample taken from them which enabled micropalaeontological
identication of the clay.
EA 72 (VAT 1712), from Rib-Hadda (?) to the King of Egypt
Reliability: High.
Matrix: Carbonatic, yellowish-tan in PPL, containing sparsely spread (~2%), well-sorted calcitic bodies
and less common calcite crystals sizing between 10m and 50m. The calcite crystals are commonly
idiomorphic or subidiomorphic at the ner fraction but the calcitic bodies tend to become rounded in
the coarser. Foraminifers appear (about 3%-4%). Quartz silt appears at about 2%. Occasional heavy
minerals appear in the silt fraction and include zircon, oxyhornblende, and twinned plagioclase.
134
Fig. 9.1: Main city-states and Egyptian administrative centres in northern Canaan.
135
Opaque to reddish-tan iron minerals (~2%) that appear at a range of sizes from few micrometers to
about 30-40m, the opaques of which are angular and the translucent minerals tend to be spherical
and rounded.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=97:3) sand made up of frequent grains of subrounded
to rounded quartz sometimes containing mineral inclusions and very rarely polycrystalline (up
to 350m). Rounded micritic limestone (up to 300m) is common and there are a few angular
fragments of anhydrite (up to 400m).
Palaeontology (LG): Some globigeriniids with spinose texture, fossils in the matrix. These include:
Globigerinoides (p), Globoquadrina (p). Age: Neogene (?).
Firing temperature: Unred, as anhydrite decomposes at 1500C.
Geological interpretation: EA 72 represents a large group of tablets from Byblos. The micro-palaeontological
examination indicates that Neogene marl was used as matrix, with the addition of occasional inclusions
that commonly contain limestone and quartz, sometimes with organic materials. Sporadic fossils of
Amphiroa corallinean algae appear in several cases (i.e., in EA 78, 79), indicating a recent coastal origin
(see Azirus EA 165 for full discussion of the role of this directing fossil in provenance determination). In
the Levant the exposure of marine Neogene marl is rst and foremost a coastal Lebanese phenomenon.
In this general area, it appears in restricted areas north of Sidon, in Beirut, near the Junyah Bay, in Byblos
and from Tripoli throug the >Akkar Plain. In the latter, it is commonly accompanied by volcanic rock
fragments and mac minerals (see above EA 100 which was sent from Irqata). Therefore a provenance
at Sidon, Beirut or Byblos could be suggested even regardless of the textual evidence.
Matrix: Clayey-carbonatic, ochre-yellow in PPL, optically active groundmass with speckled b-fabric. Few
foraminifers appear (~1%). Opaque minerals (>1%) appear as angular to subangular bodies, from
a few micrometers to 40m in size. Dark red to nearly opaque haematite particles (>1%) appear as
rounded bodies, sometimes dispersed into the matrix around them and rarely as inllings within
foraminifers. Calcite crystals and micritic calcitic bodies (~10%) ranging between 10m and 120m
appear in the matrix. The calcite crystals are rarely idiomorphic, but most of these bodies are micritic
and rounded. Quartz silt is very scarce (>1%). It is accompanied by a very few plagioclase particles.
Inclusions: Sparse mixture of minerals and rock fragments (f:c ratio{0.062mm}=98:2). No remains of vegetal
material were traced. Chalk is dominant as rounded to subrounded grains with foraminifers. They
are usually up to 400m, but a single 5 mm grain was included in the sample. Rounded to subangular
quartz (up to 200m) is frequent.
Palaeontology (LG): Globigerina (p), Globorotalia (p), Lenticulina (b). Age: Neogene.
Firing temperature: There seems to be no evidence for ring in this sample but there are not enough
indicators to determine this beyond doubt.
Reliability: Fair.
Matrix: Carbonatic, light reddish-tan in PPL, birefringent groundmass with speckled b-fabric containing
sparsely spread (~3%), well-sorted calcitic bodies and less common calcite crystals sizing between
136
10m and 50m. The calcite crystals are commonly idiomorphic or subidiomorphic at the ner
fraction but tend to become rounded sparite in the coarser. Foraminifers appear (about 2%).
Quartz silt appears at about 1%-2%. Opaque to reddish-tan iron minerals (magnetite and haematite
respectively, ~2%) appear in a range of sizes from few micrometers to about 30-40m. The opaques
are angular and the translucent minerals tend to be spherical and rounded.
Inclusions: Only a few inclusions appear in this SPA sample. They are rounded micritic limestone (up to
150m) and rounded quartz grains (up to 200m).
Firing temperature: Perhaps unred but undetermined at this meagre sample.
Geological interpretation: As EA 72 and 81.
Reliability: High.
Matrix: Carbonatic, light tan in PPL, birefringent groundmass with speckled b-fabric containing sparsely
spread (~5%), well-sorted calcitic bodies and less common calcite crystals sizing between 10m
and 60m. The calcite crystals are commonly idiomorphic or subidiomorphic at the ner fraction
but tend to become rounded sparite in the coarser. Very few green glauconite spheres appear in
the matrix. Foraminifers are scarce. Quartz silt appears at about 1%. Opaque to reddish-tan iron
minerals (~2%) appear at a range of sizes from few micrometers to about 30m. The opaques are
angular and the translucent minerals tend to be spherical and rounded.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) composed of subrounded to subangular quartz (up to 380m)
which is frequent to common. Rounded micritic limestone grains (up to 650m) are common. There
are a few grains of rounded replacement chert (up to 550m), idiomorphic calcite crystals (350m)
and insect keratin particles (up to 800m) (See EA 95).
Firing temperature: Unred, judging by the preservation of uncharred insect fragments and the lack of
colour changes in the glauconite.
EA 71 (VAT1632), from Rib-Hadda to the Egyptian ofcial aya
Matrix: Carbonatic, yellowish-tan in PPL, containing sparsely spread (~2%), well-sorted calcitic
bodies and less common calcite crystals sizing between 10m and 50m. The calcite crystals
are commonly idiomorphic or subidiomorphic at the ner fraction but the calcitic bodies tend to
become rounded in the coarser. Foraminifers are common (about 3%-4%). Quartz silt appears at
about 3%. Occasional heavy minerals appear in the silt fraction and include zircon and twinned
plagioclase. Opaque to reddish-tan iron minerals (~2%) appear at a range of sizes from few
micrometers to about 30-40m. The opaques are angular and the translucent minerals tend to be
spherical and rounded.
Inclusions: Sand composed of subrounded to rounded quartz, sometimes containing mineral inclusions
and, very rarely, polycrystalline, together with rounded micritic limestone.
Vegetal material (SLY): Partly lignied plant tissues (straw).
137

Matrix: Carbonatic, yellowish-tan in PPL, containing spread (~5%), well-sorted calcitic bodies and
less common calcite crystals sizing between 10m and 50m. The calcite crystals are commonly
idiomorphic or subidiomorphic in the ner fraction but the calcitic bodies tend to become rounded
in the coarser. Foraminifers are scarce. Quartz silt appears as about 2%. Occasional heavy minerals
appear in the silt fraction and include zircon, hornblende and epidote. Opaque to reddish-tan iron
minerals (~2%) appear in a range of sizes from few micrometers to about 30-40m. The opaques are
angular and the translucent minerals tend to be spherical and rounded.
Inclusions: Sand (f:c ratio{0.062mm}=95:5) made up of rounded micritic or sparitic limestone (up to 1mm),
rounded quartz (up to 300m) and idiomorphic calcite (up to 750m).
Firing temperature: Probably un red or very lightly red although this sample lacks indicators for
further determination.
Matrix: Carbonatic, ochre to reddish-tan in PPL, containing sparsely spread (~2%), well-sorted calcitic
bodies and calcite crystals sizing between 10m and 50m. The calcite crystals are commonly
idiomorphic at the ner fraction but the calcitic bodies tend to become rounded in the coarser.
Foraminifers are spread throughout the matrix (~4%). Quartz silt is scarce (about 1%). Very few
heavy minerals that appear in the silt fraction and include zircon and hornblende. Opaque to reddishtan iron (~1%) appear in a range of sizes from a few micrometers to about 50m. The opaques are
angular and the translucent minerals tend to be spherical and rounded. The latter appear also as
inllings of some foraminifers.
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand consisting of subrounded to rounded quartz (up to
380m), sometimes containing mineral inclusions, and rounded micritic and sparitic limestone
(up to 300m). There are also a few rounded grains (up to 300m) foraminiferous chalk with iron
mineral stains.
Firing temperature: Probably unred or very lightly red but there are not enough indicators to estimate
the temperature.
EA 77 (VAT 1635 + 1700), from Rib-Hadda to the Egyptian ofcial Amanappa
Reliability: High.
and calcite crystals sizing between 10m and 50m. The calcite crystals are commonly idiomorphic
at the ner fraction but the calcitic bodies tend to become rounded in the coarser. Foraminifers are
uncommon (1%). Quartz silt is scarce (about 1%). Very few heavy minerals that appear in the silt
fraction and include hornblende and rutile. Opaque to reddish-tan iron (~1%) appear at a range
of sizes from few micrometers to about 30-40m. The opaques are angular and the translucent
138
minerals tend to be spherical and rounded. Small bodies of charred matter (up to 50m) appear in the
matrix, perhaps crumbs from the charred vegetal material of the inclusions. Few green glauconite
concentrations appear.
Inclusions: Spread (f:c ratio{0.062mm}=94:6) sand with frequent rounded micritic and sparitic limestone (up
to 300m), a few subrounded to rounded quartz grains (up to 500m), sometimes containing mineral
inclusions, very few grains (up to 120m) of replacement chert and one (250m) rounded grain of
chalcedony. Also present is a 450m rounded coprolite without spherulites.
Vegetal material (SLY): Frequent (up to 1 mm long). Fragment of parenchymatous tissue, distorted
fragments, probably of leaves.
Firing temperature: Unred, judging by the preservation of uncharred organic material and the green
colour of the glauconite.
Reliability: High.
Matrix: Clayey, carbonatic, orange-tan to tan in PPL with scarce foraminifers. Opaque minerals are
rather common (~2%), up to 50m in size, angular in the ne fraction and subrounded at coarser
grain sizes. Red to dark red and nearly opaque haematite concentrations appear in the matrix as
single rounded bodies (up to 200m) or stains which are dispersed into the matrix. Quartz silt
(<1%) occurs together with smaller amount of heavy minerals including iddingsite and plagioclase.
The carbonate crystals within the matrix are dense (15%), usually around 10m but occasionally
reaching 20m-30m.
Inclusions: Sand (f:c ratio{0.062mm}=93:7) in which rounded fragments of fossiliferous marine limestone
(beachrock) and more commonly separate fossils derived from it (up to 800m) are frequent to
dominant. The fossils consist predominantly of the corallinean algae Amphiroa, together with mollusc
shell fragments. Rounded grains(up to 400m) of smoky to brown stained replacement chert with local
intergrowth of chalcedony, occasionally containing pseudomorphs after dolomite are frequent. Rounded
to subrounded fragments of micritic limestone (up to 870m) with localized brownish staining are
common, as are rounded to subangular quartz grains (up to 380m). There are also a few polycrystalline
fragments (up to 400m) of geode quartz with very common liquid and mineral inclusions.
Paleontology (LG): Small badly preserved foraminifers in matrix and inclusions, mostly unidentied,
Bryozoa, corallinean algae. Age: Pliocene to Pleistocene.
Firing temperature: There are not enough markers to determine the ring temperature, but slight
decomposition of the calcite suggests ring at 6000C-7000C.
Geological interpretation: EA 78 is similar in most details to EA 96, 165 and 167 (the latter two sent by
Aziru of Amurru). It should therefore be assigned to umur (see EA 165).
Conclusions: EA 78 was made at umur. This could have been mentioned in the missing part of the tablet.
Reliability: High.
bodies and less common calcite crystals sizing between 10m and 70m. Very rare little (~60m)
glauconite concentrations appear within the matrix. The calcite crystals are commonly idiomorphic
139
or subidiomorphic at the ner fraction but the calcitic bodies tend to become rounded in the coarser.
Foraminifers are few (about 1%). Quartz silt appears at about 1%. Occasional zircon appears in
the silt fraction. Opaque to reddish-tan iron minerals (~2%) appear at a range of sizes from few
spherical and rounded. The haematite appears also as dispersed stains within the matrix and rarely
as inllings within foraminifers.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=98:2) sand of predominantly subrounded to subangular
quartz (up to 380m) sometimes containing mineral inclusions. Rounded micritic limestone grains
(up to 300m) are common.
Vegetal material (SLY): Few (up to 850m), a fragment of bark of wood including abundant rhombic
crystals of calcium oxalate. No further identication is possible.
Paleontology (LG): A few fragments (up to 370m) of Amphiroa corallinean algae.
Firing temperature: There are no signs of any signicant ring in this tablet since the glauconite
preserved its green pleochroism and the organic matter is uncharred.
Reliability: Fair.
idiomorphic or subidiomorphic at the ner fraction but the calcitic bodies tend to become rounded in
the coarser. Foraminifers are common (about 3%-4%). Quartz silt appears at about 1%. Occasional
heavy minerals appear in the silt fraction and include zircon, hornblende and spinel(?). Opaque to
reddish-tan iron minerals (magnetite and haematite respectively, ~2%) that appear at a range of
sizes from few micrometers to about 30-40m, the opaques of which are angular and the translucent
Inclusions: Sparsely spread (f:c ratio could not be estimated) sand of subrounded to rounded quartz (up
to 250m), rounded micritic limestone (up to 400m) and angular anhydrite (up to 200m).
Vegetal material (SLY): Up to 250m. Minute fragments, unidentied.
Firing temperature: Unred, since anhydrite decomposes at 1500C.
Reliability: High.
Matrix: Carbonatic, tan in PPL, containing sparsely spread (~2%), well-sorted calcitic bodies and
less common calcite crystals sizing between 10m and 50m. The calcite crystals are commonly
idiomorphic or subidiomorphic in the ner fraction but the calcitic bodies tend to become rounded in
the coarser. Foraminifers appear (about 3%-4%). Quartz silt appears at about 1%. Opaque to reddishtan iron minerals (~1%) that appear in a range of sizes from few micrometers to about 30-40m, the
opaques of which are angular and the translucent minerals tend to be spherical and rounded.
Inclusions: Sand of predominantly subrounded to subangular quartz (up to 380m), sometimes containing
mineral inclusions, and rounded sparitic limestone (up to 300m). Also included was a 5 mm long
unburnt bone splinter.
140
Firing temperature: Probably unred since the bone fragment has not changed colour.
Reliability: Satisfactory to moderate.
bodies and less common calcite crystals sizing between 10m and 30m. The calcite crystals
are commonly idiomorphic or subidiomorphic in the ner fraction but the calcitic bodies tend to
become rounded in the coarser. Foraminifers appear (about 1%). Quartz silt appears at about 2%.
Opaque (angular) to reddish-tan (spherical and rounded) iron minerals appear in a range of sizes
from a few micrometers to about 30m.
Inclusions: Spread (f:c ratio{0.062mm}=90:10) sand of frequent subrounded to subangular quartz grains (up
to 420m), sometimes containing mineral inclusions. Rounded micritic limestone (up to 700m) is
common and there are a few green spheres of glauconite (up to 320m).
Firing temperature: Unred, judging by the lack of changes in the colour of glauconite.
Matrix: Clayey, ochre-yellow to greenish-yellow in PPL, optically active speckled b-fabric with strong
optical orientation. Very few foraminifers appear. Opaque minerals (>1%) appear as angular to
subangular bodies, from a few micrometers to 40m in size. Dark red to nearly opaque haematite
particles (>1%) appear as rounded bodies, sometimes dispersed into the matrix around them. Calcite
crystals and calcitic bodies (~5%), ranging between 10m and 80m appear in the matrix. The
calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and rounded. Quartz silt
is very scarce (>1%). It is accompanied by very few plagioclase particles.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2),
composed of rounded to subrounded sparitic limestone (up to 380m) and idiomorphic calcite
crystals (600m).
Vegetal material (SLY): Plant tissues (up to 700m lengthwise). A fragment with large elongated, thinwalled cells, probably a monocotyledon leaf. No further identication is possible.
Firing temperature: Unred, since vegetal material is uncharred.
EA 86 (BM 29804), from Rib-Hadda to an Egyptian ofcial named Amanappa
Reliability: High.
Matrix: Carbonatic, grey-yellowish tan in PPL, containing sparsely spread (~2%), well-sorted calcitic bodies
idiomorphic or subidiomorphic at the ner fraction but the calcitic bodies tend to become rounded in the
coarser. Foraminifers appear (about 3%-4%). Quartz silt appears at about 1%. Opaque to reddish-tan iron
minerals appear in a range of sizes from few micrometers to about 30m. The opaques are angular and
the translucent minerals tend to be spherical and rounded.
141
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand made up of subrounded to rounded quartz (up to 420m)
sometimes containing mineral inclusions, rounded sparitic biogenic limestone (up to 500m),
rounded foraminiferous chalk (up to 400m) and a single rounded grain of iddingsite (70m). There
is also a single rounded fragment (150m) of insect keratin. (See EA 95). The number of grains was
insufcient for frequency determinations.
Firing temperature: Most likely unred judging by the preservation of unred keratin.
Reliability: High. Stereoscopic examination reveals insect and plant remains as in EA 95. A seed was taken for
archaeobotanical identication but not included in the microscopically examined sample.
Matrix: Carbonatic, ochre in PPL with speckled b-fabric containing sparsely spread spherical green
glauconite concentrations. Fragments and few complete foraminifers appear (about 2%), some
are inlled by iron minerals. Quartz silt appears at less than 1%. Accessory minerals within the
silt fraction include hornblende, muscovite, and plagioclase. Opaque to reddish-tan iron minerals
appear in a range of sizes from a few micrometers to about 30m. The opaques are angular and the
translucent minerals tend to be spherical and rounded.
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand of rounded to subangular quartz (up to 220m) and
rounded replacement chert (up to 500m). The number of grains is insufcient for frequency
determinations. There are individual coarse and ne sheep wool bres. The former are 90m
in diameter and include a chambered medulla and evenly dispersed pigments. The ne ones are
translucent, 40m to 60m in diameter, with a continuous medulla.
Firing temperature: Unred due to the preservation of wool and the green pleochroism of glauconite.
Matrix: Carbonatic, ochre in PPL with speckled b-fabric containing sparsely spread spherical green
glauconite concentrations. Fragments and few complete foraminifers appear (about 2%), some
are inlled by iron minerals. Quartz silt appears at less than 1%. Accessory minerals within the
silt fraction include hornblende, muscovite, and plagioclase. Opaque to reddish-tan iron minerals
appear at a range of sizes from few micrometers to about 30m. The opaques are angular and the
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand of rounded to subangular quartz (up to 170m), rounded
replacement chert (up to 480m) and angular biosparite (1.2 mm). There are not enough grains for
frequency determinations. Mixed in are individual ne sheep wool bres, 20m to 40m in diameter
with a continuous medulla, Amphiroa corallinean algae and insect keratin fragments: (up to 300m).
(See EA 95).
Firing temperature: Unred, judging by the preservation of wool and the green pleochroism of glauconite.
Geological interpretation: EA 87 and 88 are identical and were seemingly made of the same lump of clay.
Hence they were probably sent at one and the same time: one to the Pharaoh and the other to an army
commander (Amanappa).
142

Reliability: High.
Matrix: Clayey, ochre-yellow to greenish-yellow in PPL, optically active with speckled b-fabric. Very
few foraminifers appear. Opaque minerals (>1%) appear as angular to subangular bodies, sizing
from few micrometers to 40m. Dark red to nearly opaque haematite particles (>1%) appear as
rounded bodies, sometimes dispersed into the matrix around them. Calcite crystals and calcitic
bodies (~20%), ranging between 10m and 80m appear in the matrix. The calcite crystals are
commonly idiomorphic in the smaller grain sizes and tending to become rounded in the larger.
Quartz silt is very scarce (>1%). It is accompanied by very few plagioclase particles. Other
accessory minerals are zircon, hornblende and mica.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2).
rounded to subrounded quartz is dominant (up to 890m but usually smaller), sometimes with
mineral inclusions, and rounded to subangular micritic limestone (up to 1mm) is frequent.
Vegetal material (SLY): Common (up to 600m), uncharred vegetal tissue fragments. A badly preserved
tissue fragment with a single bre, Another tissue fragment with several bres, and another with
four vessel elements.
Firing temperature: Unred, since the vegetal material is uncharred.
Reliability: High.
Matrix: Clayey, ochre-yellow to yellowish-tan in PPL, optically active with striated b-fabric and very
pronounced optical orientation. Elliptical to rounded bodies of purer clay, highly birefringent, are
sparsely spread within the matrix sizing up to 450m. Isolated foraminifers appear in the matrix.
Opaque minerals (~1%) appear as angular to subangular bodies, sizing from few micrometers to
50m. Dark red to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes
dispersed into the matrix around them. Calcite crystals and calcitic bodies (~5%), ranging between
10m and 80m appear in the matrix. The calcite crystals are commonly idiomorphic, the calcitic
bodies are micritic and rounded. Quartz silt is scarce (~1%). No heavy minerals were encountered
in the silt.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2). The
vegetal material is apparently the only added temper, as the minerals are seemingly clasts occurring
naturally within the matrix. They consist of subangular to subrounded quartz (up to 320m), rounded
to subrounded sparite (up to 750m) and idiomorphic calcite crystals (up to 200m).
Vegetal material (SLY): Plant tissues (up to 1 mm lengthwise). Straw from monocotyledon plants,
containing phytoliths, most likely Graminae = Poaceae.
Firing temperature: Unred, judging by the preservation of uncharred vegetal material.
Geological interpretation: EA 90 represents a second clay class that was used in Byblos for the
production of tablets. It is more clay-rich than EA 72, but in all other respects (palaeontology, silt,
inclusion types) they are identical. It probably represents a different deposit or facies of Neogene
marine clays. The clay type of EA 90, which dominates the Byblos tablets, is also typical of a small
group of Tell el-Yahudiyeh vessels from Byblos that we examined.
143

Reliability: High.
Matrix: Clayey, deep reddish-tan in PPL, and optically active with striated b-fabric and very pronounced
optical orientation. Elliptical to rounded bodies of purer clay, highly birefringent, are sparsely spread
within the matrix sizing up to 640m. Opaque minerals (~1%) appear as angular to subangular
bodies, from few micrometers to 50m in size. Dark red to nearly opaque haematite particles (~2%)
appear as rounded bodies, sometimes dispersed into the matrix around them. Calcite crystals and
calcitic bodies (~1%), ranging between 10m and 80m, appear in the matrix. The calcite crystals
are commonly idiomorphic, the calcitic bodies are micritic and rounded. Quartz silt is scarce (~1%).
It is accompanied by a few plagioclase and hornblende particles, reaching the ne sand category.
Inclusions: Badly sorted sand (f:c ratio{0.062mm}=95:5). Subangular to subrounded quartz (up to 450m)
is frequent, as are rounded to subrounded sparitic and micritic limestone (up to 1 mm). There is a
single angular particle of hornblende (170m).
Vegetal material (SLY): Partly to entirely lignied plant tissues (up to 1 mm long) including straw and
fewer herbal leaves.
Firing temperature: The colour change into deep reddish-tan is most likely due to the high iron but low
carbonate content of the clay. These changes occur despite the rather low ring temperature that
this tablet has undergone. It is obviously below 8000C as the hornblende clast shows no sign of
alteration. Since some of the calcite does show anomalies in their birefringence due to heating, it
may be estimated at 7000C or slightly below.
Reliability: High.
Matrix: Clayey, dark reddish-tan to dark greyish-tan in PPL, nearly isotropic. Opaque minerals (~1%)
appear as angular to subangular bodies, from a few micrometers to 50m. Dark red to nearly opaque
haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the matrix around
them. Calcite crystals and calcitic bodies (~1%), ranging between 10m and 80m, appear in the
matrix. Quartz silt is scarce (~1%).
Inclusions: Badly sorted sand (f:c ratio{0.062mm}=95:5). Subangular to rounded quartz (up to 500m),
sometimes with mineral inclusions, is frequent as is limestone (up to 650m). The latter is partially
or entirely decarbonated due to the high ring temperature, often with reaction edges in the matrix
around the particle. There is a single angular particle of hornblende measuring 170m.
Firing temperature: This tablet is an extremely highly red version of the petrofabric represented by EA 90.
The state of isotropism of the matrix and the decomposition of the carbonates indicate a ring
temperature of above 800 0C, possibly around 900 0C. It is unclear why EA 91-92 were red an
exceptional phenomenon in the Byblos correspondence.
EA 93 (VAT 1663), from Rib-Hadda to the Egyptian ofcial Amanappa
144
pronounced optical orientation. Isolated foraminifers appear in the matrix. Opaque minerals
(~1%) appear as angular to subangular bodies, sizing from few micrometers to 50m. Dark red
to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed into
the matrix around them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and
80m appear in the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies
are micritic and rounded. Quartz silt is scarce (~1%). No heavy minerals were encountered in
the silt.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=95:5) of
rounded to subrounded micritic limestone (up to 1.2 mm), sometimes with sparse quartz silt. There are
also rounded grains of quartz (up to 300m) and replacement chert (up to 380m).
Firing temperature: No evidence for ring is apparent in this sample.
pronounced optical orientation. Isolated foraminifers appear in the matrix. Opaque minerals (~1%)
appear as angular to subangular bodies, sizing from few micrometers to 50m. Dark red to nearly
opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the matrix
around them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 80m appear
in the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and
rounded. Quartz silt is scarce (~1%). No heavy minerals were encountered in the silt.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=95:5)
consisting of rounded to subrounded micritic limestone (up to 1.2 mm), sometimes with sparse
quartz silt and rounded quartz (up to 300m). Stereomicroscopic examination of this tablet reveals
a considerable amount of insect fragments. Some of these were extracted from the surface for
entomological identication. The quantity of insect fragments negates the possibility that a single
bug was accidentally included in the lump of clay before it was kneaded by the scribe. Instead, it
raises the likelihood that beetles were intentionally crushed and mixed with the clay. This oddity
appears in three other letters of Rib-Hadda EA 69, 86 and 88. Stereomicroscopic examination
suggests that the same species of beetle was used in all four tablets. We cannot suggest any
interpretation for this peculiar choice, which is meaningless technologically and hence must
represent some form of ritualistic behaviour.
Firing temperature: Unred, judging by the preservation of uncharred organic material.
145
EXCURSUS: BEETLE FRAGMENTS IN THE CLAY OF EA 95

Moshe Inbar and Vladimir Chikatunov
Several insect fragments were found and removed from EA 95, a letter of Rib-Hadda, ruler of Byblos,
to the King of Egypt. The largest fragment (Fig. 9.2a) was identied as the tibia (leg fragment) of the
darkling beetle Akis latreillei Solier, 1834 (Coleoptera: Tenebrionidae: Akidini). Additional, smaller
parts (less than 1 mm long), were found, but they appear to be parts of the same leg. Akis latreillei has
typical East Mediterranean distribution from Anatolia in the north, to Syria, Lebanon, Israel, Jordan and
Egypt in the south (e.g. Fattorini 2000). It inhabits xeric (arid) and semiarid habitats, such as sand dunes.
Unlike other beetles (i.e., Scarabaeidae Buprestidae and Elateridae), darkling had no important role in
ancient Egyptian religion and mythology (Kritsky 1991).
Little is known about the basic biology of A. latreillei. The adult beetles (Fig. 9.2b) are black and
smooth and are often ground dwellers. The larva is geophilous (living in soil). As most darkling beetles,
the adult and larvae are saprophagous, feeding on dead plant and animal matter including detritus, carrion
and dung. Interestingly, the species is predominantly synantropic; its distribution and abundance are
inuenced by anthropogenic factors. It lives in close proximity to humans, following and spreading with
human settlements and activity. Thus, it seems likely that fragments of this geophilous beetle were mixed
with soil during the preparation of the clay for the tablet. Because the species has a wide distribution in
the Levant we cannot pinpoint a specic origin of the tablet. Nevertheless, it is possible that early human
activity in the region affected the original distribution of the beetle.
Fig. 9.2: a) Leg fragment of the darkling beetle Akis latreillei from EA 95. b) Adult beetle A. latreillei.
146

Reliability: High.
Matrix: Carbonatic, greyish-tan in PPL with speckled b-fabric and foraminifers (2%). Quartz silt is
scarce (about 1%). Very few heavy minerals that appear in the silt fraction and include hornblende
and rutile. Opaque (angular) to reddish-tan (spherical and rounded) iron minerals (~2%) appear in
a range of sizes from a few micrometers to about 30-40m.
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand of subrounded quartz (up to 330m) with a few chitin
insect fragments (up to 200m).
Vegetal material (SLY): Frequent (up to 330m long), badly preserved uncharred plant tissues, rich in bres.
Conclusions: Petrography indicates that this tablet indeed belongs to the Byblos correspondence (see
Moran 1992:174).
Matrix: Carbonatic, yellowish-tan in PPL with speckled b-fabric and foraminifers (2%). Quartz silt is
about 7%, with some accessory heavy minerals including plagioclase, rutile, zircon and hornblende.
Opaque (angular) to reddish-tan (spherical and rounded) iron minerals (~2%) appear in a range of
sizes from few micrometers to about 30-40m.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) of subangular to rounded quartz: (up to 280m) and rounded
sparitic and micritic limestone (up to 650m).
Vegetal material (SLY): Common (up to 300m long), uncharred plant tissues, unidentied.
Matrix: Clayey, carbonatic, orange-tan to tan in PPL with foraminifers (~2%), many of which are inlled
with iron oxide minerals. Opaque minerals are rather common (~2%) sizing up to 100m, angular
in the ne fraction and subrounded in the coarser grain sizes. Quartz silt occurs (~1%) together with
smaller amount of plagioclase. The carbonate crystals within the matrix are dense (15%), usually sizing
around 10m but occasionally reaching 20m-30m. The larger crystals, as well as the foraminifers,
exhibit anomalous birefringence due to partial decomposition of the carbonate during ring.
Inclusions: Sand (f:c ratio{0.062mm}=~95:5) consisting of frequent rounded to subangular grains of quartz
(up to 700m) sometimes with mineral inclusions. Rounded to subrounded fragments of micritic
limestone (up to 300m) with localized brownish staining are common as are . subangular to
subrounded articulated fragments of the calcareous corallinean algae Amphiroa (up to 700m). There
are also a few ferruginous argillaceous rock fragments (up to 280m) with some quartz silt.
147
Firing temperature: Estimated at between 7000C and 8000C based on the alteration of the calcite crystals.
Geological interpretation: EA 103 is similar petrographically to EA 78, 96 and 165-167 (the latter two sent
by Aziru of Amurru), all assigned to the coastal area of the >Akkar Plain, namely to umur.
Conclusions: The petrographic data suggests that this letter was sent from umur, as indeed indicated by
the text (lines 13-16; Moran 1992:176, n. 1). This is the only case in the Amarna correspondence in
which an author explicitly states that he writes his letter from an Egyptian administrative centre.
Reliability: High.
Matrix: Clayey, dark ochre-yellow to tan in PPL, optically active with striated b-fabric and very pronounced
optical orientation. Single and very rare foraminifers appear in the matrix. Opaque minerals (~2%)
appear as angular to subangular bodies, from few micrometers to 50m in size. Dark red to nearly
opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the matrix around
them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m appear in the matrix.
The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and rounded. Quartz
silt is scarce (~1%). Scarce heavy minerals appear in the silt but range also to the sand fraction. These
include zircon and augite.
Inclusions: A mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=90:10). The vegetal
material is apparently the only added temper as the minerals seem to be clasts occurring naturally within
the matrix. They are subrounded to rounded quartz (up to 380m), rounded fragments of unidentied
fossil coralline algae (up to 350m) and rounded calcite (up to 200m).
Vegetal material (SLY): Unidentiable plant tissues (up to 400m long).
Reliability: High.
Matrix: Carbonatic, greyish-tan in PPL with speckled b-fabric and foraminifers (1%-2%). Calcite
crystals and calcitic bodies (~5%), ranging between 10m and 80m appear in the matrix.
Quartz silt is scarce (about 1%). Very few heavy minerals appear in the silt fraction and include
hornblende and zircon. Opaque to reddish-tan iron (~2%) appear in a range of sizes from a few
spherical and rounded.
Inclusions: Spread (f:c ratio{0.062mm}=95:5) sand added frequent angular replacement chert (up to 1
mm) sometimes with chalcedony intergrowths and stilloliths, angular calcite (up to 400m) and
rounded quartz (up to 450m) sometimes with mineral inclusions. There are also a few grains of
calcareous eolianite (up to 750m) engulng an Amphiroa sp. coralline alga fossil.
Vegetal material (SLY): A few (up to 320m lengthwise) tissue fragments of cereal epidermis.
148

Matrix: Clayey, dark ochre-yellow to tan in PPL, optically active with striated b-fabric and very
rounded. Quartz silt is scarce (~2%).
Inclusions: Very sparsely spread rock fragments (f:c ratio{0.062mm}=98:2), seemingly clasts of rounded
quartz (up to 200m), micritic and sparitic limestone: (up to 350m), chert (up to 350m) and
fragments of one unclear coralline alga fossil which occur naturally within the matrix. Frequencies
were too low to estimate their relative quantities.
Firing temperature: Probably unred but the evidence is insufcient.
Reliability: High.
within the matrix sizing up to 1.2 mm. Isolated foraminifers appear in the matrix. Opaque minerals (~2%)
appear as angular to subangular bodies, sizing from few micrometers to 50m. Dark red to nearly opaque
haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the matrix around them.
Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m appear in the matrix. The
calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and rounded. Quartz silt is
scarce (~1%).
Inclusions: Very sparsely spread rock fragments (f:c ratio{0.062mm}=98:2), seemingly clasts occurring
naturally within the matrix. They comprise rounded fragments of quartz (up to 330m), micritic and
sparitic limestone (up to 400m) and calcite (up to 150m). Frequencies were too low to estimate their
relative quantities.
Firing temperature: Unred judging by preservation of uncharred vegetal material.
Geological interpretation: As EA 90. Petrographically this tablet is identical to EA 108.
Reliability: High.
Matrix: Clayey, deep ochre to tan in PPL, optically active with striated b-fabric and very pronounced
optical orientation. Elliptical to rounded bodies of purer clay (up to 450m), highly birefringent,
are sparsely spread within the matrix. Isolated foraminifers appear in the matrix. Opaque minerals
(~2%) appear as angular to subangular bodies, from a few micrometers to 50m in size. Dark red
to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the
matrix around them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m
149
appear in the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic
and rounded. Quartz silt is scarce (~1%).
Inclusions: No inclusions are found in the sample except for an 800m subrounded fragment of sparitic
limestone and a 400m bone fragment with a slight increase in the birefringence due to heating.
Firing temperature: Probably lightly red due to colour change in the matrix and the bone fragment.
Reliability: High.
Matrix: Clayey, ochre-yellow to light tan in PPL, optically active with striated b-fabric and very
rounded. Quartz silt is scarce (~1%). Scarce heavy minerals appear in the silt but range also to the
sand fraction. These include zircon and augite.
material is apparently the only added temper, as the minerals are seemingly clasts occurring naturally
within the matrix. These are subrounded to rounded quartz (up to 380m) and somewhat less sparitic
and micritic limestone (up to 400m). There are some rounded fragments of unidentied coralline
alga fossils (up to 350m) and a few ne wool hairs.
Vegetal material (SLY): A fragment of a young stem or leaf petiole showing a dense group of vessel
members of the primary xylem with a dense spiral secondary cell wall thickening. No further
indications are possible.
Firing temperature: Unred, judging by preservation of uncharred vegetal material.
Reliability: High.
Matrix: Clayey, dark ochre-yellow to yellowish-tan in PPL, optically active with striated b-fabric and very
pronounced optical orientation. Rounded bodies of purer clay (up to 300m). highly birefringent, are
sparsely spread within the matrix. Isolated foraminifers appear in the matrix. Opaque minerals (1%)
appear as angular to subangular bodies, sizing from few micrometers to 50m. Dark red to nearly opaque
scarce (~1%).
Inclusions: A mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=95:5). The
organic material is apparently the only added temper, as the minerals are seemingly clasts occurring
naturally within the matrix. These are subrounded to rounded quartz grains (up to 400m), a rounded
3 mm foraminiferous chalk fragment and a completely lignied coprolite fragment (600m) from a
herbivore, no druzes or spherulites.
150
Firing temperature: Unred due to the preservation of uncharred vegetal material.

EA 115 (VAT 1630), from Rib-Hadda to the King of Egypt (?)
Reliability: Fair.
optical orientation. Rounded bodies of purer clay, highly birefringent, are sparsely spread within the
matrix sizing up to 300m. Very few foraminifers appear. Opaque minerals (>1%) appear as angular
to subangular bodies, sizing from few micrometers to 40m. Dark red to nearly opaque haematite
crystals and calcitic bodies (~5% or slightly higher), ranging between 10m and 80m appear in the
matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and rounded.
Inclusions: No inclusions were included in this meagre sample.
Reliability: High.
Matrix: Clayey, dark ochre-yellow to yellowish tan in PPL, optically active with striated b-fabric and
very pronounced optical orientation. Elliptical to rounded bodies of purer clay, highly birefringent,
are sparsely spread within the matrix sizing up to 450m. Isolated foraminifers appear in the matrix.
Opaque minerals (~2%) appear as angular to subangular bodies, ranging from few micrometers to
dispersed into the matrix around them. Calcite crystals and calcitic bodies (<5%), ranging between
bodies are micritic and rounded. Quartz silt is very scarce (>1%).
Inclusions: Occasional minerals, rock fragments and ecofacts (f:c ratio{0.062mm}=95:5). The vegetal material
is apparently the only added temper, as the minerals are seemingly clasts occurring naturally within
the matrix. They are rounded foraminiferous chalk (up to 700m) and subrounded to rounded quartz
(up to 200m). Also present are individual, very ne sheep wool or hair bres, 20m in diameter
with a continuous medulla and a rounded fragment of unred bone splinter 300m long.
Firing temperature: Unred, judging by preservation of uncharred wool.
Reliability: High.
Matrix: Clayey, greyish-yellow to greenish-yellow in PPL, optically active speckled b-fabric with strong optical
orientation. Opaque minerals (~1%-2%) appear as angular to subangular bodies, ranging from a few
micrometers to 70m. Dark red to nearly opaque haematite particles (>1%) appear as rounded bodies,
sometimes dispersed into the matrix around them. Densely spread (~10%-15%), badly sorted calcite
crystals and calcitic bodies ranging between 10m and 80m appear in the matrix. The calcite crystals
are commonly idiomorphic, the calcitic bodies are micritic and rounded. Quartz silt is very scarce (less
than1%). It is accompanied by very few accessory heavy minerals including plagioclase and apatite.
151
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2),
of frequent subrounded to subspherical micritic limestone (up to 650m), common to frequent
subangular to rounded quartz (up to 400m) sometimes with mineral or liquid inclusions and a
few subangular polycrystalline particles (up to 200m) of anhydrite.. Mixed in are partly lignied
but well-preserved fragments (up to 650m) of herbivore manure dominated by plant tissues and
phosphatic cement. The vegetal material includes well-digested plant tissue and the identiable
cell type is a small group of primary xylem vessel members with their typical spiral secondary cell
walls. The inorganic residues that are typical of herbivore animal manure, comprising of spherulites
(spherical brous crystal concentrations formed in the animals intestines), druzes (star-shaped
carbonate concentrations formed in the animals intestines), and single rhomb-shaped crystals of
calcium oxalate (Brochier 1983). Of special interest are the spherulites, which were previously
reported to exist almost exclusively in the manure of caprovines (Brochier et al. 1992). Therefore,
the presence of these residues (especially spherulites) in sediments has previously been interpreted as
evidence for the presence of sheep/goats (Brochier et al. 1992). However, in later reviews of the topic,
this interpretation has been somewhat extended by the discovery of spherulites in the manure of other
domesticated herbivores too, such as bovines (Canti 1997), and some other animals including pigeons
(Goren 1999). In any event, the complete state of preservation of a coprolite fragment in this tablet
enables its condent identication as sheep dung.
Firing temperature: Unred. Anhydrite decomposes at 1500C and the coprolites were likely to be affected
too (by shrinking) had they been red.
Reliability: High.
within the matrix sizing up to 450m. Isolated foraminifers appear in the matrix. Opaque minerals
(~2%) appear as angular to subangular bodies, ranging from a few micrometers to 50m. Dark red
to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed into the
matrix around them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m
appear in the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic
and rounded. Quartz silt is scarce (~1%). Scarce heavy minerals appear in the silt but range also to the
sand fraction. These include zircon and augite.
naturally within the matrix. They are subrounded to rounded quartz (up to 200m, rounded
foraminiferous chalk (up to 300m) and rounded calcite (up to 200m).
Vegetal material (SLY): Complete and well preserved tear-shaped and nearly rounded plant seeds, around 1.4
mm lengthwise belonging to larger seeds. Additional badly preserved tissue fragments.
Firing temperature: Unred, judging by preservation of uncharred vegetal material.
152

Reliability: High.
Matrix: Clayey, dark ochre-yellow to yellowish-tan in PPL, optically active with striated b-fabric and
very pronounced optical orientation. Elliptical to rounded bodies of purer clay, highly birefringent,
are sparsely spread within the matrix sizing up to 450m. Isolated foraminifers appear in the matrix.
Opaque minerals (~2%) appear as angular to subangular bodies, ranging from a few micrometers to
bodies are micritic and rounded. Quartz silt is scarce (~1%). Scarce heavy minerals appear in the silt
but range also to the sand fraction. These include zircon and augite.
Inclusions: The minerals (f:c ratio{0.062mm}=95:5) are seemingly clasts occurring naturally within the
matrix. They are one idiomorphic crystal of subrounded to rounded quartz (u p to 120m) and
subspherical calcite (up to 200m).
Firing temperature: Probably unred as there is no clear evidence for ring.
Reliability: High.
optical orientation. Very few foraminifers appear. Opaque minerals (>1%) appear as angular to
subangular bodies, sizing from few micrometers to 40m. Dark red to nearly opaque haematite
is very scarce (>1%). It is accompanied by very few plagioclase particles.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2), of rounded
to subrounded sparitic limestone (up to 380m) and a 600m idiomorphic calcite crystal.
Vegetal material (SLY): tiny fragments of plant tissues with no specic characteristics.
Firing temperature: Unred due to the preservation of uncharred organic materials.
pronounced optical orientation. Elliptical bodies of purer clay, highly birefringent, are sparsely
spread within the matrix sizing up to 370m. Isolated foraminifers appear in the matrix. Opaque
minerals (~2%) appear as angular to subangular bodies, sizing from few micrometers to 50m. Dark
red to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes dispersed
into the matrix around them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and
100m appear in the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are
micritic and rounded. Quartz silt is scarce (~1%).
153
Inclusions: A mixture of minerals and organic matter (f:c ratio{0.062mm}=95:5). The vegetal material is
apparently the only added temper, as the minerals are seemingly clasts occurring naturally within the
matrix. These are subrounded to angular quartz (up to 280m) and rounded calcite (up to 200m).
Vegetal material (SLY): Complete and well-preserved nearly rounded plant seeds, around 2 mm
lengthwise (but belonging to larger seeds). The seed coat, composed of sclereides, has thick
secondary cell walls with many transverse pits. Probably a seed coat of a legume.
Matrix: Clayey, dark ochre-yellow in PPL, optically active with striated b-fabric and very pronounced
optical orientation. Elliptical to rounded bodies of purer clay, highly birefringent, are sparsely
spread within the matrix sizing up to 300m. Isolated foraminifers appear in the matrix. Opaque
minerals (~2%) appear as angular to subangular bodies, sizing from few micrometers to 50m.
Dark red to nearly opaque haematite particles (~2%) appear as rounded bodies, sometimes
bodies are micritic and rounded. Quartz silt is scarce (~1%).
material is apparently the only added temper, as the subrounded to rounded quartz (up to 100m) and
calcite (up to 200m) seem to be clasts occurring naturally within the matrix.
Vegetal material: A few unidentied plant tissues representing straw.
Reliability: Fair to moderate.
Matrix: Clayey, carbonatic, dark reddish-tan to tan in PPL. Opaque minerals are spread (~2%) ranging up
to 50m, angular in the ne fraction and subrounded at coarser grain sizes. Quartz silt occurs (~1%)
together with mica laths.
Inclusions: Sand (f:c ratio{0.062mm}=~85:15) of subangular to angular quartz (up to 370m) sometimes with
mineral inclusions, subangular to subrounded fossiliferous limestone (up to 300m) containing
fragments of fossilized mollusc shells but also one fragment of a calcareous corallinean alga, and
micritic limestone (up to 600m) with common localized brownish staining.
Firing temperature: Estimated at between 7000C and 8000C judging by the alteration of calcite.
Geological interpretation: This small sample from EA 126 is apparently similar petrographically to EA
78, 96, and 165-167 (the latter two sent by Aziru of Amurru), all assigned to the coastal area of the
>Akkar Plain, namely to umur.
Conclusions: The petrographic data suggest that this letter was sent from umur, though this is not indicated
by its contents. Moran (1992:206, n.1) proposed that EA 126, 129, 137(?) and 362 were written by
the same scribe, about the same time. However, EA 129 and EA 362 are petrographically different
from EA 126. Moreover, EA 137, which we did not examine, was undoubtedly sent somewhat later,
154
from Beirut, as indicated by its text (see EA 136 and EA 138 below). The four letters could have been
written by the same scribe, but most probably sent at different times from different places.
Matrix: Clayey, dark ochre-yellow to light tan in PPL, optically active with striated b-fabric and very
sparsely spread within the matrix sizing up to 650m. Isolated foraminifers appear in the matrix,
including a well-preserved specimen of Bolivina. Opaque minerals (~2%) appear as angular to
subangular bodies, ranging from a few micrometers to 50m. Dark red to nearly opaque haematite
particles (~2%) appear as rounded bodies, sometimes dispersed into the matrix around them. Calcite
is scarce (~1%).
Inclusions: A mixture of minerals and rock fragments (f:c ratio{0.062mm}=90:10). They include usually angular
to subangular quartz (up to 600m), including a large (600m) angular grain of polycrystalline quartz
with rough inclusions, one rounded 700m concentration of chalcedony, a very few rounded grains of
chert (up to 150m) and a subrounded 700m grain of eolianite.
Firing temperature: Probably unred, but there too few indicators to determine it beyond doubt.
EA 129 (VAT 1637 + 1638), from Rib-Hadda to the King of Egypt
Reliability: High.
sparsely spread within the matrix sizing up to 450m. Isolated foraminifers appear in the matrix.
Opaque minerals (~2%) appear as angular to subangular bodies, ranging from a few micrometers to
bodies are micritic and rounded. Quartz silt is scarce (~1%). Scarce heavy minerals appear in the silt
but range also to the sand fraction. These include zircon and augite.
naturally within the matrix. They comprise subrounded to rounded quartz (up to 380m), rounded
calcite (up to 200m) and rounded fragments (up to 350m) of unidentied coralline alga fossils.
Vegetal material (SLY): Several tissue fragments, some include bres and some include fragments of
vessel members with typical spiral secondary wall thickening.
Palaeontology (LG): Bolivina (b), Globigerina (p), Globorotalia (p), Radiolaria. Age: Neogene.
Geological interpretation: As EA 90. The palaeontological analysis carried out on this sample helps in
the geological identication of all the tablets belonging to this petrographic group.
155

Matrix: Clayey, dark ochre-yellow in PPL, optically active with striated b-fabric and very pronounced
optical orientation. Isolated foraminifers appear in the matrix. Opaque minerals (~2%) appear as
angular to subangular bodies, sizing from few micrometers to 50m. Dark red to nearly opaque
them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m appear in the
Quartz silt is scarce (~1%).
Inclusions: No inclusions except for an occasional unidentied plant tissue fragment within the matrix.
Reliability: High.
Matrix: Clayey, dark ochre-yellow to light tan in PPL, optically active with striated b-fabric and very
pronounced optical orientation. Elliptical to rounded bodies (up to 650m) of purer clay, highly
birefringent, are sparsely spread within the matrix. Isolated foraminifers appear in the matrix,
including a well-preserved specimen of bolivina. Opaque minerals (~2%) appear as angular
to subangular bodies, ranginging from a few micrometers to 50m. Dark red to nearly opaque
them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m appear in the
Quartz silt is scarce (~1%).
Inclusions: A mixture of minerals and rock fragments (f:c ratio{0.062mm}=90:10). They include usually
angular to subangular quartz: (up to 600m) including a large (600m) angular grain of
polycrystalline quartz with rough inclusions, a very few rounded chert grains (up to 300m) and
subrounded eolianite (up to 800m).
Firing temperature: Probably unred, but there are not enough indicators to determine it beyond doubt.
Geological interpretation: As EA 90. EA 127 and 131 are petrographically similar and seem to have
been made of the same lump of clay. A short span of time must have separated the dispatch of these
two letters.
Reliability: High.
Matrix: Clayey, dark ochre-tan in PPL, optically active with striated b-fabric and very pronounced optical
orientation. Isolated foraminifers appear in the matrix. Opaque minerals (~5%) appear as angular
to subangular bodies, sizing from few micrometers to 120m. Calcite crystals and calcitic bodies
(2%), ranging between 10m and 100m appear in the matrix. The calcite crystals are commonly
idiomorphic, the calcitic bodies are micritic and rounded. Quartz silt is very scarce (>1%).
Inclusions: A mixture of minerals and rock fragments (f:c ratio{0.062mm}=90:10). They include usually
156
angular to subangular quartz: (up to 700m) sometimes with rough inclusions, rounded calcite
grains (up to 200m), subrounded eolianite (up to 650m) of which one grain contains a large
fragment of articulated coralline alga and also rounded fragments of the coralline alga Amphiroa
(up to 350m).
Reliability: High.
rounded. Quartz silt is scarce (>1%).
Inclusions: Very few grains of angular quartz (up to 700m) and rounded micritic limestone (up to
700m), most likely naturally occurring within the clay (f:c ratio {0.062mm}=95:5).
Reliability: Fair.
Matrix: Carbonatic, bright yellowish-tan in PPL, optically active with speckled b-fabric and very weak
optical orientation, with high calcitic component. Badly sorted calcite crystals and micritic bodies
ranging between 10m and 80m appear in the matrix together with sparsely spread foraminifers.
Iron minerals, both opaque and dark reddish-tan, are spread in the matrix. The latter occasionally
stain the clay and appear as inllings within foraminifers.
Inclusions: Very few grains of rounded micritic limestone (up to 200m) and milky fossil mollusc shell
fragments up to 230m lengthwise (in this small sample), most likely naturally occurring within the
clay (f:c ratio {0.062mm}=95:5).
Reliability: High.
Matrix: Carbonatic, bright yellowish-tan in PPL, optically active with speckled b-fabric and very weak
ranging between 10m and 80m appear in the matrix together with sparsely spread foraminifers.
Iron minerals, both opaque and dark reddish-tan, are spread in the matrix. The latter occasionally
stain the clay and appear as inllings within foraminifers.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=98:2). Subrounded
157
to angular micritic limestone (up to 1.2mm) is common as are milky fossil mollusc shell fragments (up to
1.5mm long), lacking any internal structure (hence fossilized). There are a few subangular (idiomorphic
in one case) grains of quartz (up to 350m) and idiomorphic calcite (up to 450m).
Vegetal material (SLY): Few, up to 450m lengthwise, unidentied uncharred, badly preserved tissue
fragments.
Geological interpretation: The text of EA 136, 137 and 138 indicates that Rib-Hadda sent them from
Beirut. Indeed EA 136 and 138 are similar petrographically (EA 137 has not been examined). The
matrix of these tablets resembles in its petrography and micropalaeontology that of EA 100 sent
from Irqata. Yet, the clay type used in EA 138 is obviously of Neogene age, presumably Miocene
marl, similar to the clay and inclusions of letters from Beirut (below). This geological unit (calcaires
marnoux or M2b after Dubertret 1951a) appears in the Beirut peninsula, along the Junia Bay and
near Batroun and Tripoli, but not near Byblos. These letters were therefore made in Beirut, as
specically stated in their text.
EA 139 (BM 29828), from Ili-Rapi to the King of Egypt
Sampling method: This tablet was sampled only for ICP analysis. Chemically it is very close in elemental
composition to EA 129 of Byblos (see Appendix) and hence can be readily attributed to the Byblos
correspondence.
EA 140 (VAT 1639), from Ili-Rapi to the King of Egypt
Reliability: High.
Matrix: Clayey, orange-yellow to tan in PPL, optically active with striated b-fabric and very pronounced
scarce (~1%). Scarce heavy minerals appear in the silt including epidote and zircon.
naturally within the matrix. They include subspherical to rounded quartz (up to 500m), subangular
to subspherical micritic and sparitic limestone (up to 880m) and rounded calcite (up to 230m).
Vegetal material: Up to 880m (in length), unidentied badly preserved uncharred fragments (straw).
Coralline alga: Up to 350m, rounded fragments of unidentied fossils.
Firing temperature: Unred judging by the preservation of uncharred vegetal material.
EA 362 (AO 7093), from Rib-Hadda to the King of Egypt
Reliability: Moderate to fair.
Matrix: Clayey, ochre-yellow in PPL, and optically active with striated b-fabric and very pronounced optical
orientation. Opaque minerals (~1%) appear as angular to subangular bodies, ranging from a few
micrometers to 50m. Dark red to nearly opaque haematite particles (~1%) appear as rounded bodies.
158
Calcite crystals and calcitic bodies (2%), ranging between 10m and 100m appear in the matrix. The
very scarce (>1%).
Inclusions: Occasional minerals, rock fragments and artefacts (f:c ratio{0.062mm}=95:5). The minerals,
rounded micritic and sparitic limestone (up to 400m) and subrounded to rounded quartz (up to
100m), are seemingly clasts occurring naturally within the matrix. There are also individual very
ne sheep wool bres, 20m in diameter with a continuous medulla.
Firing temperature: Unred, since the ne wool is preserved.
EA 50 (VAT 1594), from the daughter of (?) to the Queen of Egypt
Reliability: High.
Matrix: Clayey, orange-yellow to tan in PPL, optically active with striated b-fabric and very pronounced
them. Calcite crystals and calcitic bodies (~5%), ranging between 10m and 100m appear in
the matrix. The calcite crystals are commonly idiomorphic, the calcitic bodies are micritic and
rounded. Quartz silt is scarce (~1%). Scarce heavy minerals appear in the silt including epidote
and zircon.
vegetal material is apparently the only added temper, as the subspherical to rounded quartz (up to
500m) and subangular to subspherical micritic and sparitic limestone (up to 880m) are seemingly
clasts occurring naturally within the matrix.
Vegetal material (SLY): A fragment of a plant, which includes a fragment of a primary xylem vessel
member with helical cell wall thickening. It is probably a fragment of a very young stem. No further
identication is possible.
Conclusions: The provenance of this fragmented letter, sent by the daughter of [], is unknown.
Knudtzon placed it immediately after the Ugaritic correspondence, suggesting that it was sent
from this city (apparently due to its general similarity to EA 48). However, Moran (1992:121)
noted that the conjunction al-[lu-mi] is used (with the exception of EA 34:16, 50) only at Byblos
and further south. Our petrographic and chemical analyses preclude the possibility of assigning
this letter to Ugarit since the Ugaritic tablets (as well as pottery, which is well represented in
our reference collection) cluster well and differ from all the other fabrics in the Amarna archive.
Petrographically, this letter belongs to the main group of the Byblos tablets. We therefore
confidently attribute its origin to Byblos.
159
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGM
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
NGC
QCC
QCC
QCC
MCM
MCM
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
AM
MS
VM
CP
WL
IN
Senders
BF
AN
VA
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
location
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
QZ
Byblos
68
69
71
72
74
77
79
81
82
83
84
85
86
87
88
89
101
102
106
90
91
92
93
95
105
108
109
110
111
112
115
117
118
119
120
121
123
125
127
129
130
131
132
133
139
140
362
50
78
103
126
136
138
CT
Byblos
LS
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
umur
EA
TABLE 9.1: PETROGRAPHIC GROUPING OF THE BYBLOS LETTERS1

Matrix
Inclusions
Beirut
1. Matrix types: NGM = Neogene coastal marl. NGC = Neogene coastal clay. MCM = Miocene marl. QCC = Pliocene to
Pleistocene coastal clay. Inclusion types: LS = limestone, chalk and calcite, CT = chert, QZ = quartz, AM = Amphiroa algae
fossils (Pleistocene Holocene), MS = mollusc shell fragments, VM = vegetal material, CP = coprolites, WL = wool, IN =
insect fragments, BF = bone fragments, AN = anhydrite, VA = varia.
160
CONCLUSION: THE BYBLOS CORRESPONDENCE

The study of the Byblos letters is most important for its methodological aspects. As opposed to
other Canaanite cities where the number of tablets is relatively low (11 each from Gath and Gezer, 9
from Ashkelon, 8 from Tyre and less from all other cities), the Byblos file represents an assemblage
that enables quantitative interpretations (see Table 9.1). The 53 Byblos letters which were analyzed
(including EA 50) enable us to test some of the theoretical points discussed in the methodology chapter.
They prove that if suitable clay types (i.e., brightly coloured fine clays fit for tablet production) are
found within the exploitable range of a scribe, no foreign resources are expected to be employed. In
addition, they demonstrate that the variety of materials used within a single location would not extend
beyond its potential CTF.
Indeed, the Byblos results may be summarized by the following points:
1. Forty eight (90.6%) of the tablets are made of two related clay types with a single set of inclusions
(limestone, quartz and occasionally chert and coralline algae fossils), representing the only
materials near Byblos with the potential for ceramic production.
2. The ve letters dispatched from outside Byblos (only 9.4% of the assemblage) were sent either
from the Egyptian administrative centre of umur (EA 78, 103, 126) or from the neighbouring city
of Beirut (EA 136, 138).
Another aspect, which is not unique to the letters from Byblos but is more apparent here because
of the large number of sampled items, is the use of peculiar types of temper. These include herbivore
dung, hair and wool, crushed beetles, bone splinters and plant seeds. These bizarre inclusions have no
technological advantage and thus were probably selected for other reasons that may lie within the realm
of ritual behaviour.
II. BEIRUT (BIRUTA)

Beirut is located at present-day Beirut on the coast of Lebanon. The relatively modest in size Bronze and
Iron Age mound has recently been excavated near the seashore, in West Beirut (Badre 1997).

The environs of Ras Beirut are characterized by Miocene to recent calcareous coastal sand dunes, marl,
chalk and clays. Lower Cretaceous shales and marl appear immediately to the east, together with upper
Cretaceous limestone series (Dubertret 1951a). No reference material from Beirut was available, either
as published data or as collections in other laboratories. However, we supplemented our analysis with a
letter from Beirut that was found at Ugarit, now deposited in the Muse du Louvre.
CATALOGUE
EA 97 (VAT 1598), from Yapa-Hadda to the Mayor Shumu-Haddi
Reliability: High. In the light of the surprising results and due to its fragmentary state of preservation, this
tablet was sampled twice to produce a larger sample. This resampling conrmed the rst results.
Matrix: Silty (7%), carbonatic, tan in PPL, optically active with speckled b-fabric. The silt is essentially of
quartz, but also contains recognizable quantities of accessory heavy minerals, including hornblende,
161
zircon, augite, plagioclase, microcline, biotite, muscovite, epidote and rarely tourmaline, and rutile.
Opaques are relatively common (2%), ranging between few micrometers and about 50m.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal material (f:c ratio{0.062mm}=90:10).
The dominant component is rounded to subrounded quartz (up to 450m), occasionally with undulose
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions. Rounded micritic
limestone (up to 650m) is common and there are a number of fresh (unfossilized) aquatic mollusc shell
fragments (up to 900m lengthwise). Hornblende appears rarely as rounded grains up to 200m in size.
Vegetal material (SLY): Common (up to 400m), fragments of unidentied straw.
of charring. The tablet may have been lightly heated to preserve its shape, but not red.
Geological interpretation: EA 97 is another obvious example of a letter that was made on loess soil with
coastal sand from southern Palestine (see EA 168 above for details).
Conclusions: EA 97 was made of sediments from the Gaza region, hence it was probably written there.
In comparison with the ne fabric of EA 98, a tablet that was most likely made in Beirut, the coarse
and crumbly fabric of EA 97 negates any attempt to claim that the clay was imported to Beirut, or
that a letter from Gaza to Beirut had been recycled. Therefore the only possible interpretation is that
the letter of Yapa-Hadda was sent to the Egyptian court from Gaza.
EA 98 (VAT 1675), from Yapa-Hadda to the Egyptian commissioner Yanamu
Reliability: High.
Matrix: Carbonatic, bright yellowish-ochre in PPL, optically active with speckled b-fabric and very weak
ranging between 10m and 80m appear in the matrix together with common foraminifers. Iron
minerals, both opaque and dark reddish-tan, are spread in the matrix. The latter occasionally stain the
clay and appear as inllings within foraminifers.
Inclusions: Sparse mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=95:5) made up
of rounded foraminiferous chalk (up to 1.2mm) and a very few subangular quartz grains (up to 150m).
Mixed in are a very few keratin fragments of an insect (up to 150m long).
Firing temperature: Unred, judging by the preservation of organic material.
Geological interpretation: Similar to EA 136 and 138, EA 98 was probably made on Miocene marl. This
geological unit (calcaires marnoux, or M2b after Dubertret 1951a) appears in the Beirut peninsula
along the Junia Bay and near Batroun and Tripoli. Indeed, the text of EA 136 and 138 specically
mention that they were dispatched from Beirut.
Reliability: High.
Matrix: Carbonatic, bright yellowish-tan in PPL, optically active with speckled b-fabric and very weak optical
orientation, with high calcitic component. Badly sorted calcite crystals and micritic bodies ranging
between 10m and 80m appear in the matrix, the smaller grains are commonly clear and idiomorphic
and the larger ones are rounded and milky. Foraminifers are common, including mostly Neogene genera
(similar to EA 100). Iron minerals, both opaque and dark reddish-tan, are spread in the matrix. The
latter occasionally stain the clay and appear as inllings within foraminifers. Some phosphatic bodies,
most likely dung, appear in the matrix (up to 450m) but with no clear internal structure.
162
Inclusions: A mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=85:15). Subangular
to rounded quartz (up to 470m) is frequent. Subrounded to angular micritic limestone (up to
1.2mm) is common as are milky fossilized mollusc shell fragments (up to 1.5mm long) lacking any
internal structure and fresh (unfossilized) aquatic mollusc shell fragments (up to 300m). There are
a few idiomorphic calcite crystals (up to 300m).
Vegetal material (SLY): Few, up to 450m lengthwise, very badly preserved uncharred tissue fragments,
probably from coprolites.
Firing temperature: Probably unred or very lightly red since the dung is preserved.
Geological interpretation: As EA 98 but with coarser matrix and more abundant inclusions.
Reliability: High.
Matrix: Carbonatic, greyish-yellowish in PPL, optically active with speckled b-fabric and weak
optical orientation, with high calcitic component. Badly sorted calcite crystals and micritic
bodies ranging between 10m and 80m appear in the matrix, the smaller grains are commonly
clear and idiomorphic and the larger ones are rounded and milky. Foraminifers are uncommon,
including Neogene genii (as EA 100). Iron minerals, both opaque and dark reddish-tan, are spread
in the matrix. The latter occasionally stain the clay and appear as infillings within foraminifers.
Some phosphatic bodies, most likely dung, appear in the matrix (up to 250m) but with no clear
internal structure.
Inclusions: A mixture of minerals, rock fragments and organic matter (f:c ratio{0.062mm}=95: 5) in which
subrounded to angular micritic and sparitic limestone (up to 300m) is common as are milky
fossil mollusc shell fragments (up to 1.5mm long), lacking any internal structure , and fresh
(unfossilized) aquatic mollusc shell fragments (up to 300m). A few subangular quartz grains (up
to 230m) appear. There are also a few unburnt bone splinters and a complete section through a
bone, all of small fauna (rodents? birds?).
Firing temperature: Probably unred or very lightly red since the dung is preserved.
EA 143 (VAT 1584 + C 4764), from Ammunira to the King of Egypt
Reliability: Fair.
Matrix: Carbonatic, bright yellowish-ochre in PPL, optically active with speckled b-fabric and very weak optical
orientation, with high calcitic contents and ne texture. Foraminifers are uncommon. Iron minerals, both
opaque and dark reddish-tan, are spread in the matrix. The latter occasionally stain the clay and appear as
inllings within foraminifers.
Inclusions: Sparsely spread sand of rounded micritic limestone (up to 1.2 mm), subangular to rounded
quartz (up to 450m) and coralline algae (up to 400m) probably Amphiroa sp.
Firing temperature: There are no indicators to determine the ring temperature.
Geological interpretation and conclusions: Probably as EA 141 but of ner marl.
163
COMPARATIVE MATERIAL: A LETTER FROM BEIRUT IN THE UGARIT ARCHIVE

RS 11.730 (AO 29.507), a letter from Beirut to Ugarit
Matrix: Carbonatic, bright reddish-tan in PPL, optically active with speckled b-fabric (striated in places),
with high calcitic component. Badly sorted calcite crystals and micritic bodies ranging between
10m and 80m appear in the matrix, the smaller grains are commonly clear and idiomorphic and
the larger ones are rounded and milky. All calcite crystals exhibit anomalous interference colours
due to heating. Foraminifers are spread throughout the matrix. Iron minerals, both opaque and dark
reddish-tan, are spread in the matrix. The latter occasionally stain the clay and appear as inllings
within foraminifers.
Inclusions: A mixture of minerals and rock fragments (f:c ratio{0.062mm}=80:20) of subrounded to angular
micritic and sparitic limestone and calcite (up to 200m) altered by ring and slightly less subangular
to rounded quartz (up to 200m).
Firing temperature: Around 8000C judging by the partial decalcination of calcite.
Conclusions: This letter is similar to EA 141 and the rest of the letters from Beirut in the Amarna archive
(except for EA 97) but, in contrast, it was red at a relatively high temperature.
CONCLUSION: THE BEIRUT CORRESPONDENCE

In terms of their materials the letters from Beirut represent a homogeneous group. They are also
similar to a letter from Beirut that was found at Ras Shamra (PRU III - Nougayrol 1955:12-13). The
only exception is EA 97 which was written and dispatched from the Egyptian centre of Gaza. This
letter was sent to a certain Shumu-Hadda who was held in Egypt, probably accused of operating against
the Egyptian interests in Canaan. We may speculate that Yapa-Hadda travelled by ship to Gaza in an
effort to persuade the Egyptian officials of the innocence of Shumu-Hadda, and following the failure
of his mission wrote and sent EA 97.
III. SIDON (IDUNA)

Sidon is identied with present-day Saida on the Lebanese coast (for Late Bronze nds in and around the
town see Dunand 1969; Khalifeh 1997).

Sidon is located on a plain that is covered by Miocene to Pliocene marine deposits, overlain by active
Quaternary sand dunes of calcareous composition (Dubertret 1949b; Sanlaville 1977:694-701). To the
east, the plain is bordered by Senonian to Eocene sediments including chalk, marl and chert. These
include the Paleocene marls, now encompassed in the Chekka Formation (Walley 1997:102-103).
D. Grifths of the Institute of Archaeology, University College, London, kindly allowed us to
inspect his collection of thin sections made from Early Bronze Age II pottery from the University
College London excavations at Sidon. Lacking any published analytical data on the Sidon ceramics,
this was the only comparative material available. The thin sections indicate that during this period, three
main fabrics are found in the examined assemblage: Lower Cretaceous shales, alluvial clays from the
coast with inclusions dominated by littoral deposits (quartz sand and beachrock), and Taqiye marl with
164
various inclusion types. The Lower Cretaceous group is most likely imported to the site from the north
(Beirut area) or the east. The closest Lower Cretaceous deposits to Sidon appear near Jezzine on the
Chouf Mountains. In the local groups, made of coastal sediments, the quartz sand component was still
signicant (as opposed to more northerly beds such as those displayed in the Byblos letters). This may
be a good indicator for the southern Lebanese coastal provenance of this fabric.
CATALOGUE
Reliability: Fair. More samples could not be taken since the tablet was preserved intact.
Matrix: Carbonatic, bright ochre-tan in PPL, optically active with speckled b-fabric, with high calcitic
component. Silt is scarce (~1%) including quartz and some heavy minerals of which biotite and feldspar
were identied in this meager sample. Foraminifers are spread throughout the matrix. Iron minerals, both
opaque and dark reddish-tan, are spread in the matrix.
Inclusions: Sand consisting of rounded micritic limestone and subangular to rounded quartz, idiomorphic
in one case.
Palaeontology (LG): Only one well preserved bentonic foraminifer, perhaps Brizalina dilatata (b),
and a fraction of planctonic foraminifers: Globigerinoides (p). The age may be Miocene to Lower
Pleistocene, based on the rst foraminifer.
Firing temperature: Undetermined in this small sample.
Geological interpretation: This clay is identied as Miocene or Pliocene marl by its palaeontology.
Conclusions: Although a very small sample was taken, the petrography conrms the provenance of this
letter on the Lebanon coast.
Reliability: Moderate/satisfactory. The sample size was increased to become more reliable, using the
peeling method on a broken facet.
Matrix: Clayey, reddish-tan to dark tan in PPL, optically active and oriented. Opaque minerals (up to
40m) are rather common, angular in the ne fraction and subrounded at coarser grain sizes. Quartz
silt occurs (2%) together with a smaller amount of accessory heavy minerals, including zircon,
plagioclase, microcline, pyroxene and apatite.
Inclusions: Sand (f:c ratio{0.062mm}=~93:7) of predominantly rounded to subrounded quartz grains (up to
350m), sometimes exhibiting undulose extinction or mineral and liquid inclusions. There are a few
subrounded to subangular fragments of micritic limestone (up to 550m) with common localized
brownish staining, a very few subangular smoky to brown stained replacement chert grains (up to
250m) and a single rounded fragment of nely crystalline weathered alkali basalt (270m).
Firing temperature: Partial decalcication of some calcite in the limestone suggests ring to below 7000C.
Geological interpretation: The petrofabric of EA 145 is similar to the above-mentioned second
petrofabric from Sidon examined by Grifths (pers. comm.). The clay of this tabletlike EA
144is made of Miocene or Pliocene marl. The basalt inclusion is probably a clast derived from
the Lower Cretaceous (Basalte Crtac) of the Lebanon mountains, probably swept by the drainage
system of the Awali River. As mentioned above, the dominance of coastal quartz sand differentiates
between the south Lebanese coastal fabrics and those that appear up north (such as EA 165-167).
Conclusions: As EA 144, but a larger sample.
165
CONCLUSION: THE SIDON CORRESPONDENCE

The two letters from Sidon represent local materials. Despite the ostensible similarity of the land (KURHI.A)
zu-u-ri (EA 145:22) with the city of Zura (EA 334:3), Knudtzon (1914:488-489) correctly identied
Zimredda, the author of EA 145, with the king of Sidon rather than the king of Lachish who is called by
the same name (contra Riedel 1920:23-24). The petrographic analysis supports his attribution.
IV. TYRE (URRU)

urru is Biblical wr, at present-day Sur (Tyre) on the Lebanese coast (for Late Bronze remains see Bikai
1978:6-8).

The city of Tyre was located on a small island near the southern Lebanese coast. The coastal plain near
Tyre is covered by Quaternary dunes and marine deposits. Senonian to Eocene chalk, chert and marl
(the latter is of the Chekka Formation, equivalent to the Taqiye Formation of Israel) appear to its east.
North to Tyre there is an exposure of Miocene marl (Sneh et al. 1998).
The petrographic data retrieved from un red vessels found in the ceramic workshop at Sarepta
(Anderson 1987, 1988, 1990) were used as reference for the local ceramic materials around Tyre (E.
Bettles, pers. comm.). The reference material was produced of Chekka/Taqiye marl.
CATALOGUE
The tablets are presented here in order of their petrographic afnities and not according to their EA
numbers.
Reliability: High.
Matrix: Carbonatic, light yellowish-grey in PPL, silty (about 5%) with ne opaque particles (2%) ranging
between few micrometers to 50-60m. Foraminifers are few. The silt is mainly of quartz but with
accessory twinned plagioclase, microcline, hornblende, zircon and perhaps rutile.
Inclusions: These are sparsely spread spherical micritic limestone (up to 400m), rounded and spherical
quartz (up to 200m) sometimes with undulose extinction and a fragment of Amphiroa Sp. coralline
algae fossil.
Palaeontology (LG): Acarinina (p), Bolivina (b), Chilogumbelina (p). Age: Paleogene.
Firing temperature: There are not enough indicators to determine whether this tablet was red.
Geological interpretation: By its mineralogical and palaeontological afnities the clay of this tablet is readily
identied as belonging to the Paleocene marl (Chekka/Taqiye Formations) which does not expose along
the coast to its south or north of Sidon. This is also the closest clay source for Tyre. The coralline alga
indicates a recent coastal provenance. Hence the attribution of this tablet to Tyre is certain.
Reference: Similar to the Sarepta workshop unred vessels.
Conclusions: A letter made on local sediments from Tyre.
166

Reliability: Fair. Only a tiny chip was taken as the tablet is intact.
Matrix: Carbonatic, light yellowish-grey in PPL, with some silt (~1%) with ne opaque particles (2%)
ranging between few micrometers to 50m. Foraminifers are few.
Inclusions: Grains of rounded to spherical grains (up to 200m) of micritic limestone and rounded and to
subrounded quartz (up to 200m).
Geological interpretation and onclusions: As EA 151.
Reliability: High.
Matrix: Carbonatic, light yellowish-tan in PPL, silty (~5%) with ne opaque particles (2%) ranging
between a few micrometers to 50-60m. Foraminifers are few. The silt is mainly of quartz but with
accessory twinned plagioclase, mica and zircon.
Inclusions: These are sparsely spread micritic and sparitic limestone (up to 500m) and rounded to
subangular quartz (up to 250m).
Reliability: High.
between few micrometers to 50-60m. Foraminifers are few. The silt is mainly of quartz but with
accessory twinned plagioclase, mica, and zircon.
Inclusions: These are sparsely spread grains of micritic and sparitic limestone (up to 430m) and
rounded to subangular quartz (up to 200m).
Reliability: High.
Matrix: Carbonatic, light yellowish-grey in PPL, silty (about 5%) with ne opaque particles (2%)
ranging between few micrometers to 50-60m, part of which seem to be of the charred organic
material that appears in the inclusions. Foraminifers are few. The silt is mainly of quartz but with
accessory twinned plagioclase, microcline, hornblende, zircon and perhaps rutile.
Inclusions: These are sparsely spread, including frequent to dominant large fragments (reaching 2 mm
in size) of coralline algae fossils, exhibiting a network of carbonate veins in the form of bres
within a brown cryptocrystalline calcitic groundmass. The alga fossils are cut in the thin section in
various orientations, revealing their typical microstructure (Carozzi 1960:198-200). Also present
are frequent rounded and spherical grains of quartz sand (up to 200m).
Vegetal material (SLY): A non-indicative plant fragment.
167
Firing temperature: Unred or very lightly red since uncharred vegetal material is preserved.
EA 155 (BM 29814 + VAT 1872), from Abi-Milku to the King of Egypt
Reliability: High.
Inclusions: Sparsely spread, including predominantly micritic limestome fragments (up to 400 m.
Rounded to subrounded quartz ( up to 120 m) is common. There are a few coralline alga fragments
(up to 120 m) as well as uncharred vegetal tissue fragments.
EA 295 (VAT 1650), from [Ba>lu]-danu2 to the King of Egypt
Reliability: High.
Inclusions: These are sparsely spread grains of micritic and sparitic limestone (up to 430m) and
rounded to subangular quartz (up to 200m).
Conclusions: As EA 151. Knudtzon attributed EA 295 to the ruler of Gezer. Naaman (1979a:673-676)
suggested that the letter was sent by a ruler of Tyre, probably the predecessor of Abi-Milku and the
one killed in the palace revolt (EA 89). Moran accepted this proposal and noted that the scribe of EA
295 was not the same as that who wrote the other letters of Tyre (EA 146-155). He suggested that
the scribe too perished with his master. The petrographic analysis supports the attribution of this
letter to the Tyre assemblage.
Reliability: High.
Matrix: The matrix is argillaceous, ferruginous, dark reddish-tan to darker grey-tan in PPL, silty (about 10%).
The silt contains mostly quartz with the addition of few plagioclase feldspar grains, all subangular.
The matrix is also rich in silt-sized ferruginous opaque bodies that are similar to the larger sand sized
ferruginous ooliths that will be described later, with a gradual continuum between the grain-sizes.
Inclusions: The inclusions are sand of rounded and spherical grains (up to 200m) of quartz sand,:
rounded ferruginous ooliths of varying sizes (up to 200m), usually opaque or nearly opaque in thin
section, rounded grains of replacement clear to cloudy chert (up to 900m) and spherical grains of
micritic limestone (up to 600m).
2. For the name Ba>lu-danu see Van Soldt 2002.
168
Geological interpretation: Lower Cretaceous shales. The distribution of the Aptian Formations in
Lebanon is limited to the area that broadly extends between Mount Hermon in the south and Nahr
el-Kebir in the north (Dubertret 1962). A narrow band of this formation exposes along the ridge
from Marj >Ayyun northwards. The largest exposure appears in the area between Zale in the Beqa>
and Aaley on the western slopes of the ridge. In the Anti-Lebanon Mountains, a strip exposes
between Rashiya el-Fukhar and Zebedani. However, the volcanics appear signicantly only in the
Lebanon Mountains, north of the Beirut-Zale line. Consequently, we suggest that the origin of this
tablet should be sought in this general area.
Reference: See EA 60 (Chapter 7:EA 60).
Conclusions: The materials of this letter are undoubtedly foreign to the Tyre area.
CONCLUSION: THE TYRE CORRESPONDENCE

In terms of their raw materials the tablets from Tyre fall into two categories. The main one represents
local materials (Paleocene marl and coastal sand) and a single tablet (EA 146) made of Lower Cretaceous
shales. Theoretically, like EA 78, 96, 97, 103, 126, and 168, EA 146 could be a letter from an Egyptian
administrative centre. However Lower Cretaceous shales do not expose near Gaza, Beth-Shean or
umur. At Kumidi (Kamid el-Loz), another Egyptian centre, Lower Cretaceous sandstone and shales
do not expose locally, but only in the mountainous area near Aicha el-Fukhar, about 10 km to its east.
Therefore it is highly unlikely that this was the origin of the letter and this unusual phenomenon must
remain unexplained.
169
CHAPTER 10
SOUTHERN SYRIA AND THE NEIGHBOURING AREAS
I. DAMASCUS (DIMASHQU)
The city of Dimashqu (EA 53:63; 107:28; 197:21) is identied as present-day Damascus (Dimashq eshSham). Scholars have long debated the location of Biryawaza, a prominent ruler in southern Syria whose
seat is nowhere explicitly mentioned. Naaman (1988d:180, notes 4-9) suggested that he was the ruler of
Damascus (accepted by Moran 1992:381 and Liverani 1998:250, 456).

Damascus is located in the northwestern edge of the Damascus basin, an area covered by a thick colluvial
layer of upper Quaternary conglomerates, sandstones, marls, sandy loams and loams. The stream of Nahr
el-Barada drains Jurassic to Neogene sediments and rock fragments into this area from the Anti-Lebanon
ridge to the west. The debris that covers the basin is made of deposits that originate in the Miocene-torecent basalt ows of the Bashan region to the south, and the Triassic to Neogene mainly calcareous
formations of Jebel Antar and Jebel Qasyun to the west and northwest. At the slopes of Jebel Qasyun, a
few kilometers away from the centre of the modern city, Paleocene to Lower Eocene chalks, chert and
marls are widely exposed (Ponikarov 1964: Sheet I-36-XII).
We found no petrographic references for pottery from Damascus. Hence, our conclusions rely only
on data retrieved from geological maps.
CATALOGUE
EA 194 (VAT 1705), from Biryawaza to the King of Egypt
Matrix: Carbonatic, yellowish-tan in PPL, containing sparsely spread (~2%), well-sorted calcite crystals
and calcitic bodies in two size groups: around 10m and around 30m. The calcite crystals are
commonly idiomorphic or subidiomorphic. Foraminifers are uncommon. Quartz silt is scarce (2%).
Occasional heavy minerals appear in the silt fraction and contain twinned plagioclase, iddingsite,
and augite. Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~2%) appear
at a range of sizes from a few micrometers to about 30-40m. The opaques are angular and the
Inclusions: Sparsely spread sand (f:c ratio{0.062mm}=97:3) of frequent grains of spherical micritic limestone
(up to 300m) and basalt-derived minerals: crystals of angular plagioclase (up to 200m), iddingsite
(up to 100m), serpentinized pyroxene (up to 70m), and olivine (up to 80m). Subangular to
subrounded quartz is common (up to 300m).
Vegetal material (SLY): Few, up to 400m in length plant tissues including non-indicative tissue fragments.
Firing temperature: Unred or very lightly red since uncharred organic material is preserved.
Geological interpretation and conclusions: The micropalaeontological data from EA 196 indicates that
the clay used for EA 194, 196, and 197 is most likely of Neogene age. Since Neogene marls are
170
nearly nonexistent in Israel, a northern Canaanite provenance is clearly the case. The inclusions,
although indicating a nearby source of basalt, are primarily calcareous. This suits the geology around
Damascus where Neogene marls and a neighbouring basaltic province typify the local geology.
Together with the textual data, it leaves no doubt that the letters were indeed sent from Damascus.
EA 196 (VAT 1592 + 1710), from Biryawaza to the King of Egypt
Reliability: highly reliable.
and calcitic bodies at two size groups: around 10m and around 30m. The calcite crystals are
commonly idiomorphic or subidiomorphic. Foraminifers are uncommon. Quartz silt is scarce (2%).
Occasional heavy minerals appear in the silt fraction and contain twinned plagioclase, iddingsite and
augite. Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~2%) appear in a
range of sizes from few micrometers to about 30-40m. The opaques are angular and the translucent
Inclusions: Medium-sorted sand (f:c ratio{0.062mm}=95:5) of frequent to dominant rounded to subangular
quartz (up to 300m), sometimes exhibiting undulose extinction and mineral inclusions. Micritic
limestone is frequent to common (up to 380m). There are a few basalt-derived crystals of
plagioclase, olivine (partly or entirely altered into iddingsite), and augite (up to 120m), as well as
either idiomorphic or spherical and rounded calcite (up to 250m). Also present are a very few grains
of rounded replacement chert (up to 400m) and subangular mollusc shell fragments (up to 200m).
Palaeontology (LG): Globigerina (p), Globorotalia (p). Age: Neogene (?).
Vegetal material (SLY): Common, up to 1 mm long plant tissues, a large fragment of a seed coat and a
few smaller fragments.
EA 197 (BM 29826), from Biryawaza to the King of Egypt
Reliability: High.
Matrix: Carbonatic, yellowish-tan in PPL, containing spread (~5%), well-sorted calcite crystals and calcitic
bodies ranging gradually between 10m and 100m. The calcite crystals are commonly idiomorphic
or subidiomorphic. Foraminifers are rare and include mainly fragments of ostracoda. Quartz silt
is scarce (~2%-3%). Occasional heavy minerals appear in the silt fraction and contain iddingsite,
twinned plagioclase, serpentine, and augite. Opaque to reddish-tan iron minerals (magnetite and
haematite respectively, ~2%) appear in a range of sizes from a few micrometers to about 30-40m.
The opaques are angular and the translucent minerals tend to be spherical and rounded.
Inclusions: Medium-sorted sand (f:c ratio{0.062mm}=95:5) made up of frequent graoms of micritic limestone (up
to 750m but usually below 300m). Either idiomorphic or rounded calcite (up to 250m) is common
as are rounded to subangular grains of quartz (up to 300m) sometimes exhibiting undulose extinction.
There are a few basalt-derived crystals of plagioclase, olivine (partly or entirely altered into iddingsite)
and augite (up to 120m) and very few grains of rounded replacement chert (up to 700m). A rare
aggregate of anhydrite crystals (up to 500m) in a void (perhaps post-depositional) is seen.
171
II. MUSHIUNA
Mushiuna (EA 182-184) is a city of unknown location. It is commonly identied with M of Thutmose
IIIs topographical list (No. 25), which is mentioned side by side with well identied cities of the Bashan
(e.g. Buruna, Qanu, Ashtaroth; see Helck 1971:129, 184; Ahituv 1984:145-146; Liverani 1998:249).
However, Knudtzon (1915:1278) noted that in clay, EA 177-183 should not be separated from EA 174176, sent by rulers of city-states in the Beqa>. Also, the form of the numbers and of the a-sign in EA 182
is similar to that in EA 174 (from ashabu in the Beqa>). Therefore, Knudtzon inserted the Mushiuna
correspondence (EA 182-184) between the letters of the Beqa> and those of the Damascus-Bashan area.
CATALOGUE
This letter is discussed before EA 182 due to the larger sample that was extracted from it, and hence its
higher reliability.
Reliability: High.
Matrix: Carbonatic, yellowish-tan to light tan in PPL, birefringent with no optical orientation and devoid
of foraminifers; slightly silty (less than 1%). Occasional heavy minerals appear in the silt fraction
including iddingsite, twinned plagioclase, serpentine, and augite. Opaque angular iron minerals
appear, ranging from a few micrometers to about 30-40m.
Inclusions: Sparsely spread sand (f:c ratio{0.062mm}=98:2) of rounded micritic limestone, up to 1 mm but
usually by far smaller and rounded quartz (up to 800m).
Vegetal material (SLY): Up to 650m, non-indicative tissue fragments.
Geological interpretation: This tablet clearly represents a calcareous sedimentary environment, therefore
a source area in the Bashan is highly unlikely. The tablet does not supply enough petrographic data
to enable the identication of its precise place of origin.
Matrix: Carbonatic, tan in PPL, birefringent with slight optical orientation and devoid of foraminifers,
slightly silty (less than 1%). Occasional heavy minerals appear in the silt fraction, including
iddingsite and twinned plagioclase. Opaque angular iron minerals appear, ranging from few
micrometers to about 30-40m.
Inclusions: No inclusions were encountered apart from some organic material represented by a bre,
presumably partly digested, about 2 mm long. It is not clear whether its origin is oral or faunal.
172
EA 184 (Ash. 1893. 1-41: 426), from Shutarna to the King of Egypt
Reliability: High.
Matrix: Carbonatic, whitish to yellowish-grey in PPL, birefringent with no optical orientation; sparse,
badly preserved foraminifers. A few opaque angular iron minerals appear, about 20m in size.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) in which rounded micritic limestone (up to 600m) is dominant.
There are a few fossil mollusc shell fragments (up to 250m long) and a very few angular quartz
crystals with inclusions (up to 150m).
Firing temperature: Probably unred or very lightly red. There are no petrographic indicators for
clearer determination.
Geological interpretation: EA 184 is different from EA 182-183. It is probably made of marl that was
taken from an in situ deposit. However, there are no other details that may help in the denition of
its geological age and origin.
CONCLUSION: THE MUSHIUNA LETTERS

The sedimentary calcareous nature of the raw materials of EA 182-183 makes it unlikely that they were
written in the Bashan. If Mushiuna was indeed located in that region, Knudtzons observation of the
similarity in clay and script between EA 182 and 183 and those of the Beqa > may lead to the conclusion
that Shutarna sent his letters from the Egyptian centre at Kumidi. Since we have not examined EA 198
(stored in the Cairo Museum) and as we had no access to pottery from Kamid el-Loz, we have no reference
to the petrography of Kumidi (see Chapter 8: Conclusions). Knudtzon (1915:1281) noted that EA 184 is
different in script from EA 182-183, an observation now supported by the petrographic analysis. Is it
possible that this letter was written in Mushiuna, whereas the two other letters were dispatched from
Kumidi? Granted this assumption, the sedimentary calcareous nature of the raw materials of EA 184
does not t its attribution to the Bashan. Should we then dismiss the equation of Mushiuna with M
(No. 25) of Thutmose IIIs topographical list and locate the city elsewhere? Unfortunately, there is no
clue as to the exact source of the clay of EA 184. In the light of these uncertainties we avoid suggesting
identication for Mushiuna and for the origin of its three letters.
173
MICROSCOPIC VIEWS OF THIN-SECTIONS OF TABLETS ANALYZED

The Amarna Tablets are arranged in the order in which they appear in the text. Comparative material is
inserted in the appropriate position.
All photographs were taken under cross-polarized light.
Red scale bar = 200 m.
Details of each thin-section are discussed in the catalogue in the relevant chapter and section.
174
Plate I: Letters of the Great Powers. I. Egypt.
175
Plate I (contd.): Letters of the Great Powers. I. Egypt
II. atti
III. Babylonia
176
Plate I (contd.): Letters of the Geat Powers. III. Babylonia
177
Plate I (contd.): Letters of the Great Powers. IV. Mitanni
178
179
V. Arzawa
180
VI. Alashiya
Plate I (contd.): Letters of the Great Powers. VI. Alashiya
181
Plate I (contd.): Letters of the Great Powers. VI. Alashiya
VII. Unknown provenance
182
Plate II: Scholarly texts.
183
Plate II (contd.): Scholarly texts.
184
Plate II (contd.): Scholarly texts.
185
Plate III: North Syrian kingdoms. I. Ugarit
II. Nuashe
186
III. Nii (?)
Plate IV: Syrian kingdoms in the Middle Orontes area. I. Qatna
II. Qidshu
III. Labana
187
Plate V (contd.): The kingdom of Amurru and neighbouring polities. I. Amurru
188
189
II. umur
III. Tunip
IV. Irqata
V. Unspecied city
190
Plate VI: The Lebanese Beqa>.

I. Ehishasi
II. asi
III. Guddashuna
IV. A group of identical letters
V. Letters of unspecied location
191
Plate VII: The Lebanese littoral. I. Byblos
192
Plate VII (contd.): The Lebanese littoral. I. Byblos
193
194
195
196
197
II. Beirut
198
Plate VII (contd.): The Lebanese littoral. II. Beirut
III. Sidon
199
Plate VII (contd.): The Lebanese littoral. IV. Tyre
200
Plate VIII: Southern Syria and the neighbouring areas. I. Damascus
II. Mushiuna
Plate IX: The Bashan area.
201
Plate IX (contd.): The Bashan area.
202
Plate X: The Galilee, the coastal plain of Acco and the northern valleys. I. Hazor
II. Achshaph
III. Shim>on
IV. Acco
203
Plate X (contd.): The Galilee, the coastal plain of Acco and the northern valleys. V. Anaharath
VI. Megiddo
204
Plate X (contd.): The Galilee, the coastal plain of Acco and the northern valleys. VI. Megiddo
VII. Rehob
VIII. [URUx-(x)-i]G-ma-te
205
Plate X (contd.): The Galilee, the coastal plain of Acco and the northern valleys. IX. Ginti-kirmil
X. Pehel
206
Plate XI: The central hill country. I. Shechem
II. Jerusalem
207
Plate XII: The Shephelah and the southern coastal plain. I. Gezer
208
Plate XII (contd.): The Shephelah and the southern coastal plain. II. Gath
209
Plate XII (contd.): The Shephelah and the southern coastal plain.
II. Gath
III. Lachish
IV. A group of identical letters
V. Ashdod
210
Plate XII (contd.): The Shephelah and the southern coastal plain. VI. Ashkelon
VII. Yurza
211
Plate XIII: Unidentied cities in Canaan.
212
Plate XIII (contd.): Unidentied cities in Canaan.
213
Plate XIII (contd.): Unidentied cities in Canaan.
214
CHAPTER 11
THE BASHAN AREA
EA 201-206 is an assemblage of letters from cities in the Bashan, all containing a homogeneous
formulaic text (see Knudtzon 1915:1294, n.2). Moran (1992:273 n. 2; 278 n. 1) observed that the same
scribe wrote EA 195 (from Damascus) and EA 201-206. Hence the seven letters were either written at
about the same time in one place, or Biryawaza might have sent his scribe to the neighbouring Bashan
cities. Petrographic analysis is the ideal tool to decide this matter.
The identication of the cities that are mentioned in EA 201-206 is not always certain. Z/iribashani
(EA 201) may either be located in the region of Naveh in the Bashan (Ahituv 1984:181), or could be
identied with Ezra in the Hauran (Mazar 1975:187; see also Abel 1938, II:10). The place of Shasimi
(EA 203) is unknown. Qanu (EA 204) is commonly identied with Qanawat, below Jebel ed-Druz. Tubu
(EA 205) is usually identied at e-$ayibeh, between Bura esh-Sham and Der>a (Abou Assaf 1974;
Ahituv 1984:190-191; Liverani 1998:247 n. 12). Naiba (EA 206) was identied with Naib, in the
region of Der>a (Mazar 1975:187), but this identication still needs archaeological conrmation. In any
event, the cities in question are distributed in the Bashan and the Hauran, between Jebel ed-Druz and
Der>a, near the sources of the Yarmuk river.

The entire Bashan area is typied by basalt ows, hence the only possible clay source that is expected
to be found in this region is basaltic brown Mediterranean soil. The soil material derives from the
weathering of the primary basalt minerals (plagioclase, augite and olivine) which are mixed with quartz
grains from aeolian dust. The rate of basalt weathering to soil material is somewhat low (Ya>alon 1971),
but the relatively old age of the basalt ows has allowed the accumulation of deep soils, particularly in
at areas and in the lower positions of the landscape, where grumusols may occur.
Pottery made from this soil unit includes a number of petrographic types. They differ with the
presence or absence of certain skeleton grains and their distribution, and with the nature and amount
of carbonates present in the soil material or added to it. Common to all samples is the large amount of
quartz grains in the silt-size fraction. Olivine, augite, plagioclase and iddingsite grains and basalt rock
fragments occur in varying amounts (Adan-Bayewitz and Wieder 1992). Since there are no published
petrographic reports on pottery from the Bashan area, we used the geologically equivalent neighbouring
Golan area as reference. Pottery from the Golan sites was constantly produced of the local basaltic brown
soil and was dominated by basaltic inclusions (Goren 1991a; Goren and Halperin 2001; Porat 1998).
On the margins of the Bashan plateau, especially in the south where the Yarmuk river cuts into older
strata, a set of calcareous and siliceous rocks are exposed. These include Eocene chalk and limestone as
well as Miocene limestone, sandstone and marl, equivalent to the Hordos formation of the southern and
western slopes of the Golan. The latter formation extends to the east and forms part of the calcareous
exposures along the Rukkad gorge. This is the main clay source in this region, although Paleocene marl
(unnamed in Syria but equivalent to the Taqiye formation in Israel) is found in a few isolated exposures
along the Yarmuk river.
215
Several petrographic studies were carried out on pottery from sites located in the western part of this area.
These include Shaar Hagolan (Goren 1991a; 1992), Khirbet Zeraqun (Goren, unpublished), Khirbet el-utiyye
(Epstein 1998:155-157; Porat 1998) and Tel Dover (Cohen-Weinberger, pers. comm.). The pottery of these sites
is characterized by a marly matrix and a set of primarily basaltic with secondary calcareous rock fragments
which distinguish it from the pottery that was made in the basaltic highlands of the Golan (Porat 1998).
I. Z/IRIBASHANI
CATALOGUE
EA 201 (VAT 338), from Artamanya to the King of Egypt
Reliability: Fair.
Matrix: Carbonatic, yellowish-tan in PPL, containing spread calcite crystals and calcitic bodies
ranging gradually between 10m and ~60m. The calcite crystals are commonly idiomorphic or
subidiomorphic. Foraminifers are rare. Quartz silt is scarce. Occasional heavy minerals appear in the
silt fraction and contain iddingsite, plagioclase, and zircon. Opaque iron minerals (magnetite) appear
in a range of sizes from a few micrometers to about 30-40m are angular.
Inclusions: Due to the small sample size the inclusion assemblage is probably partial. No grain-sizes were
measured, as these may not reect the actual situation in the clay body. Present are rounded grains of
micritic limestone, idiomorphic crystals of calcite and a fresh but small fragment of alkali-olivine basalt
with additional iddingsite as part of the ne sand fraction. Stereomicroscopic examination of the tablets
surface indicates larger (up to millimeter size) fragments.
Firing temperature: Probably unred or very lightly red (as the other Bashan letters). There are no
petrographic data for estimating the ring temperature.
Geological interpretation and conclusions: The Bashan tablets can be divided into four petrographic
groups (see below). EA 201 belongs to the main group, probably dispatched from the southern
Bashan or Yarmuk Valley.
II. SHASIMI
CATALOGUE
EA 203 (VAT 330), from Abdi-Milki to the King of Egypt
Reliability: Fair.
Matrix: Carbonatic, yellowish-tan in PPL, containing spread (~3%), well-sorted calcite crystals and calcitic
bodies ranging gradually between 10m and 100m. The calcite crystals are commonly idiomorphic
or subidiomorphic. Foraminifers are uncommon. Quartz silt is scarce (~2%-3%). Occasional heavy
minerals appear in the silt fraction and contain iddingsite, olivine, and twinned plagioclase. Opaque
to reddish-tan iron minerals (magnetite and haematite respectively, ~4-5%) appear at a range of sizes
from a few micrometers to about 30-40m. The opaques are angular and the translucent minerals tend
to be spherical and rounded.
216
Inclusions: The inclusion assemblage in this sample is seemingly partial. Stereomicroscopic examination
revealed the presence of basalt that is represented here only by basaltic derived minerals. The
inclusions in this sample comprise idiomorphic crystals of calcite which appear also in the sand
fraction, rounded fragments of micritic limestone, subangular to subrounded grains of quartz,
basalt minerals (serpentine, olivine and serpentinized pyroxene) and a very few 350m aquatic
mollusc shell fragments.
Firing temperature: Probably unred or very lightly red (like the other Bashan letters). There are no
petrographic data to estimate the ring temperature.
III. QANU
CATALOGUE
EA 204 (VAT 328), from the ruler of Qanu to the King of Egypt
Matrix: Carbonatic, yellowish-tan in PPL, containing spread (~2%), well-sorted calcite crystals and
calcitic bodies ranging gradually between 10m and 100m. The calcite crystals are commonly
idiomorphic or subidiomorphic. Foraminifers are scarce. Quartz silt is scarce (~1%-2%). Occasional
heavy minerals appear in the silt fraction and contain iddingsite, hornblende, and plagioclase.
Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~3-4%) appear in a
range of sizes from a few micrometers to about 30m. The opaques are angular and the translucent
Inclusions: Badly sorted sand in which rounded grains of coarsely crystalline alkali-olivine basalt ranging
between 380m to 3mm are dominant. The olivine phenocrysts are partly or entirely altered into
iddingsite and the augite is sometimes partly serpentinized. Rounded grains of micritic limestone (up
to 300m) are common and there are a few rounded to subangular quartz grains (up to 200m). A very
few land snail shell fragments (up to 150m) are also present.
Vegetal material (SLY): A few (up to 350m long) minute non-indicative uncharred tissue fragments.
Firing temperature: Unred or very lightly red since the organic material is uncharred.
IV. TUBU
CATALOGUE
EA 205 (BM 29861), from the ruler of Tubu to the King of Egypt
Matrix: Carbonatic, yellowish-tan in PPL, containing spread (~2%), well-sorted calcite crystals and
idiomorphic or subidiomorphic. Foraminifers are scarce. Quartz silt is scarce (~2%-4%). Occasional
heavy minerals appear in the silt fraction and contain iddingsite, olivine, zircon, augite, and twinned
plagioclase. Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~3-4%)
217
Inclusions: Badly sorted grains in which subangular to rounded grains of micritic and sparitic limestone
(up to 500m) are dominant. Fresh coarsely crystalline alkali-olivine basalt (up to 2.2 mm) is
frequent. The olivine phenocrysts are partly or entirely altered into iddingsite and the augite is
sometimes partly serpentinized. Angular grains of quartz (up to 500m) are fairly common and
sometimes exhibits undulose extinction and/or mineral inclusions.
Vegetal material (SLY): A bundle of bres (up to 400m long) and several small non-indicative tissue
fragments.
Firing temperature: Unred or very lightly red due to the preservation of uncharred organic material.
V. ASHTAROTH (ASHTARTU)
Ashtartu (biblical Ashtaroth) was the dominant city in the Bashan. It is identied with Tell >Ashtara
(Abou Assaf 1968; 1969), a site located north of the Yarmuk River. Situated near the main caravan route
that passed from Damascus through the Bashan and the Yarmuk area to the west, Ashtaroth dominated
the overland route between Egypt and Mesopotamia. EA 364 indicates that it had a common border with
Hazor. In EA 256 Mut-Ba>lu, ruler of Piilu, states that he went to the aid of Ashtaroth when all the
cities of Garu had become hostile (lines 21-23). The cities of the land of Garu, which must have been
controlled by Ashtaroth, were located to the north of the Yarmuk river, east and west of the Rukkad
entry into the Yarmuk Valley (Albright 1943:10-15; Naaman 1975:42; for results of archaeological
surveys and possible identications in the southern Golan see Ma>oz 1986; Epstein 1993). Ayyab was
Mut-Ba>lus ally in the ght against the cities of the Land of Garu until their relations were broken off
(Moran 1992:309-310 n. 3). Hence he must have been the ruler of Ashtaroth.
CATALOGUE
EA 364 (AO 7094), from Ayyab to the King of Egypt
Reliability: Satisfactory to high.
Matrix: Carbonatic, yellowish-tan to yellowish-gray in PPL, containing spread calcite crystals and
idiomorphic or subidiomorphic. Quartz silt is scarce. Occasional heavy minerals appear in the silt
fraction and contain iddingsite, twinned plagioclase and opaques. Foraminifers occur.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) of generally rounded components including alkali-olivine
basalt and its derived minerals which are frequent, rounded to subrounded (up to 600m), including
various phases of basalt ranging between trachytic to nearly doleritic. Single mineral crystals include
plagioclase and olivine. Micritic limestone (up to 500m) and foraminiferous chalk (up to 900m)
are common and there are a few grains of rarely polycrystalline quartz (up to 850m), travertine (up
to 500m) and subangular replacement chert (up to 250m).
Firing temperature: Probably unred or very lightly red. The lack of abnormal interference colours in
calcite indicates that the ring, if occurred, had not reached 7000C.
Geological interpretation and conclusions: EA 364 belongs to a group of letters which were probably
dispatched from Ashtaroth (Table 11.1).
218
VI. ZURA
Zura is an unidentied city which is mentioned in EA 334, whose authors name is missing (lines 2-3:
Message of [], the ru[ler]? of Zura). Riedel (1920:24) suggested identifying it with Biblical Zoar,
located south of the Dead Sea (accepted by Aharoni 1967:159). However, the area of the Dead Sea was
not inhabited in the Late Bronze Age.
Knudtzon (1915:1348) noted that the clay of EA 336 and 337, sent by a ruler named Hiziru whose
city is not mentioned, is closely related to that of EA 334, which was sent from Zura, and connected the
latter to EA 335 on the basis of a mistaken interpretation of the noun ru/uru (back) in line 3. He
attributed this group of four tablets to southern Palestine. The verbal form maqta/iti (I fall) that appears
in EA 336:5 is typical of the letters of Gath, though it also appears once in a letter from Byblos (EA 138:4;
see Naaman 1979a:677). Finally, Knudtzon (1915:1282 n. 1) noted that the clay of EA 334 and 336 is
similar to that of EA 185, a letter which was sent from the Beqa> of Lebanon.
Our results indicate that EA 337 of iziru is identical to EA 364 from Ashtarot. EA 334 and 336
are nearly identical, suggesting that iziru was indeed related to Zura. Though these links are far from
cut and dried, we cautiously suggest that Zura should be linked with the Bashan area or near it. The
petrography of EA 337 clearly indicates a northern Canaanite provenance at least for this tablet.
CATALOGUE
Reliability: High.
Matrix: Carbonatic, light yellowish-tan in PPL, containing infrequent foraminifers (about 2%). Quartz
silt is rare (about 1%) and accompanied by accessory heavy minerals including zircon, augite and
twinned plagioclase. Opaque minerals (1%) appear in a range of sizes from a few micrometers
to about 70m. Under higher magnications (>X200) the matrix is brous, optically active and
displays very weak optical orientation.
Inclusions: The inclusions (f:c ratio{0.062mm}=~90:10) contain sand made up of frequent rounded to
subrounded grains of alkali-olivine basalt, basanite of phases ranging from trachytic to nearly
doleritic and single crystals of plagioclase and augite (up to 900m). Rounded to subrounded
grains of micritic and sparitic limestone (up to 650m) are common as is foraminiferous
chalk (up to 750m). There are a few rounded quartz grains (up to 650m) and subangular
grains of replacement chert (up to 550m) and a very few calcareous pisoliths with concentric
microlamination (up to 250m).
Vegetal material (SLY): A few (up to 750m long) uncharred plant tissues. Badly preserved noncharacteristic tissue fragments.
Geological interpretation and conclusions: As EA 364 of Ashtaroth.
EA 334 (VAT 1609), from (?) of Zura to the King of Egypt
Reliability: High.
Matrix: Carbonatic, light tan in PPL, birefringent but with no optical orientation. Quartz silt (5%)
appears together with accessory heavy minerals, including augite, zircon, hornblende, and muscovite.
219
The augite is usually angular, rather common and reaches 70m in size. Opaques (2%) appear in
sizes reaching 30m.
Inclusions: Sand (f:c ratio{0.062mm}=95:5) in which micritic limestone (up to 700m) is dominant together
with a few calcareous pisoliths (up to 650m) with concentric microlamination, some containing a
central hollow.
Vegetal material (SLY): Few (up to 150m long), uncharred plant tissues. Several badly preserved
unidentied tissue fragments.
Firing temperature: Un red or very lightly red since uncharred organic material is preserved.
Geological interpretation and conclusions: EA 334 belongs to a group of letters which were probably
dispatched from Zura (Table 11.1).
Reliability: High.
Matrix: Carbonatic, light tan in PPL, birefringent but with no optical orientation. Quartz silt (5%) appears
together with accessory heavy minerals, includes augite, zircon, and epidote. The augite is usually
angular, rather common and reaches 50m in size. Opaques (2%) appear in sizes reaching 50m.
Inclusions: Sand (f:c ratio{0.062mm}=97:3) in which micritic limestone (up to 700m) is dominant together
with a few calcareous pisoliths (up to 200m) with concentric microlamination.
Firing temperature: Most likely similar to EA 334, although vegetal material is not seen in this sample.
VII. SHARUNA
The text of EA 241 does not supply any clues as to the identication of Rusmanyas city. Knudtzon
placed it among the letters of rulers from northern Palestine, thus identifying Sharuna in the Galilee.
Other scholars dismissed this proposal and in the light of the Egyptian topographical lists identied
Sharuna in the Bashan (Edel 1966:13-14; Helck 1971:129, 184, 260; Ahituv 1984:170).
CATALOGUE
EA 241 (VAT 1678), from Rusmanya to the King of Egypt
and calcitic bodies at two size groups: around 10m and around 30m-50m. The calcite crystals
are commonly idiomorphic or subidiomorphic in the ner fraction but tend to become rounded in the
coarser. Foraminifers are uncommon. Quartz silt appears at about 4%. Occasional heavy minerals are
rather common in the silt fraction and include iddingsite, zircon, plagioclase, hornblende and olivine.
Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~2%) appear in a range
of sizes from a few micrometers to about 30-40m. The opaques are angular and the translucent
Inclusions: The inclusions contain badly sorted sand (f:c ratio{0.062mm}=95:5). Rounded micritic limestone
(up to 700m), and less commonly sparitic,is dominant. Coarsely crystalline alkali-olivine basalt
220
(up to 500m) is common with a few angular yellowish weathering products. Spherical grains of
quartz (up to 120m) are common.
Vegetal material (SLY): Non-indicative plant tissue fragments (up to 500m), one of parenchymatic nature.
VIII. TABLETS FROM UNSPECIFIED CITIES IN THE BASHAN

Knudtzon (1915:1293 n. 1) noted that the clay of this tablet is similar to that of EA 199, a letter sent either
by the king of Buruna, or by a neighbouring ruler who escorted the Kings caravan to Buruna. Hence the
attribution of EA 200 to the group of letters sent from the Bashan (for the text, see Naaman 2001a).
Reliability: High.
Matrix: Carbonatic, tan in PPL, silty (5%). The silt is of quartz accompanied by abundant iddingsite,
augite, twinned plagioclase, olivine, and zircon. Opaque iron minerals are common (3-4%). They
Inclusions: Badly sorted grains in which rounded and spherical micritic limestone are dominant often
containing sparse silty quartz, up to 1.5 mm. Spherical grains (up to 3mm) of coarsely crystalline
alkali-olivine basalt and dolerite where the olivine is commonly altered to iddingsite are frequent.
Vegetal material (SLY): A single fragment from the leaf blade of a cereal.
Geological interpretation and conclusions: Most likely as EA 201.
EA 202 (VAT 331), from Amawashe to the King of Egypt
Reliability: Fair.
Matrix: Carbonatic, greyish-tan in PPL, containing sparsely spread (~4%), medium-sorted calcite
crystals ranging between 10m and 250m. The calcite crystals are commonly idiomorphic or
subidiomorphic in the ner fraction, but tend to become rounded in the coarser. Quartz silt appears
at about 2%. Occasional heavy minerals are rather common in the silt fraction and include iddingsite,
zircon, plagioclase and olivine. Angular opaque iron minerals appear in a range of sizes from a few
micrometers to about 30-40m.
Inclusions: This sample does not include a representative assembly of inclusions. It contains clear,
subrounded feldspar.
Geological interpretation and conclusions: AS EA 201.
EA 207 (VAT 1593), from Ipte to the King of Egypt
The author of the letter mentions Puuru, the Egyptian commissioner of Kumidi. Hence his city
should be sought in southern Syria (for restoration and translation of the text see Naaman 1998b).
221

Reliability: High.
Matrix: Carbonatic, yellowish-tan to yellowish-gray in PPL, containing spread calcite crystals and
idiomorphic or subidiomorphic. Quartz silt is scarce. Occasional heavy minerals appear in the silt
fraction and contain iddingsite, twinned plagioclase and opaques. Foraminifers occur.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) of generally rounded components including predominantly
alkali-olivine basalt (up to 600m) ranging between trachytic to nearly doleritic and single crystals
of plagioclase and olivine. Micritic limestone (up to 500m) is common as is foraminiferous chalk
(up to 900m). There are a few grains of (rarely polycrystalline) quartz (up to 850m), travertine
(up to 500m) and subangular replacement chert (up to 250m).
Firing temperature: Probably unred or very lightly red. The lack of abnormal interference colours in
calcite indicates that if there was ring, the temperature did had not reach 7000C.
This letter also mentions Puuru, the Egyptian commissioner of Kumidi. Hence this city too should be
sought in southern Syria.
Reliability: High.
Matrix: Carbonatic, tan in PPL, birefringent but with no optical orientation. Quartz silt (5%) appears
together with accessory heavy minerals, including olivine, zircon, augite, hornblende, plagioclase
and muscovite. Opaques (2%) appear in sizes reaching 30m.
Inclusions: Sparse sand (f:c ratio{0.062mm}=97:3) in which micritic limestone (up to 650m) is dominant, a
single 500m ferruginous oolith displaying concentric microlamination and a single rounded mollusc
shell fragment (570m long).
Vegetal material (SLY): A tissue fragment which includes vessels of the primary xylem with their typical
spiral secondary cell walls, and a few non-indicative tissue fragments.
Geological interpretation and conclusions: EA 208 belongs to a group of letters which was probably
dispatched from Damascus (Table 11.1)
EA 209 (AO 2036), from Zishamimi to the King of Egypt
Reliability: Fair.
Matrix: Carbonatic to clayey with speckled b-fabric and pronounced optical orientation, yellowish-tan in
PPL. Quartz silt comprises about 5%. Opaques (2%) appear in sizes reaching 30m.
Inclusions: No inclusions are present in this meagre sample.
Firing temperature: Probably as EA 210.
EA 210 (VAT 1876), from Zishamimi to an Egyptian ofcial (?)
Reliability: High.
222
Matrix: Carbonatic to clayey with speckled b-fabric and pronounced optical orientation, yellowish-tan
in PPL. Quartz silt makes about 5%. Opaques (2%) appear in sizes reaching 30m. The matrix is
laminated and exfoliated in places.
Inclusions: Sand (f:c ratio{0.062mm}=95:5) with usually subspherical micritic limestone (up to 900m)
dominant.
Vegetal material (SLY): A badly preserved unidentiable fragment. Herbivore coprolite containing
unidentiable (digested) vegetal material and spherulites (see EA 118 for denition).
Geological interpretation and conclusions: Moran (1992:xxix n. 82) correctly noted that EA 210 was not
addressed to the king. The letter was probably dispatched to an Egyptian ofcial whose name appears
in line 1. Line 4 contains the salutation [may the god DN gua]rd you (DN li-u]-ur-ka). EA 210 was
probably dispatched from Damascus (Table 11.1).
CONCLUSION: THE BASHAN CORRESPONDENCE

The petrographic results from the examination of the Bashan letters present some difcult problems of
interpretation. Technologically the letters can be divided into four distinct groups (Table 11.1):
1. Tablets made of Paleocene marl and well-sorted river sand containing various facies of basalts,
limestone, chert and quartz (EA 364, 207, 337).
2. Letters made of loam or clay-loam containing basaltic minerals in the silt fraction, but the
inclusion assemblage is entirely sedimentary and contains mainly micritic limestone and notably
calcitic pisoliths (EA 334, 336).
3. Tablets made of carbonatic clay loam, including badly sorted inclusions of micritic limestone
and coarse grains of basalt or dolerite (EA 200-205, 241).
4. Letters similar to those belonging to the second group, but devoid of the calcareous pisoliths (EA
208-210).
The rst group should most probably be assigned to the city of Ashtaroth. EA 364 written by
Ayyab, the ruler of Ashtaroth is used as a type specimen for this group. The petrographic traits of this
group suit the geology near Tell >Ashtara, the site of ancient Ashtaroth (Ponikarov 1964: Sheet I-36VIw). Paleocene marl exposes about 5 km south of the site and forms the only available clay source
(excluding basaltic brown soils, which would create very dark tablets). The inclusion assemblage
most likely represents the local sand at Wadi el-Hrer near the site, where materials from various basalt
exposures are drained into an area dominated by Middle Eocene hard limestone and chert.
EA 207 (which does not carry the name of the sender or his city) and EA 337 (written by iziru)
are petrographically similar to EA 364. These letters were probably sent from Ashtaroth by neigbouring
rulers. This means that iziru should be assigned to a place not too far from the Bashan (contra
Knudtzon, who located it in southern Canaan). The petrographic and chemical examinations indicate
a similarity between EA 334 (sent from Zura) and EA 336 (written by iziru) both belonging to
the second group mentioned above. Accordingly, Hiziru may conceivably be identied as the ruler of
Zura. This city cannot be identied in the Bashan per se. Firstly, it does not appear in the Egyptian
topographical lists which enumerate many cities in this region and secondly there are many differences
in script and text between EA 364 and EA 334, 336-337. Furthermore EA 334 and 336 were made in a
sedimentary, calcareous area.
This seems to mean that Zura should probably be located in a calcareous area, not far from Ashtaroth.
EA 334 and 336 were written in Zura itself, while Zuras scribe must have traveled to Ashtaroth with his
223
lord and wrote letter EA 337 there.1 The most probable area which ts this description is the Gilead plateau.
Placing a city-state in the Gilead resolves a problem: this region one of the most fertile and densely
inhabited in Canaan has so far been left outside the boundary of the political-territorial system of the Late
Bronze Age. A suggestion that Piilu ruled over this vast area is difcult to accept.
Large enough sites in the Gilead which could have served as a centre of a Late Bronze city-state
include the mounds of Irbid and el-usn (G.R. 2329 2110) and possibly the site of Ramtha. The large
mound under the town of Irbid yielded Late Bronze nds (Lenzen et al. 1985; Leonard 1987b:261). It
seems logical to identify it with biblical Beth-arbel (Hosea 10:14) Arbela of Eusebius (Onomasticon
14:18). The site of Ramtha one of two candidates for the identication of biblical Ramoth-gilead2 has
never been properly investigated (Knauf 2001). Tell el-usn another prominent mound in the Gilead
and the second candidate for the identication of Ramoth-gilead (Dalman 1913:64; Albright 1925:16;
1929:11; Weippert 1997:32-33) also produced Late Bronze nds (Leonard 1987a:359). Zura can be
identied in any of these three sites, though we tend to prefer Tell el-usn (and to identify Ramoth-gilead
at Ramtha also Knauf 2001). In any event, the name of the city had been changed in the long process
of transition from the Late Bronze to the Iron Age, and disappeared from the toponyms of the Gilead.
EA 207 which is petrographically similar to EA 364 was written either at Ashtaroth, or in a
neighbouring city that used the same kind of clay as that of Ashtaroth.
The third petrographic group includes EA 201-205 (the Bashan cities; this group may also include
EA 206 that was not examined by us), EA 200 (from an unknown place) and EA 241 (from Sharuna). In
the light of Morans observation (1992:273 n. 2; 278 n. 1), that EA 195 (from Damascus) and EA 201206 were written by the same scribe, we suggest that Biryawaza of Damascus and his scribe traveled to a
certain place in the Bashan, where they met with the rulers of local cities, delivered to them instructions
from the Egyptian administration and wrote these letters on their behalf. Needless to say, this conclusion
is supported by the fact that EA 194 and 196-197 (from Damascus) are petrographically different from EA
201-206. The presence of fresh basalt and dolerite inclusions in the latter, together with limestone and clay
loam, may indicate a source on the margins of the Bashan volcanic plateau, perhaps in the Yarmuk area.
A journey of Biryawaza to the Bashan is mentioned in EA 197. We suggest that this trip was
connected to the preparations for the Egyptian campaign to Canaan. The beginning of letter EA 197
is broken; Naaman (1988c:183; Moran 1992:275) suggested restoring lines 1-2 [This is (?) what] he
said to [me when] your servant was in the city of A[dura (?)]. Adura is mentioned in the Kom el-etan
topographical list of Amenophis III and in papyrus Anastasi I and should undoubtedly be identied with
biblical Edrei (modern Der>a), in the southern Bashan (Edel 1966:11; Ahituv 1984:90-91). Biryawaza
complains to the Pharaoh that the king of Ashtaroth incited the city of Yanuammu (Yenoam), possibly
Tell esh-Shihab, on the Yarmuk river (Naaman 1977), to block its gates in front of him, and that the
kings of Ashtaroth, Buruna and alunni attempted to kill him. The journey of Biryawaza to the southern
Bashan is best explained by his role in the organization of the Egyptian campaign. The city of Adura, or
another city in the southern Bashan, could have been the gathering place of Biryawaza and the six rulers
of the Bashan the place from where the six letters were sent to the Pharaoh.
EA 200 and EA 241 are the only letters in this group that deal with matters other than those in EA
201-206. Hence they must have been written in the same area but on other occasions. The city of Sharuna
must also be sought in the southern Bashan. This location ts in very well since Sharuna is the only non1. A parallel case may be that of the Jerusalemite scribe who sent EA 291 from Gezer (Chapter 14.I).
2. Tell Ramith = Rumeith (Glueck 1943), a 0.2 ha site on a small knoll to the south of Ramtha (G.R. 247 212), is too small for
Ramoth-gilead (Lapp 1993).
224
Palestinian city that appears in Papyrus Petersburg side by side with several Canaanite cities (Epstein
1963; Helck 1971:166).
The fourth petrographic group, which includes EA 208-210, should be attributed to a place outside
the basaltic area of the Bashan, but not too far away, as basaltic minerals (including the unstable olivine
and iddingsite) appear in the wind-blown silt. This group is petrographically similar to EA 194, 196 and
197, which were sent from Damascus. There are two alternatives for their provenance: according to the
rst, they were sent by rulers who had no scribe in their court and when needed, traveled to Damascus
to employ the local scribe. According to the second, they were dispatched by rulers of cities in the
Damascus basin (for Late Bronze sites in this region see Supplement).
TABLE 11.1: PETROGRAPHIC GROUPING OF THE BASHAN AREA LETTERS
EA
Matrix
AOB
LS
PI
CT
QZ
364
PM
**
**
207
PM
**
**
337
PM
**
**
VM
CA
Origin
Ashtaroth
200
CL
**
***
201
CL
202
CL
203
CL
204
CL
***
205
CL
**
***
241
CL
**
***
**
334
CL
***
**
336
CL
***
**
208
CL
***
209
CL
@?
210
CL
***
@
@
Southern
Bashan or
Yarmuk
Valley
Zura
Damascus
Legend
Matrix types: PM = Paleocene marl; CL = clay loam.
Inclusion types: AO = Alkali-olivine basalt and dolerite (Miocene Pleistocene) and their derived minerals; LS = limestone;
PI = pisoliths; CT = chert; QZ = quartz; VM = vegetal material; CA = calcite. Inclusions with grey background are group deners.
Frequency: *** dominant, ** frequent, * scarce, @ undetermined (SPA samples).
225
CHAPTER 12
THE GALILEE, THE COASTAL PLAIN OF ACCO

AND THE NORTHERN VALLEYS
I. HAZOR (AURA)
The site of ancient Hazor (Tell el-Qeda) covers an area of about 80 hectares. Most of this areain the
lower moundwas inhabited only in the second millennium BCE. At that time Hazor was the largest
city in Canaan.
Hazor is the only Canaanite city mentioned in the Mari archive. The tablets referring to Hazor
deal with its diplomatic and commercial relations with Mesopotamian and north Syrian kingdoms
(Bonechi 1992, with earlier literature). Several tablets of the Old Babylonian period were unearthed in
the excavations of Tel Hazor, including a letter listing textiles and luxury goods probably prepared for
dispatch to Mari (Horowitz and Wasserman 2000). The excavations indicate that Hazor was the most
prominent city in Canaan in the Middle Bronze Age II and that it managed to maintain its status also after
the Egyptian conquest in 1457 BCE. Yet the Amarna letters do not refer to any dominant position of the
city in northern Canaan. The only meaningful evidence for its power comes from EA 364, in which the
ruler of Ashtartu (Ashtaroth) in the Bashan complains that the ruler of Hazor captured three of his cities.
This seems to indicate that the kingdom of Hazor had a common border with Ashtaroth.
Late Bronze Hazor probably controlled the northern Jordan Valley, the Upper Galilee, part of the
eastern Lower Galilee and the western part of the Golan Heights one of the largest territories in Late
Bronze Age Canaan. According to Finkelstein, it bordered on the territories of Akka (Acco) and urru
(Tyre) in the west, Shamuna and Anaharath in the south, and Damascus, Ashtartu (and Piilu?) in the
east. In the north it probably bordered on the territory of the Egyptian administrative centre of Kumidi.
According to Naaman, there were vast uninhabited areas on Hazors eastern, western and northern borders,
all of which must be considered as no-mans-land. Thus, Hazors territory bordered only on the territories
of Anaharath, Ashtartu and possibly Shamuna. The prominent position of Hazor in northern Canaan
was backed by its large population, its location on the international road to the north, and its large and
geographically diverse territory which provided it with access to varied natural and human resources.

Hazor is located on the southwestern edge of the Huleh Valley. The local lithology includes Neogene to
Quaternary conglomerate units. In the immediate surroundings of the mound there is a series of Senonian
to Middle Eocene chalks with isolated chert beds, Turonian dolomite and limestone and a Pliocene to
Quaternary alkali-olivine basalt series (Sneh et al. 1998).
Petrographic examinations of pottery from Hazor was undertaken by Shenhav (1964). The other
petrographic references for Hazor have been discussed elsewhere (Goren 2000a). They included thinsections that were made from large Late Bronze pithoi found in large quantities at Hazor but rarely in
other contemporary sites. Due to this factor and to their petrographic traits that suit the local geology
around Hazor, we assume that they were of local production. In addition, we collected sediments (clays
and wadi sand) around Hazor and used them to produce a set of comparative thin sections.
226
Fig. 12.1: Main city-states and Egyptian administrative centres in central and southern Canaan.
227
CATALOGUE
EA 227 (BM 29830), from the ruler of aura to the King of Egypt
Reliability: Satisfactory when taken together with the thin-section given to the British Museum.
Matrix: Carbonatic, light yellowish- brown in PPL, exhibiting partial isotropism due to the high ring
temperature that effected the calcite crystals both in the matrix and in the inclusions (estimated at
around 8000C), but with speckled b-fabric. Calcite crystals, partially decalcied due to the ring
and usually amorphous in shape, appear within the matrix (~10%), ranging between 5m to 60m
in size. Micritic limestone particles appear in varying sizes, from a few tens of micrometers to about
150m. Quartz silt is scarce (1%). Abundant (about 7%) iddingsite particles appear in various sizes
(ranging between 15m and 150m). A few of the larger still preserve their original olivine core.
Inclusions: The inclusions are very dense relatively to other tablets, including grits up to about 2 mm in
size (observed by stereomicroscope but not represented in the petrographic sample). All are rounded
and badly sorted. They include frequent coarsely crystalline olivine basalt (up to 600m). Micritic
limestone (up to 100m) partially decalcied due to the high ring temperature is common as is
clear replacement chert with common rhombic pseudomorphs after dolomite (up to 500m). There
are a few grains of chalk with with abundant foraminifers (up to 350m).
Geological interpretation: The matrix and the inclusion assemblage hint at an area in which calcareous
and basaltic formations are exposed. The spherical nature of the inclusions indicates the use of
sand from a wadi that drains an area where basalt, chalk and limestone, and chert are exposed.
This description agrees with the lithology of many parts of the eastern Galilee in general, and the
surrounding of Tel Hazor in particular (Sneh et al. 1998). Therefore, the attribution of this tablet to
the city-state that is mentioned on it is very likely.
Conclusions: Since EA 227 represents a coarse ware in terms of cuneiform tablets, it was compared
with the locally-made Late Bronze Age coarse ware of Hazor, which it resembles. The abovementioned thin-sections made from Late Bronze Age pithoi were used as references. The pithoi and
EA 227 proved to be identical in their petrographic composition. Therefore it is evident that EA 227
was produced at Hazor.
EA 228 (BM 29831), from Abdi-Tirshi to the King of Egypt
Reliability: High.
Matrix: Carbonatic, extremely rich in micrometer-size calcitic particles, birefringent with speckled bfabric, with rather abundant in foraminifers and badly sorted silty to sand-sized calcareous particles,
both micritic and sparitic. The matrix is also spotted with dark reddish-tan, iron-rich clay spheres,
sometimes containing quartz or calcite silt (~2%, up to 250m), iddingsite particles, and quartz silt
(4%). Opaque minerals, which appear in a range of sizes from a few micrometers to about 50-60m,
are relatively common (~2% of the matrix). They are angular in the ne fraction and rounded in
the coarser. These include opaque, through deep red and nearly opaque, to reddish-brown minerals
(presumably magnetite and haematite), the latter occasionally staining the clay.
Inclusions: Badly-sorted, spread (f:c ratio{0.062mm}=98:2-97:3) sand in which rounded foraminiferous
chalk (up to 400m) is dominant.Fragments of weathered alkali-olivine trachytic basalt, coarsely
crystalline basalt and a clear olivine crystal (up to 350m) are common and there are a few grains
228
of sparitic limestone: (up to 800m but usually not exceeding 300m) partially decarbonated due to
the high ring temperature.
Palaeontology (LG): Probably a mixture of small Senonian and Paleogene planctonic foraminifers in the
matrix and the inclusions: Acarinina (p), Hedbergella (p), Heterohelix (p), Morozovella (p).
Geological interpretation: To the naked eye this tablet is somewhat dissimilar to EA 227 as was already
noticed by Knudtzon (1915:1300, n. 3). Its fabric differs in colour (Munsell colour value: 7.5YR5/
6). The texture is hard and much ner grained than that of EA 227. Yet, EA 228 still indicates an
environment that suits Hazor. Most differences between the two tablets seem to be on the technical
side: better selected clay, more delicate inclusions (perhaps just the naturally occurring grains
within the soil, not intentionally mixed wadi sand) and a somewhat lower ring temperature. The
micropalaeontological data indicate a Senonian to Paleogene environment, which together with the
basalts typies the Hazor area. Therefore, we are inclined to interpret EA 228 as a more carefully
prepared Hazor tablet.
Reference: In terms of its petrofabric, EA 228 is identical to the mathematical tablet that was discovered
in the 1996 excavation season at Hazor (Horowitz 1997:190-197) and examined by Goren (2000a:
34-35). It is also similar to three other tablets that were recently unearthed at the site (ibid. :36-38).
EA 228, though later in date, was produced by a similar technique.
Conclusions: EA 227 and 228 demonstrate that cuneiform tablets may differ in both fabric and technology
although they are made in approximately the same period and at the same site. The gritty texture of
EA 227 is similar to that of the common coarse pottery that was made at the site, while EA 228 gives
the impression of a more selective and careful production.
COMPARATIVE STUDY: LATE BRONZE CUNEIFORM TEXTS FROM HAZOR

The results of the petrographic study of the Middle and Late Bronze tablets found so far in the excavations
of Hazor have been published previously (Goren 2000a). The Late Bronze tablets are presented again
here in conformity with the descriptive procedures established for this study.
IAA 67-1188/1189, an inscribed liver model
Reliability: High.
Matrix: Carbonatic, very rich in foraminifers, with abundant magnetite particles.
Inclusions: The inclusions are very coarse relative to the other tablets and contain grits over 1 mm in
size. Subrounded alkali-olivine coarse-crystalline to doleritic basalt (up to 1500 m) is dominant.
Subrounded micritic limestone (up to 450 m) is common as is foraminiferous chalk (up to 500 m).
Vegetal material (SLY): Common (up to 200 m), voids sometimes enclosing charred vegetal tissues or
phytoliths, all remnants of straw.
Firing temperature: Probably below 7000C judging by the partial decalcination of some of the calcite
Geological interpretation: The inclusion assemblage is rather like that of EA 227, although similar
inclusions (often with foraminiferous marly matrices) appear in tablets of the Bashan cities
(Chapter 11).
Conclusions: Two fragments of the same clay liver model were discovered in 1958 in Area H, Locus 2178
(IAA Reg. No. 67-1188/1189). They were collated, published and interpreted by Landsberger and
Tadmor (1964). Textually, a local provenance is possible; petrographically this item may represent
a ware similar to that of EA 227.
229
IAA 1997-3307, an economic document

Reliability: Fair. The petrographic examination was made by SPA because this tablet was perfectly
preserved and completely covered by the script. A larger sample could have made the interpretation
more conclusive, but this could not be done without damaging the inscribed area of the letter.
Matrix: The matrix is carbonatic with some foraminifera fragments and little silt (about 1%). Opaque
angular mineral grains (magnetite?) appear in the ne silt fraction. Scattered concentrations of clay
(glauconite?) appear in the matrix.
Inclusions: The inclusions contain rounded, well-sorted grains of quartz, limestone and some chert.
Elongated voids in the matrix indicate the presence of some vegetal material (straw) that vanished
in the process of ring.
Geological interpretation: The sample suggests that this tablet is similar in its petrofabric to EA 187
which was sent from Enishasi a small kingdom located in the Lebanese Beqa>. However, the small
size of the sample renders this similarity questionable.
Conclusions: This tablet was published and interpreted by Horowitz (2000:17-25). Petrographically it
is not local to Hazor. Regarding the possibility that it was produced in the Beqa of Lebanon, one
should note that the kingdom of Hazor was located on the international route that led to the Beqa.
Therefore, economic relations with a kingdom located there are plausible.
Horowitz read the writers name mAd-du-ap-di (Adduapdi). Looking at the facsimile, the name could
possibly be rendered mAd-du-um-mi (Addumi; this reading was suggested to Naaman by Elnathan
Weissert). If this reading is acceptable, the writer may be identied with Aaddumi, a mayor captured by
the Hittites when they attacked the Lebanese Beqa> (EA 170:14-18).
IAA 1997-3308, a short administrative document
Matrix: In thin section the petrofabric is characterized by argillaceous ferruginous clay, rich in shale
fragments. Abundant opaque minerals and silt (~10%) are found in the matrix. Other indicators of
this group are diversied shale fragments, some of which are ferruginous while others tend to be
more clayey.
Inclusions: The inclusions contain frequent subspherical to subangular grains of chert: (up to 900 m) with
micritic limestone (up to 500 m) being common as are weathered fragments of alkali-basalt (up to 500
m). Yellowish fragments of tuff or volcanic glass occur as secondary components.
Firing temperature: The conservator has red the tablet in order to preserve it (O. Cohen, pers. comm.).
The crumbly nature of the tablet was due to overring and the complete decalcination of the
limestone inclusions (perhaps as part of the destruction of the palace where it was found) that caused
its cracking. The original ring temperature cannot be estimated, but the secondary ring was about
9000C, as the calcite is decomposed while the matrix is not completely isotropic.
Geological interpretation: In its petrographic afnities this short docket seems to belong to the same
petrofabric as EA 165-167, namely to the coast of the Akkar Plain (a correction to Goren 2000:38).
Although the carbonates were damaged by the high ring temperature (thus Amphiroa fossils, if
included in the sample we took, were not preserved), the other afnities are typical. For the reasons
explained above (regarding EA 165-7) Tell Kazel (umur) seems to be the only likely provenance.
230
Reference: See EA 165.

Conclusions: This document, published and interpreted by Horowitz (2000:25-28), was sent from the coastal
ank of the >Akkar Plain, apparently from umur. The tablet is probably a docket that was dispatched
along with a certain merchandise (silver?) that is not mentioned in the text. This is another indication for
Hazors economic relations with kingdoms and cities located to its north.
CONCLUSION: THE HAZOR CORRESPONDENCE

From a petrographic point of view, the two Amarna tablets (EA 227 and EA 228) suit the general geology
of the Hazor area.
II. ACHSHAPH (AKSHAPA)

The city of Achshaph (Akshapa in the Amarna letters) is mentioned in the Egyptian Execration
texts, in Thutmose IIIs list of conquered Canaanite cities, in Papyrus Petersburg 1116A of the time
of Amenophis II (Epstein 1963; Helck 1971:166), in the Amarna archive, in Papyrus Anastasi I and
in the Bible (Josh 11:1; 19:25) (Ahituv 1984:48-49). The documentary evidence indicates that Acco
and Achshaph were the two most important cities in the Acco Plain in the second millennium BCE.
Two main identifications have been suggested for Achshaph: Tell Keisan on the eastern side of the
plain, southeast of Acco, and Tell Harbaj near the southern border of the plain (for discussions and
bibliography see Briend 1972; Lemaire 1991:140-141; Lipiski 1991:158-159). However, Tell Harbaj
is a relatively small mound (about 3-4 hectares in size), whereas Tell Keisan is an impressive site
of about 6 hectares which dominates the area around it. The excavations conducted at Tell Keisan
indicate that it was settled in both the second and first millennia BCE (Humbert 1993), which fits the
documentary evidence for Achshaph.
It seems that the territory of Achshaph encompassed the southern Acco plain and the hills
overlooking it in the west. According to Finkelstein, it bordered on the territories of Acco in the north,
Shamuna in the east, Tel Yokneam in the south and possibly Ginti-kirmil in the southwest (see below).
According to Naaman, a third city-state, named Mishal, existed in the Acco plain, possibly being
located on Achshaphs southern border.1
It is possible that Geba-mn, whose ruler is mentioned in a royal inscription of Amenophis II, also
served as a centre of a city-state. It was probably located at Tell el->Amr, at the southeastern end of the
valley (Aharoni 1967:156; Ahituv 1984:100-101; for the site, see Schmitt 1987:45-46). Thus, Achshaph
could not have had common borders with Tel Yokneam and Ginti-kirmil.
1. Mishal is mentioned in the Execration Texts, in the topographical list of Thutmose III, in the list of Canaanite city-state
messengers sent on ofcial mission to Egypt (Papyrus Petersburg 1116A) and in the Bible (Josh 19:26) (Epstein 1963;
Helck 1971:166; Ahituv 1984:143). Naaman compares the political-territorial disposition in the Late Bronze Age to that
of the Middle Bronze, when the Acco plain was divided between the three city-states of Acco, Achshaph and Mishal
(Aharoni 1967:132-133; Helck 1971:51). Kempinski (1986:70-72) suggested that Middle Bronze II Kabri also had the status
of a city-state, thus increasing to four the number of Middle Bronze Age city-states in the Acco plain. Finkelstein (1996:
238-239, 254) rejects this hypothesis and suggests a bipartite division of the Acco plain. Mishals identication at either
Tell Harbaj or Tell en-Nal is possible, but not certain (Ahituv 1984:143; Lemaire 1991:140-141; Lipiski 1991:160 with
earlier literature).
231

Reference materials of workshops from the Acco region were collected from several sites.
1. Wasters and locally produced pottery from the Byzantine workshop excavated at urvat >Utza,
located ca. 5 km east of Tel Acco (Getzov 1993). The petrographic analyses were carried out by
Goren but not yet published.
2. Pottery wasters from a mediaeval workshop found at the Acco courthouse (Goren 1997).
3. Petrographic data on the common pottery at Tel Keisan, published by Courtois (1980:355). For
Tell Harbaj, samples of sand collected in the channel of the nearby Kishon River were used for
comparison with the inclusions of the Akshapa tablet.
CATALOGUE
EA 223 (VAT 1870), from Endaruta to the King of Egypt
Reliability: High.
Matrix: Clayey, ferruginous, with speckled b-fabric, about 2% quartz silt and abundant silt-size calcareous
bodies. Opaque minerals are abundant in the silt fraction. The silt contains heavy minerals including
hornblende, zircon and epidote.
Inclusions: Sand with dominant badly-sorted rounded grains (up to 1.2mm) of foraminiferous chalk,
often with glauconite concentrations (see EA 59 for the de nition of glauconite and its geological
interpretation). Frequent to dominant are rounded to subangular grains of quartz (up to 300m) with
a few accessory feldspars that appear in the same grain-size and there are occasional glauconite
concentrations (up to 250m), perhaps derived from the chalk.
Palaeontology (LG): Loose foraminifers in the matrix and in the inclusions: Hedbergella (p), Heterohelix
(p). Age: Senonian.
Geological interpretation: The combination of ferruginous clay matrix with sand of the Israeli coastal
plain (see EA 97, above), including quartz and accessory minerals, indicates a coastal origin. In
this area, red to dark reddish-brown silts and sands with loams in soils stained and in part cemented
by sesquioxides of iron and aluminum appear as part of the Rehovot Formation, dated from the
Villfranchian to Recent age (Issar 1968; Sivan 1996). It is most likely that this red soil of the central
littoral areas of Israel, locally termed hamra soil, was used here. Hamra soil is spread along the
coastal plain from the Ashdod area to the north. As already pointed out (with regard to EA 168),
coastal sand of the classication described here does not extend as the dominant component north
of Rosh ha-Niqra on the northern border of Israel. Therefore, this tablet should be related to a site
located in the coastal plain, between Ashdod and Rosh ha-Niqra.
The dominance of chalk seems to hint at nearby chalk formations. Eocene or Senonian chalks occur
next to the coastal plain in the Shephelah hills, in Ramat Menasheh south of Mount Carmel and in
the hills overlooking the coastal plain of the western Galilee. In this rather large area glauconitic
chalk has been reported only from the Kabri Marl formation (Baida 1963) which occurs in the
centre of the Senonian section (Mount Scopus Group). The Senonian age of the unit from which the
inclusions were derived is also indicated by the micropalaeontological data. The Kabri formation
is exposed in the hills of the western Galilee. Therefore, the origin of EA 223 can be limited to the
coastal plain of the western Galilee, from Qiryat Atta northwards.
232
The sites of Tell Harbaj and Tell Keisan should be considered for matching with the petrographic
data. Though they are rather close, being located about 5 km from one another, they differ in their
geological background. Tell Harbaj is situated on the Kishon River, and more generally in the
opening of the Jezreel Valley to the coastal plain. Tell Keisan is situated in the Acco Plain which is
characterized by different sediments. The combination of sand derived from an area with Senonian
chalks mixed with hamra soil ts Tell Keisan better. At Tell Harbaj the sediments of the Jezreel
Valley, dragged as sand by the Kishon River, are still signicant alongside alluvial soils mixed with
coastal sediments. Moreover, the drainage system of the Kishon River includes the northern slopes
of Mount Carmel so that the sediments are expected to be more heterogeneous. Therefore Tell
Keisan should be the preferred source for this tablet.
Conclusions: In the western Galilee, two pottery assemblages that were examined petrographically are
seemingly similar to EA 223: the wasters and locally produced pottery from the Byzantine workshop
excavated at urvat >Utza (Getzov 1993, the petrographic analyses were made by Goren but not
published as yet) and the dominant group in the common pottery at Tell Keisan (Courtois 1980:355).
CONCLUSION
The identication of Achshaph at Tell Keisan has important implications for the discussion of the
territorial disposition of the Canaanite city-states in the Late Bronze Age. Achshaph is located very
close (about 5 km) to Acco the capital of another important city-state. This means that the capital of
a Canaanite city-state was not necessarily located in the centre of its territory. According to Naaman,
this fact challenges attempts to draw a map of the Canaanite city-state system on the basis of Thiessen
polygons or central place theory (Bunimovitz 1989; partially also Finkelstein 1996).
III. SHIM>ON (SHAMUNA)

The identication of Shim>on (Shamuna) at Tel Shimron (Khirbet Sammuniyeh) is accepted by all
scholars (Rainey 1976a). The Late Bronze Age site was apparently limited to the upper mound, leaving
the huge area enclosed by the Middle Bronze Age ramparts beyond its limits. The territory of Shim>on
embraced the northern margins of the Jezreel Valley and the central Lower Galilee. Shamuna must have
bordered on the territories of Achshaph and Tel Yokneam in the west, Hazor (and according to Finkelstein,
possibly also Acco) in the north, Megiddo in the south and Anaharath in the east. In the north, the land
of Shim>on included the town of innatuna (EA 8:17, 245:32 = biblical Hannathon), located at Tell elBedeiwiyeh. This means that the entire valley of Beth Netofa was under its domain.

Tel Shimron is located on the northern border of the Jezreel Valley, in an area dominated by Eocene
limestone, chalk and chert. A small outcrop of the Miocene Lower Basalt is exposed on its southern slope.
The local soils include predominantly the Jezreel Valley alluvial soils with more localized rendzina that
develops on the chalk units.
The site has never been excavated systematically although a thorough study of surface nds at the site
was carried out (Portugali 1982). In addition to the geological mapping we also conducted a limited survey
of the site, collected Late Bronze Age sherds from the surface and prepared thin-sections from them. In
order to obtain more comparative material we used a selection of Portugalis survey sherds. This is obviously
233
a very unreliable method for establishing a reference collection, as some of the sherds could have been made
elsewhere and imported to the site. Thus these thin-sections were used selectively and cautiously.
CATALOGUE
EA 224 (BM 29849), from um-Adda to the King of Egypt
Reliability: High.
Matrix: The matrix is carbonatic, light yellowish-tan in PPL, with speckled b-fabric, containing infrequent
foraminifers. It is rather silty (about 4%-5%) and opaque minerals appear in a range of sizes from a
few micrometers to about 30-40m.
Inclusions: The inclusions (f:c ratio{0.062mm}=95:5) are dominated by moderately sorted, rounded sand
particles including subangular to well-rounded fragments (up to 750m) of olivine basalt of several
types, ranging between nely crystalline to nearly doleritic. The basalt is usually rather weathered
and the olivine is partly or entirely altered into iddingsite. Single crystals of basalt-derived minerals,
including iddingsite, augite, and twinned plagioclase, appear in grain-sizes reaching 250m.
Rounded grains (up to 800m) of micritic, sparitic and biogenetic limestone are frequent. There
are a few fragments (up to 500m) of tan clay-stained travertine, a very few calcareous pisoliths
with concentric microlamination and a central hollow, a few angular to subrounded grains (up to 1
mm) of replacement chert, a few spherical sand-sized quartz grains (up to 300m) and a few fresh
(unfossilized) aquatic mollusc shell fragments (up to 500m). Also seen are a very few rounded
grains of foraminiferous chalk (up to 350m).
Palaeontology (LG): Mixture of Paleogene foraminifers (in matrix) and some Senonian foraminifer
ghosts in the inclusions: Acarinina (p), Hedbergella (p), Heterohelix (p), Subbotina (p) sp.
Vegetal material (SLY): Common (up to 600m long). A leaf fragment from a cereal and a number of
non-indicative tissue fragments. One tissue fragment includes primary xylem with its typical spiral
secondary cell wall thickening.
Geological interpretation: The inclusions of EA 224 contain an assembly of rounded or nearly rounded
rock and mineral fragments, including several types of calcareous rocks (limestone, chalk, travertine),
basalt, chert, and quartz grains and fragments of aquatic shells. From their shape and sorting it may be
concluded that they were all derived from river sand that was collected and perhaps sieved, then used
as the non-plastic component for making this tablet. The presence of recent (not fossilized) aquatic
mollusc shells indicates an active stream where water is found during signicant parts of the year.
Travertine and pisoliths imply the existence of nearby springs or even a lacustrine environment.
Noteworthy is the dominance of basalt fragments and basaltic derived minerals in the inclusions.
The basalts belong to the alkali-olivine category. The olivine phenocrysts are altered into iddingsite.
These af nities are typical of the Neogene-Pleistocene volcanics of the Galilee (Oppenheim 1959)
or the Golan Heights (Mor 1973). The basalt grains are thus younger, i.e., not the Lower Cretaceous
basalt that was described before (see EA 60), or the Upper Cretaceous volcanics that will be treated
later (see EA 259 below). The presence (and dominance) of the younger basalt types eliminates any
attribution of EA 224 to any part of southern or central Palestine. Neogene to Pleistocene alkaliolivine basalt ows are found from the Jezreel Valley and northwards, in the central Jordan Valley
from Beth-shean to the Sea of Galilee, and in the eastern Upper Galilee. Thus the origin of EA 224
should be sought in this area.
234
Other important components in the inclusions are the chert and the quartz. In northern Israel,
chert can be connected with the Mount Scopus Group but more typically with the lower and
middle Eocene sequence. The quartz sand is mature (i.e., rounded and not accompanied by any
accessory minerals such as feldspars, amphiboles or pyroxenes) and hence derived from mature
quartzitic sandstones. In the areas where younger basalts outcrop such sand may be related only
with the lower formations of the Lower Cretaceous lithological section. These formations outcrop
widely on the slopes of Mount Hermon, as well as in some limited exposures in the Ramim
ridge of the eastern Upper Galilee. They continue northwards into Lebanon (now combined into
the Chouf formation, see EA 60, above). Mimran (1969) and Shaliv (1972) presented a detailed
petrologic description of the analogous Lower Cretaceous formations in Wadi Far >ah and Wadi
Malikh in eastern Samaria, and a profile of the Hermon outcrops was presented by Heiman
(1985). The Jordan River drainage system supplies such quartz grains from the Hermon area to
the central Jordan Valley, and extinct streams may have drained the Ramim ridge (where similar
formations outcrop) into the Jordan (Goren-Inbar et al. 1992; Goren and Fischer 1999). Since it is
very unlikely that the scribe who created the tablet intentionally selected grains of specific rock
types from the sand, the proportions of the different inclusions must represent the composition of
the river sand near the location of the sender.
Needless to say, this description does not t the area of Tel Shimron (Shamuna of EA 224).
Although the site is located on a hill where Eocene chalk and a small exposure of alkali-olivine
basalt occur, other components within the inclusion assemblage are not represented in the local
geology. In the rst place, no stream that could drain such a collection of minerals and rock
fragments is found in the entire area of the central Jezreel Valley or around it. In the broader sense,
no stream in the hilly Galilee drains an area that is large enough to include such a rich variety
of rock types derived from different geological ages ranging from the Lower Cretaceous to the
Quaternary. The only possible candidate is the Jordan River, which collects sediments from the
entire area between Mount Hermon, the Rukkad and Yarmuk basins, the eastern Galilee, the Bethshean Valley and the northern Jordanian Plateau. Therefore, the origin of EA 224 should be sought
in the Jordan Valley.
As the inclusions represent sand that was collected from the Jordan conduit, they must have been
gathered from a point where the river was draining areas with Neogene to Pleistocene basalts,
Lower Cretaceous sandstone, Eocene or Senonian chert and chalk, Cretaceous limestone series
and Quaternary spring or lake deposits. Practically, only the area south of the Sea of Galilee can be
considered for such an assemblage, since Senonian or Eocene sections (the possible supplier of chert)
are not found to its north. Therefore the origin of this tablet should be sought in the central Jordan
Valley, from the Sea of Galilee southwards.
A previous study (Goren and Fischer 1999) can limit the provenance of EA 224 even more. The
primary goal was to use the preliminary data gained during on-site examination of ceramics at sites
in the Jordan Valley between the Sea of Galilee and Wadi Zarqa (exactly the area discussed here) in
order to de ne the typical ngerprint of each possible production area in terms of local river sand
composition. The central Jordan Valley was treated as a drainage system which transports minerals
and rock fragments extracted from the diverse lithologies that expose along its secondary streams.
The various rock types that characterize different parts of the Jordan basin can be petrographically
identied even when inspected as sand-sized particles. Therefore, sites located along the main channel
or along the tributaries can be characterized by the composition of the sand of their stream.
235
The study indicated that several components act as fossile directeurs to certain parts of the area
as they appear in the Jordan sand only from a certain spot and down-river. For example, travertine
and lacustrine pisoliths appear only south of the Beth-shean Valley, where they dominate the local
lithology. In the Beth-shean Valley and in the equivalent area east of the Jordan they appear as part
of the Holocene Tabgha formation (Schulmann 1962; Horowitz 1979, 1986; Kronfeld et al. 1988)
or as Qt (Quaternary travertine) in the most recent, detailed (1:50,000) geological map of the area
(Hatzor 2000). Recently, the pisolith bearing layer has been divided from the lower travertines
and defined as the Rehov formation (Horowitz 2001: 546-7). This formation exposes between
Naal Harod to the north, Naal Bezek to the south and west, and the Lisan formation in the Ghor
area to the east. From the junction of the Jordan with Wadi Zarqa and southwards the proportions
of the sand components again change drastically as quartzitic sands, derived from the widespread
Lower Cretaceous sandstones that outcrop along the Zarqa stream, become dominant. From this
point down-river the quartzitic component increases to become one of the major constituents of
the Jordan sand.
This description leads us to the conclusion that the origin of EA 224 should be sought in the central
Jordan Valley, between Beth-shean and Wadi Zarqa. This is the only area where all the components
discussed above can be found. The site should be adjacent to the Jordan River, since the secondary
streams that drain into it around the Beth-shean Valley (e.g. Naal Tabor, Naal Issachar and
Naal Bezek) reect their local nature and contain smaller varieties of rock-types (Nir 1960). Our
experiments proved that the secondary rivers east to the Jordan (e.g. Wadi Yabis) reect a similar
situation (Goren and Fischer 1999).
Reference: The general petrographic class described above is known to dominate ceramic assemblages
of the central Jordan Valley, including Pottery Neolithic Munhata (Goren 1992) and Sha>ar Hagolan
(Goren 1991a), Early Bronze IV Tel >Amal (Goren 1991d), sites with Early Bronze I Band-Slip (or
Grain-Wash) ware, Early Bronze Ib-Early Bronze II Tell Shalem (Goren, unpublished), Khirbet Kerak,
Beth-shean, Tell ash-Shunneh and Tell es-Sa>idiyeh (for the latter, A. Middleton, pers. comm.).
More specically to the Beth-shean area, our reference material relies on studies of ceramic
assemblages from Beth-shean and Tel Reov. Cohen-Weinberger (1998) investigated Egyptianstyled and local ceramics from the period of the 20th Dynasty at Beth-shean, as well as several Early
Bronze Age wares from that site. Bozaglo (2004) studied the Iron Age I and Iron Age II assemblages
from Mazars excavations at Beth-shean and Tel Reov. Examination of their samples revealed that
the petrofabric of EA 224 dominates the assemblages of both sites.
Conclusions: EA 224 is one of a group of tablets assignable to the Beth-shean Valley. In this area only
three sites Beth-shean, Tel Reov and Pella (Piilu of the Amarna archive) can be considered
as the place of origin of this letter. Excavations at Beth-shean have shown that in the Late Bronze
Age it was the major Egyptian administrative centre of northern Palestine. Tel Reov and Pella
functioned as capitals of two city-states which dominated the region west and east of the Jordan
respectively (see below). Needless to say, in the case of a letter of a distant city-state which was
sent from the central Jordan Valley the only logical place of origin is Beth-shean. In other words,
the ruler of Shamuna (or his scribe) appeared before the Egyptian authorities at Beth-shean and
the dispatched the letter from there. The same petrofabric appears in tablets of two other cities that
are located in markedly different geological environments: EA 232, 234-235 from Acco and EA 285
from Jerusalem (Chapter 13.II). The ve tablets should be treated as a group of letters made of the
Beth-shean Valley sediments.
236
Theoretically, it may be argued that this is a case of recycling and that an un red letter, sent from
Beth-shean to Shum-Adda of Shamuna, was recycled and kneaded again to make a new tablet.
Extending this hypothesis, it may even be suggested that a tablet bearing an order (for tax payment
in this case) from the local Egyptian administrator was used to produce an immediate reply to the
king himself. We consider such a theory highly unlikely. Firstly, from the technical point of view,
recycling an old tablet made of coarse mixture (with relatively large inclusions) is no less (and probably
more) complicated than preparing a new one of local and readily available clay. Secondly, since
there is no shortage of clay sources around Shamuna, why would the addressee destroy the letter
for its clay rather than le it? Thirdly, a letter from Jerusalem made of similar Beth-shean clay (EA
285 below) is the sole exception in the Jerusalem correspondence. All the other tablets sent by Abdieba were made of local Jerusalem sediments. To sum-up, the recycling option should be rejected.
One could also argue that in such cases the letters were written by a visiting scribe
(Moran 1992:273 n. 2, 278 n. 1, but see 279 n. 1). In other words, a scribe from the Egyptian
centre at Beth-shean came with his clay to Shamuna and wrote the letter there. This
hypothesis is highly unlikely in view of the petrographic investigation of the Bashan letters,
which were indeed written by a visiting scribe. We discovered that Biryawazas scribe did
not write them on Damascus clay but used local materials from the Bashan (Chapter 16).
IV. ACCO (AKKA)

Akka (=Acco, Tel Acco) is located on the northern bank of the Na>aman river, about 700 m from the
present seashore and east of the modern city of Acco (Acre, St. Jean dAcre of the Crusaders). It was
the most important harbour of the coastal area between Jaffa and Tyre. Late Bronze Age Acco bordered
on Tyre, Achshaph in the south and Shim>on in the southeast. Finkelstein believes that on the east it
bordered on the territory of Hazor, whereas Naaman suggests that the Upper Galilee was a kind of nomans-land, and that this included the hilly areas which separated the city-states of the Acco plain and
the kingdom of Hazor.

Tel Acco is located in an area of Quaternary deposits. In the east of its exploitable territory there
are exposures of Senonian chalk and marl of the Mount Scopus Group. The Quaternary sediments are
composed of alterations of calcareous sandstone (locally termed kurkar), red loams (hamra), dark clays
and loose sands. Some of the calcareous sandstones accumulated and were cemented under marine
conditions, while others are of aeolian origin (namely eolianites; for the relevant data for reconstructing
the ceramic ecology around Tel Acco see Sivan 1996 and the maps therein).
The sediments exposed near the site are a Pleistocene kurkar series of the Kurdane formation that
is seemingly equivalent to the Pleshet Formation in central Israel. To the west lies a series of marine
carbonatic sandstone, rich in distinctive fauna, of the Yasaf Member (Sivan 1996:126-133) and dark,
fat and organic-rich clay of the Nahariya Member (ibid.: 149-152). The continental deposits include the
Holocene eolianites (exposing only north of Acco) and the hamra loam of the Hadera Sands (Netser
and Gvirtzman 1994). Near Acco, the latter are reported to contain up to 10% quartz (Sivan 1996:155).
However, some of our reference material (pottery wasters made of hamra from urvat >Utza near Acco)
seems to have slightly higher proportions of quartz (about 15%).
For reference materials see the treatment of EA 223 (Section II above).
237
CATALOGUE
EA 234 (VAT 1641), from Shatatna to the King of Egypt
Matrix: The matrix is carbonatic, light yellowish-tan in PPL with speckled b-fabric, containing
infrequent foraminifers and some scattered glauconite concentrations (the later up to 60m in
size). It is rather silty (about 4%-5%) and rich in opaque minerals that appear in a range of sizes
from a few micrometers to about 30-40m. Heavy minerals in the silt fraction include plagioclase
and iddingsite. Under higher magnications (>X200) the matrix is brous, optically active and
displays very weak optical orientation.
Inclusions: The inclusions contain sand (f:c ratio{0.062mm}=90:10) in which subangular fragments (up to
800m) doleritic alkali-olivine basalt where the olivine is altered into iddingsite is common and
single crystals of iddingsite, augite, and twinned plagioclase appear in grain-sizes reaching 500m.
Also common are spherical to angular sand-sized quartz grains (up to 380m). There are a few
rounded grains of micritic limestone (up to 300m) and fresh (unfossilized) aquatic mollusc shell
fragments (up to 500m).
Palaeontology (LG): A few small planctonic foraminifers: Bolivina (b), Hedbergella (p). Age: probably
Senonian.
Firing temperature: Very lightly red if at all judging by the lack of any change in the carbonates.
Geological interpretation: From a petrographic point of view, the materials of this tablet cannot be
regarded as local to Acco. Its raw materials are different from the coastal sediments that characterize
the Acco region, and indicate an inland origin. Moreover, they do not exist within an exploitable
range from Acco. The basalt grains that dominate the inclusions are fresh and indicate a conduit
that is adjacent to basalt deposits. Naal Kishon that drains into the Haifa Bay some 12 km south
of Acco does collect basalt clasts from the Jezreel Valley, but in its terminal section these grains
are extremely eroded, as attested in sand samples collected from it near Tell Harbaj. The closest
basalt exposures to Acco appear near Megiddo, at a distance of nearly 30 km to the southeast as the
crow ies. The overall composition of the inclusion set, indicating some riverbed sand, is alien to
the Acco Plain in particular and the Israeli coastal area in general. The use of marl for clay in this
area is also unusual. On the other hand, this tablet is similar in most of its petrographic details to
EA 224 (from Shamuna).
Conclusions: EA 234 is most likely another letter of a Canaanite city-state ruler that was sent from the
Egyptian administrative centre at Beth-shean.
EA 235 (BM 29815), from Shatatna to the King of Egypt
Reliability: Satisfactor/High.
Matrix: The matrix is carbonatic, yellowish-tan in PPL with speckled b-fabric, containing infrequent
badly preserved foraminifers and their fragments. Quartz silt is making about 4%-5% of the matrix.
It is also rich in opaque minerals that appear in a range of sizes from a few micrometers up to about
30-40m. Under higher magnications (>X200) the matrix is brous, optically active and displays
Inclusions: The inclusions (f:c ratio{0.062mm}=90:10) contain sand and some vegetal material. Subangular
to rounded fragments (up to 800m) of alkali-olivine basalt of several types, ranging from nely
238
crystalline to nearly doleritic are common. The basalt is either fresh or rather weathered and the
olivine is partly or entirely altered into iddingsite. Single crystals of basalt-derived minerals appear
in grain-sizes reaching 250m, including iddingsite, augite, and twinned plagioclase. Rounded
grains of micritic limestone (up to 600m) are frequent and rounded to subrounded sand-sized
quartz grains (up to 300m) are common. There are a few fragments of tan clay-stained travertine
(up to 800m) and a very few rounded and spherical glauconite concentrations (up to 380m).
Palaeontology (LG): Acarinina (p). Age: Paleogene.
Vegetal material (SLY): Common (up to 900m lengthwise). Several non-indicative tissue fragments in
a bad state of preservation.
Firing temperature: Probably unred or very lightly red since the organic atter is uncharred.
Conclusions: Like EA 234, this tablet was very probably sent from Beth-shean.
EA 232 (VAT 1640), from Surata to the King of Egypt
Reliability: Fair but sufcient for comparison with EA 234-5.
Matrix: Carbonatic, yellowish-tan in PPL with speckled b-fabric, containing infrequent badly preserved
foraminifers and their fragments. Quartz silt makes up about 5% of the matrix. It is also rich in
opaque minerals that appear in a range of sizes from a few micrometers to about 30-40m. The
matrix is optically active and displays an optical orientation.
Inclusions: Rounded fragments of alkali-olivine basalt with the olivine phenocrysts altering into
iddingsite together with rounded grains of micritic limestone.
Firing temperature: Probably very lightly red or unred, but there is not enough data to estimate the
temperature.
Geological interpretation: Most likely as EA 234.
Conclusions: Stereomicroscopic examination of this tablet and the small sample taken for SPA reveal the
same features as EA 234-235 (marly matrix, mostly basalt and limestone inclusions). This letter too
must have been sent from the Egyptian administrative centre at Beth-shean.
CONCLUSION: THE ACCO CORRESPONDENCE

Letter EA 232 was sent by Surata, and letters EA 234-235 were sent by Shatatna, his son and heir to
the throne (see EA 8:19). According to the petrographic analysis both rulers sent their letters from the
Egyptian centre of Beth-shean. Nothing in the text of the three letters hints at this.
In EA 234 Shatatna explains why he held a person that defected from the service of Biryawaza of
Damascus. The deserter probably passed through Beth-shean on his way to Acco. The dispatch of this
letter from Beth-shean might be explained on the assumption that Shatatna rst negotiated with the
Egyptian authorities of Beth-shean (the garrison city mentioned in lines 13-17) a site located on the
way to Damascus, and wrote his own version of the episode there. The reason why the other two letters
were written at Beth-shean remains inexplicable.
239
V. ANAHARATH
Biblical Anaharath of the inheritance of Issachar (Josh 19:19) has been identied at Tel Rekhesh (Tell elMukharkhash) in Naal Tavor (Israel Grid Ref. 1940 2288; Aharoni 1967; Gal 1981; 1991:22-25). Anaharath
appears in the annals of Thutmose III and in the inscriptions of Amenophis II, but is not mentioned in the
Amarna letters. The list of booty taken from Anaharath by Amenophis II, which includes inter alia 17
maryannu, 6 sons of princes and 7 teams of horses (Edel 1953:133-135, 156-157; Wilson 1969a:247), may
indicate that it had a status of a city-state. Indeed, Naaman (1986:481-482; 1997:617) specied Anaharath
as the centre of a city-state that dominated the basalt plateaux of the eastern Lower Galilee.
CATALOGUE
EA 237 (VAT 1701), from Bayadi (?) to the King of Egypt
Reliability: High.
Matrix: Dense, yellowish-brown to pale greenish-brown in PPL, carbonatic with speckled b-fabric, with very
few foraminifers. Quartz silt is scarce (less than 1%) but opaque minerals are more common (about 2%).
Also abundant in the matrix are silt-sized laths of iddingsite, twinned plagioclase and calcite.
Inclusions: No remains of vegetal material. Badly sorted sand of rock fragments and their derived
minerals, f:c ratio{0.062mm}=95:5%-94:6. The inclusions comprise predominantly doleritic or coarsely
crystalline alkali-olivine basalt (up to 1 mm) with phenocrysts of idiomorphic or hypidiomorphic
plagioclase prisms, olivine and augite. The basalt exhibits pellicular or complete alteration of the
olivine phenocrysts into iddingsite and the pyroxene is commonly serpentinized. To these we may
add detrital basaltic minerals (iddingsite, plagioclase, pyroxene and serpentinized pyroxene) and
some volcanic glass. Rounded to subrounded micritic (common) or sparitic (rare) limestone (up to 1
mm) is common. The micrite is often stained by iron minerals and contains some quartz silt. There
are a few angular bodies of yellowish clay exhibiting speckled b-fabric to merely isotropic. The
latter consist of non-laminated, regularly laminated and cross-laminated inllings of limpid, yellowbrown clay (See EA 238). Rare subrounded grains of quartz (up to 120m) appear.
Geological interpretation and conclusions: See the conclusions on the Bayadi - Baduzana correspondence.
EA 238 (VAT 1867), from Bayadi to an Egyptian ofcial
Reliability: High.
Matrix: Dense, yellowish to orange-brown in PPL, carbonatic and close to isotropism (apart from the calcite
crystals). The isotropism is apparently due to a high ring temperature that partially affected the calcite
crystals both in the matrix and the inclusions. Micritic limestone particles appear in varying sizes, from
a few tens of micrometers to millimeter sizes. Abundant (about 7%) ferrous mineral particles (identied
as magnetite and haematite) also appear in similar sizes.
Inclusions: No remains of vegetal material. Relatively coarse-grained, badly sorted wadi sand (since
most particles are rounded), f:c ratio{0.062mm}=94:6-93:7. The inclusions comprise predominantly
doleritic or coarsely crystalline alkali-olivine basalt (up to 1 mm), with phenocrysts of idiomorphic
or hypidiomorphic plagioclase prisms, olivine and augite. The basalt exhibits pellicular or complete
alteration of the olivine phenocrysts into iddingsite. To these we may add detrital basaltic minerals
(iddingsite, plagioclase, pyroxene).
240
Firing temperature: Probably around 8000C on the basis of the isotropism of the matrix but the
preservation of limestone that would have been destroyed at higher temperatures.
EA 239 (VAT 334), from Baduzana to the King of Egypt
Reliability: High.
Matrix: Dense, yellowish-brown to pale greenish-brown in PPL, carbonatic with speckled b-fabric, with
very few foraminifers. Quartz silt appears (2%) together with opaque minerals (about 2%). Also
abundant in the matrix are silt-sized laths of iddingsite, twinned plagioclase and calcite.
Inclusions: No remains of vegetal material. Badly sorted sand of rock fragments and their derived minerals,
f:c ratio{0.062mm}=94:6-93:7. Doleritic or coarsely crystalline alkali-olivine basalt (up to 2 mm), with
phenocrysts of idiomorphic or hypidiomorphic plagioclase prisms, olivine and augite is predominant.
The basalt exhibits pellicular or complete alteration of the olivine phenocrysts into iddingsite and
the pyroxene is commonly serpentinized. To these we may add detrital basaltic minerals (iddingsite,
plagioclase, pyroxene and serpentinized pyroxene), and some volcanic glass. Rounded to subrounded
(up to 300m) micritic (common) or sparitic (rare) limestone is common. The micrite is often stained
by iron minerals. A few angular bodies of yellowish clay exhibiting speckled b-fabric to merely
isotropic. The latter consist of non-laminated, regularly laminated and cross-laminated in llings of
limpid yellow-brown clay (See EA 238). Rare subrounded grains of quartz (up to 120m) appear.
Vegetal material (SLY): A few (up to 300m long) plant tissues, all non-indicative.
CONCLUSION: THE BAYADI - BADUZANA CORRESPONDENCE

Both the petrographic and elemental data indicate that EA 238 of Bayadi, EA 237 which is attributed
to the same ruler (Knudtzon 1915:1304) and EA 239 of Baduzana are very closely related. Knudtzon
apparently noticed the similarity in the clay and hence placed EA 239 after EA 237 and 238.
The three letters also differ petrographically from all other Amarna tablets. It seems, therefore, that
they were sent from one place. In other words, Bayadi and Baduzana were the rulers of a Canaanite citystate that is not represented in the other Amarna letters. Baduzana must have been Bayadis successor
since the latter operated during the last days of Lab<ayu.
The three letters originated from a site located in an area where both basalts and calcareous
sedimentary rocks are exposed. The dominance and freshness of the basaltic component in the inclusions
indicate that the nearby area is predominantly basaltic while the limestone grains and the marly nature
of the matrix indicate a calcareous province bordering the volcanic terrain. The basalt inclusions are
homogeneous in texture and therefore may reect a single source. This is contrary, for example, to the
heterogeneous nature of the basalt particles in EA 224 (above) and EA 249 (below), which apparently
derived from the Jordan River sand and reect a wider and more geologically variegated drainage zone.
The latter is also indicated by the richer variety of rock types in the inclusions of EA 224 and 249. Due
to the dominance of fresh recent (Tertiary Quaternary) basalt inclusions, the petrographic traits of EA
237-239 suit both the eastern Galilee and the southern slopes of the Golan Heights.
In terms of the textual evidence, Bayadi, the author of EA 237 and EA 238, complained that after his
enemies captured Lab<ayu, they attacked and conquered his own towns (Naaman 1997:616). Since EA
241
237 and 238 refer to the Lab<ayu affair, which took place in the north Samaria highlands and the Jezreel
Valley, the eastern Galilee is a more likely place for the Bayadi letters than the Golan Heights. Added to
this is the fact that the latter area does not have any prominent Late Bronze Age site.
The petrography of the basalt inclusions may help pin-pointing the origin of these tablets. The
lithological section of the eastern Galilee contains three basic basalt layers, labeled the Lower Basalt,
the Intermediate Basalt and the Cover Basalt (Schulman 1962; Shaliv et al. 1992). The Lower Basalt
is of the alkali-olivine type, nely crystalline, with partially altered olivine, pyroxene and plagioclase
phenocrysts. The matrix consists of plagioclase, olivine, pyroxene and ore minerals in a trachytic texture.
The Intermediate Basalt is alkali-olivine with a seriate texture, and contains phenocrysts of plagioclase,
some olivine, and some pyroxene. The matrix consists mainly of pyroclase with ne pyroxene and olivine,
occasionally with ow textures. The Cover Basalt is coarsely crystalline, spheroidally weathered with a
dark brown patina. It is olivine basalt of glomerporphyritic texture (i.e., the phenocryst aggregates oating
in a ne matrix). The phenocrysts consist of plagioclase prisms, olivine and augite, all idiomorphic or
hypidiomorphic. The matrix consists of similar minerals with characteristic ow structures. Below the
Cover Basalt two calcareous sediments occur, overlying the Lower Basalt.
The basaltic inclusions in EA 238 agree well with the characteristics of the Cover Basalt, which
exposes in the eastern Galilee on the Ramat Issachar plateau, the Poriya ridge west of the Sea of Galilee
and the Korazim plateau to the north of the lake. In this area there are only three signicant Late
Bronze sites. The rst Tel Kinrot (Khirbet el-Oreimeh) is located on the northern side of the Sea of
Galilee. It is identied with the city of Chinnereth mentioned in the annals of Thutmose III, in Papyrus
Petersburg 1116A of the time of Amenophis II (Ahituv 1984:126; Hbner 1986:255-258), and in the Bible
(Josh. 19:35). Since the former source mentions a messenger from Chinnereth among other Canaanite
envoys (maryannu) sent on an ofcial mission to Egypt in Amenophis IIs 18th year (Redford 1965:
107-110), it had apparently been a Canaanite city-state in the Late Bronze Age I. Although a scarab of
Tiye, consort of Amenophis III, was discovered there (Hbner 1986:258-259, 264), the site was probably
abandoned (or shrank dramatically in size) in the Amarna period (Fritz 1999). The second site in the
region which may have been a Canaanite city-state in the Late Bronze II is Tel Rekhesh in Naal Tavor
(Gal 1981; 1991:22-25). A third, less likely, possibility is the site of Tel Qarnei iin, which is reported
to have Late Bronze II fortications (Gal 1981). Archaeologically, Tel Rekhesh is favoured by its size
and location, being the largest and possibly the only prominent Late Bronze II city in the basaltic plateau
of the eastern Lower Galilee. The archaeological data from Tel Qarnei iin, though preliminary and
fragmentary, suggest that it was a small fortied stronghold rather than a city. In any event, we tested
the last two options on the basis of the detailed petrographic data retrieved from EA 237-239.
The detailed description of the geology (with a 1:50,000 geological map) of Naal Tavor around
Tel Rekhesh is provided by Aharon (1997). The site is located on the Gesher formation (dated at about
5MA), featuring alternations of marl, clay, chalk and limestone. On top of the Gesher formation and
at short distance from the site appears the Cover Basalt. Near the site and downstream Naal Tavor
there are exposures of the Lower Basalt. Both basalts are of alkali-olivine composition. Therefore, the
environs of Tel Rekhesh are highly suitable for the petrography of EA 237-239.
Tel Qarnei iin is a volcano or a volcanic erosive hill, featuring an assemblage of basalt, dolerite,
microgabbro and pyroclastic rocks (Shoval 1996:11-22, with 1:10,000 geological map). Downhill there
are exposures of Eocene limestone and chalk, and quaternary alluvial soils. The only marl containing
formations that appear near the site (several kilometers west of it) are of the Senonian to Paleocene
Mount Scopus group (including the Taqiye formation).
242
Since EA 237-239 contain only a few barren, badly preserved foraminifera, we were not able to
determine the age of the marl that forms the matrix. The negative evidence may indicate that this
marl is not of the Paleocene or Senonian formations that were readily identied in this study by their
palaeontological and other petrographic features. Much as we were tempted to assign it to the PliocenePleistocene Gesher or Bireh formations, which would de nitely put the verdict in favor of Tel Rekhesh,
there are not enough indications for such a conclusion. Nevertheless, on the basis of the petrographic
composition, the textual evidence and the archaeological data we suggest that EA 237-239 originated
from Tel Rekhesh, which is identied with the city of Anaharath.
Anaharath probably cooperated with Lab<ayu in the early years of the Amarna archive and was
attacked and its villages conquered after Lab <ayus downfall. Its territory must have extended over the
eastern Galilee plateau. According to Finkelstein and Goren, it bordered on the territories of Shamuna
and Megiddo in the west, Hazor in the north, Piilu in the east and Rehob (or the Egyptian centre of
Beth-shean) in the south. Naaman is more sceptical about exact demarcation of these borders and leaves
place for no mans land areas that might have separated the above-mentioned kingdoms. Naaman also
argues for the existence of a few small, additional city-states in northern Palestine (e.g. Geba-mn; see
above the discussion of Achshaphs southern border). For a possible third ruler of this city in the Amarna
period, and the possible rendering of the name of the city in the Amarna correspondence, see EA 272.
VI. MEGIDDO (MAGIDDA)

The ancient city of Megiddo (identied at Tell el-Mutesellim = Tel Megiddo), is located at the strategic
point where the international road that led from Egypt to Syria descends into the Jezreel Valley. Biridiya,
the author of EA 242-246 and EA 365, presents himself as the ruler of Magidda in one letter (EA 242:
3-4) and mentions the name of his city in three other letters (EA 243-245).

The environs of Megiddo are typied by a rich selection of clay types (Arad 1964; Sneh et al. 1998).
Paleocene marl and shales of the Taqiye formation are exposed in Wadi Ara (the Aruna Pass) to the
southwest of the mound and a small outcrop also appears on the southeastern slopes of the site. The
valley immediately to the east is covered by alluvial soils. Exposures of late Miocene basalts covered by
basalt-derived soils are found west and south of the mound. To the east one nds broad areas of Lower
to Middle Eocene chalks of the Adullam and Maresha formations, covered by Rendzina soils. With the
addition of the Maastrichtian chalky marl of the Ghareb formation (to be discussed below), ve potential
clay types (excluding temper classications) are known in the area of Megiddo. Hence in terms of CTF
(Chapter 2) Megiddo has potentially a factor of over 10.
A rich body of data on the local pottery production at Megiddo has been used in this study for
comparison with the tablets. These include numerous samples of plain ware and other ceramics from
the Early Bronze Age through the Iron Age II strata excavated during the last decade and examined in
two petrographic studies (Goren 2000b; Bozaglo 2004).
243
CATALOGUE
Reliability: Satisfactory/High.
Matrix: Highly carbonatic (ICP analysis reveals over 36% of calcium carbonate in weight percent),
very ne textured, pale yellowish-grey in PPL with infrequent badly preserved foraminifers and
their fragments. Quartz silt is very rare (0.5%). Opaques are rare (1%). Limonitic stains appear
in the matrix, usually as yellowish-red bodies that stain their surroundings in the clay matrix by
diffusion, creating clouds of yellow stain around them. Sharp-edged, elongated voids that appear
in the matrix indicate the presence of some brous organic matter that vanished (besides the
vegetal tissues).
Inclusions: Uncommon (f:c ratio{0.062mm}=98:2), thus quantities could not be estimated. These include
rounded grains of foraminiferous chalk (up to 1 mm) with common limonitic stains and subrounded
sparitic limestone (up to 500m).
Vegetal material (SLY): Up to 150m (lengthwise), either charred material or uncharred plant tissues.
Several non-indicative minute fragments in a bad state of preservation.
Palaeontology (LG): A few planctonic foraminifers in matrix, including Hedbergella (p), Heterohelix
(p). Age: Senonian.
Firing temperature: Probably unred or very slightly red since the vegetal material is uncharred.
Geological interpretation and reference: Maastrichtian chalky marl of the Ghareb formation.
EA 243 (VAT 1669), from Biridiya to the King of Egypt,
Matrix: Highly carbonatic (ICP analysis reveals over 37% of calcium carbonate in weight percent), very
ne textured, pale greenish-grey in PPL with foraminifers (about 2%) and their fragments. Devoid of
quartz silt. Opaques are extremely rare. Limonitic stains appear in the matrix (3%), usually as dark
reddish-tan bodies with dispersed edges, staining the matrix around them. Sharp-edged chamberstructured voids appear in the matrix.
Inclusions: Common (f:c ratio{0.062mm}=90:10 85:15). Frequent angular fragments (up to 1.5 mm) of
red pottery that was crushed into the matrix are present. This grog, represented in the sample by
two fragments, is of sherds made of terra rossa soil, reddish-tan in PPL, silty (5%) with abundant
opaque minerals. The inclusions include twinned plagioclase, basalt(?) fragment and subrounded
quartz. There are a few clear calcite crystals (up to 200m), a few fragments (up to 350m) of alkaliolivine basalt where the pyroxene is entirely serpentinized and the olivine is altered into iddingsite
and a few rounded glauconite bodies (up to 280m) which are green in PPL and thus unred.
Palaeontology (LG): Few foraminifers: Acarinina (p). Age: Paleogene.
Vegetal material (SLY): Common, up to 1.50 mm (long) plant tissues, some preserving calcium oxalate
crystals within their tissues. Several non-indicative tissue fragments, a fragment of bark tissue with
calcium oxalate crystals.
Firing temperature: Unred or very slightly red, judging by the preservation of uncharred vegetal
material and the green pleochroism of the glauconite.
244
EA 245 (BM 29855), from Biridiya to the King of Egypt

Reliability: High.
Matrix: Carbonatic, very homogeneous and ne textured, light yellowish-orange in PPL, brous with
strong optical orientation. The clay fraction within the matrix tends to be almost isotropic, or to
produce very low birefringence (nearly rst order grey) with the gypsum plate inserted it has strong
optical orientation. The matrix is almost entirely devoid of foraminifers and only very poor in
quartz silt (less than 1%). The silt also includes some accessory heavy minerals, namely twinned
plagioclase, zircon, olivine and hornblende. Opaque and nearly opaque iron minerals appear, the rst
as micrometer-sized angular bodies, the second as rounded bodies up to 70m in size, sometimes
staining their surrounding matrix.
Inclusions: The basaltic minerals that were observed under the stereomicroscope were not seen in the thinsection. The inclusions (f:c ratio{0.062mm}=90:10), include predominantly coarse (up to nearly 2 mm.) grains
of sparitic limestone, locally stained by iron oxides, or spherical and rounded grains of micrite. Rounded
to subrounded grains of quartz, sometimes with mineral inclusions, are frequent.
Palaeontology (LG): A few small foraminifers: Bolivina (b), Gaudryina (b), Hedbergella (p). Age:
Senonian (?).
Firing temperature: Probably unred or very slightly red. The lack of any change in the calcite
birefringence indicates that temperature of around 7500C has not been reached.
Conclusions: Knudtzon (1915) attributed this tablet to Biridiya on the basis of the fabric, the script and
the contents. The petrographic analysis conrms his observation. EA 245 is identical to the Biridiya
letters, which are unique in their composition.
Matrix: Highly carbonatic (ICP analysis reveals nearly 42% calcium carbonate in weight percent), very
ne textured, pale greenish-grey in PPL with frequent foraminifers (about 2%) and their fragments,
devoid of quartz silt. Opaques are extremely rare. Limonitic stains appear in the matrix (about
3%-5%), usually as dark reddish-tan bodies with dispersed edges, staining the matrix around them.
Sharp-edged, elongated voids that appear in the matrix indicate the presence of some brous organic
matter that vanished, apart from the vegetal tissues described below.
Inclusions: No remains of vegetal material. Uncommon (f:c ratio{0.062mm}=95:5) but including
predominantly subrounded to subangular limestone grains (up to 1.1 mm) usually sparitic and
scarcely micritic, a few rounded grains of foraminiferous chalk (up to 900m) with limonitic stains
and a very few single grains of iddingsite, pyroxene and olivine (up to 200m).
Firing temperature: Probably unred or very slightly red, though there is no clear evidence for this. The lack of
any change in the calcite birefringence indicates that temperature of around 7500C has not been reached.
EA 365 (AO 7098), from Biridiya to the King of Egypt
Reliability: Fair.
245
Matrix: Carbonatic, very ne textured, pale grey in PPL with some foraminifers, devoid of quartz silt.
Few opaques and limonitic stains appear in the matrix.
Inclusions: In the meagre sample that could be taken from this tablet subrounded sparitic and micritic
limestone and single grains of iddingsite and plagioclase were observed:
Firing temperature: Probably unred or very slightly red, though there is no clear evidence for this.
The lack of any change in the calcite birefringence indicates that temperature of around 7500C has
not been reached.
Geological interpretation and reference: Maastrichtian chalky marl of the Ghareb formation
PETROGRAPHIC INTERPRETATION OF THE MEGIDDO LETTERS

To the naked eye, the bright white colour and very ne texture easily distinguish the Megiddo letters from
all other north Canaanite tablets. The chemical composition of the three tablets analyzed by ICP (EA 242,
243 and 246) clusters separately and remotely from any other Canaanite letter. The NAA examination of
EA 248 distinguished it from any kind of pottery from Megiddo or elsewhere in Palestine (M. Artzy, pers.
comm.). Thus, the Megiddo tablets form a unique, well-distinguished group.
The combined petrographic, palaeontological and chemical analyses provide the solution to this
riddle. The high carbonate contents together with the foraminiferal assemblage (which dates from the
Senonian to the lowermost Paleocene) indicate that the Megiddo scribe(s) used the chalky marl of the
Maastrichtian Ghareb formation for their tablets. This marl is too calcareous for ceramics, hence it has
never been used for pottery production. This may explain the reason for the lack of any equivalent in
the NAA comparative database. An outcrop of the Ghareb marl is found on the southern slope of Tel
Megiddo, bordering exposures of Upper Miocene alkali-olivine basalts. This combination, which is
uncommon, is expressed in the petrography of the Megiddo tablets, where accidental basalts or their
derived minerals appear within the otherwise almost clear calcareous marl.
CONCLUSION: THE MEGIDDO CORRESPONDENCE

The materials used for the production of the letters of Biridiya are typically local to Megiddo. The
territory of Megiddo bordered on those of the city-states located at Tel Yokneam and Shamuna in the
north, Anaharath and the Pharaonic lands in the east and Ginti-kirmil in the west. In the south it bordered
on Ta>anach (according to Naaman) or Shechem and Rehob (according to Finkelstein). Megiddo must
have dominated part of Ramat Menasheh (the lower Carmel) to its west and the western sector of the
Jezreel Valley up to the Kishon River in the east. Its strategic location along the international highway
from Egypt to Syria, the ample sources of water and the fertile valley guaranteed its wealth and gave it a
leading position among the city-states of northern Canaan.
A LETTER OF YASHDATA SENT FROM MEGIDDO

EA 248 (BM 29842), from Yashdata to the King of Egypt
Matrix: Highly carbonatic, very ne textured, yellowish-grey in PPL with badly preserved foraminifers
and their fragments (about 1%). Quartz silt is extremely rare. Opaques are rare (1%). Limonitic stains
appear in the matrix, usually as dark reddish-brown bodies, occasionally inlling foraminifers.
246
Inclusions: Uncommon (f:c ratio{0.062mm}=96:4), including subrounded sparitic limestone (up to 500m).
Palaeontology (LG): A few simple foraminifers: Hedbergella (p), Heterohelix (p). Age: Senonian or
lowermost Paleocene.
Vegetal material (SLY): Up to 850m (long), either charred amorphous material, plant bres, or uncharred
plant tissues. Several non-indicative fragments in a very bad state of preservation.
Firing temperature: Unred or very slightly red since the vegetal material is uncharred.
Geological interpretation and reference: Maastrichtian chalky marl of the Ghareb formation
Discussion: Our research conrms Yashdatas words (EA 248:18-20) that he sent this letter from
Megiddo. Our results are supported by the NAA study of this tablet, conducted by M. Artzy (pers.
comm.) and her colleagues from the Laurence Berkeley laboratory.
Yashdata, the author of EA 248, was the ruler of a city-state whose identity is not explicitly stated.
In his letter he informed the king that the men of Tan[ak]a have taken away all his possessions,
slaughtered his cattle and driven him from his city, so that he is staying with Biridiya, the ruler of
Megiddo. As recognized by all scholars, Tan[ak]a is an exact rendering of biblical Ta>anach, located
at Tell Ta>annek, a relatively large mound (about 4.5 ha) located 8 km southeast of Megiddo.
Naaman (1997:616-617) argues that Tan[ak]a = Ta>anach was Yashdatas capital. His
identication of Ta>anach as the centre of a city-state is based on the following arguments:
1. A late 15th century BCE archive of 13 cuneiform tablets was discovered at the site. Ta>anachs
ruler (Rewashur) received letters from neighbouring kings and possibly also from an Egyptian
ofcer who was stationed at Gaza (see Rainey 1999 for a recent discussion).
2. A messenger from Ta>anach is mentioned in Papyrus Petersburg 1116A among other Canaanite
envoys (maryannu) sent on ofcial mission to Egypt in Amenophis IIs 18th year (Redford 1965:
107-110). Hence there is no doubt that Taanach was a city-state until the late 15th century, about
half a century before the Amarna period.
3. Yashdata reports to the Pharaoh that the men of Taanach took all his possessions and drove
him away. Hence his capital must have been located at or near Ta >anach. Yashdatas city was
located in the area of Megiddo and Tan[ak]a = Ta >anach, and except for Tell Ta >annek, there
is no other site in this area which befits the status of a city-state.
4. Mycenaean IIIA1, MycIIIA2 and early MycIII B pottery found at the site (Glock 1992:288)
indicates that it was inhabited in the 14th century BCE.
Naaman assumes that Ta>anachs status as an independent city-state was maintained, in spite of its
decline, between the late 15th century and the Amarna period (for the identication of Yashdata as the
ruler of Ta>anach see Campbell 1965:109; Helck 1971:185; Liverani 1999:463).
Finkelstein doubts the identication of Ta>anach as Yashdatas capital (already Lapp 1967:8).
Intensive excavations at Tell Ta>annek by two different expeditions revealed that the site was signicantly
inhabited in the Late Bronze I and in the Iron I; yet they failed to unearth meaningful Late Bronze II (14th
and 13th centuries BCE) nds (Glock 1993). Hence, it is doubtful if Ta>anach could have served as a
centre of a city-state in the Amarna period. In Finkelsteins view, EA 248 and 245 do not provide enough
evidence for the hometown of Yashdata, except for him being a ruler in the north and an ally of Biridiya.
There are many possible scenarios in this case, and Tan[ak]a (be it Ta>anach or another place) could
have been a village in his territory or beyond.
247
VII. REHOB
The hometowns of several important gures in the Amarna archive are not mentioned in their letters.
In some instances cities under their command are mentioned in the correspondence, but it is not clear
whether they were the capitals or secondary cities in their territories. In northern Canaan, the most debated
cases involve the location of the capital cities of three rulers: Ba>lu-UR.SAG, Ba>lu-meir and Tagi. The
solution for the location of each of the three has consequences for the other two. A comprehensive picture
of the system of city-states in northern Canaan cannot be drawn without resolving their cases.
Ba>lu-UR.SAG is the author of two letters (EA 249-250) which do not disclose the name of his
city. The sons of Lab<ayu accused Ba>lu-UR.SAG in the following words: Why have you given into the
hand of the king, my lord, Gitti-padalla, the city which Lab<ayu has taken? (EA 250:12-14). The town
of Gitti-padalla (Gath-padalla) has accordingly been suggested as Ba>lu-UR.SAGs capital (Albright
1946; Rainey 1968) though this is not explicitly stated in the text. It could, in fact, have been one of his
secondary towns, or even a town in the territory of another ruler.
Rainey (1968:4) suggested rendering the name Ba>lu-UR.SAG, whose second part is written with
a Sumerian logogram meaning hero, warrior, as Ba>lu-meir, and identied him with the writer of EA
257-259 (west-Semitic mehir means warrior). This proposal was dismissed by other scholars because
letters EA 249-250 and EA 257-259 differ in all their elements (i.e., clay, script, epistolary style, topics
debated), and as the capital of Ba>lu-meir is called [x-(x)-I]G-ma-te (EA 257:12; see Naaman 1975:
33-34, 14* n. 25; Moran 1992:310; Liverani 1998:121-122). Rainey (1989:571) proposed that this placename should be read [URUkin-t]i ma-gal (i.e., Gitti=Gath). However, collation of this tablet (now in the
Berlin museum) indicated that this rendering is erroneous (Naaman 1998a) and that the text should be
rendered as originally suggested by Knudtzon (that is, [URUx-(x)-I]G-ma-te).
Considering Gitti-padalla as Ba>lu-UR.SAGs capital, Rainey (1968) identied it with the village
of Jatt north of Tulkarem, since in the topographical list of Shoshenq I dpr, which he rendered Gittipadalla, is mentioned before Borim (today Khirbet Burin) and >Aruna (the mound in the village >Ara).
However, rendering dpr as Gitti-padalla is unlikely since the toponym Ginti/Gitti is rendered knt (and
never d) in Egyptian transcriptions (cf. Ahituv 1984:95-97). The rendering and identication of this
toponym remain unknown.
The village of Jatt is built on a large prominent mound and is the most important Late Bronze tell along
the international highway between Aphek and Megiddo. Identifying it with Gitti-padalla and assuming that
it was Ba>lu-UR.SAGs capital would mean that this ruler controlled the Sharon plain and the coastal plain
of the Carmel ridge. Yet in EA 250 Ba>lu-UR.SAG complains against the attempt of Milkilu (the ruler of
Gezer) to allow the sons of Lab<ayu to expand to the Jezreel and Beth-shean Valleys and govern the city of
Piilu (EA 250:32-3953-55; see Naaman 1999). He also complains that they urge him to wage war against
the people of the land of Gina probably to be identied in the Jezreel Valley west of the Kishon river
(Naaman 1988c:184-185). Finally, Ba>lu-UR.SAG suggests that the king would command Biryawaza, the
ruler of Damascus, to operate against the Gezer-Shechem coalition. If Ba>lu-UR.SAGs city were located in
the Sharon plain, these events would have had little inuence on his own territory.
Indeed, our petrographic data indicate that EA 249, and perhaps also EA 250 reported by Knudtzon
(1915:1310 n. 1) to look similar,2 were sent from the central Jordan Valley. In such a case, Ba>luUR.SAGs concern about the events in the Beth-shean - Piilu area is self evident.
2. In his words, tablet EA 250 scheint ein und derselbe zu sein (though the script is not the same). EA 250 is kept in the Cairo
Museum and therefore we have not been able to examine it.
248
CATALOGUE
EA 249 (VAT 1603), from Ba>lu-UR.SAG to the King of Egypt
Reliability: High.
Matrix: Carbonatic, light yellowish-tan in PPL, containing infrequent foraminifers. It is rather silty
(about 4%-5%) and containing abundant opaque minerals that appear in a range of sizes from a few
micrometers to about 50m. Under higher magnications (>X200) the matrix is optically active with
speckled b-fabric and a few foraminifers.
Inclusions: The inclusions contain sand consisting of frequent rounded grains of sparitic, micritic
and biogenetic limestone (up to 800m) and subangular to well-rounded fragments (up to 750m)
of alkali-olivine basalt of several types, ranging between nely crystalline to nearly doleritic.
The basalt is usually rather weathered and the olivine is partly or entirely altered into iddingsite.
Single crystals of basalt-derived iddingsite, augite, and twinned plagioclase appear in grain-sizes
reaching 250m. Angular to subrounded grains (up to 500m) replacement chert are common.
There are a few fragments (up to 500m) of tan clay-stained travertine, a very few calcareous
pisoliths and rounded foraminiferous chalk grains (up to 350m). Also present are a few spherical
sand-sized quartz grains (up to 350m) as well as a very few fresh (unfossilized) aquatic mollusc
shell fragments (up to 500m).
Vegetal material (SLY): Few (up to 600m long). Non-indicative tissue fragments.
Conclusions: The petrographic data indicates that EA 249 was made of sediments from the central Jordan
valley, between Beth-shean and Wadi Zarqa (see EA 224). Two interpretations are possible. Either
Ba>lu-UR.SAG ruled a city in this area, or his capital was located somewhere else although the letter
was dispatched from the Egyptian centre of Beth-shean (like EA 224, 232, 234-235 and 285). In
the light of our dismissal of the equation Ba>lu-UR.SAG=Ba>lu-meir, the geographic background
of EA 250 and the archaeological data, we opt for the rst interpretation. Excluding Beth-shean,
which served as an Egyptian government centre and thus could not have been ruled by a Canaanite
mayor, the only city which would t this solution is Rehob, a city identied with the large mound
of Tel Reov (Tell e-arem) located 5 kms south of Beth-shean (Ahituv 1984:164-165). Rehob
does not appear in the Amarna correspondence, but is mentioned in a letter from Taanach (TT 2:22)
and in Seti Is stele discovered at Beth-shean (for references see Ahituv 1984:164-165). When Seti
describes the rebellion initiated by the leader of Hamath, he notes that he does not permit the ruler
(wr) of Rehob to go outside (Wilson 1969:253). It would seem, therefore, that Rehob was a capital of
a city-state. Recent excavations at Tel Reov have indicated that the Late Bronze Age city covered the
entire site, including the lower mound (Mazar, pers. comm.). Late Bronze Rehob extended therefore
over an area of ca. 10 hectares one of the largest settlements in Canaan at that time.
There are good reasons for placing Ba>lu-UR.SAG at Rehob. The site is located at the centre of
the theatre of operations of the events related in EA 250. It bordered on both the Egyptian centre of
Beth-shean and the kingdom of Piilu, and it was close to the towns attacked by Lab<ayu. No wonder
that its policy was of great concern for the sons of Lab<ayu.
In conclusion, we suggest that Rehob was Ba>lu-UR.SAGs capital city and that Ba>lu-UR.SAG
and Ba>lu-meir were two different rulers who were active at approximately the same time. Locating
Ba>lu-UR.SAG at Rehob adds another unmentioned polity to the list of Canaanite city-states of
249
the Amarna period. The territory of Rehob probably encompassed the entire Beth-shean Valley. If
this was the case, it bordered on the territories of Shechem and Megiddo (Shechem and Taanach
according to Naaman) in the west, Anaharath (or the Egyptian centre of Beth-shean) in the north
and Piilu in the east.3
Matrix: The clay is light tan in PPL. It is carbonatic, foraminiferous, silty (about 7%) and rich in silt to
ne sand sized opaque iron minerals.
Inclusions: The inclusions contain sand predominantly grains of both micritic (common) and sparitic (rare)
limestone (up to 650m) containing calcite cleavage crystals, frequent rounded grains of quartz (up to
300m) and alkali-olivine basalt (up to 300m) or its derived minerals. The olivine is partly or fully
altered into iddingsite. There are a few subangular grains of replacement chert (up to 400m) and a very
few aquatic mollusc shell-fragments (up to 250m).
Vegetal material (SLY): Up to 900m. A fragment of bark, probably from a tree, rich with rhombus of
calcium oxalate. The genus is not identiable.
Conclusions: EA 263 was sent from the Beth-shean area. Thus the writer was either the ruler of Rehob,
or the ruler of an unnamed city who dispatched his letter from the Egyptian centre of Beth-shean.
The writer complains that when he appeared before the Pharaoh (when I visited the house of my
lord) his city was plundered (lines 6-17). The letter ends with the words [because of] Tagi and
because of Lab<ayu (lines 33-34), who are possibly responsible for this act. The rulers of Shechem
and Ginti-kirmil were enemies of Ba>lu-UR.SAG, the ruler of Rehob. In view of the similarity of
the petrography of EA 249 and 263 and the historical background of Ba>lu-UR.SAG and the writer
of EA 263, we tend to identify EA 263 as a letter of Ba>lu-UR.SAG and to assume that it was
dispatched from Rehob.
VIII. [URUX-(X)-I]G-MA-TE
The identication of the city of Ba>lu-meir, the author of EA 257-259, is the subject of a long debate.
As noted above, Rainey (1968:4; 1989:571) suggested identifying Ba>lu-UR.SAG with the west-Semitic
name Ba>lu-meir and locating his capital at Gitti-padalla. Naaman (1975:34) suggested restoring lines
17-22 of EA 257 as follows: May the king, my lord, know that I serve him [with com]plete devotion, and
[the city of x-(x)-I]G-ma-te serves him [with complete devotio]n. The restoration of the name of Ba>lumeirs capital was accepted by Moran (1992:310) and Liverani (1998:122) who also disassociated
Ba>lu-UR.SAG from Ba>lu-meir and treated them as two separate rulers. Liverani (1998) proposed
completing the name of the city [URUMi-i]k-ma-te, and identied it with biblical Michmethath, located in
3.
According to our interpretation, the town of Gitti-padalla, which was conquered by Lab<ayu and later transferred by Ba>luUR.SAG to the Egyptian ofcials in Canaan (EA 250:12-14), must be sought in the Beth-shean Valley, near the western or
southwestern boundary of Rehob and next to the Pharaonic lands in the Jezreel and Beth-shean Valleys. According to EA
250, Lab<ayu conquered the cities of Shunama (Shunem), Burquna, arabu and Gitti-rimmunima, all located in or next to
the southern Jezreel Valley (Rainey 1968:7-8); Gitti-padalla should probably be sought close to these places.
250
the hill country south of Shechem. However, thorough archaeological investigations of this area have not
revealed any signicant Late Bronze Age site.
Our analysis of the three tablets sent by Ba>lu-meir (EA 257-259) seems to resolve the problem.
We shall argue that Ba>lu-meir ruled a city in the northwest Jezreel Valley. EA 221-222, sent by a ruler
named Wiktasu, are petrographically, palaeontologically and chemically identical to the letters of Ba>lumeir and therefore will be discussed together with them.
CATALOGUE
Reliability: High.
Matrix: Carbonatic, silty (about 10%), foraminiferous. The silt is essentially quartzitic and calcitic but with
abundant minerals that are derived from the volcanic component of the inclusions.
Inclusions: These include a mixture of sedimentary and igneous rocks (f:c ratio{0.062mm}=90:10), consisting
predominantly of badly sorted grains of foraminiferous chalk (up to 700m) and olivine basalt
and/or andesine olivine basalt. The latter is unusual rst and foremost in the sodic composition
of the plagioclase microlites, which consist of albite in contrast with the more calcic nature of the
plagioclase (usually labradorite) of the common basalts in the southern Levant. The overall texture
of the matrix is glassy and the olivine phenocrysts are usually altered into iddingsite or chlorite.
There are a few fragments of travertine (up to 500m).
Vegetal material (SLY): Several non-indicative, badly preserved tissue fragments.
Firing temperature: Probably unred or very lightly red since the vegetal material remains uncharred.
Geological interpretation: See discussion of EA 259.
Reliability: High.
Matrix: Carbonatic, silty (about 10%), foraminiferous. The silt is essentially quartzitic and calcitic but
with abundant minerals that are derived from the volcanic component of the inclusions.
Inclusions: These include a mixture of sedimentary and igneous rocks (f:c ratio{0.062mm}=): Badly sorted
grains of foraminiferous chalk (up to 700m) are dominant together with olivine basalt and/or
andesine olivine basalt (up to 1 mm). It is unusual rst and foremost in the sodic composition of the
plagioclase microlites, which consist of albite as against the more calcic nature of the plagioclase
(usually labradorite) of the common basalts in the southern Levant. The overall texture of the matrix
is glassy and the olivine phenocrysts are usually altered into iddingsite or chlorite. There are a few
fragments of travertine up to 500m.
Vegetal material (SLY): A few small, badly preserved tissue fragments.
Firing temperature: Probably unred or very lightly red since the vegetal material is uncharred.
Geological interpretation: See discussion of EA 259.
251

Reliability: High.
Inclusions: These include a mixture of sedimentary and igneous rocks (f:c ratio{0.062mm}=93:7). Badly
sorted grains of foraminiferous chalk (up to 700m) are dominant together with olivine basalt and/or
andesine olivine basalt (up to 1 mm). It is unusual rst and foremost in the sodic composition of the
plagioclase microlites, which consist of albite as against the more calcic nature of the plagioclase
(usually labradorite) of the common basalts in the southern Levant. The overall texture of the matrix
is glassy and the olivine phenocrysts are usually altered into iddingsite or chlorite. There are a few
fragments of travertine up to 500m.
Vegetal material (SLY): A large fragment with many vessel members with their indicative spiral
secondary cell walls and a few non-indicative tissue fragments.
Palaeontology (LG): Mixture of Paleogene fauna (in matrix) and Senonian fauna in the inclusions:
Acarinina (p), Cibicides (b), Hedbergella (p), Ostracoda, Rotalia (b), Subbotina (p).
Geological interpretation for EA 257-259: The matrix of the three tablets is composed of Rendzina
soil mixed with badly sorted chalk ranging between the ne and the coarse fraction. The matrix
contains high proportions of silt. All the soil materials in Israel include, to varying extents, aeolian
dust of desert origin (Ya>alon and Ganor 1973). Carbonate and basalt rocks do not contain silt-size
quartz grains, but large amounts of such grains occur in the soils that developed on these rocks. The
external source of the silt-size quartz grains is considered to be an aeolian contribution to the soil.
The largest amount of aeolian dust occurs in soils that developed on hard limestone and dolomitic
limestone, in which the residual material released from the dissolution of the rocks is only about 2%
(Adan-Bayewitz and Wieder 1992).
Two of the three tablets (EA 258 and 259) contain an assemblage of unique volcanic tuffs and rocks
and their derived minerals. These are lacking from the sample taken from EA 257, perhaps due to
its small size. The rocks are of basaltic composition, characterized by an ophitic texture and mac
minerals that are always altered into chlorite or serpentine. The large size of the phenocrysts and the
feldspar (when present is identied by its extinction angle as having calcic composition) indicate an
intrusive rock. Fragments of holocrystalline basalt with fresher appearance occur too. Independent
xenomorphic crystals of olivine, almost entirely altered into bowlingite or serpentine are common
in EA 259 and in smaller examples in EA 257. In conclusion, the inclusion assemblage of EA 257
and EA 259 reects a particular environment, where calcareous sedimentary rocks appear together
with unique volcanoclastic rocks.
Only a few areas in the southern Levant may t this description. Tuff, tholeiitic dykes,
holocrystalline basalt and calcareous sedimentary rocks occur in the slopes of Mount Hermon,
where Lower Cretaceous volcanics expose near Jurassic limestone formations containing fossil
reefs. However, the loam that forms the matrix of the two tablets, that is probably related to the
tuff and limestone inclusions or was collected nearby, hints at a much younger environment. In this
case the only possible location where these three components may possibly meet is Mount Carmel
(Sass 1957; 1980), the adjacent area of the Umm el-Fam hills (Sass 1968) and minor points in the
western Galilee (Kafri 1972) where Late Cretaceous volcanic activity took place. The dominant
252
volcanic rocks in these locations are pyroclastics of basic composition including primarily tuff.
As opposed to other volcanic occurrences in the southern Levant, basaltic ows play a secondary
role here (Sass 1980). However, some of these occurrences also contain some exposures of massive
basalts, sometimes xenomorphic, amigduloidal basalts and volcanic bombs (Sass 1957). If the
extremely meagre western Galilee exposures are excluded, only the Carmel ridge and the Umm
el-Fam area remain as possible candidates for the origin of the Ba>lu-meir tablets.
In the Carmel area, 12 pyroclastic volcanoes were identied. Their various parts are represented
by a series of vent deposits including black and massive pyroclastics, proximal anks with
variegated pyroclastics and distal anks with yellow pyroclastics. The characteristic colours of
these pyroclastics reect the different degree of their deposition and argillization, affected by the
increasing distance from the vent. Since the eruptions occurred in seawater, the pyroclastics were
deposited in a marine environment that effected their typical argillization processes. After the
volcanic activity had ceased, the pyroclastic cones were truncated to sea level and reef-building
fauna used their relatively shallow elevation below sea level to erect reefs over them.
The basalts appear with typical partial crystallization and an exceptional sodic composition of
the plagioclase microlites. The plagioclase may change from albite to labradorite in the context
of the rock type, but in the massive lavas the composition changs to range between labradorite
to andesine. These basalts can thus be de ned as spilites. The process of spilitization is related
to the eruption of these magmas in oceanic conditions, where the sodium-rich water reacts with
the plagioclase and replaces the calcium with sodium. The calcium typically re-crystallizes in the
form of amygdaloidal lling of voids and veins. As in the inclusions of EA 259, these are typied
by hypidiomorphic to xenomorphic plagioclase, xenomorphic olivine that is commonly altered into
bolingite or chlorite (and more rarely iddingsite, as opposed to the Pliocene Pleistocene basalts of
the Galilee and the Golan).
This occurrence cannot be confused with any other volcanic rock in the southern Levant. It is
fundamentally different from the Neogene - Pleistocene basalts of the Galilee which were all created
in continental environments. The only instances where undersea volcanic eruptions are recorded in the
geology of the southern Levant are the Late Cretaceous volcanic occurrences of the Carmel area (Sass
1980). All this strengthens the view that the city of Ba>lu-meir should be sought in this area.
Further investigation of the geological data for comparison with the petrography of EA 257-259
may limit the search area even more. Since the inclusions contain mostly spilitic basalts and not
only tuffs, an area with Late Cretaceous basaltic ows should exist near the site. Such outcrops
are extremely rare and appear only in a few localities, all scattered on the southern ank of Mount
Carmel. The most notable ones are found near Shefeya (north of Zikhron Ya>acov) and on the
western bank of Naal Rakefet (west of Yokneam). The last one is the most signicant, being the
only case in which three volcanic stages occur (Sass 1957; 1980). Therefore, this rather limited area
should be considered the only possibility for the provenance of Ba>lu-meirs tablets.
Within a range of 10 km from these exposures, only two sites can be considered as centres of a
Late Bronze Age city-state:
Dor: Though excavations have not disclosed yet the extent of Late Bronze Dor, it must have served as
the main harbor for the entire coastal plain of the Carmel. Dor is not located immediately adjacent to the
Mount Carmel volcanics, but the Shefeya exposure is only about 10 km away. Yet, the Kebara swamps
could have interrupted the access from Dor to the Carmel slopes. Moreover, the local soils at Dor are
coastal sands and hamra soils both absent from the tablets under discussion. Our petrographic study
253
of vast samples of pottery from Dor, examined by Cohen-Weinberger, Goren and Gilboa, reveal that
indeed, this coastal character is well expressed in the locally produced wares. Dor, then, can hardly be
considered as the seat of Ba>lu-meir.
Tel Yokneam: The prominent mound, located in the opening of Naal Yokneam into the Jezreel
Valley, is located only 2 km away from the Naal Rakefet basalt ow. Naal Rakefet drains directly
into Naal Yokneam and its basalts are supplied immediately to the foot of the mound where they
are mixed with the local alluvial soil. This was also observed in sand samples that we collected from
the channel of Naal Yokneam immediately below the site. The excavations at Tel Yokneam (Tell
Qeimun) have indicated that the site was inhabited in the Late Bronze II (Ben-Tor 1993) but came
short of shedding light on its size and nature at that time.
Reference: Comparative ceramic materials from the Carmel area came from Tel Nami north of Dor and
from Tel Yokneam (Bozaglo 2004). Further away, materials from the Jezreel Valley sites surveyed
by Portugali (1982) were examined by Goren. Although no workshop wasters were included in
these samples, they seem to be sufcient for the denition of locally produced fabrics, since they
contained relatively large amounts of plain ware (e.g. plain bowls, kraters, etc). Of all these, only
the Tel Yokneam assemblage corresponds to the tablets of Ba>lu-meir. The common petrographic
group of the Iron Age I-II assemblages at the site was identical in all respects to that of the tablets, in
both matrix and inclusions. Since the unique composition of the inclusions was not encountered in
any other ceramic assemblage within this area, it is obvious that Tel Yokneam should be considered
the best candidate for the city of Ba>lu-meir.
Conclusions: Our petrographic results seem to point to Tel Yokneam (Tell Qeimun) as the only possible
location for Ba>lu-meirs city. This seems to t the information in EA 245: Surata took Lab<ayu
from Megiddo in order to ship him to Egypt from his home town Acco, but released him and Ba>lumeir near the town of innatuna (Hannathon) to the northeast of Tel Yokneam. Placing Ba>lumeir at Tel Yokneam reopens the question of the interpretation and identication of the toponym
[ URUx-(x)-i]G-ma-te.
A place named >en qn >mu is mentioned in the topographical list of Thutmose III (No. 113) among
a group of toponyms that includes Beth-shean (110), Helkath (112), Geba> (114), Zephath (116)
and Burquna (117). These toponyms cover the Jezreel Valley and the margins of the Beth-shean
Valley and the Acco plain. The city of Geba> (114), which is mentioned in Thutmoses list next to
>en qn >mu, has been identified at Tell Abu Shusheh, half way between Megiddo and Tel Yokneam
(Giveon 1981; Siegelmann 1984). It is thus evident that >En Qan >amu (>en qn >mu) was the ancient
name of the city that is called by a slightly later form Jokneam (yqn >m) in the biblical text.
We cannot settle the difference of names between the city of [x-(x)-i]G-ma-te, mentioned in EA 257
as Ba>lu-meirs city, and >En Qan>amu, the citys name in Thutmose IIIs topographical list. Even
if one tries to restore it [ URUE-ni-i]q-ma-te, the two names are too remote for equation. Since the
textual and petrographic evidence do not go hand to hand in this case, Naaman would leave this
problem unresolved.
According to Finkelstein, the identification of [ URUx-(x)-i]G-ma-te at Tel Yokneam renders the
identification of Late Bronze Age >En Qan >amu and biblical Jokneam at that site unlikely. Biblical
Jokneam is mentioned in the list of conquered Canaanite kings (Josh 12:22), in the description
of the tribal boundary of Zebulun (Josh 19:11) and in the list of Levitic cities (Josh 21:34). >En
Qan >amu and Jokneam have both been identified with Tell Qeimun because of the geographic
logic in Josh 19:11 and a certain resemblance in the name. Yet, >En Qan >amu=Jokneam may be
254
sought in another mound in the region, such as Tell Qasis. The only problem with this proposal
is that it leaves Tel Yokneam without an Iron Age name. A more remote possibility is that En
Qanamu and biblical Jokneam were two different places. Late Bronze [ URUx-(x)-i]G-ma-te and
Iron Age Jokneam should then be placed at Tell Qeimun (Tel Yokneam), while En Qanamu
should be sought somewhere in the vicinity.
The city-state located in Tel Yokneam must have covered the northwestern Jezreel Valley and
probably the southern Mount Carmel. It bordered on Ginti-kirmil in the west (see below), Achshaph
(or Mishal and possibly Geba-mn according to Naaman) in the north, Samhuna in the east and
Megiddo in the south.
Reliability: Fair. Sampled again by peeling (from a small break on the reverse corner) to supply a
sufciently reliable sample.
Matrix: Carbonatic, silty (about 10%), foraminiferous. The silt is essentially quartzitic and calcitic but
with abundant minerals that are derived from the volcanic component of the inclusions (below).
Inclusions: These include a mixture of sedimentary and igneous rocks (f:c ratio{0.062mm}=) of which badly
sorted grains of foraminiferous chalk (up to 500m) are dominant, olivine basalt (up to 500m) is
common and there are a few fragments of travertine up to 500m.
Vegetal material (SLY): A non-indicative tissue fragment.
Reliability: High. In the light of the analytical results and the fragmentary state of the tablet, it was
sampled again to supply additional data.
Inclusions: These include a mixture of sedimentary and igneous rocks (f:c ratio{0.062mm}=): if which badly
sorted grains of foraminiferous chalk are dominant (up to 700m) with a few fragments of travertine
up to 500m.
Vegetal material (SLY): Several non-indicative fragments in a bad state of preservation.
Palaeontology (LG): Mixture of Paleogene fauna (in matrix) and Senonian fauna in the matrix and the
inclusions: Acarinina (p), Bolivina (b), Catapsydrax (p), Hedbergella (p), Heterohelix (p).
Discussion: Wiktasu was probably the successor of Ba >lu-meir at [ URUx-(x)-i]G-ma-te (Tel Yokneam).
His short letters do not supply enough data for historical analysis. Knudtzon (1915:1299 n. 1)
compared Wiktasus letters with the letter of Achshaph (EA 223). We found no justification for
this equation.
255
IX. GINTI-KIRMIL
Tagi, the author of EA 264-266, does not mention his city in his correspondence. Abdi-eba, the ruler
of Jerusalem, complains in one of his letters (EA 289:18-20) that Ginti-kirmil belongs to Tagi, and the
men of Gintu are the garrison in Beth-shean. It has generally been suggested that Ginti-kirmil was
Tagis capital, though theoretically it could have been one of his peripheral towns. In another letter (EA
288:24-27), Abdi-eba describes how desperate his situation is, by the words: I am at war as far as the
land of Sheru and as far as Ginti-kirmil. All the mayors are at peace, but I am at war. Tagi, the ruler
of Ginti-kirmil, was an ally of Milkilu of Gezer, Abdi-ebas main enemy, and Sheru and Ginti-kirmil
apparently mark the limits of the territories ruled by the enemies of Jerusalem. Almost all scholars
sought Ginti-kirmil near the Carmel ridge. Two Phoenician jar inscriptions found at Tel Shiqmona on
the southern outskirts of Haifa mention a place named Gath Carmel (Cross 1968:226-233). Pliny refers
to a place named Getta as being located north of the Carmel headland (Nat. Hist. V, 17, 75). The abovecited passage from EA 289 associates Tagi with the Egyptian centre of Beth-shean. Finally, a cylinder
sent by Tagi to Lab<ayu has recently been found at Beth-shean. These arguments are enough to reject
Aharonis proposal (1969) to equate Ginti-kirmil with Gath of the Shephelah. In addition, we have now
confirmed the identification of Gath (Tell e -afi) as the capital of Shuwardatu (below). The tablets of
Shuwardatu and Tagi are completely different petrographically, and considering the distribution of the
Taqiye formation (which appears in the Tagi tablets), there is simply no other option in the Shephelah
that can be seriously considered as a candidate for the capital of Tagi. However, the exact location of
Ginti-kirmil has been debated.
Alt (1925:48 n. 3), Jirku (1930:143) and Helck (1971:185 and n. 115) placed Ginti-kirmil at the
large mound of the village of Jatt in the Sharon plain. This proposal was weakened when Albright
(1946; followed by Rainey 1968) suggested identifying Jatt with Gitti-padalla. Naaman (1975:63-65)
sought Ginti-kirmil in one of the mounds of Ramat Menasheh (the lower Carmel). However, there
seems to be no prominent Late Bronze Age site with a large enough rural hinterland in this region.
Schmitt (1987:43-48) proposed placing Ginti-kirmil at Tell el->Amr, a site located at the narrow pass
by which the Kishon river forces its way from the Jezreel Valley to the coastal plain. Finally, Zwickel
(1994) proposed identifying it with Tell Abu Hawam in the bay of Haifa.
CATALOGUE
EA 264 (BM 29853), from Tagi to the King of Egypt
Reliability: Moderate. Due to the complete state of preservation of this tablet it was sampled by SPA,
including several inclusion fragments. In the hand specimen this tablet is identied as being made of
marly white clay (Munsell colour 10YR7/1) containing tiny dark grits and charred vegetal material
that was sparsely added to it.
Matrix: Carbonatic, light yellowish-grey in PPL with speckled b-fabric, silty (about 5%) with ne opaque
particles (2%) ranging between a few micrometers to 50-60m, part of which seem to be of the
charred organic material that appears in the inclusions. Foraminifers are few. The silt is mainly of
quartz but with accessory twinned plagioclase, microcline, hornblende, zircon and perhaps rutile.
Inclusions: These are sparsely spread and include frequent to dominant large fragments (reaching 2 mm
in size) bioclastic limestone of coralline algae fossils, exhibiting a network of carbonate veins in the
256
form of bres within a brown cryptocrystalline calcitic groundmass. The algae fossils are cut in the
thin section in various orientations, revealing their typical microstructure (Carozzi 1960:198-200).
Vegetal material (SLY): Dominant. Two categories appear: A) Entirely charred, compact (with no internal
cell structure) organic matter, probably vegetal in origin (up to 500m). B) A large mass of semidigested group of sclereides (stone cells), sizing 1.5 mm. Those sclereides are probably from a seed coat.
They most likely represent a coprolite of a small vegetarian (rodent?) or a larger plant fragment.
Geological interpretation: According to its petrofabric this tablet is obviously made of marl of the
Taqiye formation of the Paleocene to the basal Eocene Age (Bentor 1966:72-73, see EA 51 above
for a broader discussion on the context of this formation within the Paleocene of the Levant). This
formation, named after Jebel Abu-Taqiye (Hebrew: Har Kippa) in the Negev (Shaw 1947) consists
of chalky shales, green to grey in colour and locally gypsiferous, and a hard bank of silicied
chalk (the Har Member) capped by chalky shales. Limonitic concentrations, pseudomorphic after
marcasite, are abundant in its shaly parts. As stressed before (EA 51), Paleocene shales are almost
constant in their stratigraphic position and even in details of their composition throughout the
Levant. Equivalent beds appear in Egypt (the above-mentioned Esna shales, see EA 1) and even in
Morocco and Turkey (Bentor 1966:73).
In the Levant, the Taqiye formation outcrops over many separated areas including the Negev, the
Shephelah, the Judean Desert, the central Jordan Valley, the western Galilee, the Lebanese Beqa>
and some areas along the Mediterranean coast of Lebanon. However, if we regard the locations that
were suggested for the capital city of Tagi, the options can be reduced considerably. Tagis territory
must have lain west or northwest of the central hill country. In that part of the country, the Taqiye
formation appears from the area of TulkaremMetzer, along Wadi >Ara and in the contact zone
between the southern Mount Carmel and the lower Carmel (Ramat Menasheh). Several outcrops
also occur along the western Galilee hills.
Archaeologically, there are no signicant Late Bronze Age sites on the Carmel ridge, and only a few
major Late Bronze sites around it (other than those already identied as centres of Canaanite citystates, such as Megiddo and Tel Yokneam) which can be suggested as centres of a Canaanite city-state.
Khirbet Shallaleh in the gorge of Naal Oren (surveyed, but not published) is a relatively small site,
located far from the main routes and from fertile lands. It is also far from any exposure of the Taqiye
formation. Dor is too close to the coast and the data concerning its Late Bronze Age occupation are
fragmentary. Moreover, there are no exposures of Taqiye marl even in its broader vicinity. Along the
southern anks of Mount Carmel, Tell >Ara in Wadi >Ara (identied with >Aruna of the Thutmose III
list) is too small to be considered for the capital of a city-state (Zertal and Mirkam 2000:104-108).
Neither Tell el-Asawir at the western gateway to Wadi >Ara, nor Khirbet Sitt Leila a few kilometers
further north seem to be adequate by their size and archaeological record (see Supplement).
This leaves the large and prominent mound of Jatt (Porat et al. 1999), located on the international
road to the north. Placing Ba>lu-UR.SAG at Rehob and Ba>lu-meir at Tel Yokneam leaves Jatt open
for Tagi. Indeed, the mound is situated on a hill that is made of Middle Eocene chalks of the Adullam
formation and marls of the Taqiye formation (Ilani 1985; Sneh et al. 1996). Thus the materials of the
tablet are found immediately in and around it. Therefore, archaeological, historical and geological
considerations, as well as the similarity of the names Ginti-kirmil=Jatt, lead us to accept Alts (1925:
48 n. 3) identication of Jatt as the capital of Tagi. It seems that the lower Carmel (Ramat Menasheh)
was considered in ancient times to be part of the Carmel ridge hence the name Ginti-kirmil for the
city located slightly to the south of it, in the eastern Sharon plain.
257
Reference: Late Bronze Age pottery from tombs excavated around Jatt was examined petrographically
(Porat et al. 1999) as well as some raw Taqiye marl samples that were collected around the mound.
The latter were found to be petrographically similar to the matrix of the tablets. The positive denition
of the clay as Taqiye marl is enough to establish our conclusion.
Conclusions: The identification of Jatt (=Ginti-kirmil) as the capital of Tagi fits what we know about
his policies. Tagi cooperated with Gezer (to the south) and Shechem (to his east; see EA 263:
33-34). It seems that his territory covered the entire Sharon plain and large parts if not the entire
coastal plain of the Carmel ridge, with Dor serving as his main port. Ginti-kirmil bordered on the
territory of the Egyptian centre of Jaffa in the south (see below), Shechem in the east, Megiddo
and Tel Yokneam in the north (and on Ta>anach in the northeast according to Naaman).
Reliability: Moderate/Fair. Only a small sample was taken because of the complete state of preservation
of this tablet.
Matrix: Carbonatic, light yellowish-tan in PPL with speckled b-fabric, silty (about 4%), devoid of
foraminifers, with ne opaque particles (2%) ranging between a few micrometers to 50-60m, part
of which seem to be of the charred organic material that appears in the inclusions. The silt is mainly
of quartz but with accessory twinned plagioclase, microcline, hornblende, zircon and rutile.
Inclusions: These are sparsely spread and include a few fragments of micritic limestone (up to 400m)
and a rare ne purple stained unidentied bre (linen?), as in EA 264.
Vegetal material (SLY): Two categories appear: A) Entirely charred, compact (with no internal cell
structure) organic matter, probably vegetal of origin (up to 500m). B) Small fragments of badly
preserved straw.
Reliability: High. The sample was enlarged in order to obtain more inclusions. This was possible due
to the presence of two large breaks on both edges of the tablet from which larger fragments of clay
could be chipped without causing any damage to the written surfaces.
Matrix: Carbonatic, light tan in PPL with speckled b-fabric, silty (4%) with few foraminifers. The
silt contains predominantly quartz with accessory minerals, including hornblende, zircon, and
microcline. Opaque minerals (1%-2%) appear in sizes reaching 40m.
Inclusions: F:c ratio{0.062mm}=96:4-95:5. These include rounded grains of micritic limestone (up to 380m).
Vegetal material (SLY): Up to 650m (long), several small badly preserved tissue fragments, one of
which includes bres.
Geological interpretation: Reworked Taqiye marl, namely marl that was not collected in situ but rather
from a eld close to the section where the marl was mixed with aeolian dust (see EA 258).
Conclusions: Knudtzon (1915:1340) did not refer to the fabric of EA 264-265 but suggested that the
fabric and style of writing of EA 266 are similar to those of EA 296, thus they may have been written
by the same scribe. His conclusions were corroborated by Vita (2000:74) in his study of the scribe of
Gezer (Chapter 14.I). From a petrographic point of view, EA 266 and EA 296 clearly differ.
258
COMPARATIVE STUDY OF THE TAGI CYLINDER FROM BETH-SHEAN

Beth-shean 050020, an inscribed clay cylinder from Tagi to Lab<ayu
A small clay cylinder (height: 24 mm, circumference: 40 mm) was found in 1993 in Area E (Locus
05014, Basket 050020) at Beth-shean in the dump of the University of Pennsylvania excavations. It was
published, interpreted and discussed in detail by Horowitz (1996; see also Rainey 1998:239-242).
Reliability:
Matrix: Dense, yellowish-brown in PPL with speckled b-fabric, carbonatic and devoid of any
foraminifers. It contains haematite particles (1%) up to 30m in size. Very little quartz silt (less than
0.5%) appears.
Inclusions: Densely spread (f:c ratio{0.062mm}=~80:20) consisting only of badly sorted single rhombs of
idiomorphic dolomite, ranging in size between 25m and 300m.
Firing temperature: Estimated at above 7000C taking into account the partial decomposition of the dolomite.
Geological interpretation: Similar to EA 252 and 254 of Lab<ayu, the ruler of Shechem, and EA 286,
287, 289, and 290 of Abdi-eba, the ruler of Jerusalem. The cylinder belongs to a well-known
petrographic group that is derived from the Moza and >Amminadav formations, distributed in the
Judaea - Samaria anticline (see EA 252).
Conclusions: The cylinder carries a short message from a person named Tagi to his lord, Lab<ayu,
doubtless the ruler of Shechem. The clay of the cylinder is very different from the clay of the Tagi
tablets, and there is no way to link the city of Ginti-kirmil with the formations which characterize the
JudaeaSamaria anticline. Thus, the Beth-shean cylinder was most probably made in the territory
of Shechem. This conclusion raises obvious difculties: Why would Tagi, if indeed he was the ruler
of Ginti-kirmil, write this message to Lab<ayu, the ruler of Shechem, from the latters city, and why
should such a message nd its way to Beth-shean?
Naaman suggests that Tagi who dispatched this message is not Tagi the ruler of Ginti-kirmil,
but a Shechemite official who wrote to his lord when the latter appeared/served in the Egyptian
centre of Beth-shean. This would explain why he addressed Lab<ayu as my lord (line 2) and to
the king, my lord (lines 5-6), expressions which are commonly used in correspondence between
a king and his officials but do not conform with the practice between two equal rulers. Tagi is a
common Hurrian name; it appears in one of the Ta >anach tablets (TT 3: rev. 5). In the light of the
evident petrographic results and the undoubted origin of the cylinder in the Central Hill Country,
Goren adopts this view and agrees that there is no other possible explanation.
Finkelstein accepts Naamans proposal as a viable solution but is ready to consider other
possibilities, admittedly quite speculative. For instance, a message from Tagi, the ruler of Gintikirmil, to Lab<ayu could have been copied at Shechem and sent to the Egyptian administrative
centre at Beth-shean.
259
X. PEHEL (PIILU)
The city of Piilu is identied with the mound of the village of Tabqat Fahil in the eastern ank of the
central Jordan Valley the location of classical Pella (Houston Smith 1993).

Pella is situated at the foot of the Irbid plateau, bordering the Ghor plain of the central Jordan Valley.
The mound is located on marl and shales of the Maastrichtian Paleocene age. The Taqiye formation
is thus the available clay source in the immediate surrounding of the site. It is bordered by exposures of
Santonian to Campanian chalk, marl, chert and phosphorite to the east, and Eocene chalk to the north
and west. The Jordan riverbed, about 5 km west of the site, also drains basaltic terrains (such as the
Yarmuk basin and the basalts of the Beth-shean area and north of it), but this sand is not found in the
immediate area of the site (although it is within its CTF exploitable area). The further slopes of the Irbid
plateau are typied by Cenomanian Turonian limestone series.
To date we have not examined pottery from Pella. Edwards (1992) published a study of ceramic
raw materials around the site, but his report is not sufciently informative for the present study. It lacks
geological references to the samples collected since the names or even ages of the formations from
which they were taken are not given. Neither does it use the geological maps printed in Jordan nor older
references (i.e., Bender 1974). Finally, the locations of the samples selected cannot be correlated with the
data supplied by a geological map of the area. As an example, a high ridge south of Tell el-usn, below
the trees is not a practical landmark. Edwards only contribution to our investigation is in indicating that
potters at Pella employed raw materials from the immediate surroundings of the site.
In this study we used petrographic data from neighbouring sites represented in our comparative
collection, namely Late Bronze Age ceramics from Tell Abu al Kharaz near Pella (Goren and Fischer
1999). This site is located in a geological environment similar to that of Pella. The nds there included a
Late Bronze I pottery workshop with a potters wheel and a group of un red vessels, discovered in 1996.
CATALOGUE
EA 256 (BM 29847), from Mut-Ba>lu to the King of Egypt
Matrix: Carbonatic, greyish-yellow in PPL speckled b-fabric, with few foraminifers, rich in haematite
particles (3%) and fewer opaques (1%). The haematite also appears as in llings within some
foraminifers. Under higher magnications (>X200) the matrix is brous and exhibits sparkled
Inclusions: Sparsely spread (f:c ratio{0.062mm}=99:1-98:2) particles including a very large (2.5 X 0.9 mm)
clear crystal of quartz with mineral inclusions, most likely derived from a druze.
Vegetal material (SLY): Plant tissues, up to 400m long. A large fragment with many vessel members
with their indicative spiral secondary cell walls and a few non-indicative tissue fragments.
Geological interpretation: By its petrofabric, this tablet is obviously made of Taqiye shales. Indeed,
Taqiye marl is widespread around the site of Pella. The use of Taqiye marl is also common at nearby
Tell Abu al Kharaz. The group of un red Late Bronze I vessels mentioned above were all made of this
260
material and wadi sand from the adjacent channel of Wadi Yabis was used for the inclusions. In this
sand, as well as in the sand of the streams near Pella (Wadi Jirm and Wadi Malawi), the only coarse
quartz likely to be found is from druzes typical of Cenomanian - Turonian formations that appear in
the highlands to the east. Therefore, Pella ts the geological environment reected by this tablet.
Reference: The closest reference to the use of such materials near Pella is the above-mentioned unred
pottery from the Tell Abu al Kharaz workshop.
Conclusions: EA 256 of Mut-Ba>lu of Piilu (see lines 8, 13, 34) was written and sent from the authors capital.
EA 255 (VAT 333), from Mut-Ba>lu, to the King of Egypt
Matrix: Carbonatic, yellowish-tan in PPL with speckled b-fabric, containing infrequent badly preserved
foraminifers and their fragments. It is very slightly silty (below 1%) but richer in opaque minerals that
appear at a range of sizes from a few micrometers to about 30-40m. Under higher magnications
(>X200) the matrix is brous, optically active and displays very weak optical orientation.
Inclusions: Sand (f:c ratio{0.062mm}=90:10) in which subangular to well-rounded fragments (up to 500m)
of alkali-olivine basalt of several types, ranging between nely crystalline to nearly doleritic are
common. The basalt is usually rather weathered and the olivine is partly or entirely altered into
iddingsite. Single crystals of basaltic derived iddingsite, augite, and twinned plagioclase appear in
grain-sizes reaching 250m. Rounded grains of sparitic, micritic and biogenetic limestone (up to
500m) are frequent. There are a few fragments (up to 400m) of tan clay stained travertine and a
very few calcareous pisoliths with concentric microlamination. Also present are a very few grains
(up to 800m) of angular to subrounded replacement chert and rounded foraminiferous chalk (up
to 350m).
Vegetal material (SLY): A few (up to 650m long) non-indicative tissue fragments.
Geological interpretation: The material of this letter is different from that of EA 256 and similar to EA
224 and the like.
Conclusions: The letters of Mut-Ba>lu are dissimilar, although they both represent materials from the
central Jordan Valley. However, Pella is quite far from the Jordan River and therefore the materials
of EA 255 cannot be considered as local per se. Unlike Beth-shean and Rehob, the local geology at
Pella does not include basalt exposures. The closest source to Pella for basaltic sand is the Jordan
riverbed, located about 5 km away from the site. Indeed EA 256 seems to reect the local materials
in the environs of the site. Although theoretically both EA 256 and 255 could have been sent from
Piilu, using two distinct materials in the vicinity, it is more logical to assume that each letter was
sent from a different place EA 256 from Piilu and EA 255 from Beth-shean. The text of EA
255 deals with the passage of a caravan from Egypt to Mitanni (Hanigalbat) and it is reasonable to
assume that aya ordered Mut-Ba>lu to appear at Beth-shean for instructions, after which the letter
was dispatched to Egypt.
261
CHAPTER 13
THE CENTRAL HILL COUNTRY

I. SHECHEM (SHAKMU)
The Land of Shakmu (= Shechem) appears only once in the Amarna correspondence (EA 289:18-24),
while the town of Shechem is never mentioned. Yet there is no doubt that Shechem was the capital of
Lab<ayu (on Shechem in the Amarna letters see Reviv 1963; Campbell 1965; 1976; Naaman 1975:2746; Adamthwaite 1992).
Shechem must have controlled the entire northern half of the central highlands, as well as areas
in the Jordan Valley. According to Finkelstein, the border between Shechem and Jerusalem passed
near Bethel. Naaman assumes that the sparsely inhabited areas south of Shechem, between the site
of Sheikh Ab Zarad and Bethel, were a kind of no-mans-land. In the west the territory of Shechem
bordered on Ginti-kirmil and (according to Finkelstein) on Gezer. In the northwest it bordered on
Ta>anach (Naaman) or Megiddo (Finkelstein). The boundary between the territories of Shechem and
Rehob passed along the eastern slopes of the north Samaria Hills. In the southeast Shechem could have
bordered on the territory of Piilu, with the Jordan River separating them.
From EA 250 we learn that Lab<ayu conquered several cities in the Dothan Valley and in the
southern Jezreel Valley Burquna (today the village of Burqin), arabu, Gitti-rimmunima (nowadays
Rummaneh), Shunama (biblical Shunem, the village of Sulam in the Jezreel Valley) and Gitti-padalla
(for identifications, see Rainey 1968:7-8; Zori 1977:55, 57; Zertal 1992:68, 97-98 and the chapter on
Ginti-kirmil above). Lab<ayus conquests reached up to the Pharaonic lands in the Jezreel Valley.
According to Naaman, Lab<ayu took over the kingdom of Ta >anach, the territory of which included the
towns mentioned in EA 250. The city of Ta>anach did not recover from Lab <ayus attack and declined
soon afterward, as evident from the results of the excavations conducted at Tell Ta >annek. According to
Finkelstein the towns mentioned in EA 250 belonged to the kingdom of Megiddo, and attacking them
was part of Lab <ayus offensive against Biridiya.

The geology and reference materials from the central hill country anticline are discussed below. New
detailed geological mapping of the Shechem area (Cook 2000) was used for additional information.
CATALOGUE
EA 252 (BM 29844), from Lab<ayu to the King of Egypt
Reliability: Satisfactory. Although only tiny fragments could be taken because of the complete preservation of
the surface and the script, the results obtained of the petrographic examinations are very conclusive.
Matrix: In the hand examination the clay is yellowish (Munsell colour 2.5Y7/4), altering to pink in a
fresh break. In thin section the matrix is dense, yellowish-tan in PPL, carbonatic and devoid of any
foraminifers. It contains haematite particles (1%) up to 5m in size. The matrix is optically active
with speckled/striated b-fabric.
262
Inclusions: No remains of vegetal material. Densely spread (f:c ratio{0.062mm}=~80:20) homogeneous sand
particles in which badly sorted single rhombs of clear idiomorphic dolomite, ranging between 25m and
250m, predominate.
Firing temperature: Probably un red or very lightly red since dolomite is not altered into calcite which
occurs at slightly above 5000C.
Geological interpretation: Dolomitic sand is readily identied in this tablet even by the naked eye. Its most
characteristic feature is the homogeneous, densely spread, well-sorted ne whitish sand particles that
under a magnifying glass exhibit rhombic shapes. In thin-section they appear as silt to sand-sized
inclusions of euhedral dolomite crystals. The particles exhibit a typical rhomboid shape, indicating
the use of sand that was not subjected to any signicant processes of translocation. This implies an in
situ development of sand due to the dissolution of the cementing matrix of coarse crystalline dolomite
aggregates. Such phenomena are quite common in the Cenomanian section of the central mountain
ridge (Bentor 1945, Arkin et al. 1965).
Based on the extensive body of reference material, this petrographic group is identied as originating
from clay of the upper member of the Moza Formation, mixed with dolomitic sand that was quarried
from the capping >Amminadav Formation.
Reference: The Moza->Amminadav petrographic group is well known from pottery assemblages from
sites of different periods spread throughout the central hill country anticline. In the Chalcolithic
period it typies the Judaean sites (Goren 1987, 1991a, 1995; 1996b) but is rare in other regions. It
has been recorded from Early Bronze Age I sites in central Israel, extending from the Beersheba
Valley to Aphek (Porat 1989a: 47-48; Mir et al. 1992). In the Intermediate Bronze and Middle
Bronze Ages this group dominates sites in the vicinity of Jerusalem. In several assemblages
examined, including the large habitation sites of Naal Refaim (Eisenberg 1993; 1994), Manat
and Naal Zimra (Meitlis 1991), it constituted a major part of the ceramic material (Goren 1996a).
Petrographic examination of the Intermediate Bronze-Middle Bronze Age pottery collected in
the central hill country between Jerusalem and the Jezreel Valley (for the distribution of the sites
see Finkelstein 1991) revealed that this group dominated the ceramic assemblages of sites located
around Jerusalem (Goren, unpublished). At the Intermediate Bronze-Middle Bronze Age site of
Naal Refaim, an ancient quarry of dolomitic sand was discovered (Eisenberg 1994:86). It is
located in an outcrop of the lowermost unit of the >Amminadav Formation, immediately above the
uppermost member of the Moza clay. Intermediate Bronze Age burials found within the quarry
provide a terminus ante quem for its use. A Middle Bronze Age jar lled with this sand was
found at the quarry, and numerous potters wheels unearthed at the site con rm its use for pottery
production (Eisenberg, 1993:1280).
Moza clay, especially with dolomitic sand inclusions, dominates the ceramic assemblages of several
Middle and Late Bronze Age burial cave sites around Jerusalem, as well as the Middle Bronze
assemblage from Tel Moza west of the city (de Groot and Greenhot, pers. comm.). The >AmminadavMoza group is also known from the Iron Age I assemblages of Tell en-Nabeh and Khirbet Raddana,
as well as from the collared-rim pithoi unearthed at the site of Giloh (all as yet unpublished) and
from the pottery assemblage of Shiloh south of Shechem (Glass et al. 1993:78). This group was
found to be common in the Iron Age II assemblage from the City of David (Franken and Steiner
1990:79-85). In later periods Moza clay (sometimes with dolomitic sand) is known from the large
Early Roman period ceramic workshops at Binianei Hauma and Giv >at Hamivtar in Jerusalem, and
in the mediaeval workshop at Ramot (all examined petrographically by Goren, as yet unpublished).
263
Although dolomite rocks and hence dolomitic sand derived from them can be found in other parts of
the southern Levant, it seems that its use as inclusions was restricted to the central hill country.
Summing up, the Moza->Amminadav petrographic group dominates large areas of the central hill
country, especially around Jerusalem. It also appears further north, near Shechem. The latter area
is under-represented in our comparative material and hence less dominant in the discussion above.
Conclusions: The rich petrographic data mentioned above con rm that EA 252 was sent from the central
hill country. Although Moza clay is not exposed in the immediate vicinity of Tell Balaa (ancient
Shechem), it outcrops in Jebel Kebir, about 5 km to the northeast. This formation also appears in
signicant exposures to the south of Shechem. Hence, EA 252 can be considered a product of the
Shechem territory; the Shechemite scribes must have preferred this clay for producing their tablets
on the materials in their immediate vicinity.
Reliability: High.
Matrix: Carbonatic with uncommon foraminifers, somewhat silty (5%) greyish-tan in PPL with
tan lamina and sparsely spread globular concentrations of dark tan, silty clay. The silt
is essentially quartzitic but with additional opaques, hornblende, zircon and plagioclase.
Inclusions: Sparsely spread (f:c ratio{0.062mm}=~97:3) sand in which rounded, sometimes foraminiferous,
particles of chalk and nari (up to 400 m) are frequent. Quartz is subrounded to subspherical,
up to 250 m.
Vegetal Material (SLY): Very few, non-indicative tissue fragments.
Firing temperature: Probably unred or very lightly red since vegetal material remained uncharred.
Geological interpretation and reference: The matrix of EA 253 indicates a mixture of rendzina soil with
some terra rossa (see EA 259 In Chapter 12.VIII for the petrographic denition of the former and
EA 64 in Chapter 14.II for the latter). The two soil types are found in various locations in the Central
Hill Country, and particularly near Tell Balaa in Shechem where limestone of the Bi>na Formation is
found in close proximity to Eocene chalks (Cook 2000). Therefore, although petrographically EA 253
is different from EA 252 and EA 254, it still reects the local materials in the vicinity of Shechem.
Conclusions: Probably a letter from Shechem but made of different materials from EA 232 and EA 234.
Reliability: Fair.
Matrix: The matrix is dense, yellowish-tan in PPL with speckled b-fabric, carbonatic and devoid of any
foraminifers. It contains haematite and opaque particles.
Inclusions: No remains of vegetal material. Densely spread homogeneous sand particles made up of
badly sorted single rhombs of clear euhedral dolomite crystals ranging between 25m and 250m
and a few angular quartz grains (up to 250m).
Firing temperature: Probably unred or very lightly red since dolomite is not altered into calcite which
264
CONCLUSION: THE SHECHEM CORRESPONDENCE

Two letters of Lab<ayu belong to a petrographic group, characteristic of the Samaria anticline, that
is derived from the Moza and >Amminadav Formations. These formations expose at Jebel Kebir, at a
distance of a few kilometers from the site. EA 253 is different and is made of soils that appear closer to
the site. The textual evidence together with the petrographic data shows Shechem to be the only possible
location for this ruler.
II. JERUSALEM (URUSALIM)

The city-state, or rather dimorphic chiefdom, of Jerusalem (Urusalim) dominated a territory in the
southern part of the central hill country. Its extent in the north and south is debated among scholars.
Kallai and Tadmor (1969:143-145) and Finkelstein (1993; 1996:228-229, 234-235, 255) have argued that
the territory of Jerusalem extended over the entire southern hill country, including the sparsely settled
Hebron hills, down to the Beersheba Valley. Alt (1953:107-108; 1959:246, 251-252) and Naaman (1975:
104-114; 1992) suggested that Jerusalem was a fairly modest hill-country city-state. In view of the longperspective history of the hill country, Naaman (1992) proposed that Debir was the centre of a city-state
that dominated the southern Hebron hills and the Beersheba Valley. In his opinion the area between
Hebron and Beth-zur was a kind of no-mans-land, so that the southern border of Jerusalem extended
down as far south as this sparsely inhabited area. Finkelstein rejects this view on the grounds that there
is no textual clue to the existence of a second millenniu BCE polity south of Jerusalem, and that the Late
Bronze Age nds at Khirbet Rabud (= Debir) are meagre and insignicant.
Finkelstein (1996:229, 234-235) assumes that in the north, Bethel belonged to the territory of
Shechem, whereas Naaman (1992:285-286) suggested that it was either part of the territory of Jerusalem,
or a small independent city-state.
The western border of Jerusalem ran along the western slopes of the highlands, with the towns of the
longitudinal, Senonian valley of the eastern Shephelah belonging to the city-states of Gath and Lachish.
Gath ruled Qiltu (Keilah), while Aijalon (Yalo) belonged to Gezer. In the east Jerusalems territory
reached the Jordan Valley including the area around Jericho.

Jerusalem is situated in an area dominated by Cenomanian, Turonian and Senonian rocks of the Judaea
and Mount Scopus groups. Chalk and chert of the Menuha and Mishash Formations dominate the area to
its east (Arkin et al. 1976). The old city of Jerusalem and the City of David are situated on the dolomitelimestone series of the Turonian Bi>na Formation. To the west, the older Cenomanian formations of
the Judaea Group are exposed, including the dolomitic Weradim Formation, the chalky Kfar Shaul
Formation, the dolomitic >Amminadav and Beit-Meir Formations and the clay-marl Moza Formation
(Arkin et al. 1965; 1976).
Of all the above only the Moza Formation contains clay minerals of appreciable volume (TaitelGoldman et al. 1992; Taitel-Goldman 1993:18-21). Exposures of this formation outcrop at Naal
Refaim, about 5-6 km to the southwest of the City of David. Another source of clay in this area is the
terra rossa soil that develops on hard limestone and dolomite rocks. It appears alone or as part of a soil
complex together with rendzina, or as alluvial-colluvial soils of terra rossa origin (Rabikovitch 1981:
58-59). Due to the high content of iron minerals and organic matter, the deep red colour of this soil
increases with firing to become dark tan.
265
In both Jerusalem and Shechem the scribes preferred to travel a distance of a few kilometers in order to
extract clay of the Moza Formation which, being ner in texture and brighter in shade, was found more suitable.
The extensive collection of thin-sections of pottery from Jerusalem and its vicinity covering almost all
periods was used for petrographic reference.
CATALOGUE
Reliability: High.
Matrix: The clay is dense, yellowish-tan in PPL, carbonatic and devoid of any foraminifers. It contains
haematite particles (1%) up to 30m in size. The matrix is optically active and oriented with striated
b-fabric. A 1.2 mm sphere of ner clay yellowish-orange in PPL, densely packed (10%) with
elongated haematite particles (up to 120m lengthwise) appears in the matrix.
Inclusions: Densely spread (f:c ratio{0.062mm}=~80:20) homogeneous sand particles made up predominantly
of badly sorted single rhombs of clear idiomorphic dolomite ranging between 25m and 300m, a
few angular to subangular quartz grains (up to 200m) sometimes with undulose extinction and
mineral inclusions, and a very few rounded grains of micritic limestone (up to 600m).
Vegetal Material (SLY): Very few (up to 380m long) badly preserved plant tissues.
Conclusions: A local product of the central hill country anticline; no doubt a Jerusalem-made tablet.
haematite particles (1%) up to 40m in size. The matrix is optically active and oriented.
Inclusions: Densely spread (f:c ratio{0.062mm}=90:10-85:15), homogeneous sand particles made up
predominantly of badly sorted single rhombs of clear idiomorphic dolomite ranging between 25m
and 300m, a few subangular to rounded quartz grains (up to 500m), sometimes with mineral
inclusions, and rounded grains of micritic limestone (up to 500m). Also present are a very few
grains of subangular replacement chert (up to 750m).
Vegetal Material (SLY): Very few (up to 380m long) badly preserved plant tissue fragments.
Geological interpretation: Similar to EA 252 (above).
Reliability: High.
266
haematite particles (1%) sizing up to 30m. The matrix is optically active and oriented. A 1.2 mm
sphere of ner clay yellowish-orange in PPL, densely packed (10%) with elongated haematite
particles (up to 120m long) appears in the matrix.
Inclusions: Densely spread (f:c ratio{0.062mm}=~80:20), homogeneous sand particles made up
predominantly of badly sorted single rhombs of clear idiomorphic dolomite ranging between
25m and 300m, a few angular to subangular quartz grains (up to 200m) sometimes with
undulose extinction and mineral inclusions, and a very few rounded grains of micritic limestone
(up to 600m).
Vegetal Material (SLY): Very few (up to 380m lengthwise) badly preserved plant tissue fragments.
Palaeontology (LG): A few badly preserved small foraminifers, lacking outer test and heavily
recrystallized. Identied: Hedbergella (p); age undetermined.
EA 289 (VAT 1645 + 2709), from Abdi-eba to the King of Egypt
haematite particles (1%) sizing up to 30m. The matrix is optically active and oriented.
Inclusions: Densely spread (f:c ratio{0.062mm}=~80:20), homogeneous sand particles made up
predominantly of badly sorted single rhombs of clear idiomorphic dolomite ranging between
25m and 300m, a few angular to subangular quartz grains (up to 200m) sometimes with
undulose extinction and mineral inclusions, and a very few rounded grains of micritic limestone
(up to 600m).
Reliability: Moderate. Resampled by SPA to supply a satisfactorily reliable sample.
Matrix: The clay is dense, reddish-brown in PPL, carbonatic and devoid of any foraminifers. It contains
haematite particles (1%) up to 30m in size. Some quartz silt (1%) appears with accessory heavy
minerals, including twinned plagioclase, hornblende and zircon. The matrix is optically active
and oriented.
Inclusions: Densely spread (f:c ratio{0.062mm}=~90:10%-85:15), sand particles of predominantly badly
sorted single rhombs of idiomorphic dolomite, ranging between 25m and 300m. The crystals
are gloomy, showing partial decomposition of the carbonate due to ring. There are a few angular
to subrounded grains of quartz (up to 310m), sometimes with undulose extinction and mineral
inclusions. Several are polycrystalline.
Firing temperature: Estimated at above 700 0 C (dolomite decomposition) and below 800 0 C (hornblende
267
unchanged). The colour change of the matrix (as compared with EA 286 and EA 289) is due to the
firing of this tablet.
Reliability: High.
Matrix: The matrix is carbonatic, light yellowish-tan in PPL, containing infrequent foraminifers. It is
rather silty (about 4%-5%) and rich in opaque minerals (~4%) that appear at a range of sizes from
few micrometers to about 30-40m. Other minerals that appear within the silt fraction include
hornblende, zircon, augite, iddingsite, and twinned plagioclase. Under higher magnications
(>X200) the matrix is brous, optically active and displays very weak optical orientation.
Inclusions: The inclusions contain sand and some vegetal material (straw). Rounded grains of sparitic
and micritic limestone (up to 700m) are frequent. Subangular to well-rounded fragments (up
to 1.2 mm) of alkali-olivine basalt of several types, ranging between finely crystalline to nearly
doleritic, are common. The basalt is usually weathered and the olivine is partly or entirely
altered into iddingsite. Single crystals of basalt-derived minerals include iddingsite, augite, and
plagioclase. They appear in grain-sizes reaching 250m. Rounded to subrounded sand-sized
quartz grains (up to 500m) are also common. There are a few angular to subrounded grains (up
to 1 mm) of replacement chert and rounded grains (up to 350m) of foraminiferous chalk. A very
few fragments of tan clay stained travertine (up to 2 mm) and calcareous pisoliths with concentric
microlamination (up to 550m) appear and rare rounded fragments (up to 500m) of phosphorite
with spherical phosphorous concentrations, bone fragments and foraminifers.
Vegetal Material (SLY): Few (up to 750m long) plant tissues (straw). Badly preserved tissue
fragments.
Firing temperature: Probably unred or very lightly red since the vegetal material remained uncharred.
Geological interpretation: This tablet is undoubtedly alien to the central hill country anticline. It is
similar to EA 224 and likewise should be attributed to the Beth-shean group. The phosphorite
occurring in the inclusion assemblage is also a good indication for a Jordan Valley provenance.
The combination of Senonian chalks, chert and phosphorite may match either the Jordanian plateau
(e.g. the Madaba Plains or the Amman area), or the Israeli Judaean Desert. When appearing within
the Jordan sand, it directs us immediately to the central Jordan Valley since the Mishash formation
exposes along the Jordan drainage system only between Khirbet esh-Shunneh and Wadi Yabis.
Conclusions: EA 285 was made of sediments from the central Jordan Valley. It was most likely produced
at the Egyptian administrative centre of Beth-shean.
EA 291 (VAT 1713), from the ruler of Jerusalem (?)
Reliability: High. Sampled twice in order to increase the sample size, due to the fragmentary state of this
tablet.
Matrix: Carbonatic with some foraminifers (~1%-2%), dark greyish-tan in PPL, optically active and with
weak optical orientation. Opaque minerals are infrequent (below 1%), sizing up to 40m. Haematite
268
appears in similar grain sizes. Quartz silt is uncommon (below 1%).

Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=~95:5) of subangular to rounded quartz (up
to 500m) and rounded foraminiferous chalk (up to 300m).
Conclusion: Petrographically EA 291 belongs to the typical fabric of the Gezer tablets (below).
Discussion: Knudtzon attributed this letter to the Jerusalem correspondence, and was followed by all
other scholars (e.g. Schroeder 1915:No. 167; Moran 1992:334; Liverani 1998:88). The attribution
was made on the basis of (a) the script; (b) the writing on the left round side of the tablet which
appears in several letters of Jerusalem and is rare in other Amarna letters; (c) the verbal form
lumaer, which is known only from the Jerusalem correspondence (Knudtzon 1915:1344 n. 1;
Moran 1975:150-151). Since in all these elements EA 291 differs from the Gezer tablets, there is a
marked contrast between the results of the petrographic analysis and the other data related to this
letter. It seems that on a certain occasion, when peaceful relations existed between the two polities,
the ruler and scribe of Jerusalem visited Gezer and sent EA 291 from there.
CONCLUSION: THE JERUSALEM CORRESPONDENCE

Five of the seven letters of Abdi-eba, the ruler of Jerusalem belong to a petrographic group that is
derived from the Moza and >Amminadav formations distributed in the central hill country anticline
and used frequently for pottery production in the vicinity of Jerusalem. Only EA 285 and 291 are alien
to Jerusalem. The first was most likely sent from Beth-shean. In this letter, Abdi-eba complains that
Yanhamu, an important Egyptian official in Canaan, took possession of his house and settled there an
Egyptian garrison. Abdi-eba might have traveled to Beth-shean to discuss this delicate matter with an
Egyptian official, and on this occasion wrote and sent this letter to the Pharaoh. EA 291 was sent from
Gezer. This letter is fragmentary and the background of its exceptional origin remains unknown.
269
CHAPTER 14
THE SHEPHELAH AND THE SOUTHERN COASTAL PLAIN
I. GEZER (GAZRU)
Gezer (Gazru in the Amarna correspondence) is unanimously identied with Tel Gezer (Tell el-Jazari),
located between the coastal plain and the valley of Aijalon.
Ayyaluna (biblical Aijalon = the village of Yalo) and ara (biblical Zorah; for the identication
and the site see recently Niemann 1999) belonged to the territory of Gezer (EA 273), which reached the
foothills in the east. Many scholars assume that Rubutu was a town near the border of Gezer and Jerusalem
(Mazar 1957:60-63; Aharoni 1967:286-287; Kallai and Tadmor 1969:143-144; Kitchen 1973:434-435;
Ahituv 1984:165-167). Accordingly it was identied with Khirbet amideh near Larun (Aharoni 1969).
Recently, Naaman (2000) suggested identifying it with the Rubutu mentioned in a Ta>anach tablet (TT
1:26) and with the city of Aruboth the centre of the Solomonic third district (1 Kgs 4:10). According to
this proposal, Rubutu should be sought at Tell el-Muafar, the largest mound in the Dothan Valley.
The city of Aphek is mentioned in the topographical list of Thutmose III and the account of
Amenophis IIs second campaign to Canaan and then disappears from the Egyptian records (Ahituv
1984:61). Following its surrender to Amenophis II, Aphek was probably annexed to the territory of the
nearby Egyptian centre of Jaffa, which extended from the sand dunes south of the city up to the sources
of the Yarkon River near Tel Aphek (Ras el->Ain). This site was rst built on a small scale, and in the
13th century an Egyptian fortress was built on the ruins of the Canaanite palace (Kochavi 1990). It is thus
evident that in the west and northwest Gezers territory bordered on that of the Egyptian centre of Jaffa,
while in the southwest it might have reached the Mediterranean coast. The town of Muazu, mentioned
in EA 298:25, should be identied with Yavneh-Yam on the coast (see Supplement) rather than Tell esSulan located more inland (Ahituv 1984:143-145, with earlier literature), as may be inferred from its
name (mz = harbour; see Stieglitz 1974). It could have served as Gezers outlet to the sea. In the south
the territory of Gezer reached Naal Sorek and bordered on Gath and Ashdod.
Gezers control over vast areas in the central coastal plain and the northern Shephelah explains its
prominence in the arena of Canaan in the Amarna period.

Tel Gezer is situated on Eocene chalk of the Adulam formation, capped by the nari crust and bordered by
Paleocene marl and shales of the Taqiye formation. The latter supplies a good source of clay for ceramic
production. Other possible sources are the rendzina soils that develop on the Eocene chalk crusts and the
alluvial soils that appear in the valleys west and north to the site. Gezer is bordered on the west by the
coastal plain where hamra red sandy loam and alluvial soils that were mixed with coastal sand blown
inland prevail (Sneh et al. 1998).
Only one petrographic study of pottery from Gezer has been published to date (Bullard 1970:107108). The local wares include materials from the immediate area around Gezer and from the lowlands
immediately north and east of the mound.
270
CATALOGUE
Reliability: High.
Matrix: Carbonatic with some foraminifers (~2%-3%), light greyish-tan in PPL, optically active and
with weak optical orientation. Opaque minerals (up to 40m) are infrequent (below 1%). Haematite
appears in similar grain sizes, but also as stains within the matrix and occasionally as inllings of
foraminifers. Quartz silt is uncommon (below 1%) and is accompanied by some accessory heavy
minerals, including zircon, epidote, plagioclase, hornblende, garnet and biotite. Very rare glauconite
concentrations (up to 100m) appear within the matrix.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=95:5-93:7) which is predominantly subangular
to rounded quartz (up to 250m), sometimes with undulose extinction and mineral inclusions.
Microcline appears as an accessory in the same grain size. Rounded grains of foraminiferous chalk
are common (up to 250m).
Vegetal material (SLY): A few (up to 500m long), badly preserved unidentiable tissue fragments.
Palaeontology (LG): Abundant planctonic foraminifers in matrix. Identied: Acarinina (p), Subbotina
(p) sp. Age: Paleogene.
Firing temperature: Very lightly red (probably below 5000C) as the glauconite is altered to pale yellow
although the organic matter is uncharred.
Geological interpretation: Based on its textural, mineralogical and palaeontological af nities, the clay of
EA 268 is readily identied as marl of the Taqiye formation (see EA 264). The combination of Taqiye
marl with coastal sand (see EA 168) which characterizes EA 268 is uncommon in the southern Levant.
There are only a few places where outcrops of the Taqiye formation nger towards the coastal plain.
From south to north they include the Gezer area, the Western Galilee and several locations near Tyre.
As explained above (EA 165, 223), the coastal sand north of the Carmel ridge is mostly calcareous
and rich in bioclasts, while south of the Carmel it is dominated by quartz and accessory minerals of
Nilotic origin. Therefore, the only option left for this tablet is the vicinity of Tel Gezer.
Reference: Similar clay and inclusions form one of the local fabrics in the ceramic assemblage of Gezer
(Bullard 1970:107-108).
Conclusions: EA 268 was most likely prepared at and sent from Tel Gezer.
Reliability: High.
Matrix: Carbonatic with some foraminifers (up to 5% in places), light to dark greyish-tan in PPL, optically
active and with weak optical orientation. Opaque minerals (up to 40m) are infrequent (below 1%).
Haematite appears in similar grain sizes, but also as stains within the matrix and occasionally
as inllings of foraminifers. Quartz silt is uncommon (below 1%) and is accompanied by some
accessory heavy minerals, including zircon, epidote, plagioclase, hornblende, garnet and biotite. Very
rare glauconite concentrations up to 100m in size appear within the matrix.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=~93:7) made up predominantly by
subangular to rounded quartz grains (up to 300m), sometimes with undulose extinction and
mineral inclusions. Microcline and plagioclase feldspars appear as accessories in the same grain
271
sizes. Rounded foraminiferous chalk (up to 1 mm but usually up to 500m) is common, in one case
stained by ferrous minerals.
Palaeontology (LG): Acarinina (p), Subbotina (p) sp. Age: Paleogene.
Firing temperature: Estimated at around 6000C to below 7000C, as the glauconite is altered to orangeyellow and the calcite inllings of the foraminifers are slightly deformed by heating.
Geological interpretation: As EA 268, but red to somewhat higher temperature.
Conclusions: Similar to EA 268.
Matrix: Carbonatic with some foraminifers (~2%-3%), light greyish-tan in PPL, optically active and with
weak optical orientation. Opaque minerals up to 40m in size are infrequent (below 1%). Haematite
minerals, including zircon, epidote, plagioclase, hornblende, garnet and biotite. Very rare glauconite
concentrations (up to 100m) appear within the matrix.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=95:5-93:7) dominated by subangular to
Microcline and plagioclase feldspars appear as accessories in the same grain sizes. Rounded
foraminiferous chalk (up to 250m) is common. There are a very few grains of angular replacement
chert with light brown stains (up to 150m).
Vegetal material (SLY): Common (up to 500m long), consisting of two classes: charred, dense, opaque
plant tissues (more frequent) and straw. One of the charred fragments is probably a dicotyledonous tree
(cross-section). There are a few fragments of straw, some of which include typical cell wall thickening
of the vessel member. None can be identied to the genus level.
Firing temperature: Estimated at around 6000C to 7000C as the glauconite is altered to orange-yellow.
The calcite inllings of the foraminifers are slightly deformed by heating and the organic matter is
partly charred.
Reliability: Moderate. Only tiny pieces were taken because the tablet is complete.
Matrix: Carbonatic with some foraminifers (~1%-2%), light to dark greyish-tan in PPL, optically active
and with weak optical orientation. Opaque minerals (up to 40m) are infrequent (below 1%).
Haematite appears in similar grain sizes. Quartz silt is uncommon (below 1%).
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=93:7) of subangular to rounded quartz (up
to 250m) sometimes with undulose extinction and mineral inclusions, subrounded chalcedony (up
to 250m) and rounded foraminiferous chalk (up to 300m).
Vegetal material (SLY): Up to 380m in this sample, badly preserved straw fragments.
Firing temperature: Estimated at around 6000C to 7000C as the calcite inllings of the foraminifers are
slightly deformed by heating.
272
EA 292 (BM 37647), from Ba>lu-danu to the King of Egypt

Reliability: High.
Matrix: Carbonatic with some foraminifers (~1%), light greyish-tan in PPL, optically active, with
weak optical orientation. Opaque minerals up to 40m in size are infrequent (below 1%). Quartz
silt is uncommon (below 1%) and is accompanied by some accessory heavy minerals, including
hornblende, zircon, epidote and garnet.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=95:5-93:7) of predominantly subangular to
rounded quartz grains (up to 250m), sometimes with undulose extinction and mineral inclusions,
rarely polycrystalline. Rounded foraminiferous chalk (up to 500m) is common.
Vegetal material (SLY): Few (up to 350m lon) badly-preserved plant tissue fragments (straw).
Palaeontology (LG): A few badly preserved planctonic foraminifers in matrix. Identied: Acarinina (p),
Subbotina (p) sp. Age: Paleogene.
Firing temperature: Unred or very lightly red, as the organic matter is uncharred. There are not
enough indicators to make a more precise estimate.
Reliability: High.
Matrix: Carbonatic with some foraminifers (~2%), light greyish-tan to light tan in PPL, optically active and
with weak optical orientation. Opaque minerals (up to 40m) are infrequent (below 1%). Haematite
minerals, including hornblende, zircon, epidote and plagioclase. Very rare glauconite concentrations
up to 80m in size appear within the matrix.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=95:5-93:7) dominated by subangular to
rounded quartz grains (up to 350m), sometimes with undulose extinction and mineral inclusions,
rarely polycrystalline. Accessory epidote appears in the ne sand fraction. Rounded foraminiferous
chalk (up to 250m) is common.
Vegetal material (SLY): Common (up to 500m long), unidentiable badly-preserved plant tissue fragments.
Palaeontology (LG): Planctonic foraminifers in matrix. Identied: Acarinina (p), Subbotina (p) sp. Age:
Paleogene.
Firing temperature: Very lightly red (around 5000C) as the glauconite is altered to pale yellow but the
organic matter is uncharred.
orientation. The silt is essentially of quartz, but also contains recognizable quantities of accessory
heavy minerals, including hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote
and rarely tourmaline and garnet. Opaques are relatively common (3%), ranging between a few
micrometers and about 60m.
273
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal matter (f:c ratio{0.062mm}=93:
7). Rounded to subrounded quartz grains (up to 430m) predominate, occasionally with undulose
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of
plagioclase feldspar appears in the same grain size. Fresh (unfossilized) aquatic mollusc shell
fragments (up to 400m long) are common and there are a few grains of rounded micritic limestone
(up to 320m).
Vegetal material (SLY): Common (up to 1.5 mm long), uncharred plant tissues (straw), unidentiable.
Firing temperature: Probably unred or very lightly red, as the vegetal matter shows no trace of charring.
Geological interpretation: This document is one of a group of tablets that are made of loess soil with
coastal sand (see EA 168). This combination is restricted to the area between Raphia in the south,
Ashkelon in the north and the Mefalsim - Kissufim area in the east.
Conclusions: Knudtzon (1915:1346, n. 2) noticed that while EA 297 was similar in script and
fabric to EA 292-293 of Ba >lu-danu, EA 298-300 formed a different group. This observation
is confirmed by our examinations. EA 298 is one of several tablets that were written on
northwestern Negev sediments as were EA 299-300. This stands as another tribute to
Knudtzons excellent powers of observation.
In EA 298, as well as in other tablets that belong to the loess matrix category (below), the
inclusions clearly indicate a coastal environment. The specific raw materials point to the coastal
strip between Raphia and Ashkelon. The only feasible explanation is that EA 298 was sent from
Gaza, the main Egyptian administrative centre in southern Canaan.
Reliability: High.
Matrix: Silty (10%), carbonatic, yellowish-tan in PPL, optically active with very weak optical orientation.
The silt is essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals,
including hornblende, augite, zircon, plagioclase, biotite, muscovite, rutile, epidote and (rarely) garnet.
Opaques are relatively common (3%) ranging between a few micrometers and about 70m.
7). Rounded to subrounded quartz grains (up to 700m) are dominant, occasionally with undulose
limestone grains (up to 640m) are common. Augite (180m), epidote (70m), hornblende
(120m), microcline (140m), rutile (100m) appear as rare accessories in the fine sand fraction.
Vegetal material (SLY): A few (up to 900m long), uncharred plant tissues (straw). A tissue fragment
with bres, badly preserved tissue fragments, all unidentiable.
Conclusions: Like EA 298, this is a tablet of a southern coastal origin, most likely from Gaza.
EA 300 (VAT 1606), from Yapau to the King of Egypt
Knudtzon (1915:1346, n. 2) noted that this tablet is similar in script and fabric to EA 298-299. It was
not possible to sample it because of its complete state of preservation and as the script covers its entire
surface. Examination under the stereomicroscope revealed beyond any doubt that it is indeed similar
to EA 298. Therefore it should be related to Yapaus Gazatian correspondence.
274

Reliability: High.
orientation. The silt is essentially of quartz, but it also contains recognizable quantities of accessory
heavy minerals, including hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote, and
rarely rutile and garnet. Opaques are relatively common (3%), ranging between a few micrometers
and about 60m.
7) comprising predominantly rounded to subrounded quartz grains (up to 700m), sometimes with
undulose extinction, occasionally polycrystalline, rarely with mineral and liquid inclusions. A grain
of plagioclase feldspar appears in the same grain size. Rounded micritic limestone (up to 380m) is
common and there are a few (up to 400m long) fresh (unfossilized) aquatic mollusc shell fragments.
Vegetal material: Common (up to 400m lengthwise) uncharred plant tissues. Cereal straw fragments
and unidentied tissue fragments.
Conclusions: As EA 298, a tablet of a southern coastal origin sent from Gaza.
OTHER LETTERS SENT FROM GEZER

EA 272 (BM 29863), from Shum-[...] to the King of Egypt
Reliability: High.
Matrix: Carbonatic with some foraminifers up to (5% in places), light to dark greyish-tan in PPL,
optically active and with weak optical orientation. Opaque minerals in sizes up to 40m are
infrequent (below 1%). Haematite appears in similar grain sizes but also as stains within the matrix
and scarcely as inllings of foraminifers. Quartz silt is uncommon (below 1%) and is accompanied
by some accessory heavy minerals including zircon, hornblende, plagioclase and biotite. Very rare
glauconite concentrations (up to 150m) appear within the matrix, sizing.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=~93:7) dominated by subangular to
Microcline and plagioclase feldspar appear as accessories in the same grain sizes. Rounded grains
of foraminiferous chalk (up to 1 mm but usually up to 500m) are common and there are a very few
grains (up to 270m) of replacement chert.
Palaeontology (LG): Planctonic foraminifers in matrix. Identied: Acarinina (p), Subbotina (p) sp. Age:
Paleogene.
Vegetal material (SLY): A few (up to 550m long), badly preserved, uncharred plant tissues (straw).
Firing temperature: Estimated at around 6000C to below 7000C as the glauconite is altered to orangeyellow and the calcite inllings of the foraminifers are slightly deformed by heating.
Geological interpretation: As EA 268, but red to a somewhat higher temperature.
Conclusions: Knudtzon (1915:1328, n. 1) noticed that the fabric and script of this tablet are similar to
those of the Milkilu letters. His proposal has recently been reafrmed by Vita (2000:70-71). Our
investigation conrms Knudtzons hypothesis and indeed suggests that EA 272 was made at Gezer.
275
Knudtzon identified the author of EA 272 as Shum-[...] of the city of [N ]a-x-a-[x]. Naaman
(2001b) proposed deciphering the citys name as Na[]a[rat/tu] = biblical Anaharath (abandoning
his previous reading Maazi [Naaman 1975:68] after collating the text in 1978). Anaharath
is identified with Tel Rekhesh in Naal Tavor (see EA 237-239). According to this suggestion,
the author of the letter was an ally of Milkilu of Gezer, just as Bayadi, possibly his predecessor,
was an ally of Lab <ayu of Shechem. He must have come to Gezer either on his way to Gaza, or to
negotiate with Milkilu, and on that occasion asked Milkilus scribe to write a letter to the Pharaoh
on his behalf. For a similar case see EA 291 (Chapter 13.II).
Significantly, three rulers of Gezer are already known: Milkilu, Yapau, and Ba >lu-danu (for
the reading see Van Soldt 2002). Not only is there hardly any space for an additional ruler in this
sequence, but a chain of four rulers in one place has no parallel in the Amarna correspondence (in
almost all cities, only two rulers are attested).
In regard to this proposal, Finkelstein and Goren note the significant distance between Gezer
and Anaharath. All other cases in which a ruler dispatched a letter from another city, written
by the scribe of the latter, involve either a neighbouring city (as in the cases of Yashdata, EA
248; Rib-Hadda, EA 136-138; and the Bashan cities, EA 201-206), or an Egyptian administrative
centre (EA 329). Recently, Rainey (2003:201*-202*) collated tablet EA 272 and suggested that
it was written by Ba >lu-danu, the sender of EA 292-293. This suggestion, if verified, solves the
problems involved with the former deciphering of the letter.
EA 273 (VAT 1686), from Belit-nesheti to the King of Egypt
Matrix: Carbonatic with some foraminifers (up to 5% in places), light to dark greyish-tan in PPL,
optically active and with weak optical orientation. Opaque minerals are infrequent (below 1%),
sizing up to 40m. Haematite appears at similar grain sizes, but also as stains within the matrix and
scarcely as inllings of foraminifers. Quartz silt is uncommon (below 1%) and is accompanied by
some accessory heavy minerals, including zircon, hornblende, biotite, plagioclase and microcline. Very
rare glauconite concentrations (up to 150m) appear within the matrix.
Inclusions: Moderately to well-sorted sand (f:c ratio{0.062mm}=~93:7) dominated by subangular to
Angular plagioclase feldspar appears as an accessory in the same grain sizes. Rounded grains of
foraminiferous chalk (up to 400m) are common.
Vegetal material (SLY): A few (up to 750m long) badly preserved uncharred plant tissues (straw).
Palaeontology (LG): Badly preserved foraminifers in matrix. Identied: Acarinina (p), Cibicides (b) sp.
Age: Paleogene.
Firing temperature: Estimated at around 6000C to below 7000C, as the glauconite is altered to orangeyellow and the calcite inllings of the foraminifers are slightly deformed by heating.
Reference: As EA 268, that is, a Gezer-produced tablet.
Conclusions: The author is a queen or queen mother, mentioning events that took place in the eastern
territory of Gezer: a raid of the >Apiru on Ayyaluna (biblical Aijalon) and ara (biblical Zorah), in
which the two sons of Milkilu barely escaped being killed. In another letter (EA 274) she reported
that the city of ab/puma was plundered. Zadok (1986:180) suggested that ab/puma may be the
same as Saffo of Josephus (Antiquities 17:10:9-290; Wars 2:5:1-70), identied at the village of
affa (G.R. 155 146), in the foothills near Gezers eastern border. He acknowledged the lack of
276
Late Bronze remains at the site and suggested that perhaps it is to be sought in a nearby site. The
relatively large multi-period mound of el-Burj (orvat Tittora), located only 3 km east of affa,
may t this identication, though surveys which have so far been conducted at the site failed to
produce clear evidence for Late Bronze Age occupation (Hismi 1993:109; Shavit 1992:90, contra
Gophna and Porat 1972:235). In any event, ab/puma was either a town in the territory of Gezer or
a secondary town in the territory of Belit-nesheti (see below), but not her seat.
Knudtzon (1915:1328, n. 2) noticed the similarity in fabric between the tablets of Belit-nesheti
(EA 273-274) and the tablets of Milkilu and Shuwardatu. His observation has recently been
corroborated by the palaeographic study of Vita (2000). Our petrographic analysis indeed indicates
that EA 273 was produced at Gezer. Liverani (1975:336 n. 25; 1998:126 n. 56; see Moran 1992:318
n. 1) suggested that Belit-nesheti was a queen mother who ruled in the name of her young son. The
question is whether she was a queen regent at Gezer, or a queen in a neighbouring city-state. In the
light of the petrographic results, Finkelstein and Goren support the former possibility, namely, that
Belit-nesheti was Queen Regent at Gezer after Milkilus death.
Naaman, while accepting this possibility, notes the following difculties in this assumption. Firstly,
Belit-nesheti referred to Milkilus sons in the third person (the two sons of Milkilu barely escaped
being killed), rather than in the rst person as would be expected if she was their mother. Secondly,
the designation sons of Milkilu has some parallels in the Amarna letters (the sons of Lab <ayu,
the sons of Arzaya [EA 289:7] and the sons of Abdi-Ashirta). These designations are always used
by neighbouring rulers for the heirs of well-known mayors shortly after the latters death, because the
name of the deceased father was better known than the names of his heirs. Thirdly, Yapau, Milkilus
heir to the throne of Gezer, complained that his younger brother, who was already grown-up, rebelled
against him (EA 298:20-27). Yapau, who was probably an adult upon his fathers death, ruled for
only a short period of time (Naaman 1975:69-72). As an alternative to the queen regent hypothesis,
Naaman suggests that Belit-nesheti might have ruled a neighbouring city shortly after Milkilus
death. He points to Beth-shemesh as her possible seat, and proposes that in her letters Belit-nesheti
reported to the Pharaoh on the dangerous situation near her territory. Finkelstein objects to the Bethshemesh alternative, mainly because the renewed excavations at the site have thus far not supported
the notion that the site was a major settlement in the Late Bronze Age (Grant and Wright 1939:35-50).
In the southern sector, Middle Bronze Age remains were unearthed immediately below Iron Age I
debris (Bunimovitz and Lederman, pers. comm.).
Beth-Shemesh (Tell er-Rumeileh) is a fairly large mound (about 4 ha) located on Naal Sorek, near
the southeastern border of the kingdom of Gezer. It was a relatively prosperous town in the Late
Bronze Age II (Stratum IV; Bunimovitz and Lederman 1993:250). A tablet with cuneiform alphabet
dating to the 12th century BCE was discovered at the site (Sass 1991), indicating that a local court
with a professional scribe operated there at that time. Writing in Akkadian, however, was a different
matter, and having no local scribe Belit-nesheti might have found it necessary to write her letters
from the neighbouring city of Gezer. Goren and Finkelstein note that this tablet does not shed
light on the discussion. Firstly, according to the petrographic analysis it was probably produced at
Lachish and secondly, the possibility that Beth-shemesh ourished in the 12th century BCE has no
implication for its status in the Amarna period. According to Naaman, the rst point is irrelevant.
The tablet might have been produced at Lachish, but was brought to Beth-shemesh and there served
an administrative purpose. Possible continuity of administration from the 14th to the 12th century
at Beth-shemesh is self-evident .
277
Fig. 14.1: Distribution of loess soil in the southern coastal plain.
278
CONCLUSION: THE GEZER CORRESPONDENCE

All the letters dispatched by Milkilu (EA 268-271) and Ba>lu-danu (EA 292) were written at Gezer1, while
of the ve letters sent by Yapau, only one was written there (EA 297) while the others (EA 298-300, 378)
were sent from Gaza. The duration of Yapaus reign appears to have been brief and he never managed
to overcome the internal problems that faced him and to stabilize his rule (Naaman 1975:69-72). Yapau
must have gone to Gaza in an effort to gain support of Egyptian troops in order to crush the rebellion that
broke in his city-state, and on these/this occasion/s wrote and sent his letters to the Pharaoh.
Vita (2000) has recently examined a group of Canaanite letters sent from Gezer and other
neighbouring cities. He concluded that the same scribe wrote about 24 letters, which he labeled the
Gezer-Corpus. Among them are letters of Shuwardatu (EA 278-280), Shum-[..] (EA 272), Belit-nesheti
(EA 273-274), Yazib-Adda (EA 275-276), a letter of Tagi (EA 266) and a letter of Yatiru (EA 296).
According to the petrographic analysis, the letters of cities other than Gezer in this list (EA 278-280, 275276, 266, 296), except for EA 272 (and possibly EA 273-374), were not sent from Gezer. Assuming that
Vitas conclusions (which rest on observations already made by Knudtzon) are valid, one can propose
that the Gezer scribe traveled to neighbouring cities whose rulers were Gezers allies (e.g. Ginti-kirmil,
Gath=Tell e-a and possibly Ashdod see below) and wrote letters on behalf of their rulers. In this
regard, one wonders why Shuwardatu and Tagi would call on the service of a scribe from far-away
Gezer to write certain letters for them, while they used their own scribes for other messages. We leave
this debate open for future research. A remarkable exception is letter EA 291, written at Gezer by the
Jerusalemite scribe. In the light of its fragmentary nature, no solution can be suggested for this enigma.
II. GATH (GIMTU)

It is widely accepted today that Gath of the Shephelah (Tel at=Tell e-a) was the seat of Shuwardatu
(Rainey 1975a; Liverani 1998:80). Whether the Gimtu mentioned in EA 290:9 is Shuwardatus capital
is debated among scholars, and there is no certainty that his capital is mentioned by name in the Amarna
letters (see Aharoni 1969:141-145; Naaman 1979a:682-683, with earlier literature). Stagers recent
proposal (1995:343) to locate Gath at Tel Haror should be rejected on three grounds: First and foremost,
in both the Late Bronze and the Iron Age I it was a relatively insignicant site (Oren et al. 1991). Second,
Tel Haror is too far to the south to be associated with Qiltu (biblical Keilah=Kh. Qila, G.R. 1504 1135
see below), which Shuwardatu refers to as my city (EA 280:23). Third, Stagers proposal can readily
be rejected on the basis of our petrographic study of Shuwardatus letters.
Gaths eastern border should be drawn along the line dividing the highlands from the Shephelah.
The city of Qiltu, located in the longitudinal valley of the eastern Shephelah, was a bone of contention
between Gath and Jerusalem (EA 279-280; 289:25-28; 290:14-18). Shuwardatu refers to it as my city
(EA 280:23) and hence it probably marks the southeastern corner of his territory. In the southeast Gath
bordered on the territory of Lachish and on the west it bordered on Ashkelon and Ashdod. In the north,
the border between Gath and Gezer seems to have followed the course of Naal Sorek (Wadi e-arar).
If this reconstruction is correct and there were no other city-states in this area, Gath ruled over one of the
most densely settled territories in Late Bronze Age Canaan (see Supplement). This conclusion may be
1. Knudtzon (1915:1344 n. 2) noted that EA 292-294 have exactly the same script, but the clay of EA 294 (greyish chocolatebrown) differs from that of EA 292-293 (grey, or grey with some brown spots). The same script appears in EA 296 and 297
and is close to that of Milkilus and the other related letters (EA 267-280).
279
corroborated by letter EA 283, in which Shuwardatu complains of rebellion in his territory, and that 30
cities have waged war against me (lines 19-20).

Tel at (Tell e-a) is situated in the Lower Shephelah, near the course of Naal Ha-Elah. The detailed
geology of its area is supplied by Buchbinder (1969: Kfar Menahem map, scale 1:20,000). The site
itself is situated on exposures of Middle Eocene globigerinal chalk of the Maresha Member of the Zorah
Formation, typically capped by nari crust. In its close vicinity lies the calcareous pebbly sandstone of the
Pliocene Pleshet formation, which in this area is heavily naried (Buchbinder 1969:6, 9). In the nearby
channel of Naal Ha-Elah a mixture of Quaternary soil, loess and gravel covers these rocks. This mixture
also forms the local soil of the site. It is composed of dark brown grumosolic soil and brown rendzina in
the higher places and accumulative brown and reddish-brown grumosols, dark brown grumosolic soils
and residual dark brown soils in the wadi channels. The latter soils are made of mixed alluvial clay, or
silty-clay, with aeolian silty-clay sediments (Dan et al. 1976: soil map of Gedera, 1:50,000). Within the
broader circle of 10 km from the site only the Adullam Member of the Zorah Formation is added. It is
generally characterized by the same lithology as in the Maresha Member, with the addition of silicied
chalk and chert nodules (Buchbinder 1969:5). Therefore, in terms of CTF (Chapter 2) Tel at is likely
to display a low index of mainly brown grumosolic loessial soils and brown rendzina, with mainly chalk
and nari, and occasionally some chert and ne-sand quartz as inclusions.
Late Bronze Age pottery retrieved in recent excavations at the site (kindly supplied by A. Maeir) and
ceramic workshop wasters from two nearby sites provided comparative material. One of these is a Late
Roman kiln excavated in 1998 at urvat a, ca. 1 km east-northeast of the mound (G.R. 136 123) by
S. Lander and D. Varga of the Israel Antiquities Authority. The other is an Iron Age II ceramic workshop
excavated at the foot of the mound in 2001 by Y. Israel of the Israel Antiquities Authority.
CATALOGUE
Reliability: High.
Matrix: Carbonatic, tan in PPL and optically active with speckled b-fabric. The silt (~10%) contains
essentially quartz but with the addition of calcite and accessory heavy minerals, of which hornblende,
plagioclase, rutile, epidote, zircon, microcline, muscovite, tourmaline, and opaques were identied.
Foraminifers are extremely rare.
Inclusions: Moderately sorted sand (f:c ratio{0.062mm}=~95:5) in which nari containing clay and some
quartz silt (up to 800m) is dominant. Subangular to rounded quartz (up to 400m), sometimes with
undulose extinction and mineral inclusions, is common.
Vegetal material (SLY): Tissue fragments, partly digested, with typical vessel members in some.
Geological interpretation: The matrix of this tablet combines the petrographic properties of loess (siltycarbonatic wind-blown, clay loam) and brown rendzina soil. The inclusions contain nari from the
mother-rock of the rendzina soil, some wind-blown quartz sand and articially added straw.
Brown rendzina occurs together with pale rendzina in the semi-arid to subhumid Mediterranean
climate. The distribution of the two soils is related to catenary differentiation (Dan et al. 1972).
280
The brown rendzina derives from the nari crust. This soil is eroded downslope, forming colluvialalluvial soils and grumusols. The amount of soil that results from the weathering of the nari is low,
and relatively large amounts of aeolian dust contribute to the formation of the brown rendzina soil.
The Eocene chalks of Israel are comprised of more than 60% foraminifer biorelicts. In the upper
nari the foraminifera are destroyed by dissolution and recrystallization processes, while in the
lower nari about 30% foraminifer biorelicts occur. The appearance of the foraminifera is one of the
important components in the description and classication of these soils and of the pottery that is
made from them.
Reference: The combination of rendzina-loessial soil with chalk and nari inclusions is known from
several sites in the southern Shephelah. It appears in our thin-section collection of Late Bronze
pottery from Lachish (Goren and Halperin 2004), Qubeiba (urvat Kfar Lachish) and Tell arasim
(Goren, unpublished). The latter site is located only two kms to the northwest of Tel at. Pottery
of the Tel at workshops mentioned above was also made from the same soil type. Therefore, the
materials in this tablet reect the geology of the Tel at environment.
Conclusions: EA 279 was sent from the lower Shephelah, most probably from Tel at. Our study
excludes the possibility of locating the city of Shuwardatu at Tel Haror (Stager 1995), since the
latter is located in the northwestern Negev an area typied by entirely different geological
environment (see EA 315).
Matrix: Carbonatic, tan to greyish-tan in PPL and nearly isotropic. The silt (~10%) contains essentially
quartz but with the addition of calcite and accessory heavy minerals, of which hornblende, epidote,
zircon, microcline, muscovite, tourmaline, and opaques were identied.
Inclusions: Sand (f:c ratio{0.062mm}=~90:10) in which nari containing clay and some quartz silt (up to 300m
in this small sample) is frequent. Subangular to rounded quartz (up to 200m), sometimes with
undulose extinction and mineral inclusions, is also frequent. An unidentied larva shell was seen.
Reliability: High.
Matrix: Carbonatic, yellowish-tan to tan in PPL, optically active with speckled b-fabric and weak optical
orientation. The silt (7%) contains equal amounts of quartz and micritic calcite, and accessory heavy
minerals, of which zircon, hornblende, rutile, muscovite, and opaques were identied. Foraminifers
are found (3%).
Inclusions: Badly sorted sand (f:c ratio{0.062mm}=~95:5) in which rounded nari and chalk, sometimes with
foraminifers, containing clay and some quartz silt (up to 1.6mm) is dominant. Subangular to rounded
quartz (up to 230m), sometimes with undulose extinction and mineral inclusions, is common.
Firing temperature: Low if at all red judging by the absence of any effect on the calcite or clay matrix.
281

Matrix: Carbonatic, tan in PPL and optically active with very weak optical orientation. The silt (~10%)
contains essentially quartz but with the addition of calcite and accessory heavy minerals, of which
hornblende, plagioclase, rutile, epidote, zircon, microcline, muscovite, tourmaline, and opaques
were identied. Foraminifers are extremely rare.
Inclusions: Moderately sorted sand (f:c ratio{0.062mm}=~95:5) in which nari containing clay and some
quartz silt (up to 700m) is frequent as is subangular to rounded quartz (up to 320m), sometimes
with undulose extinction and mineral inclusions. Rounded microcline feldspar appears as an
accessory in the same grain sizes.
Vegetal material (SLY): Badly preserved, probably partly digested plant fragments, unidentiable but
most likely herbivore dung.
Reliability: High.
Matrix: Carbonatic, yellowish-tan to tan in PPL, optically active with speckled b-fabric and weak optical
orientation. The silt (10%) contains equal amounts of quartz and micritic calcite, and accessory
heavy minerals, of which zircon, hornblende, pyroxene, muscovite, and opaques were identied.
Foraminifers are found (3%).
Inclusions: Moderately sorted sand (f:c ratio {0.062mm}=~90:10) in which rounded nari and chalk (up to 800m)
often with foraminifers and related nari containing clay and some quartz silt are dominant. Subangular to
rounded quartz (up to 380m), sometimes with undulose extinction and mineral inclusions, is common.
Rounded microcline feldspar appears as an accessory in the same grain sizes.
Palaeontology (LG): A few small foraminifers in matrix. Identied: Acarinina (p), Bolivina (b) sp. Age:
Paleogene.
Vegetal material (SLY): Plant tissue fragments, some of which show typical spiral thickening of the
secondary cell wall. There is a large fragment of wood, in the radial longitudinal plane, showing ray
cells and axial cells but unidentiable to genus.
EA 366 (AO 7096), from Shuwardatu to the King of Egypt
Reliability: Fair/Moderate.
Matrix: Carbonatic, tan to greyish-tan in PPL. The silt (~10%) contains essentially quartz but with the
addition of calcite and accessory heavy minerals, of which hornblende, epidote, zircon, plagioclase,
muscovite, and opaques were identied.
Inclusions: Sand of nari up to 200m and subangular to rounded quartz, up to 250m, in this sample.
282

Reliability: High.
Matrix: Carbonatic, greyish-tan to tan in PPL, optically active with speckled b-fabric and very weak
optical orientation. The silt (5%) contains essentially quartz but with the addition of calcite and
accessory heavy minerals, of which hornblende, plagioclase, rutile, epidote, zircon, microcline,
muscovite, tourmaline, and opaques were identied. Foraminifers are rather abundant (over 5%).
Inclusions: Moderately sorted sand (f:c ratio{0.062mm}=~95:5) dominated by rounded chalk and nari (up to
800m), often with foraminifers, and related nari containing clay and some quartz silt. Subangular
to rounded quartz grains (up to 380m), sometimes with undulose extinction and mineral inclusions,
are common. Rounded microcline feldspar appears as an accessory in the same grain sizes. There are
a very few rounded grains of chert (up to 150m).
Palaeontology (LG): The following foraminifers were identied in the matrix: Acarinina (p),
Chiloguemelina (p), Subbotina (p) sp. Age: Paleogene.
Vegetal material and coprolites: A few (up to 750m long), isotropic, hence digested, plant tissues
in which the cellulose component of the secondary cell wall has been digested and disappeared.
Additionally, there are coprolite fragments containing spherulites and calcium oxalate crystals,
indicating herbivore dung (see EA 118 for the denition of inorganic residues in dung).
Geological interpretation: The tablet is made of reworked marl of the Taqiye formation (based on the
petrographic afnities and the age, as determined by palaeontology). The inclusion assemblage
indicates an area where chalk, nari and some wind-blown ne sand quartz appear. This suits the
higher Shephelah area.
Reference: See EA 268 for references to the Taqiye formation in the Shephelah.
Conclusions: See Section IV below.
EA 63 (BM 29817), from Abdi-Ashtarti to the King of Egypt
essentially quartz but with the addition of calcite and accessory heavy minerals, of which hornblende,
Inclusions: Well-sorted sand (f:c ratio{0.062mm}=~95:5) dominated by rounded nari grains (up to 800m),
containing clay and some quartz silt. Subangular to rounded quartz grains (up to 300m), sometimes
with undulose extinction and mineral inclusions, are common.
Vegetal material and coprolites: Common (up to 750m lengthwise), plant tissues. Fragments of
parencymathous (soft) tissues and a fragment including primary xylem. Another tissue fragment
including bres. One fragment includes spherulites, indicating dung. No further identication is
possible.
Conclusions: Knudtzon attributed EA 63-65 to the Phoenician coast, suggesting that they were sent by
Abdi-Ashirta, King of Amurru. However the three letters share a common scribal background with
283
EA 282-284 which were sent by Shuwardatu. Therefore, Naaman (1979a:676-684) suggested that
the author of EA 63-65 and EA 335 was the successor of Shuwardatu at Gath. This was accepted
by Moran (1992) and Liverani (1998). Petrographically, three of the four letters are similar to the
tablets of Shuwardatu and were probably written in the same region. Thus the petrographic results
support Naamans hypothesis.
EA 64 (BM 29816), from Abdi-dINNIN to the King of Egypt
Sampling method: SPA. Tablet completely preserved.
Reliability: Moderate (together with the BM copy thin section).
Matrix: Reddish-tan in PPL, silty (~20%), non-carbonatic, ferruginous matrix exhibiting strong optical
orientation hence kaolinitic in nature. The silt includes predominantly quartz with accessory zircon
and hornblende. Opaques are common.
Inclusions: Well-sorted sand (f:c ratio{0.062mm}=~90:10) predominantly rounded quartz grains and
accessory minerals (up to 320m), sometimes with undulose extinction and mineral inclusions. A
few subrounded or subangular calcite crystals ( up to 200m).
Firing temperature: Undetermined, but below 750 0C since the calcite is unaffected.
Geological interpretation: The matrix is readily identied as terra rossa soil that was mixed with sieved
wadi sand. This soil unit occurs in the hilly areas of Israel, where a subhumid Mediterranean climate
prevails. The parent material is hard limestone, dolomitic limestone or dolomite. The soils grade
into colluvial-alluvial or brown red Mediterranean soil on the footslopes and further into a grumusol
in the valleys. The main soil development took place during the Quaternary period when a large
amount of aeolian dust was incorporated into the soil. Two types of terra rossa soil are recognized:
red terra rossa that developed on hard limestone, and a brown-red type that developed on dolomitic
limestone. The former is rich in the kaolinitic clay mineral while the latter is smectitic (Dan et al.
1976; Koyumdjisky and Dan 1969). The rst type is relevant to the case of EA 64.
Terra rossa soil is rich in silt-size quartz grains and very ne sand of 30m-100m. The 100m value
is the upper size-limit of the aeolian dust of desert origin that settled in Israel (Wieder and Gvirtzman
1999). As hard limestone and hard dolomitic limestone do not contain silt-size quartz grains, the
microstructural pattern shows vividly the considerable contribution of the aeolian dust to the soil
material. This soil type is not local to the area of Tel at, where dark brown grumusolic soils and
brown rendzina are exposed (Dan et al. 1976; soil map of Gedera, 1:50,000). Terra rossa soils appear
only from the Elah Valley and eastwards (Dan et al. 1976; soil map of Beth-shemesh, 1:50,000), at the
edge of the exploitable area of Tel at
Reference: Terra rossa is widely exposed over the mountainous regions within the Mediterranean climatic
zones of the southern Levant, including the central highlands, Mount Carmel and the Galilee. It
also appears in the Shephelah, in wadi channels draining these regions. The exact provenance of
ceramics belonging to this petrographic group cannot be determined on the basis of their matrix
alone. However, the reference material from Levantine sites, together with the inclusions that appear
in this tablet, may indicate a more specic provenance.
The use of terra rossa as clay for ceramic vessels is known from assemblages belonging to the
central hill country or the upper Shephelah. terra rossa soil, mixed with wadi sand, crushed calcite
or grog, was frequently used by Iron Age potters in Judah as a ceramic raw material, especially
for the production of cooking-pots. In the City of David, most of the numerous clay gurines were
made locally of this soil (Goren et al. 1996). More relevant is the case of the LMLK stamped jars. A
selection of 180 items of this jar type was examined by NAA (Mommsen et al. 1984). The results
284
suggested that the jars were produced at a single site, perhaps located in the Upper Shephelah.
In a recent study carried out by Goren and Bunimovitz (unpublished), samples of these jars were
examined petrographically and discovered to have been produced of terra rossa soil and chalk,
quartz, and chert temper. Accordingly, this group could be linked with the upper Shephelah,
adjacent to the foothills of the Judaean highlands. A survey of comparative raw materials conducted
around Tel Sochoh and Tel Lachish, where numerous LMLK jars were found, revealed exposures of
similar soils and sands near the former.
Conclusions: The petrographic data suggests an Upper Shephelah origin for EA 64. If Abdi- d INNIN,
the author of EA 64, was indeed a ruler of Gath (i.e., his name would then be read Abdi-Ashtarti),
the letter was sent from a location in the eastern ank of the Gath territory. In this case, the
town of Qiltu (Keila) seems to be the most likely candidate. Indeed Shuwardatu, Abdi-Ashtartis
predecessor, called it my city (EA 280:23) and might have sent one of his letters from there
(see EA 278). In a different situation, Abdi-Ashtarti might have also sent his letter from Qiltu.
Alternatively, the author of EA 64 can be disassociated from the author of EA 63, 65 and 335 (see
EA 229). In this case the former must have ruled a city-state in the eastern Shephelah. Naaman
suggests Beth-Shemesh, Tell >Aiun and Tell Beit Mirsim as possible candidates.
EA 65 (VAT 1685), from Abdi-Ashtarti to the King of Egypt
essentially quartz, with the addition of calcite and accessory heavy minerals, of which hornblende,
Inclusions: Well-sorted sand (f:c ratio{0.062mm}=~95:5) in which rounded nari grains (up to 800m)
containing clay and some quartz silt are dominant. Subangular to rounded quartz grains (up to
300m), sometimes with undulose extinction and mineral inclusions, are common. A few terrestrial
snail shell fragments (up to 900m long) were included.
Vegetal material and coprolites: A few (up to 750m long) plant tissues. Two very small tissue fragments
with elongated cells, probably bres. No further identication is possible.
Conclusions: Knudtzon (1899:297; 1915) suggested rendering the name of the author [a]d-[r]a-INNIN. He
noted the close similarity of EA 65 to letters EA 63-64, and assumed that the same ruler sent all three.
He explained the writing [a]d-[r]a-INNIN as a metathesis of the rst two signs (ad+ra = ARDA =
>abdi). Moran (1992:136) assumed a ligature of signs and read it [a]b-[d]i-INNIN. Naaman (1998a)
suggested reading the damaged name [mZ]u?-[r]a- [u]r?, identifying him with Zur-Ashar, the author
of EA 319. However, this suggestion is not supported by the petrographic analysis.
EA 335 (VAT 1616 + 1708), from Abdi-Ashtarti (?) to the King of Egypt
Reliability: High.
Matrix: Carbonatic, yellowish-tan to tan in PPL, optically active with speckled b-fabric. The silt (~10%)
contains essentially quartz, with the addition of calcite and accessory heavy minerals, of which
hornblende, plagioclase, zircon, mica, and opaques were identied. A green glauconite sphere about
100m in size was seen in the matrix. Foraminifers are scarce, some being inlled with iron minerals.
285
Inclusions: Sparsely-spread sand (f:c ratio{0.062mm}=~98:2) predominantly rounded chalk and nari grains (up
to 500m), often with foraminifers, and related nari containing clay and some quartz silt. Subangular
to rounded quartz (up to 250m), sometimes with undulose extinction and mineral inclusions, is
common. Rounded microcline feldspar appears as an accessory in the same grain sizes.
Firing temperature: Unred, judging by the lack of alteration in colour in the glauconite sphere.
EA 229 (VAT 1689), from Abdina (?) to the King of Egypt
Reliability: High. It was possible to take a second sample from this letter fragment in order to enlarge the
sample size. This was done after the preliminary results were obtained.
Matrix: Reddish-tan in PPL, silty (~20%), non-carbonatic, ferruginous matrix exhibiting strong optical
orientation hence kaolinitic in nature. The silt includes predominantly quartz with accessory mica
and hornblende. Opaques are common.
Inclusions: Well-sorted sand (f:c ratio{0.062mm}=~90:10) in which rounded quartz grains (up to 350m),
sometimes with undulose extinction and mineral inclusions, are dominant. Accessory heavy
minerals that appear with the quartz are microcline (up to 250m), zircon (up to 120m), epidote
(up to 120m), and hornblende (up to 150m). There are a few subrounded to subangular calcite
crystals (up to 300m).
Firing temperature: Undetermined, but below 7500C since the calcite is unaffected.
Conclusions: This letter is petrographically identical to EA 64. It is therefore evident that Knudtzons
attribution of EA 229 to northern Canaan was wrong. The letter is badly broken (only part of the
introduction is preserved) and does not supply any textual clue for its provenance.
The name of the author is written Abdi-na [.]. It is unclear whether the name is complete, or its
last part broken. Moran (1992:290) restored line 3 Message of Abdina, [the ruler of ].2
CONCLUSION: THE GATH CORRESPONDENCE

The petrographic analysis connects Shuwardatu to the Shephelah, and conrms the identication of
his seat at Tel at (Tell e-a). It also supports Naamans suggestion that Abdi-Ashtarti, the sender
of EA 63, 65 and 335 was the heir of Shuwardatu on the throne of Gath. Two letters, one attributed to
Shuwardatu (EA 278) and the other to Abdi-Ashtarti his heir (EA 64) are not made of materials from Tel
ats immediate environment. They may have been sent from a town in the eastern territory of Gath,
such as Qiltu. Alternatively, the similarity between EA 64 and EA 229 may suggest that their authors
ruled in another city-state located east of Tel at.
2. The name of the author of EA 64 is written Abdi-dINNIN. The logogram INNIN stands in standard Akkadian for the
Goddess Ishtar. The equation INNIN=Ashtartu was established on the basis of the resemblance of the names Ishtar and
Ashtartu and on the assumption that EA 63 (whose author is called Abdi-Ashta(r)ti) and 64 were written by the same ruler.
If different rulers wrote the two letters (as may be inferred from the petrography), then the way is open for a different reading
of the Goddess name in EA 64. Hypothetically, one may read INNIN as >Anat and restore in EA 229 Abdi-na-[ti ], i.e.,
Abdi-(>A)nati (compare Grndahl 1967:20, 105, 375). However, there is no evidence to support the equation INNIN=>Anat
and there is no determinative before the assumed Goddess name in EA 229.
286
III. LACHISH (LAKISHA)

Three rulers of Lachish are mentioned in the Amarna archive: Zimreddi (EA 329), Yabni-Ilu (EA 328)
and Shipi-Ba>lu (EA 330-332). Lachish is identied with Tell ed-Dweir in the Shephelah. The Late
Bronze Age city-state controlled a densely settled territory in the southern Shephelah (Naaman 1988a;
Finkelstein 1996:232). Its easternmost towns were located in the fertile longitudinal valley at the foot
of the Hebron hills. Establishing its border in the south depends on whether the large Tell >Aiun and
possibly also Tell Beit Mirsim were centres of independent city-states (Naaman), or were secondary
towns within the territory of Lachish (Finkelstein). In the west Lachish bordered on Ashkelon and Yurza
and in the north on Gath.

An assemblage of unred pottery vessels from the Late Bronze Age potters workshop found at Lachish
(Tufnell 1958) was used as reference material (Magrill and Middleton 2004). It was kindly shown to us
in hand specimens in the British Museum by P. Magrill and in thin-sections by A. Middleton. Selected
wares from the Tel Aviv University excavations at Lachish, including plain wares, were also examined
(Goren and Halperin 2004).
The area of Tel Lachish is typied by dark brown grumusolic soils and brown rendzina (Dan et al.
1976; soil map of Qiryat Gath, 1:50,000). Lithologically the area of the site is unique since it is found in the
very restricted niche where the Oligocene Lachish Formation is exposed. This formation is subdivided into
two distinctive members. The lower (the Ramle Member) is composed of a conglomerate of chalk boulders
from the Beit Guvrin and Zorah Formations set in a marly-chalky matrix with abundant large foraminifers,
various molluscs and calcareous algae. The upper (the Gal<on Member) consists of hard biosparitic and
biomicritic limestone with the same fauna and bioclasts as in the Ramle Member (Buchbinder 1969:7).
CATALOGUE
EA 332 (VAT 1883), from Shipi-Ba>lu to the King of Egypt
Reliability: High.
Matrix: Silty (15%), carbonatic, tan in PPL, optically active with speckled b-fabric. The silt is
essentially of quartz but also contains recognizable quantities of accessory heavy minerals
including hornblende, zircon, augite, plagioclase, biotite, muscovite, epidote, and rarely
tourmaline. Opaques are relatively common (3%), ranging between a few micrometers and about
60m. In the matrix, clay concentrations (~2%), deep brown in PPL, appear as rounded fine
bodies between 10m to 1mm in size. Many of these spheres are surrounded by voids created by
the shrinkage of the clay that was higher than that of the matrix. Cracks appear within the clay
bodies for the same reasons. No silt appears within these spheres but they often contain very fine
opaque particles, a few micrometers in size.
Inclusions: Apart from the clay bodies, the inclusions include sparse sand (f:c ratio{0.062mm}=~95:5) of frequent
rounded chalk grains (up to 280m) including nari, spherical quartz (up to 220m) and a few, usually
rounded, fossilized mollusc shell fragments (up to 400m). There are also a very few burnt bone splinters
(up to 160m).
Vegetal material (SLY): A few (up to 500m long) uncharred plant tissues. Unidentied tissue fragments.
287
Firing temperature: No evidence for signicant ring is available. The vegetal matter shows no trace of
charring. The tablet may have been lightly heated to preserve its shape, but never red.
Geological interpretation: Petrographically, EA 332 is similar to the reference material from Tel Lachish,
namely the Late Bronze Age workshops unred vessels (Magrill and Middleton 2004). Most
signicant are the rounded mollusc shell fragments that have also been noticed (by Middleton) in
the Lachish workshop material.
Conclusions: A letter probably sent from Lachish.
EA 330 (BM 29848), from Shipi-Ba>lu to the King of Egypt
Matrix: Carbonatic, greyish-tan to tan in PPL, optically active with speckled b-fabric. The silt (10%) contains
essentially quartz, with the addition of calcite and accessory heavy minerals, of which hornblende,
Foraminifers are rather abundant (over 5%).
Inclusions: Sand with frequent spherical quartz grains (up to 400m) and common rounded chalk (up to
350m), including nari.
Vegetal material (SLY): A few (up to 280m long) uncharred plant tissues and a parenchymatous tissue
fragment.
Geological interpretation: Under the microscope the material of this tablet combines the properties of
loess (silty-carbonatic clay loam) and pale rendzina (many foraminifers and chalk particles). The
inclusions contain quartz sand and chalk particles. For the reasons that were explained above (EA
279), these materials suit the lithology of the southwestern Shephelah.
Reference: Similar to some of the examined pottery from Lachish (Goren and Halperin 2004).
Conclusions: The petrographic details are insufcient for assigning this tablet specically to Lachish.
However, a general southwestern Shephelah origin is conrmed, inferring that EA 330 was probably
sent from the Lachish territory.
EA 329 (VAT 1673), from Zimreddi (sic Moran 1992) to the King of Egypt
Reliability: Fair. Only tiny pieces were sampled because the tablet is complete.
Matrix: Silty (15%), carbonatic, tan in PPL, optically active with speckled b-fabric. The silt is essentially
of quartz, but it also contains recognizable quantities of accessory heavy minerals including
hornblende, zircon, augite, plagioclase, and biotite. Opaques are relatively common (3%), ranging
from a few micrometers to about 50m.
Inclusions: Sand of spherical quartz (up to 850m), rounded chalk (up to 350m), including nari and
mollusc fossil shell fragments (up to 260m).
Vegetal material (SLY): Up to 280m long. A badly preserved, non-indicative tissue fragment.
Geological interpretation: This tablet is dominated by quartz sand that is by far coarser than that of
EA 332 (below). Despite the small sample size, both surface examination under the stereoscopic
microscope and the petrographic analysis suggest that it contains coastal sand. Therefore, EA 329
may be another letter that was dispatched from the GazaAshkelon area.
288
Conclusions: This letter was probably sent from Gaza.
Discussion: EA 329 is identical in script and text to EA 321 (from Ashkelon), and the same scribe must
have written the two letters (Campbell 1965:113) in the same place. The similarity is clearly shown
in the facsimile published by Schroeder (1915:Pl. 180) who put them on one plate. The rulers of
Lachish and Ashkelon must have traveled to Gaza where they received verbal orders from the
Egyptian ofcial (EA 321:15-23; 329:13-20), and the Ashkelonite scribe wrote the two identical
letters there.
Reliability: High.
of quartz, though it also contains recognizable quantities of accessory heavy minerals, including
hornblende, zircon, augite, plagioclase, biotite, muscovite, epidote, and rarely tourmaline. Opaques
are relatively common (3%), ranging from a few micrometers to about 60m. In the matrix, clay
concentrations (~2%), deep brown in PPL, appear as rounded ne bodies, sizing between 10m to
1mm. Many of these spheres are surrounded by voids, created by the shrinkage of the clay body that
was higher than that of the matrix. Cracks appear within the clay bodies for the same reasons. No silt
appears within these spheres, but they often contain very ne opaque particles, a few micrometers
in size.
Inclusions: Apart from the clay bodies, the inclusions include sand (f:c ratio{0.062mm}=~90:10) of frequent
rounded chalk grains (up to 1.2 mm), including nari, and spherical quartz (up to 750m) is common.
There are a few mollusc shell fragments (up to 900m) containing some opaque iron minerals
(hence fossilized).
Vegetal material (SLY): A few (up to 500m long) uncharred plant tissues, cereal straw fragments and
unidentied fragments.
Geological interpretation: This letter is similar both petrographically and chemically to EA 332.
Conclusions: EA 311, which was previously unclassied, is most likely another Lachish tablet.
CONCLUSION: THE LACHISH CORRESPONDENCE

The petrographic analysis has conrmed the Lachish origin of EA 330 and 332, attributed EA 329 to the
group of letters sent from Gaza, and added a previously unclassied tablet (EA 311) to the Lachish group.
IV. A GROUP OF IDENTICAL LETTERS SENT BY

TWO RULERS FROM THE SHEPHELAH
EA 275-278 (see above Section II.) and EA 267, housed in the Cairo Museum and thus not examined,
form an assemblage of letters from cities in southern Canaan, all containing the same text. Because they
were identical in clay and script, Knudtzon (1915:1329 nn. 1-2) placed Yazib-Addas letters (EA 275276) and EA 277 whose writers name is broken before Shuwardatus rst letter (EA 278). Vita (2000)
289
even suggested that all letters of this group were written by the Gezer scribe at roughly the same time.
Indeed, it is clear that they were either written at about the same time in one place, or that a single scribe
traveled between several cities (see above, the Beqa> and Bashan cities, for similar cases). Petrographic
analysis is the ideal tool to decide this matter.
CATALOGUE
Reliability: High.
Matrix: Carbonatic, greyish-tan in PPL with speckled b-fabric, silty (7%). The silt contains mostly quartz
with accessory heavy minerals including hornblende, plagioclase, microcline, zircon, biotite, and
epidote. Opaque minerals are spread throughout the matrix (2%) reaching 60m in size.
Inclusions: Organic matter with some occasional sand (f:c ratio{0.062mm}=93:7) consisting of subrounded
to subangular quartz (up to 270m) and a single 960m grain of quartz and calcite sand in micritic
cement.
Palaeontology (LG): Badly preserved planctonic foraminifers in matrix. Identied: Acarinina (p),
Globorotalia(?) (p). Age: Neogene(?). However, the larger sample from EA 276 which is
petrographically identical to this tablet undoubtedly indicates Paleocene marl.
Vegetal material (SLY) and coprolites: Frequent (up to 2.4mm long) including straw from monocotyledon
plants, containing phytoliths, most likely Graminae = Poaceae. Another well-preserved botanical
structure has no xylem, thus is probably a freshwater alga. Several phosphorous bodies, including
spherulites (see EA 118 above for denition), indicate herbivore dung.
Firing temperature: Probably unred or very lightly red since the vegetal matter is uncharred.
Reliability: High.
Matrix: Carbonatic, greyish-tan in PPL with speckled b-fabric, silty (7%). The silt contains mostly quartz
with accessory heavy minerals including hornblende, plagioclase, microcline, zircon, biotite, and
epidote. Opaque minerals are spread throughout the matrix (2%) reaching 60m in size.
Inclusions: (f:c ratio{0.062mm}=). Rounded micritic limestone (up to 930m) and subrounded to subangular
quartz (up to 250m).
Vegetal material (SLY) and coprolites: Frequent (up to 1.4mm long) including straw from monocotyledon
plants, containing phytoliths, most likely Graminae = Poaceae. One tissue fragment is composed
almost entirely of bres. Some of the other tissue fragments are in an extremely degraded state of
preservation and it is impossible to evaluate their taxonomical position. Therefore it was probably
chewed. The latter contain occasional spherulites, indicating herbivore dung (see EA 118 above for
denition).
Palaeontology (LG): The identications were made on the petrographic thin section and on a washed
sample: Acarinina, Morozovella acuta, M. acqua, M. edgari, Spiroplectammina plummerae,
Tappanina selmenensis, Lenticulina, Gavelinella, Nodosaria. Age: Upper Paleocene.
Firing temperature: Probably unred or very lightly red since the vegetal matter is uncharred.
290
EA 277 (BM 29864), from Yazib-Adda (?) to the King of Egypt

Sampling method: Sampled only for elemental analysis.
Geological interpretation: The element concentrations of EA 275, 276 and 277 (Appendix, Table A1)
reveal great similarity and low standard deviations, indicating that the three tablets are made of
similar materials.
CONCLUSION: EA 275-278
Petrographically EA 278 is unique in the Shuwardatu correspondence. It was sent from the Shephelah,
but not from Tel ats immediate environment. The Taqiye marl, of which it was made, can be found in
the longitudinal valley which separates the Shephelah from the hill country (Buchbinder 1969: Sheets
13-12, 14-12; Sneh et al. 1998). The closest outcrops appear near Netiv Ha-Lamed Hei, some 12 km
east of Tel at. EA 278 may have been sent from one of the eastern towns of Gath, where the Taqiye
formation is exposed, such as Qiltu. Another possible interpretation, based on the similarity in text of
EA 278 with that of EA 267, is that EA 278 was written by the Gezer scribe (above). The Gezer area
features the Taqiye formation, though none of the Gezer tablets, which are all characterized by the
additional use of coastal sand, is similar to this one.
Petrographically, EA 275 and 276 are identical in clay and inclusions. EA 275-276 and 277 are
chemically identical. The similarity in material and text suggest that the three tablets could have been made
of the same lump of clay. While EA 278 (from Shuwardatu) is identical as to matrix and contains the same
inclusion types, it differs from the others in the proportions of the inclusions (especially the dung and straw)
which appear in it in inferior amounts and sizes. However, this may represent differences dictated by human
behaviour and it has no inference on the provenance of the four letters, which was seemingly similar.
Naaman and Goren suggest that all four letters were sent on one occasion from a location that was
outside the con nes of Gath. This could have been the capital city of Yazib-Adda, which must have
been located in the Shephelah, adjacent to Gath and presumably also to Gezer. If indeed the Gezer scribe
wrote all these letters (as Vita suggested), he must have traveled to Yazib-Addas capital and written
them there. The palaeontological and petrographic af nities of the tablets indicate that they were formed
of reworked marl of the Taqiye formation. In the Shephelah area, the latter outcrops as a narrow strip
bordering the western edge of the longitudinal valley separating the higher Shephelah and the Judaean
Highlands, between Lahav and the Gezer area. Only a few mounds along this valley can serve as
candidates for the city of Yazib-Adda: Tell Beit Mirsim, Tell >Aiun and Tel Beth-shemesh. The latter
site borders the territories of Gezer and Gath, whereas the territory of Lachish blocks the connection
between Gezer and Gath on the one hand and Tell Beit Mirsim and Tell >Aiun on the other. Naaman
and Goren contend that Beth-shemesh ts all the data presented above, and although no certainty can
be achieved, it must be considered the best candidate for the place of Yazib-Addas capital (for Bethshemesh see also EA 273).
In this case too (cf. EA 273-274), Finkelstein notes the relatively limited extent of the Late Bronze
Age remains found at Beth-shemesh. He would accept the Beth-shemesh solution (admitedly as
somewhat remote) only if the four letters were indeed sent from Yazib-Addas capital. However, there
are two more possiblilities: they were dispatched from a town on the eastern ank of the Gath territory,
such as Qiltu (see also EA 64), or else they were sent from Gezer. The latter is a less attractive option
because it would mean that the Gezer scribe chose a specic mixture for these four letters which was not
used for the other Gezer tablets. In any event the seat of Yazib-Adda remains uncertain.
291
V. ASHDOD
In the present study two previously unclassied letters have been related to Ashdod. This identication
relies on the petrographic data and the textual evidence in EA 296.
A city named Ashdod is not mentioned in the Amarna letters or in any other Late Bronze Age
Egyptian source. The identication of the Ashdadites of the Ugaritic texts as people from Ashdod was
dismissed by Naaman (1997:609-611) who suggested that the city of Ashdad was located elsewhere,
possibly in Cyprus. Naaman (1997:612-615) further proposed that a city called Tianna, which is
mentioned in three broken Amarna letters (EA 284:30-32; EA 298:28-29; EA 306:33-35) as being
located near the borders of Gezer, Gath and Ashkelon, and in Papyrus Petersburg, should be identied
at Tel Ashdod. According to this suggestion, admittedly hypothetical, the name of the Late Bronze Age
city (Tianna) was changed to Ashdod in the Iron Age. Finkelstein accepts this as a valid hypothesis,
although still preferring to keep the name Ashdod for the Late Bronze Age city of Tel Ashdod (without
the equation the Ashdadites of Ugarit=Ashdod) and to identify Tianna (as a secondary town, not as
a city-state) somewhere else. Recently Rainey (2003:193*-194*) collated tablets EA 284 and 298 and
dismissed the reading Tianna in both texts. In his opinion, Tianna is a ghost town. If this sugestion is
veried, Tianna must be omitted from the discussion.
Late Bronze Age remains were uncovered in different sectors of the upper mound at Tel Ashdod. The
city, which was not fortied, covered an area of ca. 7 hectares (Finkelstein and Singer-Avitz 2001:231-235).
CATALOGUE
EA 296 (BM 29840), from Yatiru to the King of Egypt
Reliability: High.
Matrix: Clayey, ferruginous, reddish-brown in thin-section, slightly silty (about 2%-5%), foraminiferous,
rich in small (usually below 10m) opaque bodies of iron minerals.
Inclusions: Moderately sorted sand of dense, well-sorted, rounded to subangular sand-sized quartz grains
(up to 400m) with the occasional addition of other heavy minerals, including epidote, zircon, rutile
and feldspar.
Palaeontology (LG): Abundant, badly preserved foraminifers in matrix. The original test of the
foraminifers is lacking, their lling is stained with some ferrous minerals. Identied: Acarinina (p),
Subbotina (p) sp. Age: Paleogene (?).
the ring temperature of this tablet.
Geological interpretation: In this tablet sand of the Israeli coastal plain is mixed with highly ferruginous
ne clay. The quartz sand is accompanied by grains of accessory minerals including epidote,
zircon, rutile and feldspars. These too hint at a coastal origin (see discussion in EA 97 above). In
this area, red to dark reddish-brown silts and sands with loams in soils appear as part of the Reovot
Formation. It is most likely that red soil of the central littoral area of Israel, locally termed hamra,
was used here, perhaps after some purication by dilution of the sand component. Hamra soil is
spread along the coastal plain of Israel from the Ashdod area in the south to the Carmel coastal
plain in the north (Dan et al. 1976). As seen with regard to EA 168, coastal sand of the classication
described here does not extend further to the north. In the Carmel coastal plain the hamra soil
292
contains carbonatic components that are contributed by the local lithology, a phenomenon which
is clearly represented in the pottery of Tel Dor. In this area the carbonatic component already
increases to compose about 50% of the sediment (Nir 1989: 12). Therefore, the place of origin of
this tablet should be sought in a more limited section of the coastal plain between Ashdod in the
south and the area of Caesarea in the north.
Reference: Hamra-made pottery is distributed in Israel in sites located mainly along the central coastal
plain. We refer to Chalcolithic assemblages from sites between Nizzanim and Maabarot (Goren
1991b) and to the Early Bronze Age assemblages of Palmahim (Goren 1991a: Appendix 2). In the
Middle Bronze Age this petrographic group dominates the ceramic assemblage of the Rishon LeZion cemetery (A. Cohen-Weinberger, pers. comm.). A Middle Bronze kiln site containing vessels
all belonging to this petrographic group was excavated near Yavneh-Yam (Singer-Avitz and Levy
1992, with appendix on petrographic analyses by Goren). The same group dominates a set of Middle
Bronze II workshop sites in the Tel-Aviv area and at Tel Michal (Kletter and Gorzalczany 2001).
Conclusions: Knudtzon placed this letter near the Gezer tablets noting that it is identical in script to EA
292-294 and in clay to EA 294 (1915:1346 n. 1). Yet our research indicates that there is a difference
in the clay between EA 296 and the Gezer tablets. The petrographic analysis indicates that EA 296
was sent from the central coastal plain of Israel, between Ashdod and Caesarea. Archaeologically,
only Ashdod could have functioned in this area as a Canaanite city-state (to differ from the
Egyptian administrative centre of Jaffa).
When expressing his loyalty to the Pharaoh, Yatiru says that he is guarding the city gate of
Azzatu (=Gaza) and the city gate of Yapu (=Jaffa) (EA 296:31-33). Hence it is only logical to locate
his city between these Egyptian centres. Moreover, in EA 294 (the other letter made of hamra
soil) the sender complains to the Pharaoh that the men whom he sent to serve in Yapu (Jaffa) were
seized by a certain Peya, possibly a leader of a band of >Apiru that stayed in the town of Muazu,
identied at Yavneh-Yam (Naaman 1997:613 n. 17, 615). His complaint seems to indicate that he
wrote this letter from his own city.
Another option which is viable petrographically, is that EA 296 was sent from the Egyptian centre
of Jaffa. Our study has shown that letters of Canaanite rulers were dispatched from at least four of
the six Egyptian administrative centres in Canaan (Gaza, Beth-shean, umur and Kumidi [Chapter
16]). However, this is less favourable textually because of the references to Gaza, Yapu and
Muazu. We are therefore inclined to place both Yatiru, the sender of EA 296, and the sender of
EA 294, at Ashdod.
EA 294 (BM 29854), from i-x-x-ni3 to the King of Egypt
Reliability: High.
Matrix: Clayey, ferruginous, reddish-brown PPL, slightly silty (about 2%-5%) with foraminifers, rich in
small (usually below 10m) opaque bodies of iron minerals.
Inclusions: Moderately sorted sand of dense well-sorted rounded to subangular sand-sized quartz grains
(up to 400m) with the occasional addition of other heavy minerals, including epidote, zircon, rutile
and feldspar.
3. Knudtzon and Moran (1992:336-337) deciphered the writers name as Ad[d]a-[d]anu. However, the rst sign is clearly i
and the last is ni (for the conicting readings see the literature cited by Moran 1992:335 n. 1). Hence, the author of EA 294
is most probably not the author of EA 292.
293
Palaeontology (LG): Badly preserved fauna, heavily recrystallized. Identied: Acarinina (p), Subbotina
(p) sp. Age: Paleogene (?).
the ring temperature of this tablet.
Geological interpretation: EA 294 is similar in all respects to EA 296.
Conclusions: Naaman (1997:615) suggested that i-x-x-ni of EA 294 and Yatiru of EA 296 might have
ruled in the same city (according to him at Tianna=Tel Ashdod). Since EA 294 is petrographically
similar to EA 296, and in light of our interpretation of the latter, we suggest that both tablets were
sent from Tel Ashdod.
VI. ASHKELON (AQALUNA)

Aqaluna was situated at present-day Tel Ashkelon in the southern coastal plain of Israel. The territory
of the Ashkelon city-state stretched along the coast. Its southern border must have passed to the north of
Gaza. Its northern border should be located in the area to the south of Ashdod. Ashkelons eastern border
(with Lachish and Gath) probably ran along the strip of coastal plain devoid of Late Bronze Age sites
about 10 km from the coast (see map in Finkelstein 1996:255).

Ashkelon is situated in an area dominated by Quaternary aeolian and marine coastal sediments. These
include shifting sand, alluvial clay and sand, aeolian silty-clay sediments and alluvial silt and clay (Dan
et al. 1976, 1:50,000 soil map, Ashkelon Sheet). These sediments, representing Quaternary marine
and coastal deposits, are generally included in the Pleshet Formation (Issar 1961), a term engulng a
series of calcareous sandstones transgressively deposited above the older clays of the Jaffa formation.
The sand component is made of subangular to rounded quartz, cemented by calcite. Corallinean algae,
molluscs and foraminifera are also common in several units. These can be divided into several classes:
the lowermost and oldest marine type rich in corallinean algae, the littoral type which is rich in sand and
mollusc shell fragments and referred to as beach rock and the continental eolianite (kurkar) composed of
calcite cemented sand (Nir 1970: 42-44).4
Petrographically the alluvial and aeolian silty clays are generally combined under the denition of
loess since they are all typied by a carbonatic clay matrix with high proportions of aeolian quartz silt
(10%-20%) with abundant accessory minerals. The latter include mostly hornblende, feldspar, zircon and
biotite, with the occasional occurrence of augite, garnet, rutile, tourmaline and epidote. The coastal sand
makes the most available temper in this area. This association, which is the sole combination of clay and
ecofact inclusions in the entire exploitable territory of Ashkelon makes this site a clear case of CTF=1.
A comprehensive petrographic study of the 7th century BCE pottery from Ashkelon has recently
been carried out by Master (2001). It is used here as reference for the local group. The common
petrographic group of the Middle BronzeLate Bronze Age pottery from the Ashkelon excavations
furnished additional reference material (A. Cohen-Weinberger, pers. comm.). The data from kiln wasters
from the immense Byzantine workshop excavated in the modern city by Israel (1993; 1995) was also
used. The latter also supplied the Ashkelon data for the ICP analyses of workshop wasters.
4. The depositional processes, subdivision and dating of these facies have recently been reconsidered (Sivan 1996; Sivan et al.
1999, see discussion on the ceramic ecology of Acco above), hence this general division is somewhat outdated. In any event,
this discussion is beyond the scope of our study.
294
CATALOGUE
is essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals
which include hornblende, augite, zircon, plagioclase, biotite, muscovite, rutile, epidote, and rarely
garnet. Opaques are relatively common (3%), sizing between a few micrometers and about 70m.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal matter (f:c ratio{0.062mm}=93:7)
predominantly rounded to subrounded quartz grains (up to 400m) sometimes with undulose
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions.
Vegetal material: Common (up to 400m long) uncharred plant tissues: cereal straw fragments including
cereal epidermis.
Geological interpretation: See EA 168 (Chapter 7.I) for detailed discussion on the provenance of the
loess with quartzitic coastal sand fabric. EA 321, as well as the other tablets of Yidia and Shubandu,
originated in the Gaza-Ashkelon area.
Reference: By its fabric, this letter is similar to the Ashkelon reference materials mentioned above.
Conclusions: This letter could have been considered as Ashkelon-made. However, considering the
similarity in script and text to EA 329 (Chapter 14.III), EA 321 must have been sent from the
Egyptian centre of Gaza. The petrofabrics of Gaza and Ashkelon are similar.
Sampling method: This tablet was sampled only for ICP analysis.
Conclusions: Cluster and principal component analyses place this tablet within the Ashkelon group.
Reliability: High.
Matrix: Silty (10%), carbonatic, dark reddish-tan in PPL, optically active but tends to become
isotropic in places, with very weak optical orientation. The silt is essentially of quartz, but it
contains recognizable quantities of accessory heavy minerals, which include hornblende, augite,
zircon, plagioclase, microcline, biotite, muscovite, rutile, epidote, and rarely garnet. Opaques
are relatively common (3%), sizing between a few micrometers and about 70m. Sharp-edged,
elongated voids indicate the presence of vegetal matter that vanished in this tablet due to its
relatively high firing temperature.
Inclusions: Rather coarse mixture of minerals and rock fragments (f:c ratio{0.062mm}=~95:5) in which
rounded to subrounded quartz (up to 380m) is dominant, sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. Rounded micritic limestone
(up to 640m), sometimes containing a few silty quartz grains, is common. Appearing as
rare accessories in the ne sand fraction are hornblende (up to 200m) partially altered into
oxyhornblende, biotite (120m long), plagioclase (250m), microcline (140m) and rutile (100m).
295
Firing temperature: The tablet was red at nearly 8000C as seen by the partial change of hornblende into
oxyhornblende and the slight isotropism of the matrix. However, the ring process was brief as the
limestone inclusions did not undergo any severe decalcication. The ring of a Canaanite tablet to
such temperatures is very uncommon in the Amarna archive (but see also EA 227 and EA 325).
Conclusions: While it is impossible to distinguish between Gaza and Ashkelon tablets petrographically. there
is no reason to attribute this letter to Gaza. It should thus be considered as an Ashkelon-made tablet.
Matrix: Silty (10%), carbonatic, dark reddish-tan in PPL, optically active but tends to become isotropic
in places, with very weak optical orientation. The silt is essentially of quartz, but it contains
recognizable quantities of accessory heavy minerals, which include hornblende (almost completely
altered into oxyhornblende), zircon, plagioclase and biotite. Opaques are relatively common (3%),
ranging between a few micrometers and about 70m. Sharp-edged, elongated voids indicate the
presence of vegetal matter that vanished in this tablet due to its relatively high ring temperature.
Inclusions: Rather coarse mixture of minerals and rock fragments (f:c ratio{0.062mm}=~95:5) in which
rounded to subrounded quartz (up to 380m) is dominant, sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. Appearing as rare accessories
in the ne sand fraction are zircon (120m long), plagioclase (100m), microcline (140m) and
rutile (100m). Very coarse rounded fragments (up to 3 mm) of aquatic mollusc shells, partly
decalcied by ring, are common.
Firing temperature: The tablet was red at nearly 8000C, as seen by the partial change of hornblende
into oxyhornblende, the slight isotropism of the matrix and the partial decalcication of the shell
fragments. The ring of a Canaanite tablet to such temperatures is very uncommon in the Amarna
archive (but see also EA 227 and EA 324).
Conclusions: Most likely an Ashkelon-made tablet.
Reliability: High.
Matrix: Silty (10%), carbonatic, yellowish-tan in PPL, optically active with speckled b-fabric. The silt is
essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals which
include hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote, and rarely tourmaline and
garnet. Opaques are relatively common (3%), ranging between a few micrometers and about 60m.
in which rounded to subrounded quartz (up to 430m) is dominant, sometimes with undulose
limestone (up to 1mm), sometimes containing a few silty quartz grains, is common. Appearing
as rare accessories in the ne sand fraction are hornblende (up to 150m), plagioclase (100m),
microcline (up to 350m) and rutile (100m). There are a few fresh (unfossilized) aquatic mollusc
shell fragments (up to 200m long).
296
Vegetal material (SLY): Common (up to 3 mm long) uncharred cereal straw fragments and unidentied
tissue fragments.
EA 302 (VAT 332), from Shubandu to the King of Egypt
essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals, which
include hornblende, augite, zircon, microcline, plagioclase, biotite, muscovite, epidote and tourmaline.
Opaques are relatively common (3%), ranging between a few micrometers and about 60m.
7) in which rounded to subrounded quartz (up to 470m) is dominant, sometimes with undulose
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions. Microcline appears
in the sand fraction (250m). Rounded micritic limestone (up to 800m), sometimes containing a
few silty (in one case sandy) quartz grains, is common.
Vegetal material: Common (up to 900m long) uncharred cereal straw and other cereal tissue fragments.
Conclusions: Naaman (1975:135-138) suggested that Shubandu was the ruler of Ashkelon before Yidia.
Indeed, the tablets of Yidia and Shubandu that we have examined are petrographically identical and
thus probably all were sent from the same location. Since the petrography of the Ashkelon and Gaza
letters is similar, theoretically Shubandu could have sent his letters from Gaza (and the same holds
true for Yidia). However, we have no example of a ruler who wrote all his letters in an Egyptian
centre. Thus, without ruling out the possibility that a letter or two of the Yidia-Shubandu le could
have been sent from Gaza, Naamans proposal is supported by the petrographic study.
Reliability: Fair.
include hornblende, augite, zircon, microcline, plagioclase, biotite, muscovite and epidote. Opaques
are relatively common (3%), ranging from a few micrometers to about 60m.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal matter comprising rounded to
subrounded quartz (up to 430m), sometimes with undulose extinction, or mineral and liquid inclusions
and rounded micritic limestone (up to 700m), sometimes containing a few silty quartz grains.
Vegetal material (SLY): Up to 650m long uncharred plant tissues. Several unidentied tissue fragments.
297

garnet. Opaques are relatively common (3%), between a few micrometers and about 60m in size.
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions. Microcline
appears in the sand fraction (250m). Rounded micritic limestone (up to 320m), sometimes
containing a few silty (in one case sandy) quartz grains, is common. Appearing as rare accessories
in the ne sand fraction are augite (up to 300m), plagioclase (100m) and microcline (up to 250m).
Fresh (unfossilized) aquatic mollusc shell fragments (up to 2 mm long) are common.
Vegetal material (SLY): Common (up to 350m long) uncharred cereal straw fragments.
Geological interpretation: As EA 321. The representation of heavy minerals in the sand fraction of
this sample is typical of coastal sand, and stems from the larger sample size as compared with
EA 302 and 303.
Reliability: High.
garnet. Opaques are relatively common (3%), between a few micrometers and about 60m in size.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal matter (f:c ratio{0.062mm}=93:7) in
which rounded to subrounded quartz (up to 620m) is dominant, sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. Appearing as rare accessories in
the ne sand fraction are augite (250m), and hornblende (up to 120m)
Vegetal material: Common (up to 400m long) partly lignied, uncharred cereal straw fragments
including several vessel members.
Geological interpretation: As EA 321. The representation of heavy minerals in the sand fraction of
this sample is typical of coastal sand, and stems from the larger sample size as compared with EA
302 and 303.
298
CONCLUSION: THE ASHKELON CORRESPONDENCE

One ruler of Ashkelon, Yidia, is explicitly mentioned as such in the Amarna correspondence. It has been
suggested that another ruler, Shubandu, was the king of Ashkelon before him. The petrographic analysis
seems to have con rmed this proposal. Yidia sent at least one letter (EA 321) from Gaza. Whether these
rulers sent a few of their other letters from Gaza cannot be established due to the similarity of material
between the Ashkelon and Gaza tablets.
VII. YURZA
Yurza was the place of origin of EA 315 and 316. Mazar (1952) suggested identifying it with Arza,
mentioned by Esarhaddon as a town on the border of the Brook of Egypt, and to locate both at Tell
Jemmeh on the western bank of Naal Besor (Wadi Ghazzeh), about 13 km south of Gaza (see also
Aharoni 1967; Naaman 1979b:72-73; Liverani 1998:67). With an estimated area of ca. 3 ha., Tell
Jemmeh is one of the prominent sites along Naal Besor, although compared to the major mounds of the
Shephelah (e.g. Lachish and Tell >Aiun) it is relatively small.
Assuming that Yurza is identied at Tell Jemmeh, it would be the only city-state known to have
been located in the southern coastal plain, south or southeast of Gaza. Yurzas southern border probably
corresponded to the boundary of sedentary activity facing the arid zone. Naal Besor (Wadi Ghazzeh)
should probably to be identied with the Brook of Egypt of the biblical and Assyrian sources, which
served as the southern border of the Egyptian province of Canaan (Naaman 1979b; 1986:239-243). In
the northwest Yurza bordered on the territory of the Egyptian centre of Gaza (and possibly the city-state
of Ashkelon), and in the northeast it bordered on the territory of Lachish (and the city-state(s) located at
Tell >Aiun and/or Tell Beit Mirsim [Naaman]).

The following were used as reference material for sites that have either been identied with Yurza or are
located near its borders (all having low CTF values):
1. For Tell Jemmeh wasters collected from a Late RomanEarly Byzantine kiln site found 600 m
south of the mound. The workshop has never been excavated, but the kilns and the waste dumps
are seen on the surface.
2. For Tel Haror a group of locally produced votive vessels found at a workshop in the Middle
Bronze Age sanctuary complex (Oren 1997:263-264, Fig. 8.12). In addition, pottery wasters
from a group of Byzantine workshops located at Naal Bohu (G.R. 0907 0927), about 5 km
northwest of the site, were used.
3. Tel Sera> (Tell esh-Shari>ah) wasters from a Byzantine workshop located 300 m. south of the
site (G.R. 1195 0886) and the common local Late Bronze petrographic group at the site.
Another site in the region for which we have reference material is Deir el-Bala. We used the
petrographic data on the Late Bronze Age workshop presented by Killebrew (1988) and examined similar
material collected by Beit-Arieh from other parts of the same workshop.
299
CATALOGUE
Reliability: High.
include hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote and tourmaline. Opaques
are relatively common (3%), ranging from a few micrometers to about 60m.
Inclusions: Moderately sorted sand (f:c ratio{0.062mm}=85:15) of spherical and usually rounded grains
extinction, sometimes polycrystalline, rarely with mineral and liquid inclusions. Limestone and
nari (up to 600m) are frequent including micrite (usually stained by clay). Appearing as rare
accessories in the ne sand fraction are hornblende (up to 140m) and zircon (110m).
Vegetal material (SLY): Uncharred plant tissues, cereal straw fragments and unidentied fragments.
Geological interpretation: For reasons explained above (see EA 168), the materials of this tablet
suggest a western Negev origin. The abundant nari grains seem to indicate a wadi channel that
drains an area with Mediterranean subhumid climate, dominated by chalk lithology. The nari
grains are mixed with coastal sands. The heavy minerals which appear with the sand (namely the
hornblende) were not subjected to severe chemical and mechanical weathering, a phenomenon
typical of inland sands (Slatkine and Pomerancblum 1958). Moreover, the quartz grains are
coarse, that is, they were not sorted by distant transportation processes. All these features point
to a location not far from the coast. They indicate a locality in the northwestern Negev, near
a wadi channel which drains the Shephelah region (where chalks and nari are exposed). The
best candidate is Tell Jemmeh, located in the northwestern Negev on Naal Besor. All other
sites mentioned above are found further inland, and thus their local wadi sands feature other
sedimentary characteristics (Gilead and Goren 1989:Fig. 2).
A large body of petrographic data makes the identication of the provenance of this tablet quite
easy. Pottery with similar clay (loess soil) and inclusions (wadi sand of quartz and calcareous rocks)
is known in the area of lower Naal Besor and Naal Gerar at sites from all periods: Neolithic,
Qatian sites of Naal Besor (Goren 1988; 1991c); Chalcolithic sites in Naal Gerar and Naal
Besor (Gilead and Goren 1989; 1995; Goren 1987; 1991a; 1995; Goren and Gilead 1987); En Besor
and Site H near Naal Besor both dated to the Early Bronze Age I (Porat 1989a); locally made
Middle Bronze Age II votive vessels from a workshop adjacent to the shrine at Tel Haror (Oren
1997:263-264, Fig. 8.12); Byzantine workshop wasters near Tel Sera> (on Naal Gerar), Tell Jemmeh
and Tell Far >ah South (on Naal Besor).
ICP analysis reveals that EA 315-316 are closer to the wasters of the Byzantine workshop near Tell
Jemmeh than to any other site in the region, including Tel Sera> and Tel Haror (Appendix:Tables 1-2).
Conclusions: EA 315 was sent from the lower Besor area. Of the two sites located in this region Tell
Jemmeh and Tell Far>ah the former was the more important in the 14th century BCE. The ICP study
of the wasters from the Byzantine workshop near Tell Jemmeh also points in the same direction. We
therefore support the identication of Yurza with Tell Jemmeh. .
300

is essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals,
which include hornblende, augite, zircon, plagioclase, biotite, epidote, and tourmaline. Opaques are
relatively common (3%), ranging from a few micrometers to about 80m.
Inclusions: Moderately sorted sand (f:c ratio{0.062mm}=85:15) of spherical and usually rounded grains
in which quartz (up to 540m) is dominant, sometimes with undulose extinction, sometimes
polycrystalline, rarely with mineral and liquid inclusions. Limestone and nari (up to 300m) are
frequent including micrite (usually stained by clay).
Vegetal material (SLY): Common (up to 1.5 mm long) uncharred plant tissues and several unidentied
tissue fragments.
CONCLUSION: THE YURZA CORRESPONDENCE

Both petrographic and chemical examinations point to Tell Jemmeh as the place of origin of EA 315 and
316. This location ts well the archaeological data from the site and the textual evidence on Yurza/Arza.
Yurza was the southernmost city-state in Canaan. It controlled the fringe area on both sides of Naal
Besor and Naal Gerar and possibly also the pastoral groups which lived further south.
Our petrographic results atly reject Raineys suggestion (1993) of identifying Yurza at Tell el-esi.
Another possible candidate for the location of Yurza is Tel Haror (Tell Abu Hureirah), located on the
main road which connected Gaza with the Beersheba Valley. Yet, the site is commonly identied with
biblical Gerar and its area in the 14th century BCE was quite insignicant (Oren 1993).
301
CHAPTER 15
UNIDENTIFIED CITIES IN CANAAN
I. AH$IRUNA
Knudtzon read the name of the city mentioned in EA 319 Ahtiashna/Ahtirumna (A-ti-a/rum-na) and
placed it among the south Canaanite letters. Aharoni (1967:148), Rainey (1978:105), Ahituv (1984:96)
and Liverani (1998:138) suggested deciphering the name as Ginti-ashna (gin8-ti-a-na) and identifying
it with kntw-sn (Gath-sn) mentioned in the topographic list of Thutmose III (no. 44). Naaman (1988a:
98, n. 19) dismissed this proposal on two grounds: (a) the rst sign is clearly a, and not gin8 (b) the city
mentioned by Thutmose III is located in the Jezreel Valley, whereas the vocabulary of EA 319 is typically
south Canaanite, similar to the letters of Gezer, Ashkelon and Lachish. This observation was accepted by
Moran (1992:350).
Naaman suggests rendering the city name Ahiruna (Ah--r-na) and derive its name from the verb
r (to surround) plus the Semitic sufx n. Thus it is derived from the same verb as biblical Ataroth
(>Art), crowns, which was probably allotted to places located on top of hills/mounds. Ataroth was
a common place-name in the rst millennium BCE and is known from Ephraims northeastern (Josh 16:
7) and southern (Josh 16:2, 5, 7; 18:13) borders, near Bethlehem (1 Chr 2:54) and in Moab (Num 32:3,
34, 35; lines 10-11 of Mesha inscription). Naaman suggests that the nameCanaanite in originwas
carried by the capital of a small city-state in the Amarna period.
CATALOGUE
EA 319 (VAT 1722), from Zur-Ashar to the King of Egypt
of quartz, but also contains recognizable quantities of accessory heavy minerals, which include
hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote, and rarely tourmaline and garnet.
Opaques are relatively common (3%), ranging from a few micrometers to about 60m.
Inclusions: The inclusions include predominantly dense, well-sorted spherical sand-sized quartz grains
(up to 750m), rarely polycrystalline, sometimes with undulose extinction and mineral inclusions
with the addition of straw and a few (up to 200m in the thin section but more in the tablet) sharpedged fragments of fresh (i.e., unfossilized) marine mollusc shells.
Vegetal material (SLY): Uncharred plant tissues and an unidentied tissue fragment (up to 200m in the
thin section but more in the tablet).
of charring. The tablet may have been lightly heated to preserve its shape, but never red.
Geological interpretation: This tablet was made of southern Canaan coastal sediments (the area between
Raphia and Ashkelon).
Reference: As EA 168 (Chapter 7).
302
Conclusions: There are three options for interpreting the petrographic data:
1. The letter was sent from the town of Ahiruna. In this case, this city should be identied in the
southern coastal plain, between Raphia and Ashkelon. This is hardly possible. There are only two
major Late Bronze Age sites in this area which are not securely identied. The rst is Deir el-Bala
(Dothan 1993), located south of Naal Besor which apparently served as the southern border of
the Egyptian province of Canaan. Hence it is questionable whether a Canaanite city-state could
have been located there. The second is Tell el->Ajjul. Apart from the fact that it was proposed as a
candidate for the location of Sharuhen (Kempinski 1974), it is not clear whether it was inhabited
in the Late Bronze Age IIA (Gonen 1981). In addition, the area under discussion is too limited and
sparsely settled to contain an additional Canaanite city-state besides Yurza and Ashkelon.
2. EA 319 was written in Gaza and should be included in the group of letters that Canaanite rulers
dispatched from the Egyptian centre there. In this case, Ahiruna was probably a small city-state in
southern Canaan, as indicated by the formulae and vocabulary used by the scribe.
3. The script and text of EA 319 is south Canaanite because the sender, who could have been of northern
origin, used the service (in Gaza) of a southern Canaanite scribe. This option is less favourable because
letters of north Canaanite rulers that were sent from Gaza (e.g. EA 317-318) were written in the
northern scribal tradition by scribes who escorted their rulers to Gaza.
Naaman supports the second alternative and considers the third unlikely.
II. LETTERS OF UNPROVENANCED CANAANITE RULERS

and less common calcite crystals between 10m and 70m in size. Very rare small (~60m)
glauconite concentrations appear within the matrix. The calcite crystals are commonly idiomorphic
or subidiomorphic in the ner fraction but the calcitic bodies tend to become rounded in the coarser.
Foraminifers are few (about 1%). Quartz silt appears at about 1%. Occasional zircon appears in the silt
fraction. Opaque to reddish-tan iron minerals (magnetite and haematite respectively, ~2%) appear in a
range of sizes from a few micrometers to about 30-40m. The opaques are angular and the translucent
Inclusions: Sparsely spread (f:c ratio{0.062mm}=90:10) sand in which grains of subrounded to subangular
quartz (up to 640m) sometimes containing mineral inclusions are common as are those of rounded
micritic or sparitic limestone (up to 800m). There are a few fragments (up to 1.2 mm) of alkaliolivine dolerite, showing alteration of the olivine into iddingsite, as well as fragments of corallinean
algae (up to 370m).
Vegetal material (SLY): Few (up to 850m), including a fragment of bark of wood including abundant
rhombic crystals of calcium oxalate. No further identication is possible.
Firing temperature: Unred judging by the preservation of uncharred vegetal material and the green
Conclusions: Although this letter is too fragmentary for interpretation it was denitely not sent from Byblos,
as suggested by Knudtzon who placed it within the Rib-Hadda correspondence. Connecting this letter,
303
whose script and text are identical to that of Rib-Haddas letters, with Irqata, the major city of the >Akkar
Plain, is also problematic (see EA 251). This issue remains open for further investigation.
EA 230 (BM 37646), from Yama to the King of Egypt
Matrix: Carbonatic, ochre to dark yellowish-tan in PPL, with speckled b-fabric. Silt is scarce (1%)
containing predominantly quartz with the addition of some epidote and hornblende. Opaques (~2%)
appear, ranging between 10m and 70m.
Inclusions: Sand and added organic materials (f:c ratio{0.062mm}=90:10) consisting of rounded quartz (up
to 400m) sometimes with mineral inclusions and micritic limestone (up to 240m). A 400m-long
hair, probably human, was observed.
Vegetal material (SLY): A fragment of a leaf including a vein with its typical xylem and wide zone of
parenchymatic tissue, probably a cereal leaf fragment.
Geological interpretation: Petrographically, EA 230 is not very informative. There is no diagnostic
component in it which can disclose its origin. However, the negative evidence is also important. It
is neither from Gaza, Beth-shean or Jaffa, nor from Kumidi, Ullasa or umur.
Conclusions: EA 230 is a unique tablet that raises many intriguing questions. The name of Yama appears
in the Amarna correspondence only here. The text says that Yama is guarding the kings rulers,
soldiers and cities. Naaman has suggested (1990:401) that Yama is a mistake for Maya, the name of
a well-known Egyptian commissioner (rabiu) in Canaan. This reading would explain the claim to
offer protection to the city-state. If Maya were the author of this letter, it would be the only message
of an Egyptian ofcial known to us which was written in cuneiform.
If this hypothesis is accepted, then EA 230 can be regarded as a report sent by the Egyptian
commissioner in Canaan from an unknown location outside the Egyptian administrative centres. It
could have been sent in the course of an inspection of the city-states (and their rulers) and of the
condition of the Egyptian guard units in Canaan.
EA 251 (BM 29862), from (?) to the King of Egypt
Matrix: Carbonatic, greyish-yellow tan to tan in PPL, with speckled b-fabric. Fine calcite crystals of
about 10m appear in the matrix. Haematite appears as small particles sizing 10-50m, or as stains
within the matrix. Foraminifers are common.
Inclusions: Sand and vegetal material (f:c ratio{0.062mm}=90:10) consisting of subangular to well-rounded
fragments (up to 350m) of rather weathered alkali basalt ranging between nely crystalline to
doleritic, rounded grains of sparitic, micritic and biogenetic limestone (up to 320m), idiomorphic
or subidiomorphic anhydrite (up to 240m) spherical. quartz grains (up to 220m), rounded
replacement chert (up to 280m) and idiomorphic clear calcite (up to 300m). Single crystals of
basalt-derived augite, olivine, and twinned plagioclase appear in grain-sizes reaching 250m.
Palaeontology (LG): Spinose planctonic foraminifers in the matrix and in the inclusions, also one
bentonic. Identied: Globigerinoides (p), Valvulineria (b) or Rosalina (b). Age: Neogene (?) as
suggested by the benthic foraminifers.
Vegetal material (SLY): Up to 600m long non-indicative tissue fragments.
304
Firing temperature: Unred or very lightly red since the vegetal material is uncharred.
Geological interpretation: EA 251 presents a mixture of presumably Neogene marl with sand that is
dominated by volcanic and basic intrusive rocks. Within the con nes of Canaan this combination
is restricted to a few areas. It is typical of the >Akkar Plain on the Syro-Lebanese coast. Another
possible area is the western part of the Damascus basin (Ponikarov 1964: Sheets I-36-XII, I-37-VII).
In addition, a very small exposure of Neogene marl (of the Bireh and Gesher Formations) occurs
adjacent to the basalt ows of Naal Tavor in the eastern Galilee (Aharon 1997) (cf. Chapter 12.V).
Reference: See EA 169 (Chapter 7).
Conclusions: EA 251 is the second in a two-tablet letter. While Knudtzon assigned it to a central
Canaanite ruler, we tend to reject this option for several reasons. If indeed this was the case, the
only possibility would be to assign it to Anaharath (Tel Rekhesh) and connect it with the Bayadi
Baduzana le (EA 237-239). However, the matrix of the latter is different (probably purer clay
rather than marl) and the textual traits (two-tablet letter) are unlike the other Bayadi Baduzana
letters. Alternatively, EA 251 could be assigned to the >Akkar area by its clay class and inclusion
assemblage that are comparable with EA 169, EA 100 and the like.
From the examination of the other Amurru texts it becomes clear that Aziru sent his letters from
Irqata only after he captured the city (i.e., after the dispatch of EA 100). However, EA 251 could not
have been written by Aziru as his letters are divided by lines which separate the text into passages,
whereas these dividing lines are not present in this letter. The text of EA 251 is alien to the Amurru
correspondence, though theoretically, it might have been sent by Aduna of Irqata (EA 75:25-26; 140:
10-11) before he was killed by the >Apiru/Abdi-Ashirta and the latter took over the city (EA 62:13,
17, 22). If this is the case, then the letter should be dated to the early stage of the Amarna archive.
Looking at Bezold and Budges facsimile of EA 251 (BB 79) it seems to have been written on
a tablet from which a previous letter was washed or erased. Two old lines on the edge and some
isolated signs on the reverse are clearly visible. This opens some interesting possibilities. Was the
tablet originally prepared and written in the >Akkar area, sent to another ruler who wiped the older
text and wrote his own letter, or did the local scribe in the >Akkar plain write a letter, erase it and
then write the present letter? This uncertainty still awaits further investigation.
EA 261 (BM 29858) from Dashru to the King of Egypt
Matrix: Carbonatic, very ne textured, pale greenish-grey to light tan in PPL with frequent foraminifers
(about 2%) and their fragments with very rare quartz silt. Rare heavy minerals include hornblende,
muscovite and zircon. Opaque minerals appear in the matrix (~1%) up to 60m in size but usually
below 30m. Limonitic stains appear in the matrix as dark reddish-tan to yellowish bodies with
dispersed edges, staining the matrix around them. On some occasions they also appear as inllings
of foraminifers. Fragments of charred organic matter appear in various sizes (from 10m and above)
besides the vegetal tissues to be described below.
Inclusions: Uncommon (matrix to inclusion ratio about 5%). Present are rounded and spherical grains
(up to 400m) of foraminiferous chalk (common), a few rounded to subrounded quartz grains (up
to 310m) and a very few (up to 100m) single grains of iddingsite and plagioclase. There are also
a few (up to 1 mm) densely packed coprolites containing druzes and remnants of compressed plant
tissues but no spherulites. This indicates herbivore dung. The ne texture of the coprolite indicates
a caprovine rather than a bovide or equid, providing that it belongs to a domestic animal. The lack
305
of spherulites is typical of goat rather than sheep (Brochier et al. 1992, Canti 1997, Goren 1999).
Non-indicative tissue fragments. The coprolite contains fragments of non-digested primary vessel
members with their typical spiral secondary cell walls.
Palaeontology (LG): Few small foraminifers: Hedbergella (p). Age: Upper Cretaceous (Senonian ?).
Vegetal material (SLY): Dominant, up to 350m long, containing: A) uncharred, partly lignied plant
tissues (straw), B) charred shrunken vegetal material.
Firing temperature: Unred since the organic material is uncharred.
Geological interpretation and conclusions: Taqiye marl not far from basalt exposures, originating.
probably from the Galilee. This combination may suit the environs of Shamuna (Tel Shimron) in
the northern Jezreel Valley. Since EA 224 is not local to Shamuna, and we were not able to examine
EA 225 (stored in the Cairo Museum), this hypothesis must await further investigation.
III. LETTERS OF UNPROVENANCED RULERS, SENT FROM GAZA

Based on their sign forms and text, Knudtzon placed EA 211-218, and 226 among the north Canaanite
letters. He must have noticed the similarity in clay of the former group and listed them consecutively (see
Knudtzon 1915:1297 n. 1; 1298 n. 2). However their petrographic traits indicate that they were composed
in southern Canaan on the coast, making Gaza the only possible place of origin.
EA 317-318 of Dagan-takala were placed by Knudtzon in southern Canaan, most likely due to their
similarity in clay to the Ashkelon tablets. While Artzi (1968) demonstrated that they are of northern origin,
our petrographic examinations indicate that they were most likely dispatched from Gaza.
Another group of tablets with the same petrographic characteristics is EA 307-312. These are fragmented
tablets that were placed by Knudtzon between the Ashkelon and Yurza letters, probably because of their
similarity in clay to the Ashkelon assemblage. Although theoretically some of these tablets could have been
sent by the rulers of Ashkelon, we prefer to combine them with the Gaza group of letters.
Reliability: Fairly reliable but sufcient for condent determination of the matrix class and comparable
with EA 212-213. Since only tiny crumbs were taken from this tablet, the inclusion sizes do not
indicate their full dimensions.
of quartz, but it also contains minor quantities of accessory heavy minerals including hornblende,
epidote, augite, zircon, plagioclase, biotite and muscovite. Opaques are relatively common (3%)
ranging from a few micrometers and about 60m.
Inclusions: Rather coarse mixture of minerals, rock fragments and vegetal material (f:c ratio{0.062mm}=93:7)
dominated by rounded to subrounded quartz (up to 370m), sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of plagioclase feldspar
appears in the same grain size. Fresh (unfossilized) aquatic mollusc shell fragments (up to 270m
long) are common.
Vegetal material (SLY): Uncharred plant tissues (straw) up to 200m in this sample.
Geological interpretation: As explained in detail in the case of EA 168, this tablet was made of south
Canaan coastal plain sediments, in the area between Raphia and Ashkelon.
306
Conclusions: Zitriyara wrote three letters (EA 211-213) that provide very little information about his
location and affairs. In the light of their script and text, which resemble other north Canaanite
letters, the only possible interpretation of their southern provenance is that he appeared before the
Egyptian ofcials in the administrative centre at Gaza and wrote his letters from there.
Reliability: Fairly reliable but sufcient for condent determination of the matrix class and comparable
with EA 211 and 213. Since this tablet is complete, only tiny bits were taken. Therefore, the
inclusion sizes do not indicate their full size-range.
of quartz, but it also contains signicant quantities of accessory heavy minerals including hornblende,
epidote, augite, zircon, plagioclase, biotite and muscovite. Opaques are relatively common (3%) ranging
from few micrometers to about 60m.
dominated by quartz sand with accessory minerals (up to 280m). The quartz grains are rounded to
subrounded, sometimes with undulose extinction, sometimes polycrystalline, rarely with mineral and
liquid inclusions. Grains of microcline and plagioclase appear in the same grain size. Hornblende
appears in the ner sand fraction (up to 140m). Mollusc shell fragments, visible in the stereoscopic
examination, were not included in the sample.
Vegetal material: Straw was visible in the stereoscopic examination but is not included in the sample.
Reference: As EA 168 (Chapter 7.I).
EA 213 (BM 29859), from Zitriyara to the King of Egypt
Reliability: High.
of quartz, but it also contains signicant quantities of accessory heavy minerals including hornblende,
epidote, augite, zircon, plagioclase, biotite and muscovite. Opaques are relatively common (3%) ranging
from a few micrometers and about 60m.
dominated by quartz sand with accessory minerals (up to 830m). The quartz grains are rounded to
subrounded, sometimes with undulose extinction, sometimes polycrystalline, rarely with mineral
and liquid inclusions. One 830m quartzite grain and one of augite (180m) appears in the same
fraction. Fresh (unfossilized) aquatic mollusc shell fragments (up to 400m long) are common.
Calcareous quartz sandstone (up to 500m) (kurkar, see EA 97 - Chapter 9.II) and a few fragments
of rounded micritic limestone (up to 320m) are present.
Vegetal material (SLY): A long fragment of non-woody but brous material, probably a segment of a
culm of a cereal. Several segments from cereal leaves and several fragments which include bres
and primary xylem with typical spiral secondary cell wall thickening.
307
Reference: As EA 168 (Chapter 7).

EA 215 (BM 29843), from Bayawa to the King of Egypt
augite, zircon, epidote, plagioclase, biotite and muscovite. Opaques are relatively common (3%)
ranging from few micrometers to about 60m.
7) dominated by rounded to subrounded quartz (up to 450m), sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of plagioclase feldspar
appears in the same grain size. Mollusc shell fragments, visible in the stereoscopic examination,
were not included in the sample.
Vegetal material: Straw was visible in the stereoscopic examination but is not included in the sample.
Conclusions: The correspondence of Bayawa includes two letters (EA 215-216) that belong to two
different periods. The rst mentions Yanamu as Egypts commissioner in Canaan and the second
refers to Maya. The two tablets also differ in their script, suggesting that between the days of
Yanamu and Maya a new scribe was employed (Moran 1992:284). The script and text of Bayawas
tablets resemble north Canaanite letters, but EA 215 was produced of southern Canaan coastal clay.
Therefore, it was most likely sent from Gaza. We could not check the other letter of Bayawa (EA
216) which is stored in the Cairo Museum.
EA 226 (VAT 1610), from Shipu-ria to the King of Egypt
Reliability: High.
tourmaline, augite, zircon, epidote, plagioclase, biotite and muscovite. Opaques are relatively
common (3%) ranging from a few micrometers and about 60m. Some foraminifers are sparsely
spread in the matrix.
dominated by rounded to subangular quartz (up to 400m), sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. There are a few grains of rounded
micritic limestone (up to 200m) and a very few of rounded replacement chert (up to 280m).
Vegetal material (SLY): Voids of vegetal material consumed during ring, leaving phytoliths in some cases.
Firing temperature: Probably lightly red (5000C) judging by the burning of the vegetal material but no
apparent colour change in the hornblende.
Geological interpretation and conclusions: As EA 211. Knudtzon assigned this ruler to northern Canaan.
The letter was most likely sent from Gaza.
308
EA 317 (VAT 1676), from Dagan-takala to the King of Egypt

Reliability: High.
of quartz, but it also contains recognizable quantities of accessory heavy minerals including
hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote, and rarely rutile. Opaques are
relatively common (3%) ranging from a few micrometers to about 60m.
7) dominated by rounded to subangular quartz (up to 680m), sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of microcline (250m)
and a grain of epidote (160m) appear in this fraction. A few (up to 400m long), fresh (unfossilized)
aquatic mollusc shell fragments occur.
Vegetal material (SLY): Very few uncharred plant tissues, single minute non-characteristic tissue fragment.
Firing temperature: Probably unred or very lightly red as the vegetal material shows no trace of
Conclusions: The petrographic analysis of EA 317-318 indicates that they were written in the Gaza area.
This would explain why Knudtzon placed them among the south Canaanite letters. Dagan-takala and
his scribe must have appeared before the Egyptian authorities at Gaza, and on that occasion sent the
two letters, written by the hand of a south Syrian scribe (for the north Canaanite provenance of the
scribe, see Artzi 1968).
EA 318 (BM 29857), from Dagan-takala to the King of Egypt
Reliability: High.
Matrix: Silty (10%), carbonatic, tan in PPL, optically active with speckled b-fabric. The silt is essentially of
quartz, but it also contains recognizable quantities of accessory heavy minerals including hornblende,
augite, zircon, plagioclase, biotite, muscovite, epidote, and rarely rutile. Opaques are relatively
common (3%) ranging from a few micrometers and about 60m.
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of hornblende
(130m) and a grain of augite (140m) appear in this fraction.
Vegetal material: Common (up to 1.5 mm long) uncharred cereal straw fragments, a probable root tissue
fragment and other non-characteristic tissue fragments.
Reliability: High.
309
Matrix: Stereoscopic examination in the Vorderasiatisches Museum suggested that it is made of loess.
Under the microscope the matrix is very silty (20%), carbonatic, dark tan in PPL, optically active
with speckled b-fabric. The silt is essentially of quartz, but it also contains recognizable quantities
of accessory heavy minerals including hornblende, augite, zircon, rutile, plagioclase, biotite,
muscovite, epidote, and tourmaline. Opaques are relatively common (5%) ranging from few
micrometers to about 60m.
Inclusions: No inclusions are found in this tablet but a few of the quartz grains reach the ne sand
fraction. A small fresh (unfossilized) mollusc shell fragment also appears.
the ring temperature.
Geological interpretation: This tablet is identied as being made of loess soil. Therefore, it originated
from southern Canaan, most likely the northern Negev or the Negev coastal plain. Further
identication is impossible due to the lack of any coarse fraction.
Conclusions: Knudtzon attributed this fragmented tablet to northern Canaan. However, its petrography
denitely denes it as a product of southern Canaan.
Reliability: High.
including hornblende, augite, zircon, plagioclase, microcline, biotite, muscovite, and epidote.
Opaques are relatively common (3%) ranging from a few micrometers and about 60m. A grain of
hornblende on the process of altering into oxyhornblende appears at this fraction, sizing 90m.
dominated by rounded to subangular quartz (up to 410m), sometimes with undulose extinction,
sometimes polycrystalline, rarely with mineral and liquid inclusions. A few (up to 240m long), fresh
(unfossilized) aquatic mollusc shell fragments occur.
Firing temperature: Probably red at around 7000C or slightly above judging by the partial alteration of
hornblende into oxyhornblende.
Reliability: High.
Matrix: Silty (10%), carbonatic, greyish-tan in PPL, optically active with speckled b-fabric. The silt
including hornblende, oxyhornblende, augite, zircon, plagioclase, biotite, muscovite, and epidote.
Opaques are relatively common (3%) ranging from few micrometers to about 60m.
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of microcline
(210m), a grain of hornblende (130m) and a grain of zircon (150m) appear in this fraction.
310
Vegetal material (SLY): Many fragments in a bad state of preservation, some of which include primary xylem
with its typical spiral secondary cell wall thickening. One fragment of parenchyma tissue is present.
Firing temperature: Unred or very slightly red since the vegetal material is uncharred.
Matrix: Silty (10%), carbonatic, light-tan in PPL, optically active with speckled b-fabric. The silt is essentially
of quartz, but it also contains recognizable quantities of accessory heavy minerals including hornblende,
titanaugite, zircon, plagioclase, biotite, muscovite, epidote, and rarely tourmaline. Opaques are relatively
common (~2%) ranging from few micrometers to about 60m.
(up to 600m) is common.
Vegetal material: Common (up to 1.5 mm lengthwise) various uncharred plant tissue fragments including
one which is probably a leaf fragment.
Geological interpretation: Like EA 168, this tablet was made of sediments from the Gaza-Ashkelon area.
Conclusions: A letter from Gaza or Ashkelon. Knudtzon placed it correctly, probably because of its
typical fabric, near the Ashkelon letters.
Reliability: High to satisfactory.
including hornblende, augite, zircon, plagioclase, biotite, muscovite, and epidote. Opaques appear
(~2%) ranging from few micrometers to about 40m.
Inclusions: A mixture of minerals, rock fragments and vegetal material (f:c ratio{0.062mm}=93:7) dominated
by rounded to subangular quartz (up to 340m), sometimes with undulose extinction, sometimes
polycrystalline, rarely with mineral and liquid inclusions. A grain of microcline appears in the same
grain size. Rounded micritic limestone (up to 490m) is common.
Vegetal material (SLY): Common (up to 1.5 mm long) uncharred non-characteristic tissue fragments.
Geological interpretation: This tablet was made of sediments from the Gaza-Ashkelon area.
Reference: EA 168.
Reliability: High.
311
including hornblende, augite, zircon, rutile (80m in size), plagioclase, biotite, muscovite, epidote,
and rarely tourmaline and garnet. Opaques are relatively common (3%) ranging from a few
sometimes polycrystalline, rarely with mineral and liquid inclusions. A grain of microcline (220m)
appears. A few grains of rounded micritic limestone (up to 350m) and fresh (unfossilized) aquatic
mollusc shell fragments (up to 300m long) occur.
Vegetal material (SLY): Common (up to 1.5 mm long) uncharred non-characteristic plant tissue fragments
and one cereal straw fragment.
Reference: EA 168.
Reliability: High.
Matrix: Silty (10%), carbonatic, light-tan in PPL, optically active with speckled b-fabric. The silt is
essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals
including hornblende, augite, zircon, plagioclase, biotite, muscovite, and epidote. Opaques are
(up to 350m) is common.
Vegetal material (SLY): A few (up to 1.5 mm long) uncharred non-characteristic plant tissue fragments
and a few cereal straw fragments.
Geological interpretation: This tablet was made of sediments from the Gaza-Ashkelon area.
Reference: EA 168.
EA 312 (VAT 1886 + 1709), from (?) to the King of Egypt (?)
essentially of quartz, but it also contains recognizable quantities of accessory heavy minerals including
hornblende, augite, zircon, plagioclase, biotite, muscovite, epidote, and oxyhornblende. Opaques are
Inclusions: A mixture of minerals, rock fragments and vegetal material (f:c ratio{0.062mm}=95:5) dominated
by rounded to subangular quartz (up to 360m), sometimes with undulose extinction, sometimes
polycrystalline, rarely with mineral and liquid inclusions. A grain of plagioclase feldspar appears
312
in the same grain size. Fresh (unfossilized) aquatic mollusc shell fragments (up to 1.7mm long) are
common and there are a few fragments (up to 320m) of rounded micritic limestone
Vegetal material (SLY): Common (up to 700m long) uncharred cereal straw fragments and unidentied
tissue fragments.
IV. SMALL FRAGMENTS OF LETTERS OF OTHER PROVENANCES

Reliability: High.
Matrix: Silty (10%), tan in PPL, optically active with speckled b-fabric. The silt is essentially of quartz,
but it also contains minor quantities of accessory heavy minerals including zircon, plagioclase,
biotite, muscovite, and epidote. Opaques are relatively common (3%) ranging from a few
Inclusions: Sand (f:c ratio{0.062mm}=93:7) of the following dominated by rounded to subangular quartz (up
to 160m), sometimes with undulose extinction, rarely polycrystalline or with mineral and liquid
inclusions. Rounded sparitic (larger grain sizes) or micritic (smaller fraction) limestone (up to 1 mm)
and foraminiferous chalk (up to 260m) are common.
Firing temperature: Probably very lightly red since there is no colour change in the hornblende or calcite.
Geological interpretation: Terra rossa with wind-blown quartz sand and local limestone and chalk. See
EA 64.
Conclusions: This tablet, which mentions Zitriyara (line 11), is possibly similar to EA 64. If so, an upper
Shephelah provenance may be suggested.
EA 240 (VAT 2198 + 2707)
Matrix: Silty (7%-10%), tan in PPL, optically active with speckled b-fabric. The silt is essentially of
quartz, but it also contains minor quantities of accessory heavy minerals including hornblende,
zircon, plagioclase, biotite, and muscovite. Opaques are relatively common (3%) ranging from a
few micrometers to about 60m.
Inclusions: Sand (f:c ratio{0.062mm}=95:5) in which rounded grains of chalk (up to 650m), including nari,
and subrounded to subangular quartz (up to 150m) are common.
Vegetal material (SLY): Common (up to 500m long), a bundle of bres, a group of parenchyma cells,
unidentied.
Geological interpretation, reference and conclusions: Terra rossa with wind blown quartz sand and local
limestone and chalk. See EA 64.
313

Reliability: High.
Matrix: Dense, orange to reddish-tan in PPL, carbonatic and devoid of any foraminifers. It contains haematite
particles (1%) and opaques (1%), all sizing up to 30m. The matrix is optically active and oriented with
striated b-fabric. Fine (silty) dolomite rhombs are sparsely spread in the matrix.
Inclusions: In this tablet the aplastic constituents include only clear pure crystals (f:c ratio {0.062mm}=75:25).
The crystals are clear, exhibit typical twinning and zoning features and split along their cleavage planes,
indicating that the craftsman intentionally crushed them prior to their mixture in the paste.
Firing temperature: Probably unred or very lightly red since there is no sign of alteration in the
dolomite crystals (occurring at above 5000C).
Geological interpretation: The clay of this tablet is undoubtedly taken from the Moza formation (cf.
EA 286 - Chapter 13.II). Hence this tablet demonstrates a mixture of crushed calcite inclusions
with Moza clay. This combination is rather surprising since it reects intentional and unexpected
investment in the selection and preparation of the non-plastic component. The use of such inclusions
is very well-known from ceramic vessels but is unique among the clay cuneiform tablets that we
have examined so far.
The earliest cases known till now for the mixture of crushed calcite as temper in pliable materials
have been recorded in the case of PPNB plaster objects such as the Naal Hemar beads (Goren et al.
1993). In that case the mixture of crushed calcite in the matrix did not have any apparent technical
benet over the use of any other calcareous binder. The process of mining the mineral, usually
from naturally-occurring veins in limestones, carefully crushing it and mixing it in the matrix
was probably more elaborate and time-consuming than the use of the more available wadi sand
commonly used in antiquity as temper.
In later ceramics the use of calcite as temper is known to increase the thermal-shock resistance of
the clay body and reduce its porosity. This is due to the relatively similar expansion rates of calcite
and clay (Arnold 1985:24). Pure calcite is superior to limestone because it decarbonates at higher
temperatures and thus is more resistant to heating. For these reasons crushed calcite is very common
as tempering material of cooking-pots and holemouth jars from the Early Bronze Age I to the end of
the Iron Age, when it is replaced by other tempers.
However, in the case of the tablet in question, these technical properties of calcite temper seem
to be totally irrelevant since the product was not intended to withstand repeated heating and cooling.
We suggest that the scribe who had formed this letter took some ready-mixed clay with inclusions
from the lump of the local potter who had prepared it for the production of cooking vessels.
Reference: For the use of Moza clay for ceramic production, see EA 252 (Chapter 13.I). The mixture
of this clay with crushed calcite is known from many instances, all restricted to the Central Hill
Country of Israel.
Conclusions: The clay of this tablet clearly assigns it to the central hill country of Palestine. Unfortunately
the text of EA 219 is so badly broken that we cannot attribute it with certainty to the correspondence
of either Shechem or Jerusalem.
314

Reliability: High.
Matrix: Carbonatic, Silty (7%), yellowish to yellowish-grey in PPL, optically active with speckled bfabric. The silt is essentially of quartz, but it also contains minor quantities of accessory heavy
minerals including zircon, sphene, biotite, muscovite, and epidote. Opaques are relatively common
(3%) ranging from a few micrometers to about 60m.
Inclusions: Sand (f:c ratio{0.062mm}=95:5) of the following dominated by subrounded to subangular quartz
(up to 170m), sometimes with undulose extinction, rarely polycrystalline or with mineral and
liquid inclusions. Rounded fragments of micritic limestone and separate fossils (up to 650m) of
articulated fragments of the calcareous corallinean algae Amphiroa, together with some mollusc
shell fragments are frequent. Subangular smoky to brown stained replacement chert (up to 1mm)
with local intergrowth of chalcedony is common.
Vegetal material (SLY): A few uncharred tissues (up to 350m long).
Geological interpretation: As EA 72 from Byblos (Chapter 9.I).
Conclusions: Most likely a fragment of a letter from Byblos.
EA 236 (Ash. 1893 1-41: 423) from (?)
Reliability: High.
Matrix: Highly carbonatic, very ne textured, pale greenish-grey in PPL with a few foraminifers and
some quartz silt (below 1%). Opaques are extremely rare. A few green glauconitic spheres appear
in the matrix (up to 100m). Sharp-edged, elongated voids that appear in the matrix indicate the
presence of some brous organic matter that vanished, besides the vegetal tissues to be described
below.
Inclusions: Uncommon sand grains (f:c ratio{0.062mm}=95:5) dominated by subrounded to subangular
quartz (up to 720m), sometimes with mineral inclusions. Subrounded to rounded grains (up to up
to 320m) of micritic limestone are frequent.
Firing temperature: Probably unred or very slightly red judging by the lack of colour change in the
glauconite.
Geological interpretation and conclusions: The matrix is probably Paleocene marl. However, there are
not enough petrographic data to suggest an origin for this small fragment.
315
CHAPTER 16
CONCLUSION
CONTRIBUTIONS AND LIMITATIONS OF THE ANALYTICAL METHODS
The present study launches a new analytical tool for the investigation of Near Eastern cuneiform
tablets. While the two pioneering NAA studies by Artzy et al. (1976) and Dobel et al. (1977)
introduced archaeometry into cuneiform studies, their limited scope and debated results brought about
the abandonment of such methods in this context for over two decades. We believe that the current
study opens new possibilities for future research. Yet, scholars should be also aware of the limitations
and shortcomings of this new methodology.
Cuneiform tablets can be technically dened as ceramic products but they should never be
automatically equated with pottery (Chapter 2). In other words, they cannot be subjected uncritically
to the methodology and concepts of pottery provenance studies in archaeology. First and foremost, in
contrast to most pottery vessels, clay letters usually bear a message that supplies additional data as to
their origin and addressee. Regardless, during the rst stage of the examination of each tablet which
carries such additional information, we attempted to estimate what our interpretations for each tablet
would be if we were unaware of their textual evidence.
We must emphasize, however, that pottery examinations too are never carried out blindly. Proper
archaeometric and petrographic laboratories never investigate vessels that have no archaeological,
chronological or typological context. In other words, the background data usually limits greatly the
possibilities for the conclusions. The interpretation relies on the results of the material analysis, but it
takes into consideration the archaeological and historical data. In a few cases this may lead the scholar
to a specic site where the vessel could have been made. However, usually results from analytical
examinations of pottery limit themselves to the area and seldom attempt to identify the specic site
where a vessel had been produced. Since in most historical periods certain types of pottery were massproduced, the examination of a large number of similar vessels can accumulate data to such a degree
that they eventually hint at the exact location of the workshop(s). This can be demonstrated by the study
of the Early Bronze Age II-III Metallic Ware (Greenberg and Porat 1996), to list only one of many
examples. In this sense, the examination of pottery is, in a way, more objective than the interpretation of
texts, which sometimes involves preconceptions and assumptions and is therefore a subjective analysis.
The text of a tablet supplies an exclusive piece of evidence. Even within an assemblage of documents
composed by the same individual, each tablet should be treated as a unique artefact, created under very
special and distinctive circumstances. The name of the sender and his domicile alone cannot be used as
evidence for the provenance of a document, in the same way as nowadays a writer may use his personal
letter-head even when sending a letter from abroad. At the same time, identifying the place where a letter
was composed, regardless of the identity of the sender and his residence, may supply a new and different
narrative for the text. Our study supplies many examples of this kind, e.g. the Amurru correspondence
and the letters sent from Egyptian administrative centres in Canaan.
The complexity of the study of tablets written by a well-identied ruler is best illustrated by the
rich correspondence of Rib-Hadda of Gubla. The fty analyzed letters of Rib-Hadda (out of nearly 300
tablets examined here) form a large assemblage that is statistically signicant. Petrographically, the
316
Rib-Hadda letters form four different groups. The rst two (Neogene coastal marl [NGM] and Neogene
coastal clay [NGC]) represent the local clays at Byblos, while the other two (Pliocene Pleistocene
coastal clay [QCC] and Miocene marl [MCM]) are alien to the environs of this city (Table 9). In terms of
distribution, the local groups (NGM and NCG) form 38% and 52% of the assemblage respectively. The
alien letters (10% of the assemblage) were sent by Rib-Hadda from the Egyptian administrative centre at
umur (6%) and from his exile at Beirut (4%). If the case of Rib-Hadda indeed represents the Canaanite
correspondence, it implies that there is 90% chance that a rulers letter was sent from his hometown.
Indeed, this is more or less the case with other Canaanite rulers too.
Had we not known from the text that the tablets were made at Byblos, the petrographic evidence
would direct us as follows: The micropalaeontological evidence would indicate a Neogene age for both the
NGM and NGC groups, showing that they are two related clay sources of somewhat different carbonate
contents. The presence of Amphiroa sp. alga fossils within the inclusions would point at a recent coastal
environment. This combination would denitely direct us into the coast of Lebanon between Sidon and
the >Akkar plain (see EA 165 and 72 for further details). Without further typological or archaeological
information these would be the limits of detection. As for the alien letters, petrography managed to
trace their exclusive composition. While in the case of the letters that Rib-Hadda sent from Beirut their
origin is explicitly stated in the text, in the case of the letters which he sent from umur only one text (EA
103) out of three gives evidence for its origin.
Other examples may further illustrate the issue of origin. Without the textual evidence, the letters of
Lab<ayu and Abdi-eba would be set by petrography in the Central Hill Country of Israel, but with no
specic location within this vast area (indeed, this is the case of EA 219). In fact, they could be assigned
to the same place of origin unless trace element analysis (by NAA or ICP) would have told us that they
form two separate clusters. On the other hand, without the petrographic study we would never know that
EA 285 was written by Abdi-eba in Beth-shean and EA 291 in Gezer.
In contrast, a study of the clay of Egyptian letters, ignoring their text, would fail to indicate an
Egyptian provenance for most of them since they were made of a clay type that was never used in Egypt
for pottery production. This is complicated by the fact that equivalent beds exist in the Levant, where
they have been exploited for pottery production since earliest times. Petrographic analysis of EA 1, 14,
370 and the like, would therefore suggest a Levantine provenance. In this case it is only the combination
between the text and the clay composition that reveals the complete story. However, an important
conclusion of our study is that such cases are limited and that in most instances easy-to-identify local
clays were utilized for the production of tablets.
In conclusion, we should treat the text and clay as two corresponding sources of information that
should always be conjoined. It goes without saying that the textual evidence supplies numerous other
aspects of information and therefore we do not intend to equate the two disciplines by their contribution
to the study of documents. All we want to say is that the contribution of archaeological science is
signicant enough to be applied regularely to the study of the origin of clay tablets.
TECHNOLOGICAL ASPECTS OF THE AMARNA TABLETS AND THEIR IMPLICATIONS

The second aspect that is brought up by the results of this study relates to the technology of tablet
production. So far, our knowledge about the technological aspects of this skill was merely circumstantial
and fragmentary. While a major part of the Amarna archive, namely the Canaanite correspondence, is
rather unique in the context of Near Eastern cuneiform archives, the international part of the archive may
reect on other cases too. This section will summarize the main features that result from this study.
317
Clay selection
In Chapter 2 we referred to the technological requirements that ostensibly dictated the selection of
clay and temper by the scribes who made the tablets. We assumed that these necessities would diverge
in certain cases between the clay used for tablets and that applied for pottery production in a given
location. However, the results of this study reveal only very few cases in which this discrimination is
clearly visible. Apart from the Egyptian texts, already discussed in the previous section, only in the
case of Megiddo was a specic clay source, unused for pottery, employed in order to produce very light
coloured dockets. Technological studies (by Goren and his students) of the pottery assemblages from
Megiddo have indicated that they were regularly manufactured of the alluvial soil of the Jezreel Valley,
or marl of the Taqiye formation, both appearing immediately at the foot of the site. However, the bright
marl of the Ghareb formation, inadequate for pottery manufacture since it contains over 30% calcium
carbonate (hence white in shade), was systematically employed for the production of the letters. This
case is especially noteworthy because it is the only instance in the Canaanite correspondence where a
special type of clay was chosen specically for tablet production. By contrast, at the nearby site of Tel
Yokneam, which according to our study was the seat of Ba>lu-meir, the local alluvium was the source of
clay for pottery and tablets alike. The unique technologic selectiveness of the Megiddo scribe emphasizes
the special role of his city in the Canaanite city-state module. It may even be suggested that the Megiddo
scribe was trained at the school of the Egyptian court where, as we saw, a similar attitude was exercised
both for the scholarly texts and the outgoing correspondence.
Although hardly ever supported by any analytical or textual data, it is sometimes assumed that
scribes used to carry with them their favourite clay to produce tablets wherever they were required to do
so. Theoretically, this may hold true for certain Mesopotamian courts where highly skilled scribes were
maintaining rich and ever-busy archives. However, at least in the case of the Amarna tablets our study
rules this assumption out. This point is extremely important since it indicates that in most cases of the
Canaanite correspondence, the location of the sender can be determined by establishing the provenance
of the materials of which a tablet was made. Needless to say, this is the fundamental basis for the
conclusions of the present study.
The evidence that supports this perception comes from different cases, including those in which
the identity and domicile of the sender are known but the provenance of the clay (and sometimes the
textual evidence) indicates that the letter was dispatched from elsewhere. For example, although the
rich assemblage of letters from Gubla (Byblos) designates the constant use of the local NGM-NGC
clays (above), when Rib-Hadda and his scribe dispatched letters from umur (e.g. EA 103) they were
written on the local QCC clay, collected ad hoc on the spot. Similar are the cases of EA 285 and 291,
dispatched by the Jerusalem scribe from Beth-shean and Gezer respectively. They were made of local
clays and not of the Moza marl that was used for Abdi-ebas other tablets. The assemblage of letters of
the Bashan cities (EA 201-206) that were written by Biryawazas scribe in the name of the local rulers
were made of indigenous materials and not of the clay that was used for the Damascus correspondence.
The letters sent from Gaza were made of an extremely coarse and usually crumbly mixture of silty
loess soil and coarse coastal sand, often with added chopped straw. Yet scribes systematically used this
off hand local mixture instead of bringing with them their superior clay (e.g. the letters sent by Yapau
of Gezer, Yappa-Hadda of Beirut [EA 97] and Aziru of Amurru [EA 169]). This is also true of the
letters sent by the rulers of Acco (EA 232, 234-235), Shamuna (EA 224) and Jerusalem (EA 285) from
the Egyptian centre of Beth-shean.
318
Added materials
The inclusions added to the clay of tablets have conflicting effects. Since tablets are relatively thick
ceramic objects, temper is needed in order to reduce the plasticity of the clay and prevent shrinkage
and crumbling upon drying. At the same time, too many or too large grits may hamper the inscribing
process. Therefore it is expected that carefully selected substances would be used for this task. However,
the results of this study reveal that this trend may be remarked only in the case of the Mesopotamian
letters, alongside the tendency to fire tablets which seems to result from the long tradition of the
cuneiform culture in this area. Consequently the Babylonian and Mitannian letters, made of Euphrates
sediments or those of its headwaters, differ from pottery that was made from the same deposits by the
absence of naturally occurring inclusions apart of the fine sand fraction. This implies that the clay was
refined prior to its use. The Egyptian and Canaanite tablets do not reflect this attitude.
The most common temper in the Canaanite letters is sand, usually containing rounded or partly
rounded minerals and/or rock fragments, indicating beach or wadi deposits. Also frequent is added
vegetal material (commonly referred to as straw). By this, Canaanite tablets are similar to most
Levantine pottery (Goren 2000b), suggesting that local scribes tended to obtain their materials from
the local potters without any further treatment. We already mentioned above the exceptional case of
the Megiddo letters, which only emphasizes the prevalence of this phenomenon.
In addition, there are several cases where scribes intentionally added some odd substances that
have little if any effect on the workability or technical properties of the clay. Therefore, their addition
was apparently motivated by other purposes. This includes hair, wool, herbivore (and rarely carnivore)
dung, and crushed beetles. The function of these inclusions remains unclear. Although highly
speculative, it may be suggested that some of these components (like the crushed beetles in letters EA
86, 87, 88, 95, 98, and 101 from the Lebanese coast) were included for some supernatural purposes,
such as attempt to control the response of the addressee. However, at present there is no way to confirm
this conjecture.
Firing
Our study reveals that most of the Babylonian, Mitannian, Hittite and Ugaritic tablets were fired
in kilns at about 700 0 -800 0 C, similarly to pottery, while Egyptian and Canaanite letters were
systematically unfired or only slightly heated to preserve their shape. The implications of this
dichotomy extend beyond the sheer technical sphere. Firing tablets to a degree of sintering of their
clay makes their inscription irreversible in terms of the possibility to moisten the document and reuse
it or wipe and change details within the text. Therefore the act of firing was motivated by two needs:
to ensure that the tablets would last for long and to prevent any possibility of forgery of details in the
text. These needs are the outcome of the habit of filing documents in archives for generations and using
them as references for legal or diplomatic affairs. Hence the act of firing a document must have been
perceived as a certificate for its authenticity. It is not surprising that tablets were fired in states where
there was an enduring tradition of the cuneiform culture. The assumption, raised in conversation by
some colleagues, that tablets were circulated and reused, is untenable. The negligent treatment of the
Egyptian and some Canaanite tablets in terms of clay modification and firing may be interpreted on
the background of the short-lived tradition of cuneiform communication in these places.
319
THE SYSTEM OF CANAANITE CITY-STATES

This issue has recently been debated by two of the authors (Finkelstein 1996; Naaman 1997). In fact,
this study was initiated as an attempt to resolve the dispute. The current research somewhat narrowed the
differences, although it has not eliminated them.
It is agreed that the list of Canaanite city-states should be composed, rst and foremost, on the basis
of the Amarna letters. Each person who wrote either to the Pharaoh or to his ofcials must be regarded as
a ruler of a city-state, regardless of the scope of his territory or his political power. Thus, if the archive
were complete, the task of making a list of city-states would have been relatively easy. Unfortunately, this
is not the case. Not only is the archive fragmentary, but also the seat of many rulers remains unknown.
Many tablets are broken and the names of their author and his town are missing, and the city of many
rulers whose names appear on tablets is not mentioned. In the course of our study we were able to identify
the seat of some rulers whose place was until now either unknown or disputed, but the location of many
others remains unknown (i.e., the senders of EA 57-58, 66, 173, 178-179, 181, 200, 202, 207-210, 211220, 229-231, 251, 260-261, 263, 307-313, 317-319, 336-337; Turbazu and Yapti-Hadda mentioned in
EA 288:39-47 and 335:8-18). Letters dispatched from some city-states mentioned in the archive have not
come down to us (e.g., Tushulti, Tubiu) and the exact location of some city-states mentioned by name
in the archive (e.g., Mushiuna, Sharuna, Airuna and possibly Zil) is unknown.
According to the documentary evidence, the major Canaanite centres along the coast of Palestine
were Ashkelon and Acco. Gezer, Gath and Lachish were the most important city-states in southern
Palestine. Shechem and Jerusalem were the major kingdom in the central hill country, and Hazor
dominated the Galilee. Megiddo, Rehob, Shamuna and Achshaph were the major city-states in the
northern plains and Ginti-kirmil was an important city-state in the central coastal plain. Other kingdoms
are also attested, some of which cannot be located. In the south were Yurza, Ashdod (Tianna according to
Naaman), Airuna and possibly Zil. In the north were [x-x-I]G-ma-te (Tel Yokneam) and Anaharath,
and according to Naaman also Ta>anach, Geba-mn and Mishal.
Naaman proposes that sites such as Beth-shemesh, Tell Beit Mirsim, Tell >Aiun, Khirbet Rabud,
Tell el->Amr and Tel Qarnei iin may have been the seats of city-state rulers. He argues that the archive
is incomplete, that some of the mayors whose seats are unknown may have ruled in these places, and that
some cities which are known to us by their Iron Age names were called in the Amarna period by different
names (see Eissfeldt 1968). Moreover, major identiable sites that have been excavated on large scale
produced very little Late Bronze II architectural elements. Consequently, the political status of almost all
sites cannot be established on the basis of their size or the quality of the material culture (see below). In
his opinion, the overall number of city-states in Canaan was about 22-25.
Finkelstein objects on the grounds that rstly, these places are not mentioned in the archive and
most, if not all of them are not mentioned in any Late Bronze Age source. Secondly, most of them had
no importance in the Late Bronze Age. Khirbet Rabud was a poor, small site and Tel Qarnei iin was
no more than a small fort (Gal 1981:218). Taking into consideration that some of the unprovenanced
mayors could have ruled in well-known sites (see, for instance, the case of Wiktasu at [x-(x)-I]G-ma-te)
the petrographic investigation of the Amarna tablets adds only three cities to the 14 listed in Finkelsteins
previous study (1996). These are Ashdod in the south and Anaharath and Yokneam in the north. Leaving
room for a few possible, additional city-states (e.g. Ahiruna), we reach a system of about 20 city-states
in western Canaan, or slightly less.
320
One of the most intriguing problems in the discussion of the system of Canaanite city-states is the
poverty of remains in the strata attributed to the Amarna period. We have already discussed the relatively
limited scope of the Late Bronze remains found at sites such as Jerusalem, Beth-shemesh and Ta>anach.
Even major sites that have been excavated on large scale, such as Gezer, Shechem and Tell Keisan
produced very little architectural elements of the Late Bronze II. Had Gezer or Shechem been excavated
without the control of the documentary evidence, we would have never guessed that they were two of
the most important Canaanite kingdoms in the 14th century BCE. Hazor and Megiddo alone yielded
monumental buildings and rich assemblages of artefacts, which indicate their foremost place among
the kingdoms of Canaan. Only a relatively limited number of Late Bronze II villages and farms have
been discovered in the course of the extensive surveys conducted in all parts of Palestine. The results of
the surveys fully corroborate the picture which emerges from the excavations of the major sites both
indicating the relative poverty of the urban culture in 14th century Canaan.
This problem is not restricted to Canaan. Tell >Arqa (Irqata), located northeast of Tripoli, which was
rst an independent city-state and later the seat of Abdi-Ashirta, produced very little Late Bronze II nds.
Several important conclusions may be drawn from the archaeological evidence. Some urban centres
were sparsely inhabited and the actual power of the Canaanite rulers must have been quite limited.
The weakness of the cities opened the way for the rise in power of elements that developed outside the
system of city-states, that is, the >Apiru and the Sutu, and of rulers who took advantage of the political
and demographic situation and turned it to their favor, for instance Lab<ayu of Shechem and Abdi-Ashirta
and Aziru of Amurru. It is not a coincidence that both were located in the highlands of the Levant an
environment which gave them a clear advantage: they were far enough from the Egyptian centres and
they could exploit the extra-urban elements in order to fulll their political goals.
Needless to say, it is impossible to establish the system of Canaanite city-states on the basis of the
archaeological evidence alone. According to Naaman, the size of mounds in which Late Bronze Age II
pottery was found might be misleading, as the built-up area on the site might be relatively small. Smaller
sites whose entire area was inhabited might be equal in their built-up area and population to larger sites
whose areas were partially uninhabited. There are serious doubts as to the accuracy of the population
estimates suggested so far by scholars. The built-up areas of sites such as Lachish, Tell e-a, Tel Gezer,
Shechem and Tell Keisan in the Late Bronze Age II is unknown and all calculations based on the overall
areas of these sites may be misleading (see Gonen 1984; Bunimovitz 1989:151-153; Broshi 1993:14;
Finkelstein 1996:242-247).
Finkelstein cannot accept Naamans criticism. Firstly, many of the Canaanite centres, including
places which did not yield evidence for monumental architecture, provide enough data to show that
they had a signicant size. Hazor, Megiddo, Rehov, Gezer, Lachish and Ginti-kirmil covered an area
of 10 hectares or more. Second, there is no evidence in any of the digs for the notion that they were
partially empty. To the contrary, almost every area which was excavated yielded Late Bronze remains.
Third, the method behind the population estimates for the Late Bronze Age is not different from the one
deployed for other periods in the history of Palestine, or other places in the ancient world. There is no
reliable substitute to the combination of archaeological and ethno-historical data and there is no solid
case for undermining this method as a system (to differ from a case of a single given site). According
to Finkelstein, then, the map of the Canaanite polity should be obtained by the study of both text and
archaeology.
Equally unrealistic, according to Naaman, is the estimation of the population in each polity based
on the exact drawing of borders. Borders that pass through inhabited areas dene the territory where
321
kingdoms imposed taxes, levied soldiers for war and workers for the corve. On the other hand, it is
doubtful whether exact borders could be delineated in sparsely inhabited areas, where there were no
permanent settlements. There is no evidence that Canaanite rulers effectively dominated these hilly,
forested areas, or that they considered them parts of their kingdoms. To the contrary, the Amarna letters
indicate that all the conicts between neighbouring kingdoms involved villages and towns. Therefore
it is not advisable to draw maps that divide, without any gaps, the entire area of Canaan, and give the
impression that there was a stable network of borders in which each king knew what belongs to him and
what to his neighbours. Nor is it useful to calculate kingdoms territories based of these articial borders.
Portions of the sparsely inhabited territories must be treated as a kind of no-mans-land, and should not
be considered integral parts of the Canaanite polity.
Finkelstein agrees that in cases in which there is no clear-cut textual evidence or geographical
and settlement logic, the delineation of ancient borders should be taken as no more than an illustration.
However, he rejects the idea of no-mans-land in Late Bronze Age Canaan. This was a small land and
the distance between the centres of city-states even in the highlands does not exceed 45 kms (in the
case of Shechem and Jerusalem). All textual evidence indicates that territories and villages were clearly
afliated and the assumption of no-mans-land is not supported by any source. Hence there is good
reason to argue that the entire area was divided between the Canaanite polities.
In the light of the archaeological evidence we can redene the term city-state, so common in the
discussions of the Amarna tablets. According to Naaman, elements such as size of the city and extent
of its territory must be eliminated from the denition. A city-state is an independent territorial unit held
by a local ruler who inherited his status from his ancestors and was recognized as a mayor (Egyptian
3ty-> Akkadian azannu) by the Egyptian authorities. In his relations with his subjects and neighbours
he considered himself to be king (arru). The status of independence of many city-states was established
already in the Middle Bronze Age and inherited by the rulers of the Late Bronze Age. After the Egyptian
conquest of Canaan, the Pharaohs left the array of kingdoms untouched and recognized the legitimacy of
the reigning dynasty in each kingdom. A few city-states lost their independence at the end of the Middle
Bronze Age (e.g. Sharuhen and Kabri) and others during the Late Bronze Age (e.g. Chinnereth at the
end of the Late Bronze I and Taanach during the Amarna period). Nevertheless, the system of Canaanite
polities remained very similar to that of the Middle Bronze Age.
Finkelstein agrees to this description except for two notions. In the rst place, the city-state system
of the second millennium BCE was not rigid. A few major Middle Bronze Age centres, such as Dan and
Kabri lost their status in the transition to the Late Bronze Age. The same holds true for changes that
took place in the late second millennium BCE (Finkelstein, forthcoming). Second, the size of the city,
its hinterland and the scope of its territory must be taken into consideration when drawing the territorialpolitical map of Canaan in the Late Bronze Age II.
THE EGYPTIAN ADMINISTRATION AND THE ORIGIN OF THE VASSAL LETTERS

Canaan had been divided into territorial-political entities of various sizes for centuries. The Egyptians left
the array of Canaanite kingdoms untouched and established a network of garrison cities to administer the
territory. Four were situated along the coast: Gaza and Jaffa in the south; Ullasa and umur in the north.
One city (Beth-shean) was located in northern Palestine, on the main route linking Syria with Egypt;
another (Kumidi) lay on the major crossroad of the Beqa> of Lebanon. The six garrison cities directly ruled
considerable territory in their neighbourhood and were administered by Egyptian ofcials. They served as
centres for the Egyptian garrisons and personnel, as gathering places for the tributes of the vassals, and as
centres where the local Canaanite rulers often appeared, either for service or for negotiation.
322
A few vassal letters were explicitly sent from neighbouring towns. Thus, Rib-Hadda of Byblos was
expelled from his city and wrote to the Pharaoh from the city of Beirut. Yashdata (of Ta>anach?) was
driven from his town and wrote from the neighbouring city of Megiddo. In almost all other instances
Canaanite mayors did not specify the place where their letters were written, and scholars naturally
assumed that they were dispatched from their hometowns. This was also our assumption when we began
analyzing the letters; we took it for granted that the chemical and petrographic analysis of each letter
would indicate the seat of the ruler who sent it.
Much to our surprise, this is not always the case. Letters from securely identied cities in Canaan
have, in most cases, a mineralogical and lithological composition that is in agreement with their
geological environment. However, some letters from securely identied Canaanite cities were made from
raw materials which do not t their location.
These tablets are clustered in three groups. One group of 22 letters was made from loess and sand
from the southern coastal plain. It includes: (a) six letters sent by rulers of south Canaanite cities, namely
Gezer (EA 298, 299, 300, 378), Gath (EA 281) and Lachish (EA 329); (b) a letter from the ruler of Beirut
(EA 97); (c) seven letters sent by Dagan-takala (EA 317-318), Zitriyara (EA 211-213), Bayawa (EA 215)
and Shipu-ria (EA 226), rulers whose seat was probably located in northern Canaan; and (d) eight
letters from other rulers whose names and places have not survived (EA 66, 217-218, 307-310, 312).
The second group contains elements (such as marl and sand with basalts, chert, limestone,
travertine and quartz), which are characteristic of the central Jordan Valley sediments. It includes
seven tablets: (a) four letters from the rulers of Acco (EA 232, 234-235) and Shamuna (EA 224); (b)
two tablets of north Canaanite rulers whose names and cities did not survive (EA 251, 263); and (c) a
letter from Jerusalem (EA 285).
The third group contains elements such as Quaternary beach deposits, chert and minerals of
volcanic origin which are characteristic in the coastal area of the >Akkar Plain between Tripoli and
Tartous. This group constitutes seven letters: three sent by Aziru of Amurru (EA 165-167); three
dispatched by Rib-Hadda of Byblos (EA 78, 103, 126); and one letter sent by an Egyptian commander
to Rib-Hadda (EA 96).
The only possible explanation for this phenomenon is, that these letters were sent from the
Egyptian centres of Gaza (the loess and sand group), Beth-shean (the Jordan Valley sediments) and
umur (the third group, which features >Akkar Plain sediments). There are some references in the
archive to messengers sent to Egypt from certain city-states, in particular Byblos, but in most cases
it was the Egyptian officials who delivered the vassal letters. Hence it seems logical to assume that
in the last leg of the trip, most letters, including those written in the hometowns of the local rulers,
were dispatched from the Egyptian centres. Egyptian officials either collected tablets prepared in
the hometowns of the vassals when they visited their cities (the common case), or handed over to the
Egyptian centres by messengers of the vassals. In the specific case discussed here, local vassals wrote
their letters when they visited the Egyptian centres.
The texts do not provide clear indications as to why these specic tablets were sent from Egyptian
centres. Some Canaanite rulers must have arrived at the Egyptian centres for service, to bring their tributes
and gifts, or to negotiate local matters with the Egyptian authorities. Their scribes, who served as private
secretaries and political advisers, must have traveled with them. This explains why the script and text of
these letters are similar to those of the other tablets sent from the hometowns of the respective rulers, while
they are different from the script and text of other letters dispatched from the same Egyptian centre.
There are other possible scenarios for tablets being written in Egyptian centres. Yapau, the ruler
of Gezer, was in deep trouble ever since he ascended the throne. In trying to consolidate his seat, he
323
begged repeatedly for Egyptian military aid. Since four of his letters were dispatched from Gaza, he
must have travelled there several times, imploring the local authorities for help and writing desperate
letters to the Pharaoh. The same is true for Shuwardatu of Gath, who ruled at the same time and faced
similar threats to his throne.
The background of other letters sent from Gaza is more difficult to establish. It is surprising
to find that the mayor of Beirut sent a letter from Gaza since rulers of the Lebanese coast usually
appeared at, and guarded, the Egyptian centres of umur and Ullasa. The dispatch of some north
Canaanite tablets from Gaza is also surprising, since the centre of Kumidi was much closer to their
towns. The explanation for these ostensible anomalies must be sought in the prominent place of Gaza
in the Egyptian administration of Canaan. Canaanite rulers who tried to meet and negotiate with high
Egyptian officials must have appeared before the commissioners located there.
The large number of letters sent from Gaza, some written by distant rulers, well illustrates its
centrality in the Egyptian administration of Canaan. Gaza was doubtless the most important centre in
southern Canaan, and possibly in the entire Egyptian province in Asia. It was the seat of governor whose
Egyptian title was messenger of the king to every foreign land. Unfortunately, the old town is buried
under the modern city of Gaza and hence cannot be excavated.
Part of the southern Jezreel Valley was crown land administered by the Egyptian centre of Bethshean. The rulers of Acco and Shamuna might have sent workers to cultivate the elds, or guard the
Egyptian centre and installations therein, and on these occasions might have written their letters. It is
possible that the king of Jerusalem also sent workers for seasonal work in the Egyptian elds near Bethshean, and on that occasion visited Beth-shean and wrote his letter.
Beth-shean is rarely mentioned in the Amarna letters. The 14th century BCE stratum (IX) yielded
some Egyptian artefacts, including the stele of Mekal, the great god, erected by the Egyptian ofcial
Pa-Ra-em-heb in memory of his father. The results of the excavations and the petrographic analysis
support the centrality of Beth-shean in the Egyptian administration of Canaan in the Amarna period and
compensate for the gap in the documentary evidence.
Seven tablets were written in umur (Tell Kazel), an important Egyptian centre located in the >Akkar
plain, south of Nahr el-Kebir, near the northern boundary of Canaan. Three of these letters were sent by
Aziru of Amurru after he had conquered umur and established his seat there (EA 165-167). Three letters
of Rib-Hadda of Byblos (EA 78, 103, 126) were also dispatched from umur, apparently when he sent
his men to serve in the town, or when he appeared before the Egyptian authorities located there. Finally,
a letter of an Egyptian commander to Rib-Hadda (EA 96) was also sent from umur. No evidence for
Egyptian administration has thus far been found in the excavations of Tell Kazel.
Kumidi (today Kmid el-Lz) is located in the southern Beqa> of Lebanon, on the road that connects
Damascus and Sidon. A few cuneiform tablets were unearthed at the site, including two letters sent by
the Pharaoh to the rulers of Damascus and Shazaena (Edzard 1970:55-60). Unfortunately, we could not
check these tablets, which are kept in the Damascus museum, nor could we analyze the Amarna letter
sent by the mayor of Kumidi (EA 198), which is deposited in the Cairo museum. Kumidi was the sole
Egyptian stronghold which also served as a centre of a Canaanite city-state.
The ten tablets sent from the city-states of the Beqa > (EA 174-179, 185-187, 363) represent
closely related materials. The Beqa > valley, over 150 km long, is bordered by mountain ranges which
are characterized by varied lithologies. At least some variability should therefore be expected in
the petrography of products that come from different locations along the valley. The most logical
explanation to the close similarity between the Beqa> tablets is that most if not all of them were sent
324
from a central site. The only option that comes to mind is that this was the Egyptian administrative
centre at Kumidi. It must be emphasized that in the case of the Gaza, Beth-shean and umur letters,
establishing the origin of the tablets rests on the observation that letters from securely identified cities
in Canaan may have a mineralogical and lithological composition that is not in agreement with their
geological environment. The attribution of the Beqa> letters to the Egyptian centre of Kumidi rests on
different ground, namely, the mineralogical uniformity of the clay of tablets dispatched by different
rulers situated in a relatively broad area.
The Egyptian centre of Jaffa is located south of the confluence of the Yarkon river and Ullasa is
probably buried under the city of Tripoli, near the confluence of Nahr Abu >Ali. The possibility that
some Canaanite letters were sent from these centres cannot be ruled out, but there is no unequivocal
evidence to prove it. The petrographic traits of the above letters that we assign to umur do not accord
very well with the Tripoli area, first and foremost in the content of basaltic minerals that are not found
south of the >Akkar. However, some leeway must be left for the possibility that a letter such as EA 96,
where such minerals were not observed (perhaps due to the small sample), could have been potentially
sent from Ullasa. Possibly EA 294 and 296 could have been sent from Jaffa but this option is less
favourable due to the reasons raised above regarding their origin.
In conclusion, Canaanite mayors must have arrived quite often at the Egyptian administrative
centres, whether for service or for negotiation and political manoeuvres. They held closer contacts
with these centres than suggested by their letters. Our research thus emphasizes more than ever the
importance of the Egyptian centres and their pivotal role in the administration of Canaan in the
Amarna period.
325
326
10.78
5.97
9.23
38.10
40.20
31.21
40.80
43.31
35.37
35.80
41.86
38.25
41.51
50.59
55.51
41.34
42.08
32.12
33.47
33.07
40.09
40.56
45.85
40.96
51.47
35.75
24.70
38.10
23.88
21.99
52.79
46.51
42.43
40.78
EA6
EA12
EA13
EA14
EA18
EA24
EA30
EA33
EA34
EA37
EA42
EA44
EA45
EA46
EA50
EA51
EA54
EA56
EA57
EA58
EA60
EA65
EA67
EA72
EA80
EA83
EA86
EA97
EA111
EA115
EA121
12.95
15.95
14.56
6.81
9.77
7.03
10.13
8.86
21.89
12.95
9.87
11.71
6.65
6.67
7.55
7.95
8.43
15.27
16.69
9.87
8.42
9.66
8.75
9.21
12.39
8.69
9.62
10.59
6.77
11.31
Al2O3
37.38
39.95
EA1
EA4
SiO2
12.16
5.46
3.77
8.43
27.19
22.31
16.49
27.91
16.36
7.84
<0.3
8.98
18.39
8.53
27.70
27.56
19.07
21.05
19.68
5.00
8.46
10.86
17.51
18.79
12.13
11.53
18.33
19.86
15.59
12.68
12.40
18.32
13.61
CaO
2.79
3.76
3.69
2.36
2.13
2.08
2.43
2.47
2.77
2.72
1.21
3.21
1.88
2.89
1.74
2.74
2.15
3.29
2.98
2.46
3.88
1.78
1.88
2.39
3.24
7.19
6.53
1.80
4.62
5.44
4.96
3.08
5.19
MgO
7.20
9.06
9.05
4.83
3.44
3.82
5.25
3.57
6.49
4.83
21.32
6.28
5.12
6.64
3.34
3.75
4.35
4.68
5.20
6.51
7.26
5.43
3.99
4.28
4.88
6.65
7.17
4.56
5.52
6.13
5.91
3.90
6.32
Fe2O3
0.99
1.04
1.20
0.87
0.52
0.62
0.81
0.56
0.76
1.02
3.36
0.97
0.71
0.86
0.48
0.63
0.77
0.57
0.59
0.81
0.94
0.63
0.50
0.55
0.55
0.46
0.58
0.66
0.51
0.66
0.59
0.63
0.61
0.40
0.46
0.57
0.23
<0.3
<0.7
0.19
0.46
0.18
0.34
0.30
0.43
0.10
0.17
0.24
0.38
0.50
0.40
0.18
0.10
0.14
<0.3
<0.2
0.21
<0.6
0.21
0.19
0.10
0.09
0.38
0.12
0.28
0.10
TiO2 P2O5
0.68
0.54
0.94
0.97
0.84
1.10
0.82
1.20
0.74
0.77
0.39
0.84
0.53
0.57
0.70
0.80
1.04
0.77
0.69
0.40
0.35
0.64
0.60
0.64
1.06
0.68
0.68
0.54
7.38
1.93
1.06
0.74
0.51
SO3
104.89
140.34
91.03
70.22
71.25
79.44
101.69
85.88
124.86
87.61
146.27
132.74
95.77
136.33
185.46
552.97
129.32
263.76
410.81
392.10
501.41
57.17
55.46
99.24
78.75
108.03
577.81
57.65
191.27
144.30
83.95
70.97
80.57
Zn
142.53
69.08
91.67
80.21
111.36
368.83
334.25
36.66
198.08
231.19
205.43
28.63
222.05
Ni
426.90
315.79
175.00
142.59
368.18
145.78
129.79
154.30
165.52
174.68
201.89
237.10
238.89
Ba
23.92
19.10
19.75
11.36
<15
<30
21.50
23.15
31.30
19.95
80.43
21.55
22.70
30.05
11.44
18.94
25.00
31.27
51.75
82.00
140.71
108.20
48.78
55.11
38.89
77.35
60.84
152.07
54.78
162.50
89.32
89.18
119.18
70.65
76.05
49.23
220.65
721.21
705.02
668.60
480.26
454.17
523.15
511.36
597.72
517.29
374.75
632.82
784.89
802.45
322.58
731.58
Mn
99.47
77.92
98.94
235.94 130.21
225.90 176.43
123.03 101.32
128.33
87.94
187.50 112.50
311.96 121.72
288.94 132.80
73.81
190.20 103.04
262.40 148.44
247.81 111.84
93.15 88.71
258.92 118.66
Cr
88.37
77.75
163.34 182.52
140.40 111.17
182.56 133.45
128.86
120.29
140.18 153.81
673.46
116.02 134.02
1435.93 558.70 366.35
808.93
526.82
712.38
260.72
414.81
262.75
605.23
261.36
236.11
498.50
408.02
121.11 145.36
97.82
95.00
107.70
78.13
131.39 102.36
125.46 124.19
167.36
460.94
132.44 149.36
249.52 2198.41 142.11 204.04
164.11 1112.33 168.92 194.08
233.44
115.91
150.00
776.56
38.28
362.62 1234.76 490.75 143.29
269.15
71.78
170.94
528.28
206.47
247.28
367.31
150.49
373.79 1331.11 1358.30 124.27
438.09 1046.79 1646.93 113.09
29.51 1028.93 371.82
31.83
<15
<10
16.70
<30
36.64
25.22
16.95
28.93
30.22
30.06
<10
33.13
Co
SiO2 to SO3 values are in weight percent. Zn to Lu values are in PPM.
2.02
2.49
2.44
1.50
1.31
1.67
1.65
1.48
1.83
1.68
2.10
2.23
1.70
2.01
1.23
1.89
3.25
2.02
1.55
2.30
2.61
1.71
1.38
1.72
2.05
1.22
1.68
1.33
1.32
1.90
1.55
1.21
1.53
Be
La
136.36 29.55
267.89 12.80
225.77 17.83
238.48 20.62
648.65 15.33
353.76 17.31
965.93 18.46
537.10 20.44
261.56 18.81
Sr
308.40 34.04
221.05 35.07
410.00 26.54
327.11 33.33
283.62 27.16
435.85 17.79
305.05 29.41
250.76 26.35
119.98 30.01
208.32 31.58
221.43 28.19
136.11 27.04
370.55 19.20
326.80 20.73
390.28 19.36
63.74
79.63
68.99
38.36
52.50
34.25
167.71 31.94
120.17 30.86
95.82
255.73 22.69
281.59 22.90
119.17 166.67 23.47
43.97
43.04
57.66
26.98
79.59
58.46
34.48
41.30
38.23
33.20
35.86
121.32 338.93 30.54
57.08
134.01 198.17 30.17
66.55
49.93
38.96
111.35 417.34 23.45
41.25
49.82
59.45
23.31
45.78
44.93
40.40
26.61
41.09
Cu
TABLE 1: RESULTS OF ICP ANALYSIS OF SELECTED AMARNA TABLETS.
25.28
26.00
25.56
22.25
18.47
21.67
23.95
18.86
27.01
25.79
21.31
31.09
24.77
20.53
18.17
17.52
26.43
36.41
37.96
21.55
23.85
27.37
19.50
18.10
20.68
12.86
17.11
17.82
14.94
18.21
17.51
18.35
18.25
59.30
61.86
95.76
48.61
33.31
36.04
53.96
39.86
53.52
62.19
58.52
69.37
54.66
80.26
44.64
47.60
53.57
48.21
50.59
64.58
72.34
46.24
45.15
50.33
42.19
28.94
38.83
45.01
31.27
36.44
40.86
37.02
40.66
Ce
8.43
8.66
10.78
6.73
4.37
4.78
7.35
5.42
7.25
7.85
6.18
8.65
7.21
9.79
6.04
6.49
6.85
7.56
7.68
7.87
8.78
7.01
6.06
6.82
5.18
3.80
4.94
5.78
4.05
4.58
5.18
4.78
5.22
Pr
APPENDIX: RESULTS OF THE CHEMICAL ANALYSES
Sm
Eu
Tb
Dy
Ho
Tm
31.01 6.35 1.25 0.85 4.78 0.95 0.47
31.36 6.66 1.29 0.86 4.75 0.98 0.47
39.53 8.40 1.66 1.07 5.82 1.16 0.57
24.85 5.05 1.08 0.78 4.43 0.93 0.48
16.58 3.56 0.76 0.55 3.15 0.65 0.26
18.11 4.42 0.91 0.65 3.70 0.70 0.28
27.26 5.69 1.33 0.82 4.60 0.93 0.44
19.93 4.02 0.86 0.65 3.22 0.67 0.34
27.54 5.60 1.21 0.81 4.61 1.01 0.49
30.07 5.93 1.30 0.88 5.19 1.20 0.56
23.07 4.93 1.16 0.75 4.42 1.01 0.41
33.38 6.53 1.42 1.04 5.83 1.44 0.60
27.20 5.32 1.19 0.73 4.24 0.98 0.40
36.44 7.29 1.62 0.97 5.61 1.23 0.46
23.35 4.64 0.95 0.71 4.00 0.95 0.36
24.83 4.63 1.12 0.76 4.03 0.90 0.35
27.57 5.45 1.35 0.81 4.68 1.00 0.42
29.87 5.75 1.28 0.91 5.80 1.43 0.57
30.27 6.03 1.26 0.99 5.61 1.36 0.57
29.12 5.39 1.15 0.75 4.26 1.00 0.45
32.33 6.07 1.28 0.84 4.79 1.13 0.48
26.97 5.33 1.10 0.86 4.65 0.94 0.37
22.16 4.14 0.85 0.62 3.67 0.73 0.30
25.45 4.76 0.91 0.65 3.66 0.87 0.37
19.23 4.16 0.84 0.67 3.84 0.73 0.34
14.60 2.86 0.67 0.46 2.69 0.63 0.30
18.81 3.78 0.82 0.60 3.44 0.81 0.36
22.29 4.53 0.95 0.65 3.75 0.87 0.35
15.24 3.21 0.71 0.50 2.87 0.69 0.30
18.22 3.78 0.88 0.64 3.67 0.90 0.38
19.67 3.97 0.87 0.67 3.89 0.91 0.38
17.66 3.59 1.50 0.58 3.50 0.67 0.28

19.27 4.03 0.91 0.64 3.76 0.86 0.36
Nd
2.43
2.57
2.91
2.64
1.68
1.92
2.49
1.92
2.57
3.15
2.54
3.55
2.50
2.78
2.35
2.18
2.51
3.34
3.34
2.69
2.94
2.46
2.02
2.10
2.16
1.63
2.08
2.16
1.85
2.34
2.30
1.91
2.37
Yb
0.34
0.36
0.43
0.38
0.23
0.24
0.33
0.26
0.37
0.42
0.33
0.48
0.32
0.38
0.32
0.27
0.33
0.45
0.46
0.36
0.39
0.34
0.28
0.28
0.30
0.22
0.26
0.30
0.23
0.30
0.31
0.27
0.30
Lu
327
42.51
23.01
45.50
36.31
20.66
23.97
30.66
32.71
26.12
53.90
19.79
51.78
49.42
50.36
54.28
48.61
30.41
41.47
28.75
37.00
36.34
41.22
32.13
37.25
34.28
32.31
31.88
42.20
59.50
38.99
32.63
31.56
35.28
31.24
36.22
35.84
16.21
14.45
10.92
EA129
EA138
EA139
EA141
EA142
EA149
EA151
EA152
EA154
EA155
EA156
EA157
EA163
EA165
EA167
EA168
EA171
EA172
EA173
EA178
EA179
EA186
EA187
EA192
EA193
EA196
EA197
EA210
EA218
EA222
EA224
EA234
EA235
EA237
EA238
EA239
EA242
EA243
EA246
SiO2
3.49
4.38
5.05
8.83
11.22
11.09
9.61
10.18
6.49
8.60
6.60
12.90
8.30
8.29
10.64
10.15
10.56
15.32
12.71
12.53
9.23
11.11
7.38
10.59
13.29
11.77
17.36
20.85
6.07
8.30
8.80
12.40
9.86
7.93
7.53
9.42
10.60
11.22
8.51
Al2O3
41.82
37.18
36.32
17.23
21.75
18.89
18.68
21.77
18.83
19.93
6.70
4.80
20.88
23.63
19.78
18.71
20.37
10.31
22.91
16.60
20.53
13.73
23.37
11.17
6.24
8.49
1.33
3.71
30.56
8.80
25.58
22.06
20.74
29.11
29.18
14.84
7.90
14.50
31.01
CaO
1.48
1.90
2.05
5.54
5.06
4.21
1.50
3.19
1.88
2.89
3.00
3.00
2.05
2.93
3.68
3.00
0.90
1.40
2.54
2.86
6.78
6.08
3.49
2.86
1.48
1.26
2.76
3.73
3.24
3.20
2.68
2.03
1.07
1.06
0.84
1.31
3.70
2.25
1.42
MgO
2.05
2.08
2.73
5.69
8.19
7.22
4.38
4.40
3.60
4.59
3.60
8.20
4.38
4.49
6.70
5.92
6.57
8.47
7.05
6.36
4.72
6.48
6.91
5.51
8.92
10.33
10.10
14.41
3.67
4.30
2.92
4.09
3.77
3.06
3.98
5.31
5.90
6.20
4.07
Fe2O3
0.15
0.22
0.26
0.74
1.02
0.92
0.68
0.65
0.69
0.88
0.81
1.54
0.73
0.63
0.91
1.04
0.94
1.04
0.83
0.86
0.67
0.65
1.00
0.80
1.74
1.61
1.41
1.42
0.56
1.01
0.46
0.55
0.60
0.42
0.89
0.77
0.92
0.94
0.74
0.84
1.13
1.63
0.33
0.15
0.12
0.59
0.25
<0.3
0.30
0.10
0.10
0.49
0.22
0.23
0.52
0.37
0.24
0.45
0.27
0.14
0.14
0.13
0.12
0.18
0.13
0.18
0.24
0.33
0.20
0.45
0.19
0.36
0.30
0.20
<0.3
0.20
0.30
0.10
TiO2 P2O5
1.34
1.46
1.69
1.32
1.27
1.46
1.48
1.06
0.96
1.17
0.40
1.40
0.98
0.83
0.81
0.76
1.01
0.66
0.61
0.96
3.85
0.64
1.92
0.68
0.65
0.48
0.50
0.65
1.33
0.90
1.55
1.08
0.97
0.61
0.51
0.61
0.90
0.57
0.89
SO3
15.78
13.00
22.32
12.43
<20
6.16
<7
<15
13.00
28.12
20.21
Co
100.24
106.72
112.57
130.63
596.16
73.30
89.67
103.80
59.76
86.77
41.00
98.00
83.75
89.45
96.67
117.95
88.10
119.29
127.97
104.27
75.42
84.91
75.88
76.71
74.75
62.89
109.85
15.27
12.91
13.88
41.24
40.12
38.73
9.13
19.21
<12
28.67
11.00
22.00
10.01
22.73
34.15
29.59
<10
35.42
30.76
23.58
24.08
35.70
32.58
23.45
51.88
58.09
42.49
1667.54 61.83
43.94
63.00
86.36
100.32
100.57
74.35
33.39
106.16
89.00
91.68
60.66
Zn
374.61
577.00
146.07
124.08
134.29
85.74
281.38
358.14
556.00
682.33
366.32
Mn
80.22
68.05
85.50
93.50
71.09
74.08
104.06 140.47
107.00 107.00
115.68 144.41
119.86 115.20
121.00
80.67
97.93
129.83 108.08
128.00 130.00
135.16 130.82
103.55 85.18
Cr
253.73
217.62
334.96
405.55
742.50
202.00
262.00
483.00
266.36
294.47
293.39
327.65
175.56
184.30
370.82
290.84
202.47
253.44
104.31
280.48
108.63
99.49
363.38
773.58
750.92
342.61
302.39
311.54
912.01
444.00
753.00
321.59
470.74
516.03
674.50
558.33
869.32
792.83
795.27
283.61
847.77
626.58
634.70
991.36
958.14
887.05
152.76
2291.93 207.54
640.15
3391.75 148.97
88.33
83.00
70.87
91.35
129.16 105.29
154.37 113.95
193.12 138.02
160.73 162.74
164.40 149.71
148.63 131.46
116.70
144.56 138.23
96.25
146.72 149.67
80.00
141.00 139.00
122.08
120.76 103.24
158.27 141.76
115.26 115.16
119.68 106.94
157.16 171.12
168.77 158.86
145.50 161.54
100.60 109.29
328.58 134.85
119.97 131.20
102.81 105.81
227.09 175.50
543.19 187.79
156.83 197.49
713.57 5851.09 459.67 361.16
19.84
346.00
305.10
319.18
335.14
624.22
78.98
281.86
191.00
241.69
70.88
Ba
127.85 1258.73 803.40
121.96
106.47
64.13
80.69
37.50
80.42
29.00
61.00
59.47
85.12
117.36
83.75
61.11
108.22
121.42
107.52
63.09
233.35
80.50
51.40
81.13
77.81
99.59
408.42
58.66
37.00
43.82
49.17
39.29
27.96
37.24
58.14
48.00
72.75
49.01
Ni
0.64
0.79
0.98
1.55
1.82
1.58
1.33
1.73
0.87
1.57
1.00
2.20
1.25
1.47
1.59
1.91
1.57
2.63
2.25
2.23
1.59
1.43
1.40
1.72
1.87
1.88
2.19
3.64
1.13
1.30
2.23
2.20
1.71
1.31
0.99
1.69
1.90
1.94
1.21
Be
La
448.08 16.32
248.00 28.00
484.38 26.39
446.24 33.68
360.00 35.29
450.89 28.89
176.02 21.89
278.37 30.99
150.00 33.00
285.70 31.58
273.94 21.40
Sr
108.11 42.09
243.36 36.04
173.36 33.54
845.08 24.32
641.24 18.02
363.58 23.25
252.95 28.13
132.85 41.01
331.82 37.84
163.51 32.49
80.81
95.65
90.11
61.73
52.33
41.84
34.67
42.95
32.79
41.55
16.00
24.00
39.17
44.85
50.06
52.12
978.85 12.68
893.13 17.97
973.29 20.99
511.86 22.37
459.49 26.11
516.40 22.91
385.57 31.65
394.73 26.99
435.00 24.38
303.23 28.12
232.00 20.00
149.00 42.00
321.82 32.99
398.65 24.75
430.61 24.22
249.91 32.40
421.16 169.44 36.71
49.21
52.82
38.29
32.05
52.69
34.03
31.78
51.57
42.75
83.22
251.70 224.96 56.82
34.77
63.00
53.80
34.54
66.79
49.70
18.80
41.57
58.00
46.38
26.33
Cu
24.18
28.98
34.33
22.33
22.16
19.55
27.13
22.60
20.63
29.46
19.00
32.00
28.56
22.85
22.15
24.74
31.30
32.56
32.78
28.49
18.81
18.41
22.25
24.12
31.21
30.21
30.98
57.06
19.89
25.00
21.10
25.76
26.65
21.91
15.92
27.21
27.00
26.88
15.99
15.91
20.82
21.59
43.38
53.71
48.79
48.03
50.91
38.85
70.19
40.61
86.49
52.86
44.15
47.16
61.10
62.65
78.29
60.07
65.42
47.36
36.40
46.13
58.20
93.66
85.13
70.51
84.83
35.77
51.83
47.02
62.46
58.86
47.78
37.98
55.09
70.20
67.18
39.77
Ce
3.05
3.62
3.91
6.04
7.09
6.31
6.44
6.78
5.13
7.62
5.46
10.80
7.03
6.05
6.33
8.14
8.00
10.59
8.32
8.72
6.21
4.78
6.37
7.85
10.39
9.78
9.56
12.52
4.97
7.01
6.44
8.40
7.62
6.35
4.84
6.93
8.59
8.38
5.22
Pr
Sm
Eu
Tb
Dy
Ho
Tm
11.75 2.54 0.72 0.45 2.86 0.65 0.35
14.18 3.19 0.84 0.51 3.40 0.77 0.41
14.99 3.28 0.79 0.57 3.71 0.86 0.45
22.20 4.68 1.30 0.73 3.92 0.80 0.39
26.52 5.52 1.30 0.81 4.55 0.93 0.43
23.98 4.84 1.22 0.73 4.09 0.81 0.39
24.41 4.81 1.02 0.76 4.28 0.90 0.38
24.88 4.95 1.08 0.75 4.15 0.87 0.42
19.66 3.94 0.96 0.64 3.66 0.72 0.31
28.51 5.60 1.25 0.89 5.28 1.14 0.55
20.09 4.23 0.97 0.64 3.66 0.78 0.39
39.52 8.06 1.86 1.16 6.29 1.29 0.61
26.72 5.25 1.15 0.85 4.61 0.95 0.38
22.02 4.33 1.01 0.72 4.04 0.83 0.41
23.68 4.87 1.17 0.77 4.38 0.88 0.41
29.56 6.18 1.42 0.88 5.04 1.00 0.47
30.47 6.27 1.31 1.02 5.49 1.15 0.44
38.49 7.83 1.66 1.09 6.28 1.26 0.61
30.38 6.22 1.35 0.91 5.31 1.12 0.55
32.03 6.22 1.42 0.92 5.45 1.13 0.54
23.02 4.73 1.01 0.67 3.76 0.73 0.37
17.83 3.74 0.90 0.57 3.59 0.76 0.37
24.79 5.17 1.24 0.76 4.22 0.85 0.41
29.44 6.09 1.28 0.81 4.78 1.01 0.51
36.90 7.28 1.52 1.07 6.02 1.28 0.65
36.02 7.42 1.53 1.01 6.12 1.28 0.66
36.33 8.32 1.84 1.18 6.64 1.41 0.69
47.61 9.67 2.29 1.41 8.05 1.78 0.80
18.79 3.90 0.88 0.61 3.42 0.70 0.36
26.69 5.38 1.14 0.74 4.50 1.00 0.49
23.75 4.66 1.04 0.67 3.77 0.77 0.40
30.97 6.10 1.24 0.82 4.72 0.97 0.49
28.02 5.27 1.09 0.84 4.76 0.91 0.37
23.77 4.63 0.91 0.68 3.95 0.78 0.32
18.08 3.76 0.88 0.56 3.09 0.60 0.23
26.26 5.46 1.19 0.86 4.80 0.94 0.41
31.52 6.28 1.22 0.89 5.13 1.07 0.51
30.95 6.43 1.32 0.88 4.93 1.06 0.51

19.31 3.93 0.90 0.54 3.05 0.63 0.30
Nd
1.88
2.25
2.58
2.36
2.36
2.07
2.52
2.52
2.04
3.27
2.30
3.54
2.61
2.39
2.40
2.86
3.08
3.49
2.96
3.09
2.01
2.15
2.12
2.69
3.53
3.48
3.56
4.47
1.77
2.81
1.96
2.52
2.39
2.02
1.48
2.61
2.96
2.76
1.60
Yb
0.27
0.29
0.36
0.27
0.31
0.27
0.33
0.31
0.28
0.44
0.30
0.47
0.35
0.31
0.30
0.35
0.42
0.45
0.39
0.40
0.28
0.29
0.30
0.37
0.51
0.48
0.49
0.63
0.26
0.41
0.28
0.36
0.34
0.27
0.20
0.37
0.41
0.39
0.21
Lu
328
32.10
27.90
31.90
39.30
EA275
EA276
EA277
EA279
6.70
26.90
42.94
53.38
58.45
1.50
4.20
48.50
52.34
54.05
59.10
52.73
52.40
29.30
39.30
32.40
38.30
34.50
34.70
39.64
EA291
EA295
EA298
EA303
EA306
EA307
EA309
EA311
EA315
EA316
EA323
EA325
EA332
EA334
EA335
EA336
EA356
EA357
EA358
EA370
0.60
36.19
EA266
EA288
28.22
EA264
0.70
19.89
EA261
30.40
40.73
EA259
EA286
43.92
EA257
EA281
36.64
23.00
EA249
EA251
SiO2
12.14
8.90
8.40
10.20
8.90
8.80
8.00
9.80
9.29
9.90
8.80
8.42
9.50
7.80
7.70
8.82
7.73
6.97
9.10
7.70
12.30
10.60
7.90
8.00
7.40
6.90
7.60
11.18
6.55
4.51
8.42
8.66
7.58
8.82
Al2O3
10.48
20.80
20.60
21.50
19.90
17.80
24.20
6.70
13.57
13.00
9.15
10.35
10.90
8.50
9.70
9.00
10.68
18.17
24.50
11.60
11.20
15.70
8.20
12.20
20.50
26.50
23.50
13.39
25.69
29.41
17.91
18.43
21.80
25.44
CaO
1.61
2.90
2.80
5.20
3.20
2.90
3.30
4.30
1.19
6.00
1.41
1.38
3.50
3.60
3.30
1.38
1.34
1.00
2.80
2.40
3.60
8.80
2.90
4.80
5.00
2.90
3.30
3.36
1.35
0.43
2.03
3.09
3.40
1.25
MgO
6.42
4.90
4.60
5.50
4.80
4.40
4.40
4.40
4.90
4.80
4.60
3.76
4.70
3.80
3.70
4.47
3.76
3.52
5.50
3.20
4.30
3.50
3.90
3.90
3.40
3.30
3.70
6.38
2.59
2.14
4.61
4.25
4.09
3.44
Fe2O3
0.86
0.67
0.65
0.82
0.78
0.77
0.71
0.76
0.90
1.03
0.86
0.79
0.88
0.69
0.66
0.86
0.72
0.68
0.83
0.49
0.58
0.43
0.63
0.77
0.59
0.53
0.60
0.90
0.66
0.35
0.91
0.78
0.76
0.50
<0.6
0.10
0.10
0.30
0.10
0.10
0.10
0.20
<0.2
0.80
0.48
<0.3
0.10
0.20
0.10
<0.1
<0.2
<0.2
0.10
0.20
0.20
0.20
0.10
0.30
0.30
0.50
0.40
0.08
<0.2
1.04
0.70
0.13
0.30
1.18
TiO2 P2O5
0.80
0.70
0.60
1.50
0.80
1.00
1.00
0.70
0.32
1.80
0.73
1.05
0.60
0.70
0.80
0.51
0.45
0.91
0.80
0.70
1.10
1.50
0.50
1.10
2.30
1.10
1.10
1.00
0.90
3.56
0.87
0.73
0.79
3.52
SO3
90.36
64.00
62.00
513.00
68.00
52.00
59.00
83.00
284.17
586.00
48.40
40.53
61.00
65.00
51.00
50.82
39.20
39.89
87.00
81.00
63.00
61.00
55.00
77.00
73.00
72.00
68.00
90.06
33.55
61.88
99.07
64.90
77.74
126.81
Zn
<30
14.00
14.00
14.00
11.00
15.00
12.00
12.00
14.58
39.00
16.20
<15
15.00
11.00
7.00
12.00
<7
10.56
13.00
9.00
7.00
10.00
6.00
14.00
16.00
11.00
10.00
31.08
<12
<15
38.95
23.10
25.77
<15
Co
305.39
394.87
0.3%
Ba
642.73
434.90
440.97
Mn
100.97 107.87
100.54 101.80
196.32 105.42
Cr
70.24
52.00
44.00
58.00
44.00
45.00
42.00
41.00
43.33
78.00
45.00
25.00
44.00
46.00
21.00
38.40
30.46
24.30
77.00
45.00
24.00
29.00
25.00
48.00
39.00
42.00
42.00
71.53
28.40
40.97
500.83
160.00
203.06
451.00
470.00
347.00
240.00
233.00
204.00
276.00
157.00
566.00
144.00
226.00
316.00
206.00
346.56
258.95
306.00
178.00
209.00
268.80
205.86
226.90
70.00
649.74
598.00
440.00
519.00
237.00
440.00
202.00
493.00
518.01
600.00
450.12
433.38
540.00
419.00
444.00
479.88
408.27
416.42
151.00
1320.00 414.00
390.00
1212.00 217.00
240.00
216.00
1006.00 492.00
1398.00 406.00
1190.00 491.00
275.06
211.20
2.7%
50.05
84.79
86.00
94.00
80.08
74.54
100.00
86.00
85.00
79.86
63.85
65.46
111.00
79.00
122.00
110.00
70.00
125.00
121.00
93.00
85.71
129.00
129.64 136.19
120.00 106.00
114.00 103.00
121.00 156.00
85.00
106.00 104.00
78.00
75.00
92.58
148.00 161.00
91.08
99.14
94.00
84.00
72.00
91.08
76.49
72.43
96.00
91.00
86.00
81.00
75.00
82.00
103.00 128.00
85.00
89.00
108.70 141.02
83.60
153.54
100.21 2361.19 1005.11 174.69 125.83
50.21
76.39
75.00
Ni
1.31
1.40
1.30
1.60
1.40
1.40
1.30
1.50
1.17
1.70
1.10
0.99
1.40
1.20
1.20
1.12
0.92
0.95
1.50
1.20
1.60
1.40
1.20
1.30
1.30
1.20
1.30
1.93
0.75
0.63
1.48
1.50
1.31
1.18
Be
La
390.00 28.00
388.00 28.00
240.72 31.39
459.60 20.15
696.67 18.06
482.43 32.37
204.15 27.99
580.30 22.82
611.67 28.89
Sr
262.32 27.30
232.11 26.51
222.00 30.00
199.00 25.00
221.00 16.00
276.48 28.34
206.21 25.78
561.55 25.63
131.00 27.00
263.00 23.00
160.00 21.00
150.00 18.00
204.00 21.00
201.00 27.00
272.00 28.00
277.00 30.00
288.00 25.00
173.00 25.00
245.00 22.00
55.00
244.29 32.50
109.00 272.00 24.00
34.00
321.00 288.00 30.00
97.00
51.00
28.00
29.00
104.96 353.00 29.25
1041.00 301.00 34.00
33.44
34.74
22.00
26.00
17.00
33.00
24.66
39.51
27.00
17.00
12.00
23.00
11.00
23.00
164.00 376.00 29.00
23.00
19.00
54.12
43.45
65.69
44.82
35.54
30.57
70.28
Cu
26.31
19.00
18.00
23.00
21.00
24.00
20.00
23.00
26.54
27.00
24.70
23.09
24.00
18.00
20.00
26.52
23.16
20.60
18.00
23.00
22.00
14.00
22.00
20.00
31.00
30.00
31.00
24.25
17.55
25.28
36.61
21.62
21.22
31.60
56.11
42.56
40.69
49.90
47.72
48.04
38.89
45.85
47.50
52.21
45.05
40.98
46.46
35.59
35.85
45.95
41.39
43.71
40.57
32.49
39.75
28.70
42.31
47.20
40.91
40.13
45.88
62.82
37.90
23.74
63.43
52.85
46.84
38.39
Ce
7.37
5.38
5.15
6.25
6.36
6.15
5.46
5.89
6.62
6.91
6.25
5.68
6.12
4.72
4.73
6.27
5.75
5.53
5.38
4.67
5.25
3.92
5.42
5.94
6.12
6.16
6.53
8.38
4.97
3.43
7.98
6.92
6.34
5.74
Pr
Sm
Eu
Tb
Dy
Ho
Tm
26.80 5.68 1.27 0.87 4.62 0.91 0.42
20.45 3.95 0.91 0.64 3.50 0.76 0.37
19.59 3.81 0.90 0.63 3.35 0.72 0.36
23.68 4.69 1.11 0.79 4.27 0.84 0.42
23.76 4.55 1.01 0.78 3.96 0.88 0.42
22.73 4.54 1.06 0.77 4.37 0.92 0.47
20.59 4.05 0.88 0.69 3.82 0.79 0.39
22.00 4.26 0.97 0.78 4.32 0.91 0.46
24.27 5.02 2.77 0.75 4.18 0.88 0.42
26.83 4.75 1.20 0.94 4.99 1.01 0.56
22.80 4.65 1.06 0.71 4.00 0.85 0.42
20.59 4.08 0.95 0.65 3.75 0.77 0.39
23.25 4.69 1.10 0.83 4.66 1.01 0.52
17.93 3.29 0.76 0.61 3.22 0.66 0.36
17.76 3.53 0.84 0.64 3.43 0.74 0.37
22.82 4.54 1.01 0.69 3.92 0.84 0.43
20.80 4.31 0.92 0.66 3.61 0.80 0.39
21.27 4.24 0.90 0.68 3.72 0.77 0.30
20.30 3.98 0.87 0.60 3.30 0.70 0.34
17.94 3.48 0.70 0.63 3.42 0.78 0.41
20.37 3.98 0.95 0.70 3.73 0.83 0.43
14.82 2.89 0.86 0.51 2.78 0.57 0.28
21.03 4.10 0.94 0.72 3.83 0.85 0.45
21.68 4.20 0.95 0.74 3.90 0.87 0.43
23.75 4.96 1.17 0.80 4.61 1.06 0.52
23.96 4.57 1.07 0.81 4.64 1.03 0.51
25.51 4.72 1.02 0.83 4.76 1.11 0.56
30.93 6.33 1.39 0.86 4.94 1.01 0.48
19.09 3.87 0.86 0.63 3.30 0.66 0.27
13.26 2.83 0.85 0.53 3.13 0.62 0.28
29.88 6.25 1.34 0.95 5.59 1.24 0.60
25.16 5.12 1.11 0.74 4.10 0.88 0.44
23.56 4.86 1.14 0.71 4.01 0.84 0.40

21.63 4.45 0.97 0.75 4.26 0.92 0.38
Nd
2.41
2.07
1.97
2.27
2.30
2.68
2.13
2.61
2.35
3.14
2.30
2.22
2.83
1.98
2.09
2.40
2.15
2.05
1.81
2.26
2.34
1.58
2.43
2.60
2.85
2.85
3.01
2.67
1.89
1.88
3.82
2.29
2.31
2.52
Yb
0.34
0.27
0.27
0.30
0.32
0.37
0.30
0.38
0.33
0.40
0.32
0.30
0.39
0.29
0.29
0.33
0.29
0.29
0.23
0.30
0.30
0.23
0.34
0.35
0.42
0.38
0.41
0.34
0.25
0.28
0.48
0.32
0.29
0.35
Lu
329
SiO2
68.1
60.5
52.7
63.2
53.7
56.3
63.1
60.2
66.6
59.1
56.6
63.1
54.6
55.9
52.4
61.7
55.7
56.9
54.5
62.9
45.9
53.9
51.6
54.0
49.2
49.7
52.6
54.6
61.7
63.3
58.3
54.0
62.0
64.2
67.6
61.1
55.7
65.0
50.1
54.1
51.2
47.7
48.6
50.9
44.1
47.5
50.5
50.9
56.1
57.5
Sample
RY-1
RY-2
RY-3
RY-4
RY-5
RY-6
RY-7
RY-8
RY-9
RY-10
ZF-1
ZF-2
ZF-3
ZF-4
ZF-5
ZF-6
ZF-7
ZF-8
ZF-9
ZF-10
JM-1
JM-2
JM-3
JM-4
JM-5
JM-6
JM-7
JM-8
AS-1
AS-2
AS-3
AS-4
AS-5
AS-6
AS-7
AS-8
AS-9
AS-10
BS-1
BS-2
BS-3
BS-4
BS-5
BS-6
BS-7
BS-8
BS-9
BS-10
HR-1
HR-2
8.8
9.1
9.3
6.4
10.1
10.8
10.8
10.5
8.7
9.6
10.1
10.2
9.6
8.2
9.6
8.5
8.8
8.9
9.4
9.9
10.3
11.1
9.8
10.7
9.3
10.1
9.8
10.0
9.5
10.9
8.3
9.5
11.0
9.5
10.1
11.4
11.0
10.2
11.6
9.4
12.0
13.0
11.3
9.8
11.2
9.6
9.0
8.8
8.4
9.8
Al2O3
1.5
12.2
7.6
14.6
16.3
10.0
9.8
2.5
15.4
14.4
8.3
15.4
14.4
15.9
13.5
15.6
15.2
15.6
14.7
14.1
15.0
15.6
15.9
16.8
16.0
16.9
19.1
8.8
7.9
9.0
12.0
10.8
10.9
6.3
7.5
8.1
7.9
21.0
19.4
22.9
20.6
20.5
21.8
25.9
18.3
18.7
18.2
12.2
9.6
9.8
CaO
0.8
2.4
2.4
0.9
1.9
3.0
2.5
2.3
1.5
2.2
2.5
2.3
2.5
2.1
2.4
2.0
2.2
2.1
2.3
2.2
3.7
3.2
4.8
3.1
2.1
3.1
2.7
2.2
1.8
2.0
1.7
2.1
2.3
1.3
1.5
1.9
2.2
2.7
3.2
4.5
3.8
4.0
2.4
2.5
4.6
3.0
3.6
2.9
3.2
1.1
MgO
5.8
5.8
5.7
3.7
5.1
6.2
6.1
9.9
4.5
4.9
5.4
5.2
5.4
4.5
4.9
5.2
4.8
4.8
5.1
5.2
5.4
5.9
4.9
5.8
5.1
5.5
5.4
5.5
5.2
6.1
4.4
5.1
6.1
5.3
5.7
6.2
6.3
5.2
5.5
4.4
5.8
6.4
5.4
4.7
6.2
5.9
5.7
6.3
6.8
5.3
Fe2O3
1.07
1.07
1.13
0.63
0.78
1.20
1.16
1.45
0.71
1.10
1.13
1.04
1.09
0.96
1.11
1.01
1.11
1.03
1.17
0.97
1.02
1.07
0.91
1.05
0.97
1.04
1.06
1.04
1.01
1.22
0.88
1.02
1.15
1.01
1.07
1.14
1.18
0.83
0.84
0.80
0.90
1.00
0.88
0.75
0.99
0.95
0.91
1.26
1.27
0.85
TiO2
0.2
0.2
0.1
0.2
0.2
0.1
0.2
0.3
0.1
0.1
<0.1
0.1
<0.1
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.2
0.1
0.1
0.2
0.1
0.2
0.1
0.2
0.5
0.4
0.4
0.6
0.6
0.5
0.4
0.6
0.4
0.5
0.2
0.2
0.2
P2O5
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
SO3
122
552
112
86
743
688
209
221
215
96
3426
467
1230
1491
4892
704
96
54
68
69
26
78
78
83
69
46
77
86
56
109
51
44
69
71
71
68
66
83
109
80
102
112
103
66
96
107
93
71
75
36
Zn
16
25
22
<10
27
18
22
24
<10
12
18
19
19
<10
18
26
18
14
20
22
13
19
23
22
18
16
17
16
19
25
14
19
23
25
16
16
16
18
25
24
29
30
26
23
29
30
26
25
26
10
Co
112
1163
133
48
137
142
106
119
147
81
1345
92
72
72
266
5393
179
42
61
48
45
48
47
55
47
50
47
44
43
52
40
43
46
45
47
48
49
67
80
71
84
96
77
71
84
74
75
48
60
248
Ni
516
316
328
453
1315
452
539
444
643
410
363
362
399
415
408
410
426
397
435
539
730
587
535
627
1064
555
493
469
465
807
395
331
444
323
709
614
416
2922
2208
2233
2524
3270
2002
2343
2483
2093
2277
399
313
360
Ba
488
610
650
259
424
596
550
787
281
455
463
416
513
417
479
452
464
438
530
443
458
493
447
504
458
485
467
514
524
631
412
516
621
521
536
648
641
501
663
528
717
743
695
552
666
664
584
594
610
404
Mn
Cr
91
141
104
97
109
104
121
147
179
101
556
178
98
99
106
738
99
95
107
87
95
102
92
110
96
98
97
86
83
101
80
93
100
86
89
100
105
137
156
136
147
171
125
134
153
137
138
110
116
97
SiO2 to SO3 values are in weight percent. Zn to Lu values are in PPM.
93
98
87
64
102
103
93
158
86
80
98
96
83
91
104
88
91
94
94
94
105
99
94
101
97
105
107
89
81
92
77
89
102
80
80
91
103
113
114
102
118
139
101
108
131
114
118
110
117
97
1.8
1.7
1.8
1.4
1.9
1.9
1.8
3.1
1.5
1.5
1.8
1.7
1.8
1.8
2.0
1.6
1.5
1.6
1.6
1.5
1.6
1.6
1.5
1.7
1.6
1.7
1.6
1.6
1.5
1.7
1.4
1.5
1.7
1.5
1.6
1.7
1.7
1.7
1.79
1.40
1.79
1.90
1.65
1.50
1.88
1.84
1.79
1.78
1.96
1.7
Be
49
477
56
35
540
491
82
80
120
38
2662
318
908
1111
3788
1146
114
18
24
17
15
14
13
13
13
16
17
11
10
14
12
8
15
32
14
13
14
20
46
28
43
43
39
16
23
21
20
15
14
14
Cu
92
229
167
218
377
220
253
117
298
382
288
374
394
368
365
377
385
376
374
397
551
442
480
448
552
541
534
338
294
271
257
319
342
282
234
312
322
757
622
794
688
752
575
674
620
536
545
373
284
285
Sr
TABLE 2: ICP ANALYSIS OF SOUTHERN PALESTINIAN WORKSHOP WASTERS
La
27
27
27
18
23
28
28
40
20
22
22
22
23
20
24
21
31
22
24
21
23
23
21
23
21
23
23
23
22
25
20
22
26
21
23
24
25
30
36
31
37
34
34
32
39
33
33
27
28
25
34
43
43
32
39
55
60
82
45
46
45
47
46
42
45
43
45
43
47
41
42
44
38
33
25
25
24
21
20
23
19
21
23
18
20
21
21
30
33
31
34
44
33
35
48
50
52
41
47
19
67
70
75
44
60
76
72
90
52
67
42
61
48
38
31
42
72
57
64
59
60
60
53
63
52
57
57
59
59
68
48
56
65
54
54
63
65
57
60
52
63
66
50
59
65
63
61
66
66
61
Ce
Pr
8.2
8.5
8.6
5.7
7.2
9.0
8.9
10.8
6.9
7.6
4.8
7.2
5.5
3.8
3.5
5.0
7.3
6.7
7.4
7.1
7.2
7.2
6.7
7.9
6.3
7.0
7.0
7.0
6.9
7.7
5.8
6.6
7.7
6.0
6.5
7.4
7.7
8.1
8.4
7.5
8.9
9.9
7.4
8.1
9.5
8.6
8.6
7.8
8.0
7.7
33
35
35
22
29
36
35
43
27
30
19
29
22
15
14
19
29
27
29
29
29
29
26
31
26
28
28
28
27
30
23
27
31
24
26
29
30
33
33
30
35
40
30
33
38
35
35
31
31
31
Nd
Sm
6.8
7.0
7.2
4.6
5.7
7.5
7.3
8.7
5.6
6.1
3.8
6.1
4.4
3.1
2.9
4.1
5.8
5.5
6.0
6.0
6.0
5.8
5.4
6.3
5.4
5.8
5.7
5.7
5.6
6.2
4.8
5.4
6.2
4.9
5.2
6.3
6.3
6.8
6.8
6.1
7.4
8.3
6.4
6.8
7.8
7.2
7.2
6.4
6.6
6.3
Eu
1.70
1.68
1.65
1.13
1.47
1.80
1.76
2.06
1.39
1.41
0.96
1.44
1.06
0.73
0.66
0.93
1.32
1.31
1.42
1.50
1.50
1.48
1.36
1.67
1.41
1.47
1.35
1.38
1.32
1.56
1.18
1.26
1.54
1.21
1.36
1.49
1.51
1.30
1.40
1.30
1.50
1.30
1.40
1.50
1.70
1.60
1.65
1.56
1.55
1.52
Tb
0.90
0.96
0.97
0.61
0.74
1.00
0.98
1.23
0.73
0.84
0.54
0.83
0.62
0.43
0.39
0.56
0.82
0.79
0.86
0.86
0.85
0.86
0.78
0.90
0.78
0.84
0.80
0.83
0.77
0.89
0.65
0.80
0.94
0.76
0.80
0.92
0.88
1.05
1.06
0.90
1.08
1.31
0.95
1.00
1.20
1.08
1.06
0.90
0.91
0.86
Dy
5.0
5.6
5.4
3.5
4.2
5.7
5.7
6.7
4.1
4.7
3.1
4.7
3.6
2.5
2.2
3.3
4.7
4.4
4.9
5.0
4.8
5.0
4.6
5.0
4.6
4.8
4.9
4.7
4.5
5.2
3.9
4.4
5.2
4.2
4.3
5.0
5.1
5.7
6.2
5.5
6.3
7.6
5.7
5.9
7.2
6.2
6.1
5.0
5.1
4.8
Ho
Tm
0.96
1.09
1.08
0.70
0.84
1.15
1.13
1.34
0.84
0.96
0.62
0.95
0.69
0.49
0.44
0.67
0.94
0.87
1.00
0.98
0.97
0.99
0.88
0.97
0.89
0.96
0.96
0.95
0.83
1.03
0.78
0.87
1.03
0.83
0.87
1.01
1.02
1.23
1.24
1.18
1.33
1.61
1.17
1.22
1.48
1.24
1.24
1.00
1.04
0.39
0.46
0.44
0.28
0.33
0.47
0.48
0.54
0.32
0.41
0.27
0.40
0.30
0.23
0.19
0.28
0.40
0.38
0.42
0.41
0.41
0.40
0.38
0.38
0.38
0.42
0.41
0.38
0.36
0.44
0.33
0.37
0.42
0.33
0.35
0.41
0.43
0.50
0.50
0.52
0.54
0.67
0.48
0.53
0.60
0.54
0.51
0.42
0.43
0.94 0.38
Yb
2.6
3.0
3.0
1.9
2.1
3.1
3.2
3.5
2.2
2.7
1.8
2.6
2.0
1.4
1.3
1.8
2.8
2.4
2.9
2.8
2.6
2.7
2.5
2.7
2.5
2.8
2.6
2.4
2.4
2.9
2.2
2.5
2.8
2.3
2.4
2.8
2.7
3.3
3.3
3.3
3.4
4.4
3.1
3.3
3.9
3.2
3.3
2.7
2.8
2.5
Lu
0.37
0.45
0.45
0.27
0.31
0.48
0.48
0.51
0.34
0.42
0.25
0.39
0.31
0.21
0.20
0.30
0.41
0.39
0.46
0.39
0.39
0.39
0.39
0.38
0.38
0.40
0.40
0.38
0.36
0.43
0.32
0.37
0.42
0.34
0.36
0.41
0.43
0.52
0.53
0.49
0.54
0.65
0.48
0.53
0.60
0.51
0.48
0.43
0.42
0.36
330
SiO2
54.8
56.3
50.6
54.1
49.3
45.5
41.7
45.4
44.9
44.3
48.7
45.1
53.5
61.1
67.2
64.6
63.2
59.4
65.3
69.9
71.1
74.1
65.3
64.4
67.8
62.4
56.2
61.2
62.8
60.6
56.2
71.1
66.1
52.4
53.7
62.1
52.1
49.4
46.8
46.7
46.7
47.9
47.4
38.7
38.6
44.1
47.0
47.2
40.9
45.9
48.9
47.6
47.7
52.3
50.6
48.2
Sample
HR-3
HR-4
HR-5
HR-6
HR-7
SR-1
SR-2
SR-3
SR-4
SR-5
SR-6
SR-7
SR-8
ASD-1
ASD-2
ASD-3
ASD-4
ASD-5
ASD-6
ASD-7
ASD-8
ASD-9
ASD-10
MEF-1
MEF-2
MEF-3
MEF-4
TUV-1
TUV-2
TUV-3
TUV-4
TUV-5
TUV-6
SUF-1
SUF-2
SUF-3
SUF-4
SUF-5
SUF-6
SUF-7
SUF-8
SUF-9
SUF-10
BSM1
BSM2
BSM3
BSM4
BSM5
BSM6
BSM7
BSM8
BSM9
BSM10
HLU-1
HLU-2
HLU-3
7.3
6.7
5.7
5.8
6.6
6.4
7.1
6.6
5.7
6.4
5.9
6.5
5.6
5.5
6.7
6.8
6.8
7.9
9.4
7.2
7.9
10.0
11.6
11.2
10.5
10.0
10.4
12.1
10.7
11.2
12.0
12.0
9.0
10.3
9.0
10.7
10.6
10.9
11.3
10.2
11.6
9.5
8.6
8.8
11.2
10.9
9.9
8.3
8.9
11.6
10.1
10.5
11.4
10.0
10.3
8.9
Al2O3
CaO
10.3
15.4
15.0
11.1
21.9
25.4
19.2
21.7
22.1
19.8
24.6
18.7
6.9
7.5
5.4
6.9
10.2
6.0
6.4
7.0
7.7
6.0
14.2
12.2
10.9
11.1
10.9
14.3
9.0
10.8
7.7
12.0
23.0
28.0
15.6
26.0
25.1
25.8
23.0
28.4
21.1
24.8
27.1
26.9
21.7
24.8
28.0
20.8
20.1
28.7
27.6
26.3
18.2
20.1
19.1
10.1
3.5
3.4
4.1
3.5
3.3
2.7
3.2
3.6
3.2
3.5
3.0
3.6
2.1
2.2
1.9
1.9
1.9
1.7
2.1
1.3
1.4
2.0
2.8
2.3
2.2
2.5
2.2
2.5
2.1
1.5
2.0
2.4
1.9
2.3
1.7
2.1
2.5
2.7
2.5
2.6
2.6
3.1
2.2
2.7
4.4
4.0
3.0
2.4
3.5
4.1
3.2
3.4
2.6
2.7
2.9
3.8
MgO
6.4
5.7
4.9
5.2
5.4
5.0
5.6
5.3
5.0
5.6
5.0
5.8
5.1
4.7
5.7
5.5
5.0
5.8
6.6
4.6
5.0
6.2
6.1
5.7
6.0
5.7
5.7
6.7
5.9
5.5
6.5
6.4
4.1
4.8
4.3
5.1
4.9
5.2
5.5
4.7
5.6
4.84
4.62
4.61
5.85
5.58
5.07
4.11
4.36
5.49
5.00
5.15
5.64
4.85
5.07
7.5
Fe2O3
TiO2
1.18
1.08
1.03
1.04
0.94
0.96
1.02
1.07
0.95
1.08
0.96
1.11
1.05
1.07
1.13
1.06
0.96
1.08
1.25
1.04
1.00
1.17
1.14
1.12
1.17
1.08
1.11
1.28
1.07
1.07
1.20
1.19
0.78
0.86
0.94
0.87
0.89
0.89
0.92
0.82
0.95
0.88
0.80
0.70
0.91
0.92
0.81
0.71
0.74
0.90
0.80
0.87
0.91
0.83
0.86
1.39
0.2
0.2
0.4
0.2
0.2
0.2
0.2
0.2
0.3
0.1
0.2
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.2
0.2
0.1
0.2
0.3
0.3
0.2
0.3
0.3
0.3
0.3
0.3
0.4
0.3
0.4
0.4
0.4
0.5
0.5
0.4
0.4
0.5
0.5
0.5
0.3
0.4
0.4
0.3
P2O5
SO3
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
0.1
<0.1
0.4
<0.1
<0.1
<0.1
<0.1
<0.1
0.4
<0.1
0.4
<0.1
0.4
0.4
0.6
0.2
0.9
0.7
0.8
<0.1
1.2
0.4
0.2
0.1
0.4
<0.1
Zn
86
76
86
75
88
34
101
76
85
84
70
64
54
52
61
55
59
64
80
50
70
72
690
294
551
538
75
52
100
140
143
79
85
79
61
85
83
75
83
78
90
68
73
121
79
79
60
50
53
73
62
60
56
51
60
73
Co
24
22
20
20
21
20
29
21
26
39
20
23
20
18
21
22
21
23
26
20
20
26
27
27
27
25
26
29
27
28
29
31
21
22
23
25
24
28
26
23
31
17
17
18
22
28
21
15
16
23
22
22
20
18
20
29
Ni
54
47
43
43
45
45
52
45
41
77
54
49
41
43
44
43
44
46
54
45
43
56
59
55
57
54
49
57
57
57
60
64
46
52
48
77
52
52
57
52
60
50
51
64
81
77
76
57
60
74
76
85
54
50
52
63
Ba
379
373
352
239
377
375
785
588
377
414
347
341
335
269
389
328
414
342
301
315
409
442
398
424
443
315
791
411
434
934
303
393
673
640
492
548
543
819
757
742
825
692
2253
1729
1968
1890
2079
1538
1796
2591
2050
1770
495
563
847
791
Mn
566
486
465
412
429
413
473
411
361
419
359
450
488
461
546
580
511
603
700
684
618
752
765
730
849
763
792
723
926
922
796
803
455
530
600
577
528
599
647
543
676
492
506
540
677
634
559
474
470
606
585
612
711
634
677
640
Cr
107
98
95
86
108
105
114
99
102
104
110
106
94
89
93
91
79
94
107
84
87
100
115
108
108
97
101
115
97
100
113
115
78
98
95
95
97
102
101
93
108
93
95
101
136
138
134
93
111
149
130
122
93
87
88
130
106
90
89
90
94
84
98
104
103
107
96
103
72
81
80
82
77
84
107
68
83
88
106
94
96
90
93
119
87
83
107
113
64
91
80
90
95
87
89
86
92
90
82
87
122
121
116
89
80
127
114
114
85
85
93
130
Be
1.97
1.89
1.64
1.71
1.64
1.53
1.74
1.63
1.71
1.79
1.70
1.78
1.69
1.45
1.71
1.61
1.52
1.61
1.78
1.35
1.39
1.72
1.68
1.62
1.77
1.61
1.66
1.75
1.66
1.56
1.77
1.71
1.28
1.49
1.23
1.53
1.49
1.53
1.56
1.48
1.58
1.5
1.3
1.4
1.7
1.6
1.5
1.2
1.3
1.7
1.5
1.5
1.7
1.5
1.5
2.12
12
11
12
14
14
16
18
11
13
<10
<10
<10
<10
<10
<10
15
28
18
20
25
16
23
31
31
34
29
64
28
34
32
29
33
29
26
21
34
23
29
18
27
56
18
30
32
39
61
40
29
31
37
33
31
32
37
32
10
Cu
Sr
283
367
362
204
505
881
491
549
525
399
520
400
124
145
138
165
222
170
198
147
190
183
398
394
263
248
288
335
214
301
238
410
617
678
456
658
612
690
973
716
950
601
1018
896
852
912
718
646
768
972
843
1027
541
522
621
273
La
25
23
22
21
24
23
26
22
22
23
22
24
21
21
24
24
23
24
29
23
24
28
29
29
30
28
29
30
29
29
32
31
24
27
23
28
27
29
28
27
29
25
23
29
35
34
32
25
29
34
32
32
28
25
26
29
51
49
48
51
52
44
49
45
49
49
51
54
49
46
43
41
35
35
38
29
27
30
24
23
29
26
27
27
25
21
25
24
20
22
18
23
20
22
21
21
21
24
22
31
36
35
35
26
30
32
33
32
22
19
20
56
61
56
52
48
52
52
59
52
49
55
52
58
57
55
65
57
56
58
68
57
56
57
59
54
66
60
60
63
57
61
64
60
45
51
45
51
50
52
52
48
56
51
37
47
55
55
49
41
45
56
50
50
54
47
51
75
Ce
Pr
7.2
6.6
6.3
5.7
6.5
6.2
7.2
6.3
6.1
6.5
6.5
7.0
6.5
6.4
7.4
6.9
6.6
6.9
8.0
6.3
6.6
6.7
7.0
6.5
7.6
7.0
7.0
7.6
6.9
6.9
7.6
7.1
5.6
6.4
5.4
6.4
6.2
6.4
6.4
5.9
6.8
6.3
4.8
6.7
7.8
7.8
7.3
5.9
6.6
7.9
7.3
7.1
6.5
5.7
6.2
9.0
29
27
25
23
26
25
29
25
24
26
26
28
27
25
29
27
26
27
32
25
26
27
28
26
30
28
27
30
27
27
31
29
22
25
21
25
24
26
25
24
27
25
19
27
31
32
29
24
27
32
29
29
26
22
24
36
Nd
Sm
5.8
5.3
4.9
4.8
5.2
5.1
5.9
5.1
4.9
5.3
5.2
5.6
5.3
5.1
6.1
5.5
5.4
5.5
6.6
5.0
5.2
5.4
5.7
5.4
6.2
5.9
5.5
6.3
5.6
5.7
6.1
5.9
4.5
5.0
4.3
5.3
5.0
5.4
5.3
4.7
5.6
5.1
3.8
5.6
6.4
6.7
6.2
4.8
5.5
6.3
6.1
6.1
5.0
4.8
5.2
7.4
Eu
1.51
1.32
1.17
1.14
1.31
1.21
1.43
1.29
1.25
1.33
1.28
1.38
1.29
1.23
1.40
1.37
1.35
1.43
1.60
1.23
1.23
1.35
1.34
1.30
1.49
1.32
1.37
1.53
1.30
1.44
1.49
1.42
1.08
1.27
1.08
1.31
1.22
1.30
1.26
1.20
1.34
1.25
1.14
1.51
1.73
1.77
1.71
1.36
1.51
1,64
1.53
1.50
1.24
1.12
1.21
1.87
Tb
0.85
0.79
0.74
0.68
0.72
0.72
0.80
0.74
0.69
0.77
0.74
0.81
0.76
0.71
0.83
0.74
0.72
0.79
0.88
0.69
0.75
0.72
0.80
0.73
0.87
0.75
0.79
0.87
0.76
0.78
0.86
0.79
0.62
0.70
0.62
0.71
0.69
0.71
0.74
0.71
0.76
0.70
0.50
0.79
0.91
0.91
0.87
0.70
0.76
0.90
0.89
0.85
0.73
0.60
0.66
1.00
Dy
4.7
4.5
4.4
3.8
4.3
4.1
4.6
4.2
4.0
4.3
4.3
4.7
4.4
3.9
4.6
4.3
4.2
4.4
5.0
4.1
4.1
4.2
4.4
4.2
4.8
4.4
4.5
4.8
4.2
4.5
4.8
4.5
3.5
4.0
3.6
4.2
3.9
4.1
3.9
4.0
4.3
3.9
2.8
4.6
5.3
5.3
5.2
4.1
4.6
5.2
5.1
5.1
4.0
3.7
3.7
5.8
Ho
Tm
0.92
0.90
0.84
0.74
0.87
0.80
0.91
0.82
0.82
0.87
0.86
0.92
0.84
0.80
0.90
0.85
0.81
0.87
0.98
0.80
0.78
0.82
0.87
0.81
0.96
0.85
0.88
0.96
0.86
0.92
0.94
0.88
0.74
0.81
0.71
0.83
0.79
0.81
0.79
0.81
0.82
0.79
0.56
0.91
1.07
1.09
1.10
0.88
0.95
1.03
1.05
1.02
0.80
0.72
0.74
0.38
0.39
0.36
0.33
0.36
0.33
0.36
0.34
0.32
0.37
0.35
0.38
0.38
0.33
0.38
0.36
0.35
0.37
0.40
0.32
0.34
0.36
0.37
0.35
0.39
0.37
0.38
0.41
0.36
0.38
0.40
0.38
0.30
0.34
0.31
0.36
0.33
0.34
0.33
0.34
0.34
0.33
0.24
0.41
0.44
0.45
0.44
0.37
0.39
0.44
0.44
0.44
0.34
0.30
0.31
1.13 0.48
Yb
2.6
2.5
2.5
2.2
2.3
2.2
2.4
2.3
2.2
2.4
2.3
2.5
2.4
2.3
2.5
2.4
2.2
2.3
2.7
2.2
2.4
2.3
2.4
2.3
2.6
2.4
2.5
2.7
2.2
2.5
2.6
2.5
2.0
2.2
2.0
2.3
2.1
2.1
2.1
2.2
2.2
2.2
1.6
2.5
2.9
3.0
3.0
2.3
2.6
2.9
2.8
2.8
2.1
2.0
2.0
3.0
Lu
0.39
0.38
0.39
0.33
0.36
0.33
0.37
0.34
0.34
0.35
0.36
0.39
0.37
0.36
0.39
0.35
0.33
0.35
0.41
0.35
0.34
0.35
0.37
0.36
0.39
0.37
0.38
0.39
0.36
0.38
0.38
0.36
0.31
0.34
0.31
0.34
0.33
0.32
0.33
0.33
0.33
0.33
0.24
0.41
0.44
0.45
0.45
0.36
0.39
0.45
0.44
0.45
0.34
0.29
0.30
0.47
331
SiO2
56.2
52.9
58.4
54.3
51.0
54.5
51.8
63.2
69.9
69.5
57.6
61.1
54.0
54.1
56.1
60.7
61.6
52.8
54.3
55.1
52.9
53.3
61.9
61.8
Sample
HLU-4
HLU-5
HLU-6
HLU-7
HLU-8
HLU-9
HLU-10
LSN-1
LSN-2
LSN-3
LSN-4
LSN-5
QTR-1
QTR-2
QTR-3
QTR-4
QTR-5
MHJ-1
MHJ-2
MHJ-3
MHJ-4
MHJ-5
GIV-1
GIV-2
12.3
11.9
11.0
12.8
10.1
12.8
13.2
12.1
13.3
12.2
14.2
10.7
11.8
12.9
12.1
12.7
11.1
12.3
12.5
12.2
11.0
11.9
13.4
10.7
Al2O3
CaO
20.8
18.9
18.5
21.1
18.7
19.7
12.0
7.8
6.3
16.8
13.6
14.0
12.8
10.5
8.6
12.1
13.5
10.5
9.5
11.7
13.9
14.7
13.7
15.4
3.2
3.1
2.9
3.5
2.2
4.0
2.7
2.2
1.8
2.7
3.0
2.3
2.2
2.4
1.9
2.6
2.2
2.4
2.2
2.2
1.9
2.7
3.0
2.8
MgO
5.85
5.70
5.20
6.12
4.65
5.92
6.70
6.24
6.80
6.03
7.14
5.28
5.95
6.57
6.05
6.15
5.47
6.16
6.44
6.09
5.51
6.02
6.72
5.06
Fe2O3
TiO2
0.99
0.98
0.86
0.98
0.78
0.91
1.17
1.12
1.28
1.11
1.26
0.97
1.07
1.18
1.12
1.15
1.03
1.11
1.15
1.10
1.04
1.09
1.18
0.89
0.4
0.2
0.4
0.3
0.2
0.4
0.3
0.2
0.2
0.2
0.2
0.1
0.2
0.1
0.1
0.2
0.1
0.1
0.1
0.1
0.2
0.1
0.1
0.4
P2O5
SO3
<0.1
<0.1
<0.1
0.1
0.1
0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
<0.1
Zn
71
79
57
85
62
89
67
64
66
66
34
48
53
59
60
55
51
56
56
58
51
51
56
60
Co
23
21
20
23
18
21
27
25
23
25
30
21
24
28
25
25
22
25
26
24
26
24
28
18
Ni
60
60
54
64
48
82
60
59
57
57
67
52
58
63
60
61
54
60
62
56
56
60
64
53
Ba
487
504
521
456
425
2165
398
395
369
412
424
465
911
659
695
437
583
630
675
624
552
410
412
416
Mn
741
723
673
754
611
713
786
759
856
739
669
702
701
783
775
788
620
734
781
781
588
883
913
655
Cr
98
103
91
102
80
152
117
108
121
115
126
91
100
117
105
111
100
110
114
113
103
109
121
92
88
101
89
100
88
121
116
108
98
101
126
74
100
101
87
111
80
93
93
88
79
95
99
78
Be
1.7
1.8
1.6
1.8
1.4
1.9
1.8
1.7
1.7
1.6
1.9
1.5
1.5
1.7
1.7
1.7
1.5
1.7
1.7
1.7
1.5
1.7
1.9
1.5
35
33
40
32
28
38
37
33
34
37
28
28
38
38
36
41
35
38
36
37
36
25
31
29
Cu
Sr
538
400
481
605
499
771
388
363
369
486
416
286
431
308
346
390
292
335
308
280
284
336
356
427
La
30
30
26
31
24
37
31
29
30
29
33
27
26
30
30
30
27
29
28
28
27
30
32
25
23
22
20
22
18
32
22
21
22
20
24
19
20
20
20
21
20
20
21
20
20
21
22
20
61
60
53
62
50
62
63
62
62
59
65
53
52
59
57
59
52
58
59
58
53
62
67
52
Ce
Pr
7.3
7.0
6.4
7.5
6.0
8.6
7.6
7.5
7.6
7.1
7.7
6.3
6.3
7.2
6.9
7.1
6.4
6.9
7.0
7.0
6.4
7.3
7.9
6.3
29
28
26
30
24
35
30
30
30
28
31
25
25
28
27
28
25
27
28
28
25
29
31
24
Nd
Sm
5.9
5.6
5.2
6.1
4.7
7.0
6.3
6.0
6.1
6.0
6.3
5.1
5.1
5.9
5.6
5.7
5.2
5.6
5.4
5.8
5.2
5.9
6.4
5.1
Eu
1.41
1.34
1.25
1.42
1.14
1.30
1.52
1.51
1.55
1.42
1.50
1.27
1.36
1.45
1.38
1.37
1.26
1.41
1.40
1.42
1.22
1.36
1.55
1.24
Tb
0.79
0.73
0.69
0.80
0.61
0.94
0.78
0.75
0.77
0.72
0.85
0.65
0.68
0.71
0.71
0.71
0.64
0.71
0.72
0.69
0.66
0.70
0.80
0.65
Dy
4.2
4.2
3.7
4.4
3.5
5.5
4.3
4.3
4.4
4.0
4.8
3.7
3.9
4.2
4.0
4.3
3.7
4.0
4.1
4.0
3.7
4.0
4.5
3.7
Ho
Tm
0.87
0.82
0.74
0.84
0.68
1.16
0.86
0.81
0.87
0.79
0.93
0.71
0.76
0.81
0.79
0.81
0.76
0.79
0.80
0.75
0.73
0.82
0.91
0.35
0.35
0.32
0.35
0.30
0.46
0.35
0.35
0.35
0.33
0.38
0.31
0.32
0.34
0.34
0.33
0.32
0.33
0.33
0.33
0.32
0.33
0.37
0.75 0.30
Yb
2.4
2.3
2.0
2.3
1.9
3.1
2.3
2.3
2.4
2.2
2.6
2.0
2.2
2.2
2.3
2.3
2.1
2.2
2.2
2.1
2.0
2.3
2.5
2.1
Lu
0.35
0.34
0.31
0.33
0.30
0.46
0.35
0.35
0.35
0.33
0.38
0.31
0.31
0.33
0.32
0.34
0.31
0.34
0.34
0.33
0.30
0.35
0.37
0.32
332
Sample
ELAM 13
ELAM 15
ELAM 23
ELAM 24
ELAM 25
ELAM 26
ELAM 27
ELAM 28
ELAM 29
ELAM 30
ELAM 31
ELAM 32
ELAM 33
ELAM 34a
ELAM 34b
ELAM 35
ELAM 36
ELAM 37
ELAM 38
ELAM 39
ELAM 40
EA
EA 34
EA 33
EA 32
EA 38
EA 234
EA 246
EA 243
EA 242
EA 285
EA 287
EA 289
EA 290
EA 93
EA 121
EA 77
EA 312
EA 307
EA 183
EA 308
EA 309
EA 310
1.798
3.351
5.445
7.183
7.782
4.321
4.730
2.190
2.796
2.970
.340
.328
.273
.294
3.06
.385
.254
2.46
1.710
2.061
1.488
1.362
1.502
1.415
1.348
1.825
3.241
4.035
5.041
4.490
4.804
4.541
4.450
5.078
.391
.357
.398
.373
.364
.378
.269
.232
2.72
2.55
2.69
2.58
2.41
2.59
1.85
1.52
.278
.369
4.78.05 2.505 .313.026

5.722
2.92.03 2.791 .211.015
7.199
3.89.04 4.166 .290.018
.770
.740
.628
1.154
1.389
Eu
1.39
1.21
1.13
3.11
3.05
3.98
3.19
2.91
Hf
.705.011
.964.013
3.82.09
2.31.08
1.099
.971
.982
1.010
.945
1.070
.784
.648
1.369
1.319
1.191
1.155
1.335
1.219
1.324
1.048
.883
1.777
8.62
8.49
10.29
8.15
8.55
7.42
3.07
2.71
2.60
.863.009 1.137.019 5.54.14
.460.005
.707.007
.946.009 1.271.014 5.14.10
.264
.168
.745
.828
.909
.865
.790
Ta
2.18
1.61
1.97
2.03
2.61
2.04
Yb
3.896
6.424
2.93
.278
Lu
4.29.04 2.086 .274.017
1.66
1.650
4.633
Sm
.551
.480
.490
.510
.486
.491
.264
.310
.756
.442
.278
.374
.344
.146
.144
.093
.404
.347
.374
.351
.328
Ti
%
6.28
7.73
8.85
9.11
8.72
8.82
21.10
18.73
3.09
13.49
13.36
14.16
11.97
28.03
28.96
31.09
16.61
10.27
8.87
12.59
12.3
Ca %
7.94
7.36
Th
.511
6.11
5.33
5.61
5.67
6.00
6.45
4.48
4.06
.679
.601
.648
.687
.622
.674
.219
.391
.371
6.32.09 1.032
4.29.07 .162
13.73
17.21
12.74
27.17
26.80
37.85
28.45
27.0
La
26.75
23.14
24.41
25.27
23.79
25.33
23.98
18.38
36.17
23.30
24.30
20.55
24.73
.215 20.37.97
.197
.283
.321
1.022
.524
.537
.507
Na
%
5.15.07 .351
6.27.06
2.55
2.42
1.61
7.30
7.29
14.48
Values are in PPM. For elements marked with % values are in weight percent.
96
91
96
93
89
99
110
123
150
1052
672
1052
1262
194
157
128
126
88
133
97
93
Cr
1.85
1.18
1.55
1.49
1.17
1.47
2.72
2.26
11.22
3.26.24
3.54.16
2.36.14
2.36.15
1.3
1.2
.7
3.5
4.79
14.2
4.89
4.52
Cs
7.73
6.97
5.86
10.76
11.29
16.41
12.04
11.45
Sc
56
146
15
44
79
143
70
90
Rb
6.69
6.71
8.37
10.37
15.08
21.35
15.46
14.16
Co
1.82
1.72
1.70
2.95
3.07
5.13
3.26
3.01
Fe %
2.30
2.11
1.59
4.60
5.15
8.81
5.27
5.18
Al
%
4.40
9.5.2 2.34.04 4.61
4711
3210
4310
6011
3310
5811
48
78
107
12.20
10.17
10.36
11.23
10.29
11.95
9.76
9.06
13.88
418
507
427
288
30
388
40
40
115
15.97
13.77
14.17
16.04
14.07
17.13
10.77
11.44
10.27
3.69
3.14
3.17
3.54
3.16
3.71
2.94
2.56
5.99
475
216
381
579
119
154
193
277
592
8.87
650
559
Mn
4.99
4.23
4.42
4.80
4.21
4.89
6.15
3.27
645
595
608
632
627
726
417
432
7.08 1191
3511 12.49.12 616 13.8.4 3.25.05 7.19
208 10.72.11 676
638 10.65.11 485 12.7.3 2.8.04
577 12.44.12 454 16.1.3 3.36.05 5.22
55
55
82
50
71
111
86
73
Ni
TABLE 3: RESULTS OF NEUTRON ACTIVATION ANALYSIS OF SELECTED AMARNA TABLETS by Michal Artzy
SUPPLEMENT
CATALOGUE OF LATE BRONZE AGE II SITES IN THE EGYPTIAN PROVINCE OF CANAAN
INTRODUCTION
All recorded sites which were inhabited in the Late Bronze Age II (LBA II) are listed. The data at hand,
especially in the case of survey sherds, does not permit a clear distinction to be made between the LBA
IIA and the LBA IIB (14th and 13th centuries BCE respectively). Late Bronze Age I or Late Bronze Age
III (12th century) sites, which were possibly inhabited in the LBA II were also included, but were marked
Q (Questionable).1
Geographically, the list covers the territory of the Egyptian province of Canaan: from the desert
fringe in the south to the basin of Nahar el-Kebir and the Orontes near Homs in the north2 and from the
desert in the east to the Mediterranean Sea in the west. This area constitutes the present states of Israel,
Jordan and Lebanon and the southwestern part of Syria.
There are signicant differences in our knowledge of the settlement patterns in the different parts of
this vast territory.
1. Israel is almost fully covered by intensive archaeological surveys, many of them of the full coverage
type, which include pedestrian combing. We have managed to reach much, if not all, of the available
information, including little-known Hebrew publications. Therefore it is reasonable to assume that
the list includes all major sites (Categories E and D below) and most of the large sites (Category
C) of the Late Bronze Age. Additional small sites, belonging to Categories B and A, may yet be
discovered in the future, especially under the coastal dunes and in areas of the highlands which
were intensively terraces in post-Late Bronze periods.3 As a result of this thorough coverage, the
data from Israel constitute almost 80% of the list presented below. This detailed list is extremely
important for the study of the territorial disposition in Late Bronze Age Canaan. It must be
included in every discussion of the Late Bronze Age city-states system: territorial extension of the
units, their population, their socio-economic and political strength, settlement hierarchy in their
territories, relationship between the centre and the agricultural hinterland, etc.
2. The data for Jordan is also quite reliable. The eastern Jordan valley and large parts of the
Transjordanian plateau have been surveyed and many major sites have been excavated. We believe
that we reached most of the publications relating to the Jordanian sites. Regarding excavations, it
is noteworthy that scholars working in Jordan tend to describe material which we would label as
Iron I as Late Bronze/Iron I. Hence, only sites which yielded genuine Late Bronze Age material
were included in the list.4 As for the surveys, most were not conducted as full coverage pedestrian
surveys and therefore our knowledge of the rural countryside is far from being complete. It is
also noteworthy that most surveys carried out in Jordan do not provide information on the size of
1.
2.
3.
4.
Note sites with LB III (but no LB II) sherds surveyed by Zertal in northern Samaria. They may date to the Iron I
(see Zertals report on LB III sherds at the excavated, uni-period, Iron I Bull Site 1992:169-170).
A few sites in the northern >Akkar plain, to the north of Nahr el-Kabir, are also included in the list in order to
incorporate Tell Kazel (ancient umur) in it.
For the coastal dunes see the case of Deir el-Balah; for the highlands see the case of Sataf, where the Chalcolithic
and Early Bronze I remains were buried under later terraces (Gibson et al. 1991).
Among the sites which were omitted for this reason, one should note Balu> and Lehun in Moab.
333
3.
4.
the sites or a breakdown of the nds by period, making the estimate of the size of a multi-period
site during a given period of occupation extremely difcult (Finkelstein 1998). An additional
difculty must also be taken into consideration. The quantity of Aegean imports one of the keys
for identifying Late Bronze Age sites decreases as one goes east, away from the Mediterranean
ports. In any event, the paucity of Late Bronze sites in Jordan seems to reect a genuine situation
of a demographically depleted region. This is supported by the Amarna archive, which mentions
only two city-states east of the Jordan river and south of the Yarmuk river Pihilu and Zura.5 The
reasons for this situation are beyond the scope of this short introduction.
In Lebanon, the data for the Beqa> valley and the northern >Akkar plain are quite comprehensive, though
explorations in these regions did not include full coverage pedestrian surveys. The information on the
coast beyond the major centres and on the hilly regions to the east of the coastal strip is insufcient.
There is no doubt that many more sites will be found, especially in the former area.
Our knowledge of southwestern Syria is very limited both for the Damascus basin and for the
basalt region of the Bashan. Many more sites are expected to be discovered there. It is noteworthy
that the number of city-states in the Bashan mentioned in the Amarna archive exceeds the number
of Late Bronze Age sites known in this region.
Finally we should note that for both Lebanon and Syria we may have overlooked some of the less
accessible publications.
GENERAL NOTES
The data from excavations in major sites are usually reliable, though the reader must remember that very
few sites in Israel and the neighbouring countries revealed a full, dense stratigraphic sequence for the
Late Bronze Age. In addition, exact dating of the strata is sometimes difcult, especially in cases where
the study relies on local pottery (rather than Aegean imports and Egyptian nds). Surveys are even more
problematic. They seldom yield Aegean or Egyptian nds so that tracing the nuances of the occupational
history is extremely difcult. Many of the local sherds collected in surveys have parallels in the Middle
Bronze Age III and in the Iron Age I and therefore tracing the Late Bronze phase may be challenging.
In addition, in some of the surveys, painted Mediaeval sherds were mistakenly dated to the Late Bronze
Age.6 Finally, the reader must note that the list presented here takes into account only the sedentary
portion of the population, though in the Late Bronze Age the non-sedentary element was quite signicant
(Finkelstein 1992). This sector of the society was active mainly in the highlands and on the margins of
the arid zones. Burial elds are not included in the list either.
Names of the sites are given according to their spelling on the maps. In the case of Israel, the names are
transcribed from Hebrew, with the transcription of the Arabic name, if available, appearing in brackets.
Nicknames are given in quotation marks.
5. At least in one case related to Late Bronze Jordan, archaeology and the texts clash. Intensive excavations at the site of
ancient Dibon (Dhiban in Moab), which is supposed to be mentioned in Egyptian texts of the New Kingdom period (Kitchen
1964; Redford 1982), failed to yield Late Bronze Age material.
6. Miller (1991) reported an inated number of over 100 Late Bronze sites in the marginal area of the Kerak plateau more
than found in the entire coastal plain of Israel. As a result, the Middle BronzeLate BronzeIron I sequence there is the
complete opposite of what we know from all other areas of Palestine, on both sides of the Jordan. It seems that the surveyors
mistakenly dated Medieval painted sherds to the Late Bronze Age. This had apparently been the case in the Shechem area
survey (Campbell 1968; 1991). When revisited by the team of the Southern Samaria Survey (Finkelstein et al. 1997), a few
sites from which Late Bronze pottery was reported yielded no evidence for pre-Roman occupation; yet they were rich in
painted Islamic pottery.
334
Grid references. All references are given in the Israel Grid system, unless indicated otherwise (UTM grid
for the Lebanese coast, the >Akkar plain and the Orontes; Levant grid for the Beqa> valley in Lebanon).
If possible, grid references are listed in eight digits (accuracy of 100 m). In some cases the reference is
given in six digits only.
Size of the site in the Late Bronze Age. The sites are classied into ve size- categories:
A up to 0.3 ha. (mean 0.2); B 0.31-1 ha. (mean 0.7); C 1.1-5 ha. (mean 3); D 5.1-10 ha. (mean
7); E over 10 ha.; Q questionable Late Bronze site; ? size of the site in the Late Bronze Age cannot
be established.
Excavated sites, most of which are mounds inhabited in other periods as well, were classied
according to the combination of the overall size of site and the data revealed in the excavations. Multiperiod survey sites were classied according to the combination between the overall size of the site and
the share of the Late Bronze Age nds in the pottery assemblage collected at the site, taking into account
the difculty of identifying Late Bronze sherds in survey material (see Finkelstein 1997). Sites for which
detailed pottery data are not available are classied according to the general description of the remains.
Needless to say, classication into categories is more reliable than an attempt to estimate the precise size
of a site, but even in applying this method errors are unavoidable.
Proposed territorial afliation with a Canaanite city-state is given only for the area of present day Israel
and southern Lebanon, up to the territories of Acco and Hazor in the north. In the rest of the area, sites
were afliated regionally, with the sign # appearing before the name of the region. Abbreviations used
are: JH = Jordanian highlands; JV = Jordan Valley; LC = Lebanese coast; LH = Lebanese highlands.
The Memo eld includes a few words on the location of the site and whether excavated or surveyed.
Bibliography is given to the main and most easily accessible reference to the occupation of the site in the
Late Bronze Age, not necessarily to excavation reports. Abbreviations used are: AJFR Homs Fredericq
and Hennessy 1989; ESI Excavations and Surveys in Israel; Had. Arkh. Hadashot Arkheologiyot;
NEAEHL New Encyclopedia of Archaeological Excavations in the Holy Land.
335
336
779 842 UTM
773 849 UTM
241 233
243 244
289 214
245 239
# >Akkar
# >Akkar
# >Akkar
# >Akkar
# >Akkar
# >Akkar
# >Akkar
# >Akkar
# >Akkar
#Bashan
#Bashan
#Bashan
#Bashan
#Bashan
#Bashan
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
Bseise
Laha
Yahudiyeh
Abu Abid
Zbib
Arka, Tell
Kazel, Tell
Chas
Abu Ali (Simiriyan,
Tell)
Shihab, Tell esh-
>Ashtarah, Tell
Busra
Ash>ari, Tell
Debbeh, Tell
Sheikh Saad
Maqne, Tell
Labwa, Tell
Hosn, Tell elHizzin, Tell
Sugha, Tell
Naba>a, Tell en-
Ayyun, Tell elAin Ahle, Tell

Talbaya, Tell
2007 2463 Levant

2018 2439 Levant
1635 2080 Levant
297 248
247 249
1949 2376 Levant
2071 2503 Levant
2000 2453 Levant
1848 2248 Levant
2065 2534 Levant
2006 2440 Levant
773 845 UTM
777 840 UTM

772 842 UTM
782 844 UTM
777 844 UTM
770 847 UTM
779 825 UTM
784 845 UTM

786 841 UTM
# >Akkar
# >Akkar
Frach
Jamous, Tell
GR
Territorial
afliation
Name
C
A
B
C
?
?
C
B
C
A
B
D
?
Q
C
C
C
B
C
Q
C
C
Q
Remarks
Marfoe 1995:264-265
Marfoe 1995:266-267
Marfoe 1995:227
Braemer 1984:242-246
Weinstein 1981:20; Giveon 1965
Marfoe 1995:257
Marfoe 1995:271-272
Marfoe 1995:266
Marfoe 1995:241
Marfoe 1995:270-271
Marfoe 1995:265-266
Stubbings 1951:83
Smith 1901:344-350; Weinstein

1981:20
Abou Assaf 1968; 1969
Seeden 1983; 1986; 1988
Large tell in the Bashan, excavated

Village on a large tell; mostly classical remains,
excavated.
Tell in the Bashan, on a tributary of the Yarmuk
river, excavated.
Tell in the Hawran, surveyed..
Rock stele of Ramses II.
Site in the mid-Beqa>, surveyed.
Tell in the northern Beqa>, surveyed.
Site in the northern Beqa>, surveyed.
Tell in the Beqa>, excavated
Small site in the northern Beqa>, surveyed.
Partially destroyed tell in the northern Beqa>,
surveyed.
Small tell in the Beqa>, surveyed.
Fragment of a stele of Seti I
Medium-sized site in the Akkar plain, surveyed.

One of the largest mounds in the Akkar plain,
surveyed.
Thalman 1993:214
Surveyed
Thalman 1993:214
Site in the Akkar plain, surveyed.
Thalman 1993:214
Site in the Akkar plain, surveyed.
Thalman 1993:214
Surveyed.
Thalman 1993:214
Small site in the Akkar plain, surveyed.
Thalman 1991
One of the three largest tells of the Akkar plain;
site of Irqata, excavated
Bardre et al. 1990; Badre et al. 1994 Largest tell in the Akkar plain; site of Sumur,
excavated
Thalman 1993:214
Surveyed.
Braidwood 1940; Thalman 1993:214 Site in the Akkar plain, excavated and surveyed.
Thalman 1993:214
Thalman 1993:214
Size
Bibliography
in LB
THE SITES
337
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Damascus
area
#Damascus
area
#Damascus
area
Haql el-Gami, Tell

Deir Zenoun, Tell
Marh el-Ouassa
Bseibis, Kh.
Hirmil elNeba>a Sha>ad, Tell
Haql el-Baida
Doulab, Tell
#Damascus
area
#JH
#JH
Jalul, Tell el-
Sahab
Sakka, Tell
Deir Khabiyah, Tell
Damascus
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
#Beqa>
Zeitoun, Tell ezKamid el-Loz

Jisr, Tell elBir Dhakwa, Tell
Deir, Tell edAin Khanziri, Tell
Tanayil, Tell
Qabb Elias, Tell
Ain es-Saouda, Tell
Sirhan, Tell
Naba>a Litani, Tell
Bar Elias, Tell
Delhamiye, Tell
Ain Sofar, Tell
Abu Sherif, Tell
ashbe, Tell
Mathane, Tell elGhassil, Tell
245 142
2312 1254
287 316
258 307
272 324
2076 2557 Levant

1672 2014 Levant
2085 2595 Levant
2152 2737 Levant
2110 2724 Levant
1968 2434 Levant
2066 2521 Levant
296 309
1510 1704 Levant

1581 1872 Levant
1541 1891 Levant
1632 1960 Levant
1558 1949 Levant
1667 1980 Levant
1626 2066 Levant
1584 2061 Levant
1844 2284 Levant
1696 2056 Levant
1845 2302 Levant
1660 2040 Levant
1712 2087 Levant
1662 2093 Levant
1775 2168 Levant
1801 2229 Levant
2005 2443 Levant
1817 2199 Levant
A
C
A
A
C
C
A
Q
B
D
C
C
C
Q
B
Q
B
D
Q
B
C
B
D
C
C
C
Ibach 1987:13-14; Herr et al. 1994:

161; Younker and Merling 1999
Ibrahim 1987:76-77; AJFR: 519
al-Maqdissi 1988; Taraqji 1993.
Site included in this list solely

according to textual information
al-Maqdissi 1993
Marfoe 1995:188
Marfoe 1995:121-157
Marfoe 1995:197
Marfoe 1995:209
Marfoe 1995:209
Marfoe 1995:217-218
Marfoe 1995:222-223
Marfoe 1995:223
Marfoe 1995:243-244
Mafroe 1995:226-227
Marfoe 1995:250
Marfoe 1995:227
Marfoe 1995:227-228
Marfoe 1995:231
Marfoe 1995:235
Marfoe 1995:240-241
Marfoe 1995:265
Baramki 1961; 1964; 1966; Marfoe
1995:241-242
Marfoe 1995:273-274
Marfoe 1995:225-226
Marfoe 1995:277
Marfoe 1995:280
Marfoe 1995:280
Marfoe 1995:260-261
Marfoe 1995:274
Al-Maqdissi 1990:463
Large tell southeast of Amman, excavated
An 8 ha tell southeast of Damascus, 13 kms

north of Zakiyeh, 11 kms southeast of Katana,
surveyed.
Tell southeast of Damascus, surveyed. and
excavated
Large tell east of Madaba, excavated

Medium sized tell in the Beqa>, surveyed.
A small tell ca. 30 kms to the SE of Damascus,
surveyed.
Tell south of the Beqa>, surveyed.

Large tell in the southern Beqa>, excavated
Tell in the southern Beq>a, surveyed.
Tell in the southern Beqa>, surveyed.
Small site in the southen Beqa>, surveyed.
Site in the southern Beqa>, surveyed.
Large tell in the Beqa>, surveyed.
Tell in the Beqa> west of Baalbek, surveyed.
Round, high tell in the Beqa>, surveyed.
High tell in the Beqa>, surveyed.
Large tell in the mid-Beqa>, surveyed.
Large tell dominating the Beqa>, surveyed.
Tell in the Beqa>, excavated
338
2062 2006
205 175
205 179
204 178
205 182
2378 2364
2046 1861
2075 2065
2073 1810
203 174
(approximated)
189 327
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#JH
#LC
Abila, Tell (Qweilbeh)
Husn, Tell el-
el-Mishna
Amman Airport
Fukhar, Tell el>Umeiri, Tell el-
Amman
Dahr el-Medineh
Jarash
Rihab
Madaba, Tell
Sakhineh, Tell esDeir >Alla, Tell
Abu Kharaz, Tell

Argadat, Tell elAbu Nijrah,Tell
Qa>adan, Tell
Kharabeh, Tell elel-Baidar
Sa>idiyeh, Tell esPella (Fahl, Kh.)

Mazar, Tell elKataret es-Samra
Qraye
225 124
2070 2218
2088 1782
2530 1925
234 187
228 192
238 151
2397 2219
2343 1418
223 076
2429 1534
2329 2110
231 232
2272 1660
#JH
Umm ed-Dananir
228 160
229 218
#JH
#JH
Safut, Tell
Irbid
Q
D
A
B
B
?
?
?
?
C
B
?
C
C
Q
?
A
B
D
Fischer 1993:282-283; 1997:132-135

Yassine, et al. 1988
Kerestes et al. 1977-78:122; Yassine
et al. 1988:222
NEAEHL 4:1298-1300
Smith and Potts 1992:39
AJFR: 38; Ibrahimet al. 1976
Leonard 1979:63
Harrison et al. 2000:220

Ibrahimet al. 1976
NEAEHL I:339-340
AJFR: 488; Mittmann 1970:120
Braemer 1987
Strange 2000
Herr, personal communication on the
season of 1995-96
Bennett 1979:159; Dornemann 1983:
22
AJFR: 198-200
Worschech and Ninow 1999:172

AJFR:167-178
Leonard 1987a:359
Mare 1991:203, 205; 1996:261;

Kafa 1984
Wimmer 1987a:280; 1987b:162-165

Lenzen et al. 1985; Leonard 1987b:
261
McGovern 1986:61-63, 1989
Large tell in the Jordan valley, excavated

Large tell in the Jordan valley, excavated
Small tell in the Jordan valley, excavated
Small tell overlooking the Jordan river,
surveyed.
Cemetery southeast of Sidon, excavated
Small tell to the north of Amman, excavated

Large tell under the present town of Irbid,
excavated
Tell in the Beqah valley to the north of Amman,
excavated
Tell in the northern Jordanian plateau; Late
Bronze sherds and Late Bronze tombs nearby,
excavated
Large tell in the Irbid region, surveyed. and
excavated (soundings)
Site near Balu in Moab
The Amman airport temple; a single isolated
building, excavated
Large tell in northern Jordan
A medium-sized mound near Amman,
excavated
The Amman citadel, surveyed. (surface
collections)
A single building to the northwest of Jarash,
excavated
The tell of Jarash; MB/LB I and LB/IA I
occupations, excavated
A village on a site with mainly later remains,
surveyed.
Tell in the town of Madaba; stray LB sherds
Tell in the Jordan valley, surveyed.
Medium-sized tell in the Jordan valley,
excavated
Tell in the Jordan valley, excavated
Site in the Yarmuk valley, surveyed.
339
1592 2371
1614 2282
#LC
#LH
#Orontes
#Orontes
[URU x-x-i
]G-ma-te
[URU x-xi]G-ma-te
Khalde
Arde
Homs
Nebi Mend, Tell
Geva Shemen, Tel

(Amr, Tell el-)
Qiri, Tel
Parur, Tel (Fureir, Kh.)
[URU x-xi]G-ma-te
Yoqneam, Tel (Qeimun, [URU x-xi]G-ma-te
Tell)
URU
[
x-xBeer Tiv>on (Bir, Kh.
i]G-ma-te
El-)
Zariq, Tel (Abu Zureiq, [URU x-xTell)
i]G-ma-te
Qashish, Tel (Qassis,
[URU x-xTell el-)
i]G-ma-te
Shush, Tel (Abu
[URU x-xi]G-ma-te
Shusheh)
Bira, Tel (Bir el-Gharbi, Acco
Tell)
822 829 UTM
#LC
#LC
#LC
Q
C
B
B
B
C
C
1605 2300
1637 2354
1622 2265
1606 2323
1634 2246
1661 2563
D or
E
A
C
?
?
C
1599 2260
840 848 UTM
?
768 811 UTM
713 703 UTM

704 683 UTM
758 815 UTM
D?
Sarepta (Sarafand)
Tyre
Tripoli
719 716 UTM
#LC
D
C
Sidon
744 778 UTM

729 754 UTM
#LC
#LC
Byblos
Beirut
Bunimovitz 1989:123; NEAEHL 1:

262-263
Raban 1999:61-62
Anati et al. 1973:26, 51; Raban

1999:50
NEAEHL 4:1203
Raban 1982:24-29
NEAEHL III:805-811
Olami 1981:39-40
Ben-Tor et al. 1987
Lehmann, forthcoming
Parr 1983
Al-Moussli 1985
Saidah 1969:130
Salame-Sarkis 1973:99-100
Anderson 1988; Khalifeh 1988

Bikai 1978
Salme-Sarkis 1973:94
Dunand 1969
Montet 1928
Badre 1997
Small tell on the western margin of the Jezreel

valley, completely destroyed by modern
activity, excavated
Small tell in a small valley in the hills of Ramat
Menasheh, surveyed.
Medium size tell on the western edge of the
Jezreel valley, excavated
Small tell in the northeastern sector of the
Small tell on the western margin of the Jezreel
valley, surveyed.
Tell in the northeastern sector of the Jezreel
valley, excavated
Medium-sized tell on the western margin of the
Jezreel valley, surveyed.
Medium-sized tell in the coastal plain east of
Acco, excavated
Large mound on the coast, excavated

Mound on the coast in the modern city of
Beyrouth, recently uncovered, excavated
The site is covered by later occupations and by
the modern city
Large tell, excavated
Site under the city of Tyre, excavated
The mound of Tripoli, under the Chateau de
Raymond de Saint-Gilles, excavated
Small site south of Beyrouth, excavated
Tell in the hilly area to the east of Tripoli,
Lebanon; probably location of Ardata,
excavated
The tell of the city of Homs
Sounding and survey
Size depeing on whether the lower mound was
included in the Late Bronze city. Upper tell
measures 450x200 m; lower tell adds 450x300
m., excavated
Tell to the southeast of Haifa, surveyed.
340
1585 2584
1612 2765
1590 2629
1599 2725
1569 2449
1606 2500
1604 2449
1521 2452
1644 2532
1602 2419
1648 2419
1584 2466
1587 2405
1617 2530
1910 2312
2009 2409
Acco
Acco
Acco
Acco
Achshaph
Achshaph
Achshaph
Achshaph
Achshaph
Achshaph
Anaharath
Anaharath
Abu Hawam, Tell
Kison, Tel (Keisan, Tell) Achshaph

Achshaph
Achshaph
Achshaph
Naal, Tel (Nahl, Tell

el-)
Aphek, Tel (Kurdana,
Tell)
Qiryat Ata
Par, Tel (Far, Tell el-)

>Alil, Tel (Ras >Ali, Kh.)
Zavat, Tel (es-Subat,
Tell)
Regev, Tel (Harbaj, Kh.
el-)
Da>okh, Tel (Da>uk,
Kh.)
Maseh, orvat (Masha,
Kh.)
Menorim, orvat
1645 2634
Acco, Tel
Kefar Rosh ha-Niqra
(Musheirefeh, Kh. el-)
Giv>at Yesef (Sumeiriya,
Tell)
Akhziv (Zib, ez-)
1643 2576
>Uza, orvat (>Aiyadiya, Acco

Kh. el-)
Mimas, Tell
Acco
B
B
B
D
A
1633 2682
Acco
Kabri
1588 2674
Acco
Nahariya
Garstang 1922; 1924; Had. Arkh.

109:20
IAA data base; Lehmann,
forthcoming
Zori 1977:146; Gal 1980:42-43; The
Megiddo countryside survey
ESI 6:110
Gal 2000:86
Gal 1992:21
Bunimovitz 1989:123
NEAEHL 3:864
NEAEHL 1:7-14
ESI 13:30
Bunimovitz 1989:123; ESI 110: 89
A small site to the south of Tiberias, excavated
Site in the eastern Lower Galilee , surveyed.
Medium-sized tell to the southeast of Haifa,

surveyed.
Tell with Medieval remains, surveyed.
Medium-sized tell in the coastal plain southeast

of Acco, surveyed.
A few Late Bronze sherds in an Early Bronze
site located to the southeast of Haifa, excavated
Small to Medium-sized tell on the coast of the
bay of Haifa, excavated
Large tell in the coastal plain of Acco,
excavated
Small tell to the southeast of Haifa, excavated
Tell in Naal Sippori, surveyed.
Small tell to the east of Haifa
Medium-to-large tell near the coast, in the town

of Nahariya, excavated
Very large Middle Bronze site with a small tell
Kempinski et al. 1990:39
on its southwestern side, located in the coastal
plain of the Galilee, excavated
ESI 13:20
Tell in the coastal plain of AccoMulti-period
tell, excavated
Medium-sized tell in the coastal plain of the
Frankl et al. 2001:13
Galilee, surveyed.
NEAEHL 1:20-21
Excavated
Tadmor and Prausnitz 1959; Frankel Small tell in the coastal plain north of Acco,
and Getzov 1997:72-73
excavated
ESI 4:103-104
Galilee, excavated
Prausnitz 1975; NEAEHL 1:32
Arab village on prominent tell on the coastline,
excavated
Ronen and Olami 1983:21-22
Small tell to the southeast of Haifa, surveyed.
NEAEHL 3:1089
341
1869 2276
1959 2318
Anaharath
Anaharath
1189 1301
1175 1368
1136 1298
1196 1338
1178 1295
1166 1358
Anaharath
Ashdod
Ashdod
Duq, ed-
North of Tel Ashdod

Mor, Tel
South of Ashdod Yam Ashdod
East of Ashdod
Ashdod, Tel
Mizpe Yonah (Nebi
Yunis)
Ashdod
Ashdod
Ashdod
2066 2300
Anaharath
2040 2222
2028 2324
Anaharath
Hamizre Ha-zarua
(Kuhwani)
Halukim, orvat
Qishyon, Tel (Kasyun,

Tell)
Ein Hayadid
1996 2218
2093 2322
1940 2288
Anaharath
Anaharath
Anaharath
1983 2354
Anaharath
Yin>am, Tel (Na>am,

Tell en-)
Dover, Tel
Rekhesh, Tel
(Muqarqash, Tell)
1871 2296
2019 2328
2035 2288
Anaharath
Anaharath
Ubeidiyeh, Tell
Delhemiyeh, ed-
Anaharath
1967 2356
Anaharath
Beit Jann, Kh.
Ein Dor, orvat

(Safsafeh, Kh. es-)
Ein Hadda, Tel
2005 2295
C
D
Q
B
B
C
C
B
A
2092 2322
Dover, Tel (Duweir, Tell Anaharath

ed-)
Deir, Kh. EdAnaharath
2023 2378
Anaharath
Qadish, Kh.
Had. Arkh. 20:19

NEAEHL 1:95-98
Had. Arkh. 11:4
ESI 15:131; ESI 113:107-108
Had. Arkh. 20:19; 34-35:32

NEAEHL 3:1073-1074
Had. Arkh. 17:23-24; Maeir 1997

Appendix I:5
Zori 1962:194
Maeir 1997 Appendix II:41
Zori 1977:98-99; Gal 1991:54-55
Gal 1998:65
Gal 1992:33
Small site to the south of the Sea of Galilee,

excavated
Small, multi-period site in the Jordan valley,
surveyed.
Site to the north of Tel Ashdod, surveyed.
Small tell near the coast north of the modern
city of Ashdod, excavated
Site in the dunes near the coast south of Ashdod
Yam, excavated
Site east of Tel Ashdod, surveyed.
Large tell in the coastal plain, excavated
Small site on a hill overlooking the modern city
of Ashdod, surveyed.
Medium-sized tell in the eastern Lower Galilee,

surveyed.
Medium-sized tell in the Jezreel valley near Mt.
Tabor, excavated
Small, multi-period site in the eastern Lower
Galilee, surveyed.
Surveyed.
Kochavi 1963; Maeir 1997 Appensix Medium size site to the south of Tiberias,
I:1
surveyed.
Maeir 1997 Appendix I:3
Tell on the Yarmuk south of the Sea of Galelee,
surveyed.
Gal 1992:32
Small tell in the eastern Lower Galilee,
surveyed.
ESI 14:53
Small site in the eastern Lower Galilee,
excavated
Aharoni 1957:126-127
Small tell in the Jordan valley, surveyed.
Had. Arkh. 33:14; Bunimovitz 1989: Small tell in the Jordan valley, excavated
103
NEAEHL 4:1515-1516
Small tell in the valley of Yavneel, eastern
Lower Galilee, excavated
ESI 113:19
Tell on the Yarmuk river
Gal 1981; 1992:31-32
Medium-sized tell in Naal Tavor, eastern
Lower Galilee, surveyed.
Zori 1977:113-114; Gal 1998:67
Multi-period site in the Jezreel valley, surveyed.
342
1109 1118
1070 1190
113 124 (?)
1069 1093
122 124
1425 1407
1374 1404
1440 1408
1489 1418
1525 1388
1492 1371
1397 1456
1531 1336
1487 1314
?
1334 1658
1397 1707
Beit Jirja
Ashkelon
Ashkelon, Tel
Ashkelon
Southeast of Tel Poran Ashkelon
Ashkelon
Ashkelon
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Netiv Ha->Asara
Shar (Sawar eshShamaliya)

Gezer, Tel
Malot, Tel (Melat, Tell)
Ein Yarad
East of Gezer
Sha>albim, Tel (Salbit)
Daneb el-Kalb, Kh.
Yalu
Bir el-ilu
amid, Tel (Ras Abu

amid)
South of Sha>ar Hagi

or>a, Tel (Sar>ah)
Rishon Le-Ziyyon
Bene Beraq
Qana, Tel (el-Mukhmar, Gezer

Tell)
1540 1420
1437 1404
116 123 (?)
Ashkelon
South of Nizzanim
110 125 (?)
Ashkelon
Nizzanim beach
B
Q
A
D
B
A
D
B
Gophna and Ayalon 1998:44-51
Had. Arkh. 111:37
Bunimovitz 1989:122
Lehman et al. 1996 Taf. 3:5.24
Had. Arkh. 25:15
Shavit 1992:102-103
Aharoni 1969; Shavit 1992:95-96
Shavit 1992:100-101
Shavit 1992:93
Shavit 1992:91-92
Shavit 1992:100
Shavit 1992:97-98
NEAEHL 2:502-504
ESI 12:49-50
Had. Arkh. 17:4
ESI 110:80-81.
ESI 113:110
NEAEHL 1:107; Stager 1991
Had. Arkh. 41-42:33
Had. Arkh. 41-42:33
Had. Arkh. 16:17 (LB I)
Site in the coastal dunes north of Ashkelon,

surveyed.
Site in the coastal plain northeast of Ashkelon,
surveyed.
Small tell near south of Ashkelon
Large tell on the coastline, excavated
surveyed.
Small settlement in the coastal plain south of
Ashkelon, excavated
surveyed.
Large tell in the coastal plain, excavated
Small tell in the coastal plain west of Gezer,
excavated
Large multi-period settlement east of Gezer,
surveyed.
Small site in the coastal plain east of Gezer,
surveyed.
Medium-sized tell on a hill, on the margin of
the valley of Ayalon, surveyed.
Multi-period site on the margin of the valley of
Ayalon, surveyed.
Medium-sized tell on the southern margin of
the valley of Aiyalon, surveyed.
Small tell adjacent to the Tel Aviv-Jerusalem
highway, surveyed.
Medium-sized tell in the coastal plain
near Ramla, surveyed. (the western slope,
excavated)
Surveyed.
Medium-sized tell in the Shephelah, surveyed.
Site in the coastal plain south of Tel Aviv,
surveyed.
Site in the modern town of Bene Beraq, on the
outskirts of Tel Aviv, excavated
Small tell in the Yarkon basin, surveyed.
343
Q
C
126 156
1401 1459
1258 1475
1211 1479
1247 1453
128 147
Gezer
Gezer
Gezer
Gezer
Gezer
Gezer
Azor (Yazur)
Abu Zeitun, Tell
Rishon Le-Ziyyon
dunes
Yad Rambam
Maoz, Tel (Sulan, Tell Gezer

es-)
Yavneh Yam
Gezer
Gezer
Sarsara, Kh.
Yavneh dunes
>Ayanot
1435 1678
Gezer
Gezer
Aphek, Tel
Hadid, Tel (Haditheh,

el-)
1455 1523
1406 1518
1282 1445
Shalaf, Tel (Shallaf, Tell Gezer

esh-)
Lod (Ludd, el-)
Gezer
134 167
1316 1592
1524 1629
1398 1600
Gezer
Yehud
1441 1626
Gezer
1337 1605
Gezer
1267 1625
Gezer
Mesubbim junction
(Kheiriya; Ibn Ibreiq)
Nahshonim
1320 1665
Gezer
Gerisa, Tel (Jerishe,

Tell)
Jaffa
131 174
Gezer
Michal, Tel
Had. Arkh. 54-55:45
Ram Gophna, personal

communication; ESI 20:65
Kochavi 1989
Kaplan 1957
Had. Arkh. 25:15
ESI 2:50
Singer-Avitz and Levy 1992
Dothan 1952

communication
communication
Shavit 1992:103
Had. Arkh. 21:7-8
Finkelstein et al. 1997:239
ESI 16:84
Kochavi and Beit-Arieh 1994:95
Finkelstein 1990
Kaplan 1972
NEAEHL 2:482-484
Herzog 1989
Small site in the coastal plain north of Ramla,

surveyed.
Tell in the dunes southwest of Rishon LeZiyyon, surveyed.
A large Middle Bronze site; part of it -- the
tell on the promontary and around it -- was
inhabited also in the Late Bronze. Many LB
burials in the vicinity., excavated
Site in the coastal plain south of Tel Aviv,
surveyed.
Site in the coastal plain, south of Tel Aviv,
surveyed.
Small tell in the coastal plain near Yavneh,
surveyed.
Tell covered by the old town of Lod, surveyed.
and excavated
Large tell in the Yarkon basin. Egyptian
residency, excavated
Medium-sized tell on a hill overlooking
the coastal plain, to the northeast of Ramla,
surveyed. (lower terrace, excavated)
Surveyed
Small tell on the coastline of the Sharon plain,

north of Tel Aviv, excavated
Medium-sized tell in the Yarkon basin,
excavated
Medium-sized tell covered by the old core of the
city of Jaffa. Egyptian stronghold, excavated
Small, low-prole tell in the coastal plain east
of Tel Aviv, excavated
Small site on a hill east of Tel Aviv, overlooking
the coastal plain, surveyed.
Small site in the coastal plain east of Tel Aviv,
excavated
Small settlement on the western ank of
southern Samaria hills, surveyed.
Tell in the coastal plain on the outskirts of Tel
Aviv, surveyed. (tombs, excavated)
Small tell in the Yarkon basin, surveyed.
344
1503 1175
1399 1241
1361 1277
Gath
Gath
Gath
Gath
East of Tel Zat
Sheikh Madkur, Kh.

eshShaha, orvat (eshShah, Kh.)
North of Tel Zat
1249 1181
1503 1253
1475 1240
1476 1211
1323 1346
Gath
Gath
Gath
Gath
Gath
Gath
(Gath)
Socoh, Tel (>Abbad,

Kh.)
Beth ilqiyya
South of Azekah
Beth-shemesh, Tel
1439 1227
1477 1286
1380 1153
1298 1133
1326 1177
Gath
Gath
1339 1153
Burna, Tel (Burnat, Tell)

>Erani, Tel (Sheikh
Amed el->Areini, Tell
esh-)
Zippor, Tel (Tuyur, Tell
et-)
Zanoah, Tel (Zanu>,
Kh.)
Yarmut, Tel (Yarmuk,
Kh. el-)
1456 1166
Lavnin, orvat (Tell el- Gath

Beida, Kh.)
Zeit, Tel (Zeita elGath
Kharab, Kh.)
Rujm ed-Darbi, Kh.
Gath
A
C
C
C
1504 1135
Gath
1416 1324
Gath
Batash, Tel (Batashi,

Tell el-)
Qeila, Kh.
1278 1567
Gezer
1363 1233
1491 1485
Gezer
Hatarsi, orvat (Rujm,

Kh. el-)
Holon
Dagan 1992a:117
NEAEHL 1:250
ESI 2:92
Dagan 1992a:134
NEAEHL 2:665
Dagan 1992a:92
NEAEHL 4:1526-1527
Dagan 1992a:154
NEAEHL 2:417-419
Dagan 1992a:145-146
Dagan 1992a:153
Dagan 1992a:151-152
Kochavi 1972:49; Dagan 1992a:161
NEAEHL 1:152-157
Bunimovitz 1989:122
Dagan 1992a:94-95
Dagan 1992a:149-150
Dagan 1992a:104
IAA data base
Shavit 1992:87-88
Small tell in the coastal plain east of Ashkelon,

excavated
Small-to-medim size tell in the higher
Shephelah, surveyed.
Large Early Bronze site in the upper Shephelah,
with a small Late Bronze-Iron Age tell on its
eastern end, excavated
Medium-sized tell in the upper Shephelah,
surveyed.
Site on the coastal plain east of Ashdod,
surveyed.
A small Khirbeh in the Shephelah, surveyed.
Medium-sized tell in the Shephelah, excavated
A medium-sized tell in the lower Shephelah,

surveyed.
Medium-sized site in the lower Shephelah,
surveyed.
Medium-sized tell in the Shephelah, surveyed.
Tell in the lower Shephelah near Kiriath Gat,
excavated
A large Khirbeh in the Lower Shephelah,

surveyed.
Small site in the lower Shephelah, north of Tel
Zat, surveyed.
Medium-sized tell in the lower Shephelah,
excavated
surveyed.
A large tell in the upper Shephelah, surveyed.
Multi-period site in the foothills east of Lod,

surveyed.
Small settlement in the modern town of Holon,
surveyed.
Small to medium-sized site near Tell es-Sa,
surveyed.
Tell in the upper Shephelah, surveyed.
345
Aharoni 1959; ESI 10:106-107
B
A
C
A
D
A
A
A
A
A
C
1381 1295
1356 1233
Gath
Man>am, orvat (Deir
en-Nu>man, Kh.)
Zat, Tel (a, Tell es-) Gath
Gath
1487 1248
Gath
128 126 (?)
Ginti-kirmil 1476 2038
Jat
Girit, Tel (Tell el Jariya) Ginti-kirmil 1379 2018

>Aliya, Kh. elTimmorim

Zeror, Tel (Tell Dhurur,
Kh. et-)
Hama>apil
Bahan
Esur, Tel (Asawir, Tell)
Ein el-Arais
Nami, Tel (Jezirat enGinti-kirmil 1438 2287
Nami)
Se, Tel (Sitt Leila, Tell) Ginti-kirmil 1505 2156
Hefer, Tel (Ifshar, Tell

el-)
Tel Megadim
Bunimovitz 1989:124
NEAEHL 3:1095-1098
1338 1279
Gath
C
B

NEAEHL 3:1001; ESI 20:23
NEAEHL 2:612
Neeman 1974
Porat et al. 1985:124-125; Neeman

et al. 2000:29
Porat et al. 1985:221-223
Porath et al. 1999
Porat et al. 1985:177-179
NEAEHL 4:1522-1524; ESI 20:114115

Dagan 1992a:98; ESI 13:95
Y. Dagan, personal communication
NEAEHL 4:1525
1440 1231
Gath
Azekah, Tel (Zakariya,

Tell)
Harasim, Tel
1331 1336
Tell near Beth-shemesh, surveyed.

Surveyed
Sharon, excavated
Site in the coastal plain of the Sharon,
surveyed.
Arab village on a large tell on the eastern
margin of the coastal plain, excavated (limited
salvage soundings)
Small tell on the coastline of the Sharon plain,
surveyed.
surveyed.
Medium-sized tell on the western gate of the
Aruna pass, surveyed.
Site in Ramat Menasheh, surveyed.
Small tell on the coastline south of Haifa,
excavated
Medium-sized tell in the western ank of
Ramot Menasheh, excavated
Sharon, excavated
Tell in the coastal plain to the south of Haifa,
excavated
Large tell in the lower Shephelah, excavated
Kaplan 1953:138
Small tell in the coastal plain, surveyed.
NEAEHL 3:1052-1053; Gittlen 1992 Large tell in the lower Shephelah, excavated
ESI 2:92
Site in the coastal plain east of Ashdod,
surveyed.
ESI 2:92
Site in the coastal plain east of Ashdod,
surveyed.
NEAEHL 1:123-124
Medium-sized tell on a ridge in the higher
Shephelah, excavated
Givon 1992
Large, low-prole tell in the Lower Shephelah,
excavated
Dagan 1992a:79
A large site in the Lower Shephelah, surveyed.
Gath
A
C
B
Yad Binyamin
1325 1365
1359 1318
1330 1326
Gath
Gath
Gath
Ful, Tell elMiqne, Tel

North of Revadim
346
Poratet al. 1985:150-154
A
A
B
Q
B
B
A
?

Ginti-kirmil
Ginti-kirmil
Ginti-kirmil
Ginti-kirmil
Attaisi, Kh.
Shitri, Tel (Burgeta)
Shiqmona, Tel
Afrin, Tell
Poleg, Tel
Rujum, Kh. El-
Shweikat er-Ras, Kh.
Shalaleh, Kh.
Zeita
Abhariya, Kh.
Funeitir, Kh.
Ara, Kh.
Eran, Tel
Burin, Kh.
Migdar, Horvat (elMajdal, Kh.)
Abu Shukeir, Kh.
Yoach, Horvat
Zomera, Tel (esh-Sheikh
Abu Faraj)
1539 1995
1565 2165
1581 2147
1579 2123
NEAEHL 3:1043
Q
A
Q
C
Stern 1984
Mevorakh, Tel
(Mubarak, Tell)
Mikhmoret, Tel (Minet Ginti-kirmil 1376 2009
Abu Zaburah)
Nurit, Tel (Nurieh, Tell) Ginti-kirmil 1458 1984
Small tell on the coastline, excavated
Large tell on the coastline of the Sharon,

excavated
Arab village on a tell on the eastern margin of
the coastal plain, surveyed.
Small tell in the coastal plain, excavated
Small tell in the coastal plain of the Sharon,

surveyed.
Bunimovitz 1989:124; IAA data base A small settlement in the coastal plain of Mt.
Carmel, surveyed.
Small, low-prole tell in the coastal plain of the
Porat et al. 1985:243-245
Sharon, surveyed.
NEAEHL 4:1373-1374
Small-to-Medium-sized tell in the coastal plain
of Mt. Carmel, excavated
Bunimovitz 1989:128
surveyed.
NEAEHL 4:1193
excavated
Neeman 1969
Multi-period settlement on the eastern margin
of the coastal plain of the Sharon, surveyed.
Gophna and Porat 1972:215; Zertal Medium-sized tell on a hill overlooking the
and Mirkam 2000:407-408
coastal plain of the Sharon, surveyed.
Bunimovitz 1989:125
Medium-sized site on a dominating hill in the
Carmel ridge, surveyed.
Zertal and Mirkam 2000:349-351
Possible Late Bronze site
Bunimovitz 1989:124
Gadot 1999:74
Possible LB site
Neeman 1974; Gadot 1999:71-72;
Small to medium-sized tell in Wadi >Ara (the
>Aruna pass), surveyed.
Giveon 1970
Surveyed
Zertal and Mirkam 2000:292
Possible Late Bronze site
Yannai 1993
surveyed. (burials, excavated)
Olami 1981:96
Small site in Ramat Menasheh, surveyed.
Olami 1981:45-46
Neeman et al. 2000:46
surveyed.
Porat et al. 1985:183-184
NEAEHL 1:358
Yemma, Kh.
Dor, Tel (Burj, Kh. el-)
347
2069 2742
2190 2653
2138 2804
2135 2513
2264 2463
2147 2374
2109 2338
2117 2396
2144 2423
1933 2447
2128 2927
2054 2873
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Dibbin, Tell
Shahaf, Tel (Abalis,

Tell)
Qusibiyya el-Jadida
Rawiyeh
Shuqayif (Mashrafawi)
Ateret, Tel
Abu Mudawar, Tell
Dabsha, Kh. elKhirbeh, el-
Ein el-Hariri, Tell
Ein Taruq, Khirbet
Wadi Mas>ud
Thuraya, Tell eth-
Kfar aruv
Soreg, Tel
Qarnei iin, Tel
(Horns of Hittin)
Ein Azzaziat, Tell
Ashan, orvat
2260 2584
1757 2672
1907 2757
2186 2477
2090 2677
2054 3054
2085 2930
1892 2814
Hazor
Hazor
Hasas, Tel
Shelabun, Kh.
2115 2747
Hazor
Dardara
A
A
A
B
A
C
A
Q
A
A
Q
Ilan 1999:164
Dayan 1962:25
NEAEHL 4:1410
Gal 1981; 1992:44
Epstein 1993:85
Epstein and Gutman 1972:292
Had. Arkh. 54-55:3; personal

communication from Ronnie
Ellenblum
IAA data base; Ma>oz 1986:148;
Epstein 1993:86
Frankel et al. 2001:25
Finkelstein and Frankel, unpublished
information
Epstein and Gutman 1972; Maoz
1986:149; Epstein 1993:87
Epstein and Gutman 1972:286;
Ma>oz 1986:150; Epstein 1993:90
Epstein and Gutman 1972:291
IAA data base

IAA data base; Hartal 1989:101-102
Epstein and Gutman 1972:281-282;
Ma>oz 1986:151; Epstein 1993:88
Multi-period settlement in the southern Golan,

surveyed.
Small tell in the southern tip of the Golan,
surveyed.
Small site in a wadi in the southern Golan,
surveyed.
Small tell in the southern Golan, excavated
Small tell on top of the Horns of Hittin in the
eastern Lower Galilee, excavated
Small site in the Huleh valley, surveyed.
Small tell in the southern Golan, surveyed.
Large tell in upper Galilee of southern Lebanon,

surveyed.
Multi-period settlement in the Golan, surveyed.
Small tell in the Golan, surveyed.

Small tell in the Golan, surveyed.
Arab village on a small tell in the southern
Golan, overlooking the Sea of Galilee,
surveyed.
Frankish fort on small tell in the Huleh valley,
overlooking the gorge of the river Jordan,
excavated (the Frankish castle)
Small tell in the southern Golan, surveyed.
Small tell on the eastern margin of the Huleh

valley, surveyed.
Ilan 1999:163
Finkelstein and Frankel, unpublished Multi-period settlement in the Upper Galileeof
information
southern Lebanon, surveyed.
Finkelstein and Frankel, unpublished Large tell in the valley of Marj Ayun in
information; Marfoe1995:185 and
southern Lebanon, surveyed.
bibliography
Stepansky 1999:21
Small tell in the Huleh valley, surveyed.
IAA data base; Hartal 1989:106-107
348
2047 2772
1815 2718
1996 2797
2110 2817
1258 1037
1271 1013
1244 1063
1265 1025
128 111
1276 1055
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Jish
Teitaba
Mi>ilia
Rosh, Tel (Ruweisa, Tell

er-)
Qedesh, Tel (Qades,
Tell)
Sheikh Mahmoud
Northwest of Tel
Nagila
Nagila, Tel (Najila, Tell
en-)
Hesi, Tell
West of Tel Nagila
Uza
Qeshet, Tel
East of Tarqumiya
148 109
1940 2680
1747 2699
1910 2700
2037 2848
1990 2751
2050 2910
2105 2869
2102 2915
2032 2691
2040 2893
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
Hazor
2059 2868
Hazor
Ein Avazim
Naal Dishon
Wawiyat, Tell elAnafa, Tel
Taunat el-Tabgha
Hazor, Tel
Qiryat Shemona
(south)
Mallaa, Tell
2008 2528
Hazor
1794 2795
Hazor
Chinnereth, Tel
(>Oreimeh, Tell el-)
Naama, Tel
2045 2962
Hazor
Abel Beth Maacha, Tel

(Abil el-Qamh, Tell)
Ramia
2112 2948
Hazor
Dan, Tel (Qadi, Tell el-)
C
B
B
A
A
C
A
B
A
A
A
B
A
E
A
NEAEHL 3:1081; Amiran and Eitan

1965
NEAEHL 2:632
Lamdan et al. 1977:191
Bunimovitz 1989:129
Medium-sized tell in the lower Shephelah,

excavated
Tell in the southern lower Shephelah, excavated
Site in the southern coastal plain, surveyed.
Small tell in the southern lower Shephelah,
surveyed.
Site in the upper Shephelah, surveyed.
Large tell in the northern Huleh valley,

excavated
Dever 1986
Large tell overlooking the Huleh valley,
surveyed.
Finkelstein and Frankel, unpublished Arab village on a tell in a small valley in the
information
Upper Galilee of southern Lebanon, surveyed.
NEAEHL 1:300
Tell on a high hill, to the north of the sea of
Galilee, excavated
ESI 7-8:140
Medium-sized tell in the Huleh valley,
excavated
Ilan 1999:164
Ilan 1999:164
ESI 113:1
Small tell in the Huleh valley
Had. Arkh. 48-49:24; 67-68:8
Multi-period tell in the Huleh valley, excavated
Ilan 1999:164
Yadin 1972
Largest site in Late Bronze Canaan, excavated
Had. Arkh. 56:5-6
A small tell in the northern Jordan valley,
excavated
Stepansky 1999:20; Ilan 1999:163
Tell in the Huleh valley (probably same site as
Tel Reemim of Dayan 1963:25), surveyed.
Aharoni 1957:14; Had. Arkh. 69-71: Arab village on tell in upper Galilee, surveyed.
30
(burials and ancient synagogue, excavated)
Frankel 1994:25
Site in the Upper Galilee, surveyed.
Aharoni 1957:18; Frankel 1994:25
Arab village on tell in the Upper Galilee,
surveyed.
Aharoni 1957:15; Frankel 1994:25
Medium-sized tell in the Upper Galilee,
surveyed.
Aharoni 1957:10-13
Large tell on the margin of a small fertile valley
in the upper Galilee, surveyed. and excavated
Ilan 1999:164
Site in the lower Shephelah, surveyed.
Biran 1994:105-123
349
122 088
1196 0889
1288 0968
1373 0879
1379 1011
1433 0998
1748 2069
1700 2174
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Lachish
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Sera>, Tel (Shari>a, Tell

esh-)
Mila, Tel (el-Maliha,
Tell)
alif, Tel (Khuweilifeh,
Tell)
Agra, Tel (el-Agra, Tell)
>Eton, Tel (>Aiun, Tell)
Sheqef, Tel
Ras, Kh. ElUmm el-Baqar, Kh.

Rasm Bir Jubarat
North of Tel Lachish
Fuqeiqis, Kh.
Shoraqa, orvat
Beit Mirsim, Tell
Lachish, Tel
>Afula
Hariqet er-Ras
Burqin
Giveat >Oz
Midrakh Oz
1653 2225
1709 2096
1357 1083
1775 2237
1485 0996
1208 1079
1415 0960
1459 1075
1300 1042
1325 1015
1351 1085
1230 1075
1336 1036
Lachish
araqim, Tel (elKhiraqa, Tell)

Maaravim, Tel
D
Q
Q
B
C
C
B
A
Q
Raban 1999:82-86
Zori 1977:51; The Megiddo

countryside survey
Zertal 1992:97
Ussishkin 1985
NEAEHL I:37-39
Ofer 1993:2A, 16
NEAEHL 1:177-180
Dagan 1992a:178-179
Dagan 1992b:118
Dagan 1992b:146
Dagan 1992b:68
Dagan 1992a: 235; for a LB burial

see Tzaferis and Hess 1992
Dagan 1992b:147
NEAEHL 2:556-557
Dagan 1992a:237-238
Oren 1984; NEAEHL 4:1330-1331
Oren and Mazar 1974
Dagan 1992b:129
A medium-sized tell in the upper Shephelah,

surveyed.
Large tell in the upper Shephelah, surveyed.
(and test soundings)
Small tell in the southern coastal plain,
surveyed.
A large tell in the upper Shephelah, surveyed.
Small tell in the Lower Shephelah, surveyed.
Small tell in the lower Shephelah, surveyed.
Site in the Shephelah immediately to the north
of Tel Lachish, surveyed.
Small site in the upper Shephelah, surveyed.
excavated
Large tell in the upper Shephelah, excavated
Tell in the Jezreel valley, almost completely
destroyed by modern activity, excavated
Small site in the hills between the Jezreel and
the Dothan valleys, surveyed.
Arab village on tell in the hilly area between
the Jezreel and Dothan valleys, surveyed.
A large Middle Bronze site in the elds
of Giveat >Oz, the western Jezreel valley,
surveyed.
Samll tell on the western margin of the Jezreel
valley, compeletely destroyed by modern
activity, excavated
Large tell in the upper Shephelah, excavated
Small site in the eastern ank of the southern

coastal plain, excavated
Medium-sized tell in the eastern ank of the
southern coastal plain, excavated
Small tell in the lower Shephelah, surveyed.
A small tell near Lachish, surveyed.
350
B
Q
B
1816 2236
1706 2184
1708 2142
1810 2182
Megiddo
Megiddo
Sulam
Tel Kedesh (Abu

Megiddo
Qudeis, Tell)
Ta>anach, Tel (Ta>annek, Megiddo
Tell)
Megiddo
Jezreel, Tel (Zer>in)
1770 2266
1675 2213
1777 2058
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Megiddo
Rehob
Rehob
Rehob
Dothan, Tell
Qitneh, el-
Megiddo, Tel
Bel>ameh, Kh.
Muhaffar, Tell el-
Meshattah, el-
Khrab, el-
Rujjam, Kh.
Deir, Tell edMasad, Tel

Trumot, Tel (Humra,
Kh. el-)
1992 1995
2027 2073
1965 2052
1661 2049
1645 2047
1725 2031
1707 2054
1727 2021
1763 2041
1841 2147
Megiddo
Mezarim, Horvat
(Mazar, el-)
Yifar, Tel (Far, Tell el-)
1727 2121
Q
A
A
C
Q
Yannun, Kh.
1783 2074
Megiddo
Jenin, Tell
1782 2056
Megiddo
Najjar, Kh. En-
Medium-sized tell on the main road connecting

the Jezreel and the Dothan valleys, surveyed.
Had. Arkh. 45:16; 65-66:26-27;
Medium-sized tell in the southern tip of the
Small site on the southwestern edge of the
(LB I)
Zori 1977:55, 57-58; Gal 1998:82
Arab village on a small tell in the Jezreel valley,
surveyed.
Stern and Beit-Arieh 1979
Small tell in the Jezreel valley south of
Megiddo, excavated
NEAEHL 4:1432
Large tell on the southwestern edge of the
Jezreel valley; Late Bronze I nds, excavated
Ussishkin and Woodhead 1994:42-43 Medium-sized tell on the southern margin of
the Jezreel valley, excavated
Zori 1977:6-7
Arab village on tell in the western ridge of Mt.
Gilboa, surveyed.
The Megiddo countryside survey
Small, multi-period site on a low hill in the
NEAEHL 1:372-374
Large tell in the Dothan valley, excavated
Zertal 1992:127-129 (LB I)
Small multi-period settlement on the southern
margin of the Dothan valley, surveyed.
Loud 1948; Ussishkin 1995
Large tell (the Late Bronze settlement included
lower terrace and covered ca. 11 hectares),
excavated
Zertal 1992:114-116 (LB III)
Large tell in a hilly area, on the main road
connecting the Jezreel and the Dothan valleys,
surveyed. and excavated
Zertal 1992:108-111 (LB III)
Large tell on the norther margin of the Dothan
valley, surveyed.
Zertal 1992:132-133 (LB III)
Small, multi-period settlement in the Dothan
valley, surveyed.
Zertal 1992:117-119
Small tell in the hills west of the Dothanvalley,
surveyed.
Zertal 1992:120-122
Multi-period settlement on the northern margin
of the Dothan valley, surveyed.
Mittmann 1970:132
Small tell in the Beth-shean valley, surveyed.
Zori 1962:159-161
Zori 1962:174-175
Zertal 1992:152-154
351
1962 1895
1943 1923
1978 1926
2032 2111
2025 2017
1967 2054
2030 2076
1937 2114
1933 2115
1970 2070
1991 2049
2026 2076
1929 2131
1966 2077
1973 1977
1989 2105
1994 2035
2027 2055
2026 2109
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
amud, Tas
ilu, Tell el-
Zom, Tel
Gema, Tel (Jema>in,

Tell)
Teomim, Tel (Thum,
Tell eth-)
Artal, Tel (Sheikh Daud,
Tell esh-)
Shoqeq, Tel (Shemdin,
Tell esh-)
Socoh, Tel (Shokh, Tell
esh-)
Rehov, Tel (arem, Tell
es-)
Roeh, Tel (Ru>yan, er-)
Qataf, Tel
Zahra, Tell ezRehob

Parve, orvat (Farwana, Rehob
Kh.)
Rehob
amamat, Kh.
ammah, Tell el-
Nisa, Tel (Manshiya,

Tell el-)
Menora, Tel (Abu Faraj,
Tell)
Maluach, Tel (Qitaf,
Tell)
Madras, Tel (Madrasa,
el-)
A
A
B
Small tell overlooking the Jordan river
Small site in the eastern ank of the north

Samaria hills, surveyed.
Medium-size, multi-period site in the eastern
slopes of the north Samaria hills, surveyed.
Small tell in Wadi Malih in the eastern ank of
the north Samaria hills, surveyed.
Zori 1962:170
Zori 1962:165; Gal 1979:145

(mainly LB/Iron I transition)
Zori 1962:161-162

surveyed.
Zroi 1962:185
surveyed.
Zori 1962:176-178; Mazar 1999 and Large tell in the Jordan valley, both upper and
personal communication
lower tell -- 11 hectares -- inhabited in Late
Bronze Age, excavated
Zori 1962:167
Small tell in the Jordan valley, excavated and
surveyed.
Zori 1962:158-159
Small multi-period site in the Jordan valley,
surveyed.
Zori 1962:184
Small tell west of Beth-shean
Zori 1962:178
Arab village on a multi-period site in the Jordan
valley, immediately to the north of Tel Rehov,
surveyed.
NEAEHL 2:561
Small tell in the Jordan valley south of Bethshean, excavated
Zori 1962:171-172
Zori 1962:185-186
Zori 1962:156
Zori 1962:175
Zori 1962:155; Maeir 1997,

Appendix I:23
Zori 1962:162
Zertal 1996:283-285
Zertal 1996:273-276
Zertal 1996:314-316
352
1885 2161
1995 2093
1975 2123
1882 2152
2034 2069
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Rehob
Eshtori, Tel (Malha, el-)

Sheikh Saleh
Sheikh Hasan, Tell

(Old Tel Yosef)
Zarom, orvat
1730 1480
2021 1791
1732 1817
1773 1804
1668 1992
1893 1849
1805 1726
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Bethel (Beitin)
Sali, Kh. EsSufan, Tell
Shechem (Balaa, Tell)

Batn Umm Nari
Miski, Tell
Marjama, Kh. el-
Khelayel, el-
>Urma, Kh. el-
Deir, Kh. ed-
Miqwaq, el-
1862 1934
1865 1906
1816 1554
1873 1824
1948 1879
Shechem
Yusef, Kh.
1983 2124
Rehob
Iztabba, Tel (Mastubeh,

Tell el-)
1994 2113
2008 2139
2028 2079
Rehob
Karpas, Tel (Qarantina,

el-)
Slavim, Tel (Firr, Tell
el-)
Zemed, Tel (Sheikh esSimad, Tell)
Beth-shean, Tel
C
Q
A
A
A
A
Zertal 1996:198-199 (LB I)
Zertal 1996:207-208
Finkelstein et al. 1997:805-808
Zertal 1996:373-376 (LB I)
NEAEHL 4:1352
(LB III)
Zertal 1996:412-415
Gophna and Porat 1972:226

Kallai 1972:164; Campbell 1991:
77-83
Kelso 1968
Zori 1962:152, Late Bronze sherds

in a survey; no Late Bronze nds
reported in excavation (ESI 11:5051). Tombs only?
Zertal 1996:317-319 (LB III)
Zori 1962:161
Zori 1977:26-27
ESI 19:28-29
Zori 1962:142
Mazar 1997
Zori 1962:172
Zori 1977:83
Zori 1962:155-156
Large multi-period site in the eastern ank of

the north Samaria hills, surveyed.
Arab village on a tell in southern Samaria,
excavated
Small site in the Jordan valley, surveyed.
Small tell in the valley of Naal Shechem.
Campbell (1991:77) names it Tell Sofar,
surveyed.
Medium-sized tell, excavated
Medium-sized tell in the hills of northern
Samaria, surveyed.
Small to Medium-sized tell in wadi Farah,
northern Samaria, surveyed.
Medium-sized tell in the eastern ank of the
southern Samaria hills, excavated
Small, multi-period site on a peak in the eastern
ank of the north Samaria hills, surveyed.
Small tell on a dominating ridge in southern
Samaria, surveyed.
Small, multi-period site in the northern Samaria
hills, surveyed.
Small settlement in the northern Samaria hills,
surveyed.
Late Bronze occupation only in the upper part

of the mound, excavated
Small tell near Beth-shean, excavated
Small, multi-period site in the Jordan valley
near Beth-shean, surveyed.
Small tell in the eastern Jezreel valley,
surveyed.
Small site overlooking the Jordan river,
surveyed.
Tell near the mound of Beth-shean, excavated
Small tell in the eastern Jezreel valley,

surveyed.
353
A
C
C
1821 1881
1764 1954
1724 1998
1776 1626
1861 1669
1698 1874
1695 1899
1851 1615
1733 1726
1789 2449
1801 2383
1743 2434
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shechem
Shim>on
Shim>on
Shim>on
Shim>on
Qumy, Kh.
Kheibar, Kh.
Bir el-Jadu>
Abu Zarad, Tell
Kheir-Allah
Shiloh (Seilun, Tell)
Southeast of Shechem Shechem

Shechem
Far>ah, Tell el-
Shreim, Kh.
Ras el->Ain
Rahaya, Kh. er-
>Einabus
Kebarrah, el-
Mizpe Zevulun
(Musheirifeh, Kh. El-)
Wawiyat, Tell elGath-efer, Tel

(Mashhad)
annathon, Tel
(Bedeiwiyeh, Tell)
1696 2391
1793 1967
1703 1958
1719 1679
1707 1832
1669 1905
Shechem
Sirtassa, es-
1834 1902
Shechem
1815 2007
Shechem
Sheikh Sariyan, Kh.

EshMuntar, el-
Medium-sized multi-period settlement in the

northern Samaria hills, surveyed.
Zertal 1992:329-333
Small multi-period site in the northern Samaria
hills, surveyed.
Zertal 1992:296-298 (LB III)
Small to Medium-sized tell in the northern
NEAEHL 2:439
Large tell in upper wadi el-Far>ah, northern
Samaria, excavated
Zertal 1992:377-378 (LB I)
Medium-sized tell in the Samaria hills west of
Shechem, surveyed.
Zertal 1992:227-229 (LB III)
Medium-sized tell on the southern margin of
the Sanur valley, northern Samaria, surveyed.
Zertal 1992:177-178
Small site in the northern Samaria hills,
surveyed.
Medium-sized tell in southern Samaria,
surveyed.
Zertal 1992:247-249
Medium-size, multi-period settlement in a
valley in northern Samaria, surveyed.
Finkelstein et al. 1993 - a cult place; Medium-sized tell in southern Samaria,
excavated
not inhabited
Small site in the desert fringe southeast of
Shechem, surveyed.
Zertal 1992:415-416
Small multi-period settlement in the northern
Zertal 1992:393-395
Small site in the northern Samaria hills,
surveyed.
Small, multi-period site in the desert fringe
southeast of Shechem, surveyed.
Arab village on tell in southern Samaria,
surveyed.
Zertal 1992:224-225
Medium-sized tell on the eastern margin of the
valley of Sanur, northern Samaria, surveyed.
Raban 1982:63; The Megiddo
A large Early Bronze site and a small, multiCountry side survey
period tell nearby. The Late Bronze pottery was
collected in the former, though the actual site
was probably the latter, surveyed.
NEAEHL 4:1500-1501; Dessel 1999 Small tell in the Beth Netofa valley , excavated
Gal 1992:18; 1998:25-27; ESI 14:54 Large tell in the hills north of Nazareth,
surveyed (plus test excavation)
Gal 1992:23
Large tell in the western Beit Netofa valley,
surveyed.
Zertal 1996:117-119
354
B
A
A
A
C
1688 2316
1650 2339
1666 2321
1649 2306
1699 2340
Seifan, orvat (Beida,

Shim>on
Tell el-)
Risim, Tel (Rish, Tell
Shim>on
er-)
Re>ala, Tel (Ghalta, Tell) Shim>on
Shim>on
Shim>on
B
B
1761 2374
1668 2311
Shim>on
Shim>on
Shim>on
Shim>on
Shim>on
Shim>on
Shim>on
Ein ippori
Kfar Yehoshua
Below Har Kdumim
Khudeira, Tell el-
Shor, Tel (Thora, Tell)
Kfar Baruch
>Adashim, Horvat
(>Adas, Tell el-)
>Adashim, Tel
A
Q
1637 2325
1590 1108
1724 1310
Zeror, orvat (Musrara, Shim>on

Kh.)
Beth-Zur (Tubeiqa, Kh.) Jerusalem
Jerusalem
1794 2287
Shim>on
Jerusalem
Q
Q
1688 2277
1792 2271
1661 2287
1684 2363
1761 2325
Shim>on
1785 2311
1722 2295
Shim>on
Shadud, Tel (Shaddud,

Tell)
Yaa>
Shem, Tel (Shammam,

Tell esh-)
Shimron, Tel
(Sammuniya, Kh.)
1816 2270
Tevet, Horvat (Bir Tibis, Shim>on

Kh.)
Shiloh 1984
Zori 1977:104; The Megiddo

countryside survey
countryside survey
Raban 1999:35-36; The Megiddo
countryside survey
Zori 1977:54-55; Questionable
LB sherds only in the Megiddo
countryside survey
Zori 1977:54; LB sherds only in the
Megiddo countryside survey
countryside survey
Finkelstein 1988:48
Gal (1992:13) found no Late Bronze

sherds; Late Bronze occupation
deducted only from EA 365 and from
the fact that the site was occupied in
both the Middle Bronze and in the
Iron I
Dessel 1999
Druks 1966; Raban 1982:43-45
Raban 1982:37; The Meggiddo

countryside survey
countryside survey
coutryside survey
Raban 1982:69-76
Zori 1977:55; Questionable Late

Bronze sherds only in the Megiddo
countryside survey
Raban 1982:58-61
Small tell in the Judean highlands north of

Hebron, excavated
The City of David, excavated
Low-prole, small tell in the Jezreel valley,

surveyed
Low-prole tell in the Jezreel valley, surveyed.
Site in the Jezreel valley, surveyed.

Multi-period site on a low hill in the Jezreel
valley, surveyed.
Low prole tell in the Lower Galilee, excavated

Small tell in the Jezreel valley, excavated
(burial, excavated)
Scatter of sherds in the northern margin of the
Jezreel valley, below Har Kdumim, surveyed.
Small tell in a northern branch of the Jezreel
valley, surveyed.
Small tell in the Jezreel valley, surveyed.
Arab village on tell in the Nazareth hills,

surveyed.
Small tell, completely destroyed; large quantity

of sherds still visible at the site, surveyed.
Very large Middle Bronze site on the northern
margin of the Jezreel valley, with a smaller tell
at its highest point, surveyed.
Small tell overlooking the western Jezreel

valley, surveyed.
Small tell in the Jezreel valley, excavated
Site in the Jezreel valley near Afula, surveyed.
355
Q
A
1655 1334
1666 1280
1597 1036
1582 1446
1515 0934
0934 0975
114 088
1007 0769
Jerusalem
Jerusalem
Hebron (Rumeideh, Tell Jerusalem

er-)
Jerusalem
Beit >Ur et-Tata
Jerusalem
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Yurza
Rabud, Kh.
>Ajjul, Tell el-
Mugrabi, Tell el-
Gerar 100
Far>ah, Tell el-
Haror, Tel (Abu

Hureireh, Tell)
West of Tel Haror

Jemmeh, Tell
Gaza
Bureij, el-
Tel Ridan
el-Moghraqa
es-Sanam, Tell
Qubur el-Walaida
Netivot
112 092
1011 0827
0913 0972
0822 0822
0945 0981
094 094 (?)
0990 1015
112 087
0972 0888
1125 0877
101 088
A
A
A
C
A
A
Moza (Qaluniya)
Emeq Refaim
1921 1419
Jerusalem
Jericho, Tell
1558 0991
Jerusalem
Fawar
Dan Gazit, personal communication
Had. Arkh. 65-66:50-51
NEAEHL 4:1283
Personal communication from
Joanne Clarke and Moain Sadeq
Oren, personal communication
Dan Gazit, personal communication
Phythian-Adams 1923
Gazit and Gophna 1993:17

NEAEHL 2:668-670
Oren et al. 1991
Ram Gopnna, personal

communication
Had. Arkh. 78-79:86; Gazit and
Gophna 1993
NEAEHL 2:441-442
Gonen 1981
Kochavi 1974
Ofer 1989
Feldstein et al. 1993:221; ESI 15:84

ESI 9:149
Bienkowski 1986
Ofer 1993:2A, 40
Small tell in Wadi Gaza near the coast,

surveyed.
Small site in the southern coastal plain,
excavated
Site in the southern coastal plain, near Netivot,
surveyed.
A hill southwest of Hebron, either a settlement

or a cemetery, surveyed.
Small to Medium-sized tell in the Jodan valley,
excavated
Medium size tell west of Jerusalem, excavated
Large Middle Bronze settlement in a wadi,
in the southern part of modern Jerusalem,
excavated
Small to Medium-sized tell in Hebron,
excavated
Arab village on a tell northwest of Jerusalem,
surveyed.
Medium to large tell in the southern Judean
highlands, excavated
Large tell near the coast, south of Gaza,
excavated
Site in the southern coastal plain, near Wadi
Gaza, surveyed.
A small site in the southern coastal plain,
surveyed.
Medium-sized tell in the southern coastal plain,
excavated
Large Middle Bronze tell, with a limited Late
Bronze occupation in the southern coastal plain,
excavated
Medium-sized tell in the southern coastal plain,
excavated
Large tell under the city of Gaza
Excavatged
Small site in the southern coastal plain,
surveyed.
A small tell south of Gaza
North of Wadi Gaza, near Tell el->Ajjul
REFERENCES
Abel, F.-M. 1938. Gographie de la Palestine. Paris.
Abou Assaf, A. 1968. Tell Aschtara in Sd Syrien. Erst Kampagne 1966. Annales Archologiques Arabes Syriennes
18:103-122.
Abou Assaf, A. 1969. Tell >Aschtara 2. Kampagne 1967. Annales Archologiques Arabes Syriennes 19:101-108.
Abou Assaf, A. 1974. Ein Mittelbronzezeitliches Grab in a-$aiybih und die Gleichsetzung von a-$aiybih mit Tub-ja. Baghdader Mitteilungen 7:13-20.
Abou Assaf, A. Bordreuil, P. and Millard, A.R. 1982. La statue de Tell Fekherye et son inscription bilingue assyroaramenne. Paris.
Adamthwaite, M.R. 1992. Lab<ayas connection with Shechem reassessed. Abr-Nahrain 30:1-19.
Adan-Bayewitz, D. and Wieder, M., 1992. Ceramics from Roman Galilee: A comparison of several techniques for
fabric characterization. Journal of Field Archaeology 19:189-205.
Aharon, L. 1997. A Shallow Magmatic System of Miocene Age, the Tavor River Volcanic Vent. (Report GSI/13/97).
Jerusalem.
Aharoni, Y. 1957. The Settlement of the Israelite Tribes in Upper Galilee. Jerusalem (Hebrew).
Aharoni, Y. 1959. Zephat of Thutmose. Israel Exploration Journal 9:110-122.
Aharoni, Y. 1967. Anaharath. Journal of Near Eastern Studies 26:212-215.
Aharoni, Y. 1969. Rubute and Ginti-kirmil. Vetus Testamentum 19:141-144.
Ahituv, S. 1984. Canaanite Toponyms in Ancient Egyptian Documents. Jerusalem.
Ahmed, A.A., Mansour, H.H., El-Younsy, A.R. and Shahin, M. 1992. Sedimentology and mineralogy of the Upper
Cretaceous Lower Eocene sequence, east of Beris Oasis, Western Desert, Egypt. Bulletin of the Faculty of
Science, Assiut University 21:109-137.
Akurgal, M., Kerschner, M., Mommsen, H., Niemeier, W.-D. 2002. Tferzentren der Ostgis, Archmetrische und
archlogische Untersuchungen zur mykenischen, geometrischen und archaischen Keramik aus Fundorten in
Westkleinasien. 3. Ergzungsheft der Jahreshefte des sterreichischen Archlogischen Institutes. Wien.
Albright, W.F. 1925. Bronze Age mounds of northern Palestine and the Hauran: The spring trip of the school of
Jerusalem. Bulletin of the American Schools of Oriental Research 19:5-19.
Albright, W.F. 1929. New Israelite and pre-Israelite sites: The spring trip of 1929. Bulletin of the Schools of
Oriental Research 3:1-3.
Albright, W.F. 1943. Two little understood Amarna letters from the middle Jordan Valley. Bulletin of the American
Schools of Oriental Research 89:7-17.
Albright, W.F. 1946. The Late Bronze town at modern Djett. Bulletin of the American Schools of Oriental Research
104:25-26.
Albright, W.F. 1975. The Amarna letters from Palestine. Cambridge Ancient History II/2A. Cambridge. pp. 98-116.
Allen, R.O., Rogers, M.S., Mitchell, R.S. and Hoffman, M.A. 1982. A geochemical approach to the understanding
of ceramic technology in Predynastic Egypt. Archaeometry 24: 199-212.
Allen, R.O., Hamroush, H. and Hoffman, M.A. 1989. Archaeological implications of differences in the composition
of Nile sediments. In: Allen, R.O., ed. Archaeological Chemistry IV. Washington DC. pp. 33-56.
Almagor, G. and Hall, J.K. 1980. Morphology of the continental margin of northern Israel and southern Lebanon.
Israel Journal of Earth-Sciences 29:245-252.
Al-Maqdissi, M. 1988. Tell Saqqa (rgion de Damas). Syria 65:410.
Al-Maqdissi, M. 1990. Tell Doulab (rgion de Damas). Syria 67:463.
Al-Maqdissi, M. 1993. Tell Deir Khabiyah. Syria 70:481-483.
356
Al-Moussli, M.M. 1985. Chronique archologique: Tell Homs. Syria 62:133.

Alt, A. 1925. Das Institut im Jahre 1924. Palstinajahrbuch 21: 5-58.
Alt, A. 1953. Die Landnahme der Israeliten in Palstina. In: Kleine Schriften zur Geschichte des Volkes Israel I.
Mnchen. pp. 89-125.
Amiran, R. and Porat, N. 1984. The basalt vessels of the Chalcolithic period and Early Bronze Age I. Tel Aviv 11:11-19.
Anati, E., Avnimelech, M., Haas, N. and Meyerhof, E. 1973. Hazorea I. Capo di Ponte.
Anderson, W.P. 1987. The kilns and workshops of Sarepta (Sarafand, Lebanon): Remnants of a Phoenician ceramic
workshop. Berytus 35:41-66.
Anderson, W.P. 1988. Sarepta I, The Late Bronze and Iron Age Strata of Area II. Beirut.
Anderson, W.P. 1990. The beginnings of Phoenician pottery: vessel shape, style and ceramic technology in the early
phases of the Phoenician Iron Age. Bulletin of the American Schools of Oriental Research 279:35-54.
Arad, A. 1964. The Geology and Hydrology of the Lower Cretaceous Section of the Northern Negev and the Judean
Mountains. Jerusalem.
Arkin, Y., Braun, M. and Starinsky, A. 1965. Type sections of Cretaceous formations in the Jerusalem - Beit
Shemesh area. Israel Geological Survey Stratigraphic Section 1:1-59.
Arkin, Y., Braun, M., Buchbinder, B., Diamant, E., Itzhaki, Y., Lasman, N., Rot, I. and Shachnai, E. 1976 (ed.
1997). Jerusalem and Vicinity, Geological Map 1:50,000. Jerusalem.
Arnold, D.E. 1985. Ceramic Theory and Cultural Process. Cambridge.
Arnold, D. and Bourriau, J., eds. 1993. An Introduction to Ancient Egyptian Pottery. Mainz am Rhein.
Artzi, P. 1968. Some unrecognized Syrian Amarna letters (EA 260, 317, 318). Journal of Near Eastern Studies
27:163-171.
Artzy, M., Perlman, I. and Asaro, F. 1976. Alashiya of the Amarna Letters. Journal of Near Eastern Studies 2:171-182.
Asaro, F., Perlman, I. and Dothan, M. 1971. An introductory study of Mycenaean IIIC Ware from Tell Ashdod.
Archaeometry 13:169-175.
Astour, M. 1969. The partition of the confederacy of Muki-Nuae-Nii by uppiluliuma. A study in political
geography of the Amarna Age. Orientalia 38: 381-414.
Astour, M.C. 1977. Tunip-Hamath and its region, a contribution to the historical geography of central Syria.
Orientalia 46:51-64.
strm, P. and Herscher, E., eds. 1996. Late Bronze Age Settlement in Cyprus: Function and Relationship. (Studies
in Mediterranean Archaeology and Literature, Pocket-Book 126). Jonsered.
Badre, L. 1997. Excavations of the American University of Beirut Museum 1993-1996. BAAL 2.
Badre, L., Gubel, ., al-Maqdissi, M. and Sader, H. 1990. Tell Kazel Syria: Excavations of the AUB Museum,
1985-1987 preliminary report. Berytus 38:9-124.
Badre, L., Gubel, ., Capet, E. and Panayot, N. 1994. Tell Kazel (Syrie), Rapport prliminaire sur les 4e-8e
campagnes de fouilles (1988-1992). Syria 71:259-346.
Badre, L. and Gubel, E. 1999-2000. Tell Kazel Syria: Excavations of the AUB Museum, 1993-1998, third
preliminary report. Berytus 44:123-203.
Baida, U. 1963. On the geology of the Gaaton-Hilazon area, western Galilee, Israel. Israel Journal of EarthSciences 13:1-15.
Bagnall, P.S. 1960. The geology and mineral resources of the Pano Lefkara-Larnaca area. Memoirs of the
Geological Survey of Cyprus 5:1-116.
Ballard, R.D., Stager, L.E., Master, D., Yoerger, D., Mindell, D., Whitcomb, L.L., Singh, H. and Piechota, D. 2002.
Iron Age shipwrecks in deep water off Ashkelon, Israel. American Journal of Archaeology 106:151-168.
Bateman, C.A. 1966. The treatment of cuneiform tablets in the British Museum. In: Bateman, C.A. et al., eds
Preservation and Reproduction of Clay Tablets and the Conservation of Wall Paintings. London. pp. 12-17.
357
Baramki, D. 1961. Preliminary report on the excavations at Tell el Ghassil. Bulletin du Muse de Beyrouth
16:87-102.
Baramki, D.C. 1964. Second preliminary report on the excavations at Tell el Ghassil. Bulletin du Muse de
Beyrouth 17:47-103.
Baramki, D.C. 1966. Third preliminary report on the excavations at Tell el Ghassil. Bulletin du Muse de Beyrouth
19:29-49.
Barclay, K. 2001. Scientic Analysis of Archaeological Ceramics, a Handbook of Resources. Oxford.
Barley, N. 1994. Smashing Pots, Feats of Clay from Africa. London.
Bartov, Y. 1994. Geological Photomap of Israel and Adjacent Areas. 1:750,000. (2nd edition) Geological Survey
of Israel. Jerusalem.
Basta, E.Z., Abdel Maksoud, M. and Talaat Abdel Aziz, A. 1979. Mineralogical constitution of some Egyptian
montmorillonitic clays using different X-ray diffraction techniques. Egyptian Journal of Geology 23:41-56.
Beier, Th. and Mommsen, H. 1994. Modied Mahalanobis lters for grouping pottery by chemical composition.
Bender, F. 1974. Geology of Jordan. Berlin-Stuttgart.
Ben-Gai, Y., Buchbinder, B. and Ben-Avraham, Z. 1996. Sequence stratigraphy of the Yaffo Formation (PlioceneLower Pleistocene) in the coastal plain of central Israel. Israel Geological Survey Current Research 10:1-9.
Bennett, C.-M. 1979. Excavations at the Citadel (al Qal>a) Amman, 1977. Annual of the Department of Antiquities
of Jordan 23:151-159.
Ben-Tor, A. 1993. Jokneam. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the Holy
Land Vol. 2. Jerusalem. pp. 805-811.
Ben-Tor, A., Portugali, Y., Avissar, M., Baruch, U. and Hunt, M. 1987. Tell Qiri A Village in the Jezreel Valley
(Qedem 24). Jerusalem.
Bentor, Y.K. 1945. Petrographic Investigations of the Upper Albian - Lower Cenomanian near Jerusalem and
Contribution to the Problem of Dolomitization and Quartzication. (Unpublished Ph.D. dissertation, The
Hebrew University) Jerusalem.
Bentor, Y.K. 1966. The Clays of Israel. Jerusalem.
Bienkowski, P. 1986. Jericho in the Late Bronze Age. Warminster.
Bikai, P. 1978. The Pottery of Tyre. Warminster.
Bingl, A.F. 1978. Ptrologie du massif ophiolitique de Pozanti-Karsanti (Taurus cilicien, Turquie): tude de la
partie orientale. (Ph.D. dissertation, Universit Strasbourg) Strasbourg.
Biran, A. 1994. Biblical Dan. Jerusalem.
Bonechi, M. 1992. Relations amicales syro-palestiniennes: Mari et Haor au XVIIIe sicle av. J.C. In: Durand,
J.-M., ed. Florelagium Marianum. Recueil dtudes en lhonneur de Michael Fleury. (Mmoires de
NABU No 1). Paris.
Bourke, S.J. 1993. The transition from the Middle to the Late Bronze Age in Syria: The evidence from Tell Nebi
Mend. Levant XXV:155-195.
Bourriau, J.D. and Nicholson, P.T. 1992. Marl clay pottery fabrics of the New Kingdom from Memphis, Saqqara
and Amarna. Journal of Egyptian Archaeology 78:29-91.
Bourriau, J.D., Smith, L.M.V. and Nicholson, P.T. 2000. New Kingdom Pottery Fabrics, Nile Clay and Mixed Nile/
Marl Clay Fabrics from Memphis and Amarna. London.
Bozaglo, E. 2004. Petrographic Investigation of Iron Age Pottery Assemblages from Megiddo and the North. (MA
thesis, Tel Aviv University). Tel Aviv.
Braemer, F. 1984. Prospections archologiques dans le Hauran (Syrie). Syria 61:219-250.
Braemer, F. 1987. Two campaigns of excavations on the ancient tell of Jarash. Annual of the Department of
Antiquities of Jordan 31:525-530.
358
Braidwood, R. 1940. Report on two sondages on the coast of Syria, south of Tartous. Syria 21:183-221.
Braidwood, R.J. and Braidwood, L. 1960. Excavations in the Plain of Antioch I. Chicago.
Briend, J. 1972. Akshaph et sa localisation Tell Keisan. Revue Biblique 79:239-246.
Brochier, J.E. 1983. Bergeries et feux de bois nolithiques dans le midi de la France. Quartr 33/34:181-193.
Brochier, J.E., Villa, P. and Giacommarra, M. 1992. Shepherds and sediments: Geo-ethnoarchaeology of pastoral
sites. Journal of Anthropological Archaeology 11:47-102.
Bronitsky, G. and Hamer, R. 1986. Experiments in ceramic technology: the effects of various tempering materials
on impact and thermal-shock resistance. American Antiquity 51:89-101.
Broshi, M. 1993. The population of Iron Age Palestine. In: Biran, A. and Aviram, J., eds. Biblical Archaeology
Today 1990, Pre-Congress Symposium Supplement (Proceedings of the Second International Congress on
Biblical Archaeology, Jerusalem 1990) Jerusalem. pp. 14-18.
Bryce, T.R. 1989. Ahhiyawans and Mycenaeans - An Anatolian viewpoint. Oxford Journal of Archaeology 8:
297-310.
Buchbinder, B. 1969. Geological Map of Hashephela Region, Israel, with Explanatory Notes, Scale: 1:20,000.
(The Institute for Petroleum Research and Geophysics, Report 1030, The Geological Survey of Israel, Report
OD/1/68) Jerusalem.
Buchbinder, B. 1975. Stratigraphic signicance of the alga Amphiroa in Neogene - Quaternary bioclastic sediments
from Israel. Israel Journal of Earth-Sciences 24:44-48.
Budd, P., Pollard, A.M., Scaife, B. and Thomas, R.G. 1995. Oxhide ingots, recycling and the Mediterranean metal
trade. Journal of Mediterranean Archaeology 8:1-32.
Budd, P., Haggerty, R., Pollard, A.M., Scaife, B., and Thomas, R.G. 1996. Rethinking the quest for provenance.
Antiquity 70:168-174.
Bullard, R. 1970. Geological studies in eld archaeology. Biblical Archaeologist 33:98-132.
Bunimovitz, S. 1989. The Land of Israel in the Late Bronze Age: A Case Study of Socio-Cultural Change in a
Complex Society. (Unpublished Ph.D. thesis, Tel Aviv University) Tel Aviv. (Hebrew with English abstract)
Bunimovitz, S. and Lederman, Z. 1993. Beth-Shemesh. In: Stern, E., ed. The New Encyclopedia of Archaeological
Excavations in the Holy Land Vol. 1. Jerusalem. pp. 249-253.
Cadogan, G. 1996. Maroni: Change in Late Bronze Age Cyprus. In: strm, P. and Herscher, E., eds. Late Bronze
Age Settlement in Cyprus: Function and Relationship. (Studies in Mediterranean Archaeology and Literature
Pocket-Book 126). Jonsered. pp.15-22.
Campbell, E.F. 1964. The Chronology of the Amarna Letters. Baltimore.
Campbell, E.F. 1965. Shechem in the Amarna Archive. In: Wright, G.E. Shechem: The Biography of a Biblical City.
London. pp. 191-207.
Campbell, E.F. 1968. The Shechem area survey. Bulletin of the American Schools of Oriental Research 190:19-41.
Campbell, E.F. 1976. Two Amarna notes: The Shechem city-state and Amarna administrative terminology. In:
Cross, F.M. et al., eds. Magnalia Dei The Mighty Acts of God: Essays on the Bible and Archaeology in
Memory of G. Ernest Wright. New York. pp. 39-54.
Campbell, E.F. 1991. Shechem II: Portrait of a Hill Country Vale. Atlanta.
Canti, M.G. 1997. An investigation of microscopic calcareous spherulites from herbivore dungs. Journal of
Archaeological Science 24:219-231.
Carozzi, A.V. 1960. Microscopic Sedimentary Petrography. New York and London.
Catling, H. 1962. Patterns of settlement in Bronze Age Cyprus. Opuscula Atheniensia 4:129-169.
Catling, H. 1964. Cypriote Bronzework in the Mycenaean World. Oxford.
Catling, H. 1975. Cyprus in the Late Bronze Age. The Cambridge Ancient History (3rd edition) II/2A. London. pp.
188-216.
359
Chevenoy, M. 1959. Le substratum mtamorphique des roches vertes dans Le Bar et Le Bassit (Syrie
septentrionale). Notes et Mmoires sur le Moyen-Orient 7:1-18.
Christaller, W. 1966. Central Places in Southern Germany (translated by C.W. Baskin from the 1933 edition).
Englewood Cliffs.
Christodoulou, D. 1959. The Evolution of Rural Land Use Pattern in Cyprus. (The World Land Use Survey
Regional Monograph No. 2) London.
Cochavi-Rainey, Z. 1993. The style and syntax of EA 1. Ugarit-Forschungen 25:75-84.
Cohen-Weinberger, A. 1988. Petrographic analysis of the Egyptian forms from Stratum IV at Tel Beth-Shean. In: Gitin,
S, Mazar, A. and Stern, E., eds. Mediterranean Peoples in Transition: Thirteenth to Early Tenth Centuries BCE.
Jerusalem. pp. 406-412.
Cohen-Weinberger, A. and Goren, Y. 1996. Petrographic analysis of Iron I Age pithoi from Sasa. >Atiqot 28: 77-83.
Cook, P. 2000. Geological Map of Israel 1:50,000, Sheet 5-IV, Shechem. Jerusalem.
Cooper, L. and Fortin M. 2001. Analyse ptrographique de la cramique de Tell >Acharneh, en Syrie (3000-700 av.
J.C.): rsultats prliminaires. Cahiers darchologie du CELAT 10:137-148.
Courtois, L. 1980. Examen ptrographique et caractrisation matrielle de quelques cramiques de Tell Keisan. In:
Briend, J. and Humbert, J.B., eds. Tel Keisan (1971-1976), une cit phnicienne en Galile. Paris. pp. 353360.
Courtois, J-C. 1973. Prospection archologique dans la moyenne valle de lOronte. Syria 50:53-99.
Courtois, J-C. 1986. propos des apports orientaux dans la civilisation du Bronze Rcent Chypre. In:
Karageorghis, V., ed. Acts of the International Archaeological Symposium Cyprus between the Orient and
the Occident. Nicosia. pp. 69-87.
Cross, F.M. 1968. Jar inscriptions from Shiqmona. Israel Exploration Journal 18: 226-233.
CTH = Laroche, E. 1971. Catalogue des textes hittites. Paris.
Dagan, Y. 1992a. The Shephelah during the Period of the Monarchy in Light of Archaeological Excavations and
Survey. (MA thesis, Tel Aviv University) Tel Aviv. (Hebrew)
Dagan, Y. 1992b. Archaeological Survey of Israel, Map of Lahish (98). Jerusalem.
Dalman, G. 1913. Jahresbericht des Instituts fr das Arbeitsjahr 1912/13. Palstinajahrbuch 9:1-75.
Dan, J., Yaalon, D.H., Koyumdjisky, H. and Raz, Z. 1972. The soil association map of Israel, 1: 1,000,000. Israel
Journal of Earth-Sciences 21:29-49.
Dan, J., Yaalon, D.H., Koyumdjisky, H. and Raz, Z. 1976. The Soils of Israel (with map 1:50,000 pamphlet). (ISA,
Division of Scientic Publications, The Volcani Institute of Soil, Water and Environmental Sciences) Beit
Dagan.
Davidson, T.E. and McKerrell, H. 1976. Pottery analysis and Halaf trade in the Khabur headwaters region. Iraq
38:45-56.
Dayan, Y. 1962. An Archaeological Survey in the Huleh Valley. (Hebrew)
Deer, W.A., Howie, R.A. and Zussman, J. 1971. Rock Forming Minerals. New York.
del Monte, G.F. and Tischler, J. 1978. Repertoire Gographique des Textes Cuniformes. Band 6/I. Die Orts- und
Gewssernamen der hethitischen Texte. (Beihefte zum Tbinger Atlas des Vorderen Orients. Reihe B Nr. 7/6)
Wiesbaden.
del Monte, G.F. and Tischler, J. 1992. Repertoire Gographique des Textes Cuniformes. Band 6/II. Die Orts- und
Gewssernamen der hethitischen Texte. (Beihefte zum Tbinger Atlas des Vorderen Orients. Reihe B Nr. 7/6)
Wiesbaden.
Dessel, J.P. 1999. Tell >Ein Zippori and the lower Galilee in the Late Bronze and Iron Ages: A village perspective.
In: Meyers, E.M., ed. Galilee Through the Centuries. Winona Lake. pp. 1-32.
360
Dever, W.G. 1986. Abel-beth-ma`acah: Northern Gateway of Ancient Israel. In: Geraty, L.T. and Herr, L.G., eds.
The Archaeology of Jordan and other Studies. Berrien Springs. pp. 207-222.
Dietrich, M. 2001. Der Brief des Kommandeurs umaynu an den ugaritischen Knig Niqmepa> (RS. 20.33). Ein
Bericht ber Aktivitten nacht der Schlacht bei Qade 1275 v. Chr. Ugarit-Forschungen 33:117-191.
Dietrich, M. and Loretz, O. 1997. Der Vertrag zwischen Ir-Addu von Tunip und Niqmepa> von Muki. In: Young,
G.D., Chavalas, M.W. and Averbeck, R.E., eds. Crossing Boundaries and Linking Horizons. Studies in Honor
of Michael C. Astour on His 80th Birthday. Bethesda. pp. 211-242.
Dikaios, P. 1969. Enkomi Excavations 1948-1958 Vol. I. Mainz am Rhein.
Dikaios, P. 1971. Enkomi Excavations 1948-1958 Vol. II. Mainz am Rhein.
Dobel, A., Asaro, F. and Michel, H.V. 1977. Neutron Activation Analysis and the location of Waukanni. Orientalia
46:375-382.
Dornemann, R.H. 1983. The Archaeology of the Transjordan in the Bronze and Iron Ages. Milwaukee.
Dothan, M. 1952. An archaeological survey of the lower Rubin River. Israel Exploration Journal 2:104-117.
Dothan, T. 1978. Excavations at the Cemetery of Deir el Balah. (Qedem 10) Jerusalem.
Dothan, T. 1987. The impact of Egypt on Canaan during the 18th and 19th Dynasties in the light of the excavations
at Deir el-Balah. In: Rainey, A.F., ed. Egypt, Israel, Sinai. Tel-Aviv. pp. 121-135.
Dothan, T. 1993. Deir el-Balah. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the Holy
Druks, A. 1966. A Hittite burial near Kefar Yehoshua. Bulletin of the Israel Exploration Society 213-220.
(Hebrew)
Dubertret, L. 1945. Carte gologique au 50.000e Feuille de Hamdiye. Beirut.
Dubertret, L. 1949a. Carte gologique au 200.000e Feuille de Tripoli. Beirut.
Dubertret, L. 1949b. Carte gologique au 50.000e Feuille de Saida. Beirut.
Dubertret, L. 1950a. Carte gologique au 50.000e Feuille de Baalbek. Beirut.
Dubertret, L. 1950b. Carte gologique au 50.000e Feuille de Rayak. Beirut.
Dubertret, L. 1951a. Carte gologique au 50.000e Feuille de Beyrouth. Beirut.
Dubertret, L. 1951b. Carte gologique au 50.000e Feuille de Tripoli. Beirut.
Dubertret, L. 1955. Gologie des roches vertes du nord-ouest de la Syrie et du Hatay (Turquie). (Notes et Mmoires
sur le Moyen-Orient VI). Beirut.
Dubertret, L. 1956. Carte Gologique au 50.000e Feuille de Jbail. Beirut.
Dubertret, L. 1962. Carte Gologique du Liban, Syrie et bordure des pays voisins. Paris.
Dunand, M. 1969. Rapport sur les fouilles de Sidon en 1967-1968. Bulletin du Muse de Beyrouth 22:101-107.
Dussaud, R. 1927. Topographie historique de la Syrie antique et mdivale. Paris.
Edel, E. 1953. Die Stelen Amenophis II. Aus Karnak und Memphis mit dem Bericht ber die asiatischen Feldzge
des Knigs. Zeitschrift des Deutschen Palstina-Vereins 69: 97-176.
Edel, E. 1966. Die Ortsnamenlisten aus dem Totentempel Amenophis III. Bonn.
Edlestein, G. and Glass, J. 1973. The origin of Philistine pottery based on petrographic analysis. In: Aharoni, Y., ed.
Excavations and Studies, Essays in Honor of Professor S. Yeivin. (Publications of the Institute of Archaeolgy
No. 1) Tel-Aviv. pp. 125-132.
Edwards, I. 1992. A progress report on the Pella pottery studies. In: A. McNicoll, Edwards, P.C., Hanbury-Tenison,
J., Hennessy, J.B., Potts, T.F., Smith, R.H., Walmsley, A. and Watson, P., eds. Pella in Jordan 2 (Appendix
10). Sidney.
Edwards, W.I. and Segnit, E.R. 1984. Pottery technology at the Chalcolithic site of Teleilat Ghassul (Jordan).
361
Edzard, D.O. 1970. Die Tontafeln von Kmid el-Lz, In: Edzard, D.O., Hachmann, R., Maiberger, P. and Mansfeld,
G., eds. Kamid el-Loz Kumidi. Bonn. pp. 50-62.
Eissfeldt, O. 1968. Renaming in the Old Testament. In: Ackroyd, P.R. and Lindars, B., eds. Words and Meanings.
Essays Presented to D. Winton Thomas. Cambridge. pp. 69-79.
Eizenberg, E. 1993. Naal Refaim. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the
Holy Land Vol. 4. Jerusalem. pp. 1277-1280.
Eizenberg, E. 1994. Naal Refaim - a Bronze Age village in southwestern Jerusalem. Qadmoniot 103-104: 82-95.
(Hebrew)
Epstein, C. 1963. A new appraisal of some lines from a long-known papyrus. Journal of Egyptian Archaeology 49:49-56.
Epstein, C. 1993. The cities of the land of Garu-Geshur mentioned in EA 256 reconsidered. In: Heltzer, M., Segal,
A. and Kaufman, D., eds. Studies in the Archaeology and History of Ancient Israel in Honour of Moshe
Dothan. Haifa. pp. 83-90. (Hebrew)
Epstein, C. 1998. The Chalcolithic Culture of the Golan. Jerusalem.
Epstein, C. and Gutman, S. 1972. The Golan. In: Kochavi, M., ed. Judaea, Samaria and the Golan Archaeological
Survey 1967-1968. Jerusalem. pp. 244-295. (Hebrew)
Erendil, M. 1984. Petrology and structure of the Upper Crustal Units of the Kizilda ophiolite. In: Tekeli, O. and
Cncolu, M.C., eds. Geology of the Taurus Belt: Proceedings. Ankara. pp. 269-284.
Fabian, P. and Goren, Y. 2002. A new type of Late Roman storage jar from the Negev. In: Humphrey, J.H., ed. The
Roman and Byzantine Near East Vol. 3. Rhode Island. pp. 145-153.
Fattorini, S. 2000. Dispersal, vicariance and refuges in the Anatolian Pimeliinae (Coleoptera, Tenebrionidae):
remarks on some biogeographical tenets. Biogeographia 21:355-398.
Feldstein, A., Kidron, G., Hanin, N., Kamaisky, Y. and Eitam, D. 1993. Southern part of the maps of Ramallah and
el-Bireh and northern part of the map of >Ein Kerem. In: Finkelstein, I. and Magen, Y., eds. Archaeological
Survey of the Hill Country of Benjamin. Jerusalem. pp. 133-264. (Hebrew)
Fevert, M. and Sanlaville, P. 1966. Lutilisation des vermets dans la dtermination des anciens niveaux marins.
Mditerrane 4:357-364.
Finkelstein, I. 1981. The Shephelah of Israel. Tel Aviv 8:84-94.
Finkelstein, I. 1988. The Archaeology of the Israelite Settlement. Jerusalem.
Finkelstein, I. 1990. Soundings at ancient Bene-Beraq. >Atiqot 10:4-29. (Hebrew)
Finkelstein, I. 1991. The Central Hill Country in the Intermediate Bronze Age. Israel Exploration Journal 41:19-45.
Finkelstein, I. 1992. Pastoralism in the highlands of Canaan in the third and second millennia B.C.E. In: Bar
Yosef, O. and Khazanov, A., eds. Pastoralism in the Levant: Archaeological Materials in Anthropological
Perspective. Madison. pp. 133-142.
Finkelstein, I. 1993. The sociopolitical organization of the Central Hill Country in the second millennium B.C.E.
In: Biran, A. and Aviram, J., eds. Biblical Archaeology Today, 1990, Pre-congress Symposium: Population,
Production and Power, Jerusalem, June 1990, Supplement. Jerusalem. pp. 119-131.
Finkelstein, I. 1996. The territorial-political system of Canaan in the Late Bronze Age. Ugarit-Forschungen 28:
221-255.
Finkelstein, I. 1997. Methods of the eld survey and data recording. In: Finkelstein, I., Lederman, Z. and
Bunimovitz, S. Highlands of Many Cultures: The Southern Samaria Survey, The Sites. (Monograph Series of
the Institute of Archaeology of Tel Aviv University No. 14) Tel Aviv. pp. 11-24.
Finkelstein, I. 1998. From sherds to history: Review article. Israel Exploration Journal 48:120-131.
Finkelstein, I. 2003. City-states to states: Polity dynamics in the 10th-9th centuries B.C.E. In: Dever, W.G. and Gitin,
S., eds. Symbiosis, Symbolism and the Power of the Past: Canaan, Ancient Israel and their Neighbors from the
Late Bronze Age through Roman Palestine. Winona Lake. pp. 75-83.
362
Finkelstein, I., Bunimovitz, S. and Lederman, Z. 1993. Shiloh: The Archaeology of a Biblical Site. (Monograph
Series of the Institute of Archaeology No. 10) Tel Aviv.
Finkelstein, I., Lederman, Z. and Bunimovitz, S. 1997. Highlands of Many Cultures: The Southern Samaria Survey,
The Sites. (Monograph Series of the Institute of Archaeology of Tel Aviv University No. 14) Tel Aviv.
Finkelstein, I. and Singer-Avitz, L. 2001. Ashdod revisited. Tel Aviv 28:231-259.
Fischer, P.M. 1993. Tell Abu al-Kharaz: The Swedish Jordan Expedition 1991, second season preliminary
excavation project. Annual of the Department of Antiquities of Jordan 37:279-305.
Fischer, P.M. 1997. Tell Abu al-Kharaz: The Swedish Jordan Expedition 1995-1996, sixth and seventh season
preliminary excavation project. Annual of the Department of Antiquities of Jordan 41:129-144.
Flint, A. 1980. A preliminary thin-section analysis of Terqa and Dilbat ceramics. Syro-Mesopotamian Studies 3:
174-178.
Foraminiferi Padani. 1982. Atlante Iconograco e Distribuzione Stratigraca, secunda edizione. AGIP S.p.A
Frankel, R. 1994. Upper Galilee in the Late Bronze-Iron I transition. In: Finkelstein, I. and Naaman, N., eds. From
Nomadism to Monarchy: Archaeological and Historical Aspects of Early Israel. Jerusalem. pp. 18-34.
Frankel, R. and Getzov, N. 1997. Archaeological Survey of Israel Map of Akhziv (1) Map of Hanita (2).
Jerusalem.
Frankel, R., Getzov, N., Aviam, M. and Degani, A. 2001. Settlement Dynamics and Regional Diversity in Ancient
Upper Galilee: Archaeological Survey of Upper Galilee. (IAA Report 14) Jerusalem.
Franken, H.J. and Steiner, M.L. 1990. Excavations in Jerusalem 1961-1967, Vol. II. Oxford.
Franken, H.J. and Van As, A. 1994. Potters who used Euphrates clay. In: Gasche, H. Mesopotamian History and
Environment Occasional Publications II. Leuven. pp. 507-518.
Friedman, G.M. 1959. Identication of carbonate minerals by staining methods. Journal of Sedimentary Petrology
29:87-97.
Fritz, V. 1999. Kinneret: Excavations at Tell el-Oreimeh (Tel Kinrot): Preliminary report on the 1994-1997 seasons.
Tel Aviv 26:92-115.
Fritz, W. 1991. Bemerkungen zum Datierungsvermerk auf der Amarnatafel Kn. 27. In: Studien zur Altgyptischen
Kultur 18: 207-214.
Gadot, Y. 1999. The Wadi >Ara Pass as an International Highway during the Bronze Age, Iron Age and the Persian
Period, in the Light of the Settlement Patterns. (MA thesis, Tel Aviv University) Tel Aviv.
Gal, Z. 1979. An Early Iron Age site near Tel Menorah in the Beth-shan valley. Tel Aviv 6:138-145.
Gal, Z. 1980. Ramat Issachar. Tel Aviv. (Hebrew)
Gal, Z. 1981. Tel Rekhesh and Tel Qarney Hittin. Eretz-Israel 15:213-221. (Hebrew)
Gal, Z. 1991. Archaeological Survey of Israel Map of Gazit (46). Jerusalem.
Gal, Z. 1992. Lower Galilee during the Iron Age. Winona Lake.
Gal, Z. 1998. Archaeological Survey of Israel Map of Har Tavor (41), Map of En Dor (45). Jerusalem.
Gal, Z. 2000. A Sounding at Tel Par. >Atiqot 39:83-103.
Gale, N. 1989. Archaeometallurgical studies of Late Bronze Age oxhide copper ingots from the Mediterranean
region. In: Hauptmann, A., Pernicka, E. and Wagner, G.A., eds. Archaomettalurgie der Alten Welt. (Der
Anschnitt 7) Bochum. pp. 247-268.
Galling, K. 1953. Der tell izzin. Zeitschrift des Deutschen Palstina-Vereins 69:88-90.
Garstang, J. 1922. Tell Amr. Bulletin of the British School of Archaeology in Jerusalem 2:14-15.
Garstang, J. 1924. El Harbaj. Bulletin of the British School of Archaeology in Jerusalem 4:45-46.
Gass, I.G. 1980. The Troodos Massif: its role in the unraveling of the ophiolite problem and its signicance in
the understanding of constructive plate margin process. In: Panayiotou, A., ed. Ophiolites: Proceedings
International Ophiolite Symposium, Cyprus 1979. Nicosia: pp. 23-35.
363
Gass, I.G., MacLeod, C.J., Murton, B.J., Panayotou, A., Simonian, K.O. and Xenophontos, C. 1994. The Geology of
the Southern Troodos Transform Fault Zone. (Cyprus Geological Survey Department Memoir No. 9) Nicosia.
Gazit, D. and Gophna, R. 1993. An unfortied Late Bronze Age site at Gerar: Survey nds. >Atiqot 22:15-19.
(Hebrew)
Getzov, N. 1993. Hurvat Utza. Excavations and Surveys in Israel 13:19-21.
Gibson, S. 1994. The Tell ej-Judeideh (Tel Goded) excavations: A re-appraisal based on archival records in the
Palestine Exploration Fund. Tel Aviv 21:194-234.
Gibson, S., Ibbs, B. and Kloner, A. 1991. The Sataf project of landscape archaeology in the Judaean Hills: A
preliminary report on four seasons of survey and excavation (1987-1989). Levant 23:29-54.
Gilead, I. and Goren, Y. 1989. Petrographic analyses of fourth millennium B.C. pottery and stone vessels from the
northern Negev, Israel. Bulletin of the American Schools of Oriental Research 275:5-14.
Gilead, I. and Goren, Y. 1995. The pottery assemblages from Grar. In: Gilead I., ed. Grar, a Chalcolithic Site in the
Northern Negev. (Studies by the Department of Bible and the Ancient Near East) Beersheba. pp. 137-221.
Gittlen, B.M. 1992. The Late Bronze Age city at Tel Miqne/Ekron. Eretz-Israel 23:50*-53*.
Giveon, R. 1965. Two Egyptian documents concerning Bashan from the time of Ramses II. Rivista degli Studi
Orientali 40:197-202.
Giveon, R. 1970. Tel Eran. The Manasseh Region 3:1-4. (Hebrew)
Giveon, R. 1981. Geba> (Urk. IV, 786, 114). Gttinger Miszellen 49: 33-36.
Givon, S. 1992. History of the site during the LB Period. In: idem, ed. The Second Season of Excavation at Tel
Harasim 1991. Tel Aviv. pp. 14-19 (Hebrew)
Glass, J., Goren, Y., Bunimovitz, S.and Finkelstein, I. 1993. Petrographic analyses of Middle Bronze III, Late
Bronze and Iron Age I ceramic assemblages. In: Finkelstein, I., Bunimovitz, S. and Lederman, Z., eds. Shiloh,
The Archaeology of a Biblical Site. (Monograph Series of the Institute of Archaeology, Tel-Aviv University
No. 10) Tel Aviv. pp. 271-286.
Glascock, M.D. 1992. Characterization of archaeological ceramics at MURR by Neutron Activation Analysis
and multivariate statistics. In: Neff, H., ed. Chemical Characterization of Ceramic Pastes in Archaeology.
(Monographs in World Archaeology No. 7) Madison Wisconsin. pp. 11-26.
Glock, A. 1992. Ta>anach. The Anchor Bible Dictionary 6. New York. pp. 287-290.
Glock, A. 1993. Ta>anach. In: Stern, E., ed. The The New Encyclopedia of Archaeological Excavations in the Holy
Glueck, N. 1943. Ramoth-Gilead. Bulletin of the American Schools of Oriental Research 92:10-16.
Goldberg, P. 1987. Soils, sediments and Acheulian artifacts at Berekhat Ram, Golan Heights. Soil Micromorphology
1987:583-589.
Goldberg, P., Gould, B., Killebrew, A. and Yellin, J. 1986. Comparison of neutron activation and thin-section
analyses on Late Bronze Age ceramics from Deir el-Balah. In: Olin, J.S. and Blackman, M.J., eds. Proceedings
of the 24th International Archaeometry Symposium. Washington D.C. pp. 341-351.
Goldberg, P., Lev-Yadun, S. and Bar-Yosef, O. 1994. Petrographic thin sections of archaeological sediments: A new
method for palaeobotanical studies. Geoarchaeology 9:243-257.
Gonen, R. 1981. Tell el-Ajjul in the Late Bronze Age city or cemetery? Eretz-Israel 15:69-78. (Hebrew)
Gonen, R. 1984. Urban Canaan in the Late Bronze Period. Bulletin of the American Schools of Oriental Research
253:61-73.
Gophna, R. and Ayalon, E. 1998. Archaeological Survey of Israel, Map of Herzliyya (69). Jerusalem.
Gophna, R. and Porath, Y. 1972. The Land of Ephraim and Manasseh. In: Kochavi, M., ed. Judaea, Samaria and
the Golan Archaeological Survey 1967-1968. Jerusalem. pp. 196-241. (Hebrew)
364
Goren, Y. 1987. Petrography of Chalcolithic Period Pottery Assemblages from Southern Israel. (Unpublished M.A.
thesis, the Hebrew University) Jerusalem. (Hebrew)
Goren, Y. 1988. A petrographic analysis of pottery from sites P14 and DII. Journal of the Israel Prehistoric Society
21:131*-137*.
Goren, Y. 1991a. The Beginnings of Pottery Production in Israel, Technology and Typology of Proto-Historic
Ceramic Assemblages in Eretz-Israel (6th-4th Millenia B.C.E.). (Unpublished Ph.D. dissertation, the Hebrew
University) Jerusalem. (Hebrew)
Goren, Y. 1991b. Pottery from Chalcolithic and Early Bronze Age I mortuary sites in Israel: New aspects of the
development of ceramic technology. Eretz-Israel 21:119-126. (Hebrew with English summary)
Goren, Y. 1991c. The Qatian culture in southern Israel and the Trans-Jordan: New aspects for its denition.
Journal of the Israel Prehistoric Society 23:100-112.
Goren, Y. 1991d. Petrographic examination of the ceramic assemblage from Tel >Amal. Atiqot 20:129-130.
Goren, Y. 1992. Petrographic study of the pottery assemblage from Munhata. In: Garnkel, Y. The Pottery
Assemblage of the Sha>ar Hagolan and Rabah Stages of Munhata (Israel). Paris. pp. 329-360.
Goren, Y. 1995. Shrines and ceramics in Chalcolithic Israel: The view through the petrographic microscope.
Goren, Y. 1996a. The southern Levant during the Early Bronze Age IV: The petrographic perspective. Bulletin of
the American Schools of Oriental Research 303:33-72.
Goren, Y. 1996b. Petrographic study of the pottery assemblage. In: Gopher, A., ed. The Naal Qanah Cave, Earliest
Gold in the Southern Levant. (Monograph Series of the Institute of Archaeology of Tel Aviv University No.12)
Tel-Aviv. pp. 147-154.
Goren, Y. 1997. Excavation of the Courthouse Site at >Akko: Preliminary petrographic analyses of the ceramic
assemblage. >Atiqot, 31:71-74.
Goren, Y. 1999. On determining use of pastoral cave sites: a critical assessment of spherulites in archaeology.
Journal of the Israel Prehistoric Society 29:123-128.
Goren, Y. 2000a. Provenance study of the cuneiform texts from Hazor. Israel Exploration Journal 50:29-42.
Goren, Y. 2000b. Petrographic characteristics of several key southern Levantine ceramic materials. Internet
Archaeology, http://interarch.ac.uk/journal/issue9/index.html.
Goren, Y. 2003. Book Review: McGovern, P.E. 2000. The Foreign Relations of the Hyksos. Bibliotheca
Orientalis LX/1-2:105-110.
Goren, Y. and Fischer, P. 1999. Petrographic study of ceramic assemblages as a regional project: The Early and Late
Bronze Ages in the Central Jordan Valley. In: Gitin, S. and Vaughn, S., eds. The Practical Impact of Science
on Field Archaeology. Athens. pp. 143-146.
Goren, Y. and Gilead, I. 1987. Petrographic analysis of pottery from Shiqmim: A preliminary report. In: Levy, T.E.,
ed. Shiqmim I, Studies Concerning Chalcolithic Societies in the northern Negev Desert, Israel (1982-1984).
(British Archaeological Reports International series 356) Oxford. pp. 411-418.
Goren, Y. and Halperin, N. 2001. Pottery technology. In: Eisenberg, E., Gopher A. and Greenberg, R., eds. Tel Teo,
a Neolithic, Chalcolithic, and Early Bronze Age Site in the Hula Valley. (Israel Antiquities Authority Report
13) Jerusalem. pp. 157-162.
Goren, Y. and Halperin, N. 2004. Selected petrographic analyses. In: Ussishkin, D. The Renewed Archaeological
Excavations at Lachish 1973-1994. (Monograph Series of the Institute of Archaeology of Tel Aviv University
No. 22) Tel Aviv. pp. 2555-2570.
Goren, Y. and Zuckermann, S. 2000. The typology, provenance and technology of the Early Bronze Age I Gray
Burnished Ware: An overview. In: Philip, G. and Baird, D., eds. Breaking the Past: Ceramics and Change in the
Early Bronze Age of the Southern Levant. (Levantine Archaeology Monograph Series) Shefeld.
365
Goren, Y., Segal, I. and Bar-Yosef, O. 1993. Plaster artifacts and the interpretation of Naal Hemar Cave. Journal
of the Israel Prehistoric Society 25:120-131.
Goren, Y., Kletter, R. and Kamaiski, E. 1996. The technology and provenience of the Iron Age gurines from the
City of David: Petrographic analysis. In: Ariel, D.Z. and de Groot, A., eds. City of David Excavations, Final
Report IV. (Qedem 35) Jerusalem.
Goren-Inbar, N., Belitzky, S., Goren, Y., Rabinovich, R. and Saragusti, I. 1992. Gesher Benot Ya>akov - the Bar:
An Acheulian assemblage. Geoarchaeology 7:27-39.
Grant, E. and Wright, E.G. 1939. Ain Shems Excavations Part V. Haverford.
Greenberg, M. 1955. The ab/piru. (American Oriental Series 39). New Haven.
Greenberg, R. and Porat, N. 1996. A third millennium Levantine pottery production center: Typology, petrography,
and provenance of the Metallic Ware of Northern Israel and adjacent regions. Bulletin of the American Schools
of Oriental Research 301:5-24.
Grndahl, F. 1967. Die Personennamen der Texte aus Ugarit. (Studia Pohl 1). Rome.
Hachmann, R. 1989. Kamid el-Loz 1963-1981. German excavations in Lebanon, part one. Berytus 37:5-187.
Hadjisavvas, S. 1989. A Late Cypriote community at Alassa. In: Peltenberg, E., ed. Early Society in Cyprus.
Edinburgh. pp. 32-42.
Hadjisavvas, S. 1991. LC IIC to LC IIIA without intruders: The case of Alassa-Pano Mandilaris. In: Barlow, J., Bolger,
D.L. and Kling, B., eds. Cypriot Ceramics: Reading the Prehistoric Record. Philadelphia. pp. 173-180.
Hadjisavvas, S. 1996. Alassa: A regional center of Alasia? In: strm, P. and Herscher, E., eds. Late Bronze Age
Settlement in Cyprus: Function and Relationship. (Studies in Mediterranean Archaeology and Literature
Pocket-Book 126). Jonsered. pp. 23-38.
Hall, H.R. 1913. The land of Alashiya and the relations of Egypt and Cyprus under the Empire (1500-1100 B.C.).
(Journal of the Manchester Egyptian and Oriental Society) Manchester.
Hampe, R. and Winter, A. 1962. Bei Tpfen und Tpferinnen in Kreta, und Zypern. Mainz.
Harrison, T.P., Hesse, B., Savage, S.H. and Schnurrenberger, D.W. 2000. Urban life in the highlands of central
Jordan: A preliminary report of the 1996 Tall Madaba excavations. Annual of the Department of Antiquities of
Jordan 44:211-228.
Hartal, M. 1989. Northern Golan Heights: The Archaeological Survey as a Source of Regional History. Jerusalem.
(Hebrew)
Hassan, F. 1976. Heavy minerals and the evolution of the modern Nile. Quaternary Research 6:425-444.
Hatzor, Y.H. 2000. Geological Map of Israel 1:50,000, Sheet 6-I, II, Bet-She<an. Jerusalem.
Hawkins, J.D. 1998. Tarkasnawa King of Mira Tarkondemos, Boazky sealings and Karabel. Anatolian Studies
48:1-31.
Heiman, A. 1985. The Geology of the Banias Plateau and the Northern Huleh Valley with Emphasis on the Travertine
Deposits. (Unpublished M.Sc. thesis, The Hebrew University) Jerusalem. (Hebrew with English summary).
Helck, W. 1971. Die Beziehungen gyptens zu Vorderasien im 3. und 2. Jahrtausend v. Chr. Wiesbaden.
Helck, W. 1973. Die Lage der Stadt Tunip. Ugarit-Forschungen 5:286-288.
Hellbing, L. 1979. Alasia Problems. (Studies in Mediterranean Archaeology LVII) Gteborg.
Hemsley, L. M. 1991. Techniques of village pottery production. In: Barlow, J.A. et al., eds. Cypriot Ceramics:
Reading the Prehistoric Period. (University Museum Monograph 74, University Museum Symposium Series
Volume II ) Philadelphia. pp. 215-220.
Herr, L.G., Geraty, L.T., LaBianca, O.S. and Younker, R.W. 1994. Madaba Plains Project: The 1992 excavations at
Tell el-Umeiri, Tell Jalul, and vicinity. Annual of the Department of Antiquities of Jordan 38:147-172.
Herzog, Z. 1989. Middle and Late Bronze Age settlements (Strata XVII-XV). In: Herzog, Z., Rapp, G. and Negbi,
O., eds. Excavations at Tel Michal, Israel. (Monograph Series of the Institute of Archaeology of Tel Aviv
University No. 8) Tel Aviv. pp. 29-42.
366
Hismi, H. 1993. Southern part of the map of Beit Sira. In: Finkelstein, I. and Magen, Y., eds. Archaeological Survey
of the Hill Country of Benjamin. Jerusalem. pp. 97-131 (Hebrew with English summary)
Holms, Y.L. 1971. The location of Alashiya. Journal of the American Oriental Society 91:426-429.
Homs-Fredericq, D. and Hennessy, J.B., eds. 1989. Archaeology of Jordan II2 Field Reports. (Akkadica
Supplementum 8) Leuven.
Hope, C.A., Blauer, H.M. and Riederer, J. 1981. Recent analyses of 18th Dynasty pottery. In: Arnold, D., ed. Studien
zur altegyptischen Keramik. Meinz am Rhein. pp. 139-166.
Horowitz, A. 1979. The Quaternary of Israel. New-York.
Horowitz, A. 2001. The Jordan Rift Valley. Lisse.
Horowitz, W. 1996. An inscribed clay cylinder from Amarna Age Beth Shean. Israel Exploration Journal 46:208-217.
Horowitz, W. 1997. A combined multiplication table on a prism fragment from Hazor. Israel Exploration Journal
47:190-197.
Horowitz, W. and Shaffer, A. 1992. A fragment of a letter from Hazor. Israel Exploration Journal 42:165-167.
Horowitz, W. and Wasserman, N. 2000. An Old Babylonian letter from Hazor with mention of Mari and Ekalltum.
Israel Exploration Journal 50:169-174.
Houston Smith, R. 1993. Pella. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the Holy
Hbner, U. 1986. Aegyptiaca vom Tell el->Orme. Liber Annuus 36: 253-264.
Humbert, J.-B. 1993. Keisan, Tell. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the
Ibach, R.D. 1987. Hesban 5: Archaeological Survey of the Hesban Region. Berrien Springs.
Ibrahim, M.M. 1987. Sahab and its foreign relations. Studies in the History and Archaeology of Jordan 3:73-81.
Ibrahim, M., Sauer, J. and Yassine, K. 1976. The east Jordan Valley survey, 1975. Bulletin of the American Schools
of Oriental Research 222:41-66.
Ilan, D. 1999. Northeastern Israel in the Iron Age I: Cultural, Economic and Political Structures and
Transformations. (Ph.D. dissertation, Tel Aviv University) Tel Aviv.
Ilani, S. 1985. Geological Map of Israel 1:50,000, Sheet 5-III, Netanya. Jerusalem.
Ionas, I. 1998. Pottery in the Cyprus Tradition. Nicosia.
Israel, I. 1993. A survey of ceramic workshops, Naal Lachish - Naal Besor. Hadashot Arkheologiyot 100:91-93.
(Hebrew)
Israel, I. 1995. The economy of the Gaza-Ashkelon region in the Byzantine period in the light of the archaeological
survey and excavations of the 3rd Mile Estate near Ashkelon. Michmanim 8:119-132. (Hebrew)
Issar, A. 1961. The plio-pleistocene geology of the Ashdod area. Bulletin of the Resource Council of Israel 10G:
173-184.
Issar, A. 1968. Geology of the Central Coastal Plain of Israel. Israel Journal of Earth-Sciences 17:16-29.
Izreel, S. 1997. The Amarna Scholarly Tablets. (Cuneiform Monographs 9). Groningen.
Izreel, S. 2001. Adapa and the South Wind, Language Has the Power of Life and Death. Winona Lake.
Izreel, S. and Singer, I. 1990. The Generals Letter from Ugarit. A Linguistic and Historical Reevaluation of RS
20.33 (Ugaritica V, No. 20). Tel Aviv.
Jirku, A. 1930. Durch Palstina und Syrien. Bericht ber eine Forschungsreise in Frhjahr 1929. Zeitschrift des
Deutschen Palstina-Vereins 53:136-166.
Kafa, Z. 1984. Late Bronze Age pottery from Qwelbe (Jordan). Zeitschrift des Deutschen Palstina-Vereins 100:
12-29.
Kafri, U. 1972. Lithostratigraphy and Environments of Deposition, Judea Group, Western and Central Galilee,
Israel. (Geological Survey of Israel Bulletin 54) Jerusalem.
367
Kallai, Z. 1972. The Land of Benjamin and Mt. Ephraim. In: Kochavi, M., ed. Judaea, Samaria and the Golan
Archaeological Survey 1967-1968. Jerusalem. pp. 153-193. (Hebrew)
Kallai, Z. and Tadmor, H. 1969. Bit Ninurta = Beth Horon - on the history of the kingdom of Jerusalem in the
Amarna period. Eretz-Israel 9:138-147. (Hebrew)
Kaplan, J. 1953. Researches in the Gederah el-Mughar area. Bulletin of the Israel Exploration Society 17:138143. (Hebrew)
Kaplan, J. 1957. Archaeological survey of the Jibne district. Bulletin of the Israel Exploration Society 21:199-207.
(Hebrew)
Kaplan, J. 1972. The archaeology and history of Tel Aviv-Jaffa. Biblical Archaeologist 35:66-95.
Karageorghis, V. 1970. Chronique des fouilles et decouvrtes archologique Chypre en 1969. Bulletin de
Correspondance Hellnique 94:191-300.
Karageorghis, V., Asaro, F. and Perlman, I. 1972. Concerning two Mycenaean pictorial sherds from Kouklia
(Palaepaphos), Cyprus. Archologischer Anzeiger 188-197.
Kelso, J.L. 1968. The Excavation of Bethel (1934-1960). (Annual of the American Schools of Oriental Research
39). Cambridge, MA.
Kempinski, A. 1974. Tell el->Ajjul Beth-Aglayim or Sharuen? Israel Exploration Journal 24:145-152.
Kempinski, A. 1986. Kabri and its environment in the Middle Bronze II. In: Yadaya, M., ed. The Western Galilee
Antiquities. Tel Aviv. pp. 66-72. (Hebrew)
Kempinski, A., Miron, E. and Niemeier, W.-D. 1990. Four seasons of excavation at Tel Kabri. Qadmoniot 89-90:
31-41. (Hebrew)
Kerestes, T.M., Lundquist, J.M., Wood, B.G. and Yassine, K. 1977-1978. An archaeological survey of three
reservoir areas in Northern Jordan, 1978. Annual of the Department of Antiquities of Jordan 22:108-135.
Kerschner, M. Forthcoming. Das Keramikbild von Ephesos im 7. und 6. Jh. v. Chr. In: Cobet, J., Graeve, V.v.,
Niemeier, W-D., Zimmermann, K., eds. Fres Ionien: Eine Bestandsaufnahme. (Akten des Symposions
am Panionion 26.9-1.10 1999, Milesische Forschungen 4).
Keswani, P.S. 1989. The pithoi and other plain ware vessels. In: Todd, I.A., ed. Vasilikos Valley Project 3:
Kalavasos-Ayios Dhimitrios II, Ceramics, Objects, Tombs, Specialist Studies. Gteborg. pp. 12-21.
Keswani, P.S. 1993. Models of local exchange in Late Bronze Age Cyprus. Bulletin of the American Schools of
Keswani, P. 1996. Hierarchies, heterarchies, and urbanization processes: The view from Bronze Age Cyprus.
Journal of Mediterranean Archaeology 9:211-250.
Ketin, I. 1962. 1:500,000 lekli Trkiye Jeoloji Haritalsi, Explanatory Text of the Geological Map of Turkey,
Sinop. Ankara.
Ketin, I. 1963. 1:500,000 lekli Trkiye Jeoloji Haritalsi, Explanatory Text of the Geological Map of Turkey,
Kayseri. Ankara.
Khalifeh, I.A. 1988. Sarepta II, The Late Bronze and Iron Age Periods of Area II, X. Beirut.
Khalifeh, I.A. 1997. Sidon. The Oxford Encyclopedia of Archaeology in the Near East V. New York. pp. 38-41.
Killebrew, A. 1996. Pottery kilns from Deir el-Balah and Tel Miqne-Ekron. In: Seger, J.D., ed. Retrieving the
Past, Essays on Archaeological Research and Methodology in Honor of Gus W. Van Beek. Winona Lake.
pp. 135-162.
Killebrew, A. 1998. Ceramic typology and technology of the Late Bronze II and Iron I assemblages from Tel
Miqne-Ekron: The transition from Canaanite to Early Philistine culture. In: Gitin, S, Mazar, A. and Stern,
E., eds. Mediterranean Peoples in Transition: Thirteenth to Early Tenth Centuries BCE. Jerusalem. pp.
379-405.
368
Kitchen, K.A. 1964. Some new light on the Asiatic wars of Ramesses II. Journal of Egyptian Archaeology 50:47-70.
Kitchen, K.A. 1973. The Third Intermediate Period in Egypt (1100-650 B.C.). Warminster.
Kitchen, K.A. 1982. Pharaoh Trimphant. The Life and Times of Ramesses II. Warminster.
Klengel, H. 1963. Der Schiedsspruch der Murili II hinsichtlich Barga und seine bereinkunft mit Duppi-Teup
von Amurru (Kbo III 3). Orientalia 32: 32-55.
Klengel, H. 1965. Geschichte Syriens im 2. Jahrtausend v.u.Z., I. Berlin.
Klengel, H. 1969 Geschichte Syriens im 2. Jahrtausend v.u.Z., II. Berlin.
Klengel, H. 1970 Geschichte Syriens im 2. Jahrtausend v.u.Z., III. Berlin.
Klengel, H. 1984. Sumu/Simyra und die Eleutheros Ebene in der Geschichte Syriene. Klio 66:5-18.
Klengel, H. 1992. Syria 3000 to 300 B.C., A Handbook of Political History. Berlin.
Klengel, K. 1995. Tunip und andere Probleme der historischen Geographie Mittelsyriens. In: van Lerberghe, K. and
Schoors, A., eds. Immigration and Emigration within the Ancient Near East. (Festschrift E. Lipiski.) Leuven.
pp. 125-134.
Kletter, R. and Gorzalczany, A. 2001. A Middle Bronze Age II type of pottery kiln from the coastal plain of Israel.
Levant 33:95-104.
Klinger, J. 1995. Synchronismen in der Epoche vor uppiluliuma I. einige Anmerkungen zur Chronologie
der mittelhethitischen Geschichte. In: Carruba, O., Giorgieri, M. and Mora, C., eds. Atti del II Congresso
Internazionale di Hittitologia. Pavia. pp. 235-247.
Knapp, A.B. 1985. Alashiya, Caphtor/Keftiu and eastern Mediterranean trade: Recent studies in Cypriote
archaeology and history. Journal of Field Archaeology 12:231-250.
Knapp, A.B. 1990. Review of R.S. Merrillees, Alashiya Revisited (Paris, 1987). Bibliotheca Orientalis 47:795-801.
Knapp, A.B. 1993a. Society and Polity at Bronze Age Pella. Shefeld.
Knapp, A.B. 1993b. Social complexity: incipience, emergence, and development on prehistoric Cyprus. Bulletin of
Knapp, A.B. 1994. The prehistory of Cyprus: problems and prospects. Journal of World Prehistory 8:377-453.
Knapp, A.B. ed. 1996a. Near Eastern and Aegean Texts from the Third to the First Millennia BC. (Sources for the
History of Cyprus Vol. II) Altamont, N.Y.
Knapp, A.B. 1996b. Settlement and society on Late Bronze Age Cyprus: dynamics and development. In: strm,
P. and Herscher, E., eds. Late Bronze Age Settlement in Cyprus: Function and Relationship. (Studies in
Mediterranean Archaeology and Literature, Pocket-Book 126) Jonsered. pp. 54-80.
Knapp, A.B. 1997. The Archaeology of Late Bronze Age Cypriot Society: The Study of Settlement, Survey and
Landscape. Glasgow.
Knauf, E.A. 2001. The mists of Ramthalon, or: How Ramoth-Gilead disappeared from the archaeological record.
Biblische Notizen 110: 33-36.
Knudtzon, J.A. 1899. Weitere Studien zu den el-Amarna-Tafeln. Beitrge zur semitischen Sprachwissenschaft 4:
279-337.
Knudtzon, J.A. 1914. Zu den el-Amarna-Tafeln. Orientalistische Literaturzeiting 17:483-494.
Knudtzon, J.A. 1915 (ed. 1964). Die el-Amarna-Tafeln I-II. Leipzig.
Kochavi, M. 1963. Khirbet Kadesh Kadesh Naftali. Bulletin of the Israel Exploration Society 27:165-172.
(Hebrew)
Kochavi, M. 1972. The Land of Judah. In: Kochavi, M., ed. Judaea, Samaria and the Golan Archaeological Survey
1967-1968. Jerusalem. pp. 19-89 (Hebrew).
Kochavi, M. 1974. Khirbet Rabud = Debir. Tel Aviv 1:2-33.
Kochavi, M. 1989. Aphek Antipatris Five Thousand Years of History. Tel Aviv. (Hebrew)
369
Kochavi, M. 1990. Aphek in Canaan: The Egyptian Governors Residence and its Finds (Catalogue, The Israel
Museum). Jerusalem.
Kochavi, M. and Beit-Arieh, I. 1994. Archaeological Survey of Israel Map of Rosh Haayin (78). Jerusalem.
Koyumdjisky, H. and Dan, J. 1969. Forms of Availability of Soil Phosphorus as Related to Genetic Characteristics
of Israeli Soils. (Volcani Center Publication, No. 131) Rehovot.
Kozlov, V.V., Artyemov, A.V. and Kalis, A.F. 1966. The Geological Map of Syria, Scale 1:200,000, Sheets I-36XVIII, I-37,XIII (Trablus, Homs), Explanatory Notes. Damascus.
Krasheninnikov, V.A., Golovin, D.I. and Miravyov, V.I. 1996. The Paleogene of Syria Stratigraphy, Lithology,
Geochronology (Geologisches Jahrbuch 86) Hannover.
Kritsky, G 1991. Beetle gods of ancient Egypt. American Entomologist 37:85-89.
Kronfeld, J., Vogel, J.C., Rosenthal, E. and Weinstein-Evron, M. 1988. Age and paleoclimatic implications of the
Bet Shean travertines. Quaternary Research 30:298-303.
Kuschke, A. 1954. Beitrge zur Siedlungsgeschichte der Biqa>. Zeitschrift des Deutschen Palstina-Vereins 70:
104-129.
Kuschke, A. 1958. Beitrge zur Siedlungsgeschichte der Biqa>. Zeitschrift des Deutschen Palstina-Vereins 74:
81-120.
Kuschke, A. 1979. Das terrain der schlacht bei Qade und Aumarschwege Ramses II. Zeitschrift des Deutschen
Palstina-Vereins 95:7-35.
Kuschke, A., Mittmann, S., Mller, U. and Azouri, I. 1976. Archologischer Survey in der nrdlischen Biqa.
(Beihefte zum Tbinger Atlas des Vorderen Orients, Reihe B, Nr. 11). Wiesbaden.
Lamdan, M., Zipher, D., Huster, Y. and Ronen, A. 1977. A Survey of Prehistoric Sites in Naal Shiqma. (Hebrew)
Landsberger, B. and Tadmor, H. 1964. Fragments of clay liver models from Hazor. Israel Exploration Journal 14:
201-218.
Lapp, P.W. 1967. Taanach by the Waters of Megiddo. Biblical Archaeologist 30:2-27.
Lapp, N.L. 1993. Rumeith, Tell er-. Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the Holy
Lartet, L. 1877. Exploration gologique de la Mer Morte, de la Palestine et de lIdume. Paris.
Lehmann, G. 2001. Phoenicians in western Galilee: First results of an archaeological survey in the hinterland
of Akko. In: Mazar, A., ed. Studies in the Archaeology of the Iron Age in Israel and Jordan. (Journal for
the Study of the Old Testament Supplement Series 331) Shefeld. pp. 65-112.
Lehmann, G., Niemann, H.M. and Zwickel, W. 1996. Zora und Eschtaol: Ein archologischer Oberchensurvey
im Gebiet nrdlich von Bet Schemesch. Ugarit-Forschungen 28:343-442.
Lemaire, A. 1991. Asher et le royaume de Tyr. In: Lipiski, E., ed. Phoenicia and the Bible. (Studia Phoenicia XI)
Leuven. pp. 135-152.
Lenzen, C.J., Gordon, R.L. and McQuitty, A.M. 1985. Excavations at Tell Irbid and Beit Ras, 1985. Annual of the
Department of Antiquities of Jordan 29:151-159.
Leonard, A. 1979. Kataret Es-Samra: A Late Bronze Age cemetery in Transjordan? Bulletin of the American
Schools of Oriental Research 234:53-65.
Leonard, A. 1987a. Tell Jarash Tell el-Husn highway survey. Annual of the Department of Antiquities of Jordan
31:343-390.
Leonard, A. 1987b. The signicance of the Mycenaean pottery found east of the Jordan River. Studies in the History
and Archaeology of Jordan 3:261-266.
Lipiski, E. 1991. The territory of Tyre and the tribe of Asher. In: Lipiski, E., ed. Phoenicia and the Bible. (Studia
Phoenicia XI). Leuven. pp. 153-166.
Liverani, M. 1965. Il fuoruscitismo in Siria nella tarda eta del bronzo. Rivista Storica Italiana 77:315-336.
370
Liverani, M. 1975. La royaut syrienne de lge du bronze rcent. In: Garelli, P., ed. Le palais et la royaut
(Rencontre Assyriologique Internationale, XIX). Paris. pp. 329-356.
Liverani, M. 1998. Le lettere di el-Amarna 1: Le lettere dei Piccoli Re. (Testi del Vicino Oriente antico). Brescia.
Liverani, M. 1999. Le lettere di el-Amarna 2: Le lettere dei Grandi Re. (Testi del Vicino Oriente antico). Brescia.
Lombard, J. P. 1987. Provenance of sand temper in Hohokam ceramics, Arizona. Geoarchaeology 2:91-119.
London, G. 1981. Dung-tempered clay. Journal of Field Archaeology 8:189-196.
London, G. 1985. Decoding Designs: The Late Third Millennium B.C. Pottery from Jebel Qacaqir. (Unpublished
Ph.D. dissertation, The University of Arizona) Tucson.
London, G. 1986. Ceramic ethnoarchaeology in Cyprus. American Schools of Oriental Research Newsletter
38:9-10.
London, G. 1989. A comparison of two contemporaneous lifestyles of the late second millenium B.C. Bulletin of
London, G. 1991. Ethnoarchaeological evidence of variation in Cypriot ceramics and its implications for the
taxonomy of ancient pottery. In: Barlow, J.A. et al., eds. Cypriot Ceramics: Reading the Prehistoric Period.
(University Museum Monograph 74, University Museum Symposium Series Volume II, ) Philadelphia. pp.
221-236.
Loud, G. 1948. Megiddo II: Seasons of 1935-39. Chicago.
Luterbacher, H. P. 1986. Paleogene. In: Planktonic Foraminifera as Stratigraphic Tools, Work-desk on Planktonic
Foraminifera. Tbingen, Feb. 17th to 22nd, 1986.
Mir, A.M., Yellin, J. and Goren, Y. 1992. A re-evaluation of the red and black bowl from Parkers excavations in
Jerusalem. Oxford Journal of Archaeology 11:39-53.
Mir, A.M. 1997. The Material Culture of the Central Jordan Valley during the Middle Bronze II Period: Pottery
and Settlement Pattern. (Ph.D. dissertation, the Hebrew University) Jerusalem.
Magrill, P. and Middleton, A. 2004. Petrographic analysis of Late Bronze Age pottery. In: Ussishkin, D. The Renewed
Archaeological Excavations at Lachich 1973-1994. (Monograph Series of the Institute of Archaeology of Tel
Aviv University No. 22) Tel Aviv. pp. 2554.
Malbran-Labat, F. 1999. Nouvelles donnes pigraphiques sur Chypre et Ougarit. Report of the Department of
Antiquities of Cyprus 1999:121-123.
Malinov, A. and Fati, M. 1985. Projet de dveloppement des rseaux dirrigation et de drainage Tar al Ula-Al
>Asharneh (Al Ghab); Rapport gologique du km 26,938 au km 50,708. Rono (with 1:50,000 and 1:25,000
maps).
Mallory-Greenough, L.M., Greenough, J.D. and Owen, V. 1998. New data for old pots: Trace-element
characterization of ancient Egyptian pottery using ICP-MS. Journal of Archaeological Science 25:85-97.
Ma>oz, Z.U. 1986. The Golan Heights in Antiquity: Geographical History. (Unpublished M.A. thesis, the Hebrew
Mare, W.H. 1991. The 1988 season of excavation at Abila of the Decapolis. Annual of the Department of Antiquities
of Jordan 35:203-221.
Mare, W.H. 1996. The 1994 and 1995 seasons of excavation at Abila of the Decapolis. Annual of the Department
of Antiquities of Jordan 40:259-269.
Marfoe, L. 1979. The integrative transformation: Patterns of sociopolitical organization in southern Syria. Bulletin
of the American Schools of Oriental Research 234:1-42.
Marfoe, L. 1995. Kamid el-Loz 13. The Prehistoric and Early Historic Context of the Site. Bonn.
Mason, R.B. 1994. Islamic Glazed Pottery 700-1250. (Unpublished Ph.D. dissertation) Oxford.
Mason, R.B. and L. Cooper. 1999. Petrographic analysis of Bronze Age pottery from Tell Hadidi, Syria. Levant.
31:135-147.
371
Masson, E. 1971. Rouleau inscrit chypro-minoen trouv Enkomi en 1967. In: Schaeffer, C.F.A., ed. Alasia I.
Paris. pp. 457-477.
Masson, E. 1983. Premiers documents chypro-minoens du site Kalavasos-Ayios Dhimitrios. Report of the
Department of Antiquities of Cyprus 1983:131-141.
Masson, E. 1989. Vestiges crits trouvs sur le site de Kalavasos-Ayios Dhimitrios. In: South, A., Russell, P.A.
and Keswani, P.S., eds. Vasilikos Valley Project 3: Kalavasos-Ayios Dhimitrios II Ceramics, Objects, Tombs,
Specialist Studies. (Studies in Mediterranean Archaeology LXXXI:3) Gteborg. pp. 38-40.
Masson, O. 1973. propos de lle dAlasia. Kadmos 12:98-99.
Master, D. 2001. The Seaport of Ashkelon in the Seventh Century BCE: A Petrographic Study. (Unpublished Ph.D.
dissertation, Harvard University) Cambridge, MA.
Matson, F.R. 1965. Ceramic ecology: an approach to the study of the early cultures of the Near East. In: Matson,
F.R., ed. Ceramics and Man. Chicago. pp. 202-217.
Mazar, A. 1994. Four thousand years of history at Beth-Shean. Qadmoniot 27:66-83 (Hebrew)
Mazar, A. 1997. Four thousand years of history at Tel Beth-Shean: An account of the renewed excavations. Biblical
Archaeologist 60:62-76.
Mazar, A. 1999. The 1997-1998 excavations at Tel Rehov: Preliminary report. Israel Exploration Journal 49:1-42.
Mazar, B. 1952 Yurza-Tell Jemmeh. Palestine Exploration Quarterly: 48-51.
Mazar, B. 1957. The campaign of Pharaoh Shishak to Palestine. Supplement to Vetus Testamentum 4: 57-66.
Mazar, B. 1961. Geshur and Maachah. Journal of Biblical Literature 80:16-28.
Mazar, B. 1975. Cities and Districts in Eretz-Israel. Jerusalem (Hebrew).
Mayerson, P. 1994. The Gaza wine jar (Gazition) and the lost Ashkelon jar (Askalonion). In Mayerson, P., ed.
Monks, Martyrs, Soldiers and Saracens. Jerusalem. pp. 346-351.
Mayerson, P. 1996. Another unreported Ascalonian jar: The Sabitha/Sapation. Israel Exploration Journal
46:258 - 261.
McGovern, P.E. 1986. The Late Bronze and Early Iron Ages of Central Transjordan: The Baq>ah Valley Project,
1977-1981. Philadelphia.
McGovern, P.E. 1989. The Baq>ah valley project 1987, Khirbet Umm ad-Dananir and al-Qesir. Annual of the
Department of Antiquities of Jordan 33:123-136.
McGovern, P.E. 2000. The Foreign Relations of the Hyksos. (British Archaeological Reports International Series
888). Oxford.
Meitlis, I. 1991. Jerusalem, Wadi Zimra. Excavations and Surveys in Israel 10:125-127.
Mendenhall, G.E. 1973. The Tenth Generation: The Origins of the Biblical Tradition. Baltimore and London.
Merrillees, R.S. 1972. Alasia. In: Proceedings of the First International Cypriological Congress. Nicosia. pp. 111-119.
Merrillees, R.S. 1987. Alashiya Revisited. (Cahiers de la Reveue Biblique 22) Paris.
Merrillees, R.S. 1992. The government of Cyprus in the Late Bronze Age. In: strm, P., ed. Acta Cypria. Acts of
the International Congress on Cypriote Archaeology Held in Gteborg on 22-24 August 1991. Part 3. (Studies
in Mediterranean Archaeology and Literature, Pocket-Book 120). Jonsered. pp. 310-328.
Michaelidou-Nicolaou, I. 1980. Regroupement de deux fragments de tablettes dEnkomi avec criture chyprominoenne. Studi Micenei ed Egeo-Anatolici 21:7-16.
Miksa, E. and Heidke, J.M. 1995. Drawing a line in the sands: Models of ceramic temper provenance. In: Heidke,
J.M. and Stark, M.T., eds. The Roosevelt Community Development Study, Volume 2: Ceramic Chronology,
Technology, and Economics. (Center for Desert Archaeology Anthropological Papers No. 14) Tucson, Arizona.
pp. 133-205.
Miller, M.J., ed. 1991. Archaeological Survey of the Kerak Plateau. Atlanta.
372
Mimran, I. 1969. The Geology of the Wadi Malikh Area. (Unpublished MSc. thesis, The Hebrew University) Jerusalem.
(Hebrew)
Mimran, Y. 1972. The Tayasir volcanics, a Lower Cretaceous formation in the Shomeron, central Israel. Geological
Survey of Israel Bulletin 52:1-9.
Mittmann, S. 1970. Beitrge zur Siedlungs- und Territorialgeschichte des nrdlichen Ostjordanlandes. Wiesbaden.
Mommsen, H. 2001. Provenance determination of pottery by trace element analysis: Problems, solutions and
applications. Journal of Radioanalytical and Nuclear Chemistry 247: 657-662.
Mommsen H., Beier, Th., and Hein, A. 2002. A complete chemical grouping of the Berkeley neutron activation
analysis data on Mycenaean pottery. Journal of Archaeological Science 29:613-37.
Mommsen, H., Kreuser, A., Lewandowski, E., and Weber, J., 1991. Provenancing of pottery: Status report
and grouping. In: Hughes, M., Cowell, M. and Hook, D., eds. Neutron Activation and Plasma Emission
Spectrometric Analysis in Archaeology. (British Museum, Occ. Paper 82) London. pp. 57-65.
Mommsen, H., Perlman, I. and Yellin, Y. 1984. The provenience of the LMLK jars. Israel Exploration Journal
34:89-113.
Montet, P. 1928. Byblos et lEgypte, Quatre campagnes de fouilles Gebeil 1921, 1922, 1923, 1924. Paris.
Mor, D. 1973. The Volcanism of the Central Golan Heights. (Unpublished M.Sc. thesis, the Hebrew University).
Jerusalem. (Hebrew)
Moran, W.L. 1975. The Syrian scribe of the Jerusalem Amarna letters. In: Goedicke, H. and Roberts, J.J.M., eds.
Unity and Diversity. Essays in the History, Literature and Religion of the Ancient Near East. Baltimore and
Moran, W.L. 1992. The Amarna Letters. Baltimore.
Muhly, J.D. 1972. The land of Alashiya. Refernces to Alashiya in the texts of the second millennium BC and the
history of Cyprus in the Late Bronze Age. In: Karageorghis, V., ed. Acts of the First International Cyprological
Congress. Nicosia. pp. 201-219.
Muhly, J.D. 1989. The organization of the copper industry in Late Bronze Age Cyprus. In: Peltenburg, E.J., ed.
Early Society in Cyprus. Edinburgh. pp. 298-314.
Muhly, J.D. 1996. The signicance of metals in the Late Bronze Age economy of Cyprus. In: Karageorghis, V. and
Michalides, D., eds. The Development of the Cypriot Economy from the Prehistoric period to the Present Day.
Nicosia. pp. 45-59.
Myers, E.M., ed. 1997. The Oxford Encyclopedia of Archaeology in the Near East. Oxford.
Mynors, H.S. 1986. Mesopotamian Ceramics of the Third Millennium BC With Analysis of Pottery From Abu
Salabikh, Kish and Ur. (Unpublished Ph.D. dissertation, University of Southampton)
Naaman, N. 1975. The Political Disposition and Historical Development of Eretz-Israel According to the Amarna
Letters. (Unpublished Ph.D dissertation, Tel Aviv University) Tel Aviv. (Hebrew with English abstract)
Naaman, N. 1977. Yenoam. Tel Aviv 4:168-177.
Naaman, N. 1979a. The origin and historical background of several Amarna letters. Ugarit-Forschungen 11:
673-684.
Naaman, N. 1979b. The Brook of Egypt and Assyrian policy on the border of Egypt. Tel Aviv 6:68-90.
Naaman, N. 1986. The Canaanite city-states in the Late Bronze Age and the inheritances of the Israelite tribes.
Tarbiz 55:463-488. (Hebrew)
Naaman, N. 1988a. The Southern Shephelah during the Late Bronze Age according to the cuneiform documents.
In: Stern, E. and Urman, D., eds. Man and Environment in the Southern Shephelah. Beersheba. pp. 93-98.
(Hebrew)
Naaman, N. 1988b. Historical-geographical aspects of the Amarna Tablets. In: Proceedings of the Ninth World
Congress of Jewish Studies, Bible Studies and Ancient Near East, Jerusalem 1985. Jerusalem. pp. 17-26.
373
Naaman, N. 1988c. Pharaonic lands in the Jezreel Valley in the Late Bronze Age. In: Heltzer, M. and Lipiski, E.,
eds. Society and Economy in the Eastern Mediterranean (c. 1500-1000 B.C.). Leuven. pp. 177-186.
Naaman, N. 1988d. Biryawaza of Damascus and the date of the Kmid el-Lz >Apiru letters. Ugarit-Forschungen
20:179-193.
Naaman, N. 1990. On gods and scribal traditions in the Amarna letters. Ugarit-Forschungen 22:247-255.
Naaman, N. 1992. Canaanite Jerusalem and its Central Hill Country neighbours in the second millennium B.C.E.
Ugarit-Forschungen 24:275-291.
Naaman, N. 1994. The Canaanites and their land - a rejoinder. Ugarit-Forschungen 26:397-418.
Naaman, N. 1996. Looking for the King of Egypts judgment. Revue dAssyriologie et dArchologie Orientale
90:145-159.
Naaman, N. 1997. The network of Canaanite kingdoms and the city of Ashdod. Ugarit-Forschungen 29:599-612.
Naaman, N. 1998a. Collations of some Amarna tablets in the Berlin Museum. Nouvelles Assyriologiques Brves
et Utilitaires 1998: No. 50.
Naaman, N. 1998b. EA 207. Nouvelles Assyriologiques Brves et Utilitaires 1998: No. 97.
Naaman, N. 1999. Milkilus messenger and the sons of Lab>ayu (EA 250:28-34). Nouvelles Assyriologiques
Brves et Utilitaires 1999: No. 27.
Naaman, N. 2000. Rubutu/Aruboth. Ugarit-Forschungen 32:373-383.
Naaman, N. 2001a. EA 200. Nouvelles Assyriologiques Brves et Utilitaires 2001: No. 4.
Naaman, N. 2001b. The city of Anaharath in an Amarna letter. Nouvelles Assyriologiques Brves et Utilitaires
2001: No. 31.
Naaman, N. 2002. Dispatching Canaanite maidservants to the Pharaoh. Ancient Near Eastern Studies 39:76-82.
Nahmias, J. 1969. Source rocks of the Saqiye Group sediments in the Coastal Plain of Israel a heavy mineral
study. Israel Journal of Earth-Sciences 18:1-16.
Naggar, Z.R. 1966. Stratigraphy and classication of Type Esna Group of Egypt. Bulletin of the American
Association of Petroleum Geologists 50: 1455-1477.
Neeman, Y. 1969. A summary of the results of the archaeological survey in the Manasseh region, Part one. The
Manasseh Region 1. (Hebrew)
Neeman, Y. 1974. An archaeological survey of the Manasseh region, Part two. The Manasseh Region 6. (Hebrew)
Neeman, Y., Sender, S. and Oren, E. 2000. Archaeological Survey of Israel, Map of Mikhmoret (52), Map of Hadera
(53). Jerusalem.
Negbi, O. 1986. The climax of urban development in Bronze Age Cyprus. Report of the Department of Antiquities
of Cyprus 1986:97-121.
Netser, M. and Gvirtzman, G. 1994. Geological Map of the Quaternary of Gush Dan (Central Coastal Plain,
Israel). (Geological Survey of Israel Annual Meeting, Nof Ginosar) Jerualem. p. 75.
Nicklin, K. 1979. The location of pottery manufacture. Man 14: 436-458.
Niemann, M.N. 1999. Zorah, Eshtaol, Beth-shemesh and Dans migration to the south: A region and its tradition in
the Late Bronze and Iron Ages. Journal for the Study of the Old Testament 86:25-48.
Nir, D. 1960. Recherches granulomtriques dans les oueds de la rgion de Beth-Chean. (Bulletin of the Research
Council of Israel 9G) Jerusalem. pp. 192-206.
Nir, D. 1970. Geomorphology of Eretz-Israel. Jerusalem. (Hebrew)
Nir, Y. 1989. Sedimentological Aspects of the Israel and Sinai Mediterranean Coasts. (Geological Survey of Israel
Internal Report). Jerusalem. (Hebrew)
Noll, W. 1976. Mineralogy and techniques of the early ceramics of Great Mesopotamia. Neues Jahrbuch fr
Mineralogie Abhandlung 127(3):261-288.
374
Nougayrol, J. 1955. Le palais royal dUgarit III - Textes accadiens des archives sud. (Mission de Ras
Shamra 6). Paris.
Nougayrol, J., Laroche, E., Virolleaud, C. and Schaeffer, F.A. 1968. Ugaritica V. Nouveaux textes accadiens,
hourrites et ugaritiques des archives et bibliothques prives dUgarit. Commentaires des textes historiques
(Premire Partie). Paris.
Novak, M. and Pflzner, P. 2000. Ausgrabungen in Tall Misrife-Qatna 1999: Vorbericht der deutschen Komponente des
internationalen Kooperationprojektes. Mitteilungen der deutschen Orient-Gesellschaft zu Berlin 132:253-295.
Ofer, A. 1989. Excavations at biblical Hebron. Qadmoniot 87-88:88-93. (Hebrew)
Ofer, A. 1993. The Highland of Judah during the Biblical Period. (Ph.D. dissertation, Tel Aviv University) Tel Aviv.
(Hebrew)
Olami, Y. 1981. Archaeological Survey of Israel Daliya Map (31) 15-22. Jerusalem.
Oppenheim, M. 1959. The Vulcanological Phenomena of South-Eastern Lower Galilee. (Unpublished Ph.D.
dissertation, the Hebrew University). Jerusalem. (Hebrew)
Oren, E.D. 1984. Governors residencies in Canaan under the New Kingdom: A case of Egyptian administration.
Journal of the Society for the Study of Egyptian Antiquities 14:37-56.
Oren, E.D. 1987. The Way of Horus in North Sinai. In: Rainey, A.F., ed. Egypt, Israel, Sinai. Tel-Aviv. pp. 69-120.
Oren, E.D. 1993. Haror, Tel. In: Stern, E., ed. The New Encyclopedia of Archaeological Excavations in the
Oren, E.D. 1997. The Kingdom of Sharuhen and the Hyksos Kingdom. In: Oren, E.D., ed. The Hyksos: New
Historical and Archaeological Perspectives. Philadelphia. pp. 253-283.
Oren, E.D. and Mazar, A. 1974. Notes and News, Tel Maaravim. Israel Exploration Journal 24:269-270.
Oren, E., Yekutieli, Y., Nahshoni, P. and Feinstein, R. 1991. Tel Haror - after six seasons. Qadmoniot 93-94:1-19.
(Hebrew)
Pantazis, T.M. 1967 (ed. 1978). The Geology and Mineral Resources of the Pharmakas-Kalavasos Area. (Geological
Survey Department of Cyprus Memoir No. 8). Nicosia.
Parr, P. 1983. The Tell Nebi Mend Project. Annales Archologiques Arabes Syriennes 33(2):99-117.
Parr, P.J. 1997. Nebi Mend, Tell. The Oxford Encyclopedia of Archaeology in the Near East IV. New York. pp. 114-115.
Parrot, J.F. 1974. Lassemblage ophiolithique du Bar-Bassit (nord ouest de la Syrie): tude ptrographique et
gochimique du complexe lonien, des laves en coussins qui lui sont associes, et dune partie des formations
effusives du volcano-sdimentaire. Cahier ORSTOM, Service Gologique 6:97-126.
Parrot, J.F. 1977. Assemblage ophiolitique du Bar-Bassit et termes effusifs du volcano-sdimentaire. Paris.
Parrot, J.F. 1980. The Bar-Bassit (northwestern Syria) ophiolitic area. Ooliti 2:279-295.
Peltenburg, E.J. 1996. From isolation to state formation in Cyprus, c. 3500-1500 B.C. In: Karageorghis, V. and
Michalides, D., eds. The Development of the Cypriot Economy from the Prehistoric to the Present Day.
Nicosia. pp. 17-43.
Perlman, I., and Asaro, F. 1969. Pottery analysis by neutron activation. Archaeometry 11: 21-52.
Petrie, W.M.F. 1894. Tell el Amarna. London.
Philip, G., Jabour, F., Beck, A., Bshesh, M., Grove, J., Kirk, A. and Millard, A. 2002. Settlement and landscape
development in the Homs Region, Syria: Research questions, preliminary results 1999-2000 and future
potential. Levant 34:1-23.
Pomerancblum, M. 1966. The distribution of heavy minerals and their hydraulic equivalent in sediments of the
Mediterranean continental shelf of Israel. Journal of Sedimentary Petrology 36/1:162-174.
Ponikarov, V.P., (ed.) 1964. The Geological Map of Syria, 1:200,000 (20 sheets with geological cross sections and
explanatory notes). Moscow.
375
Ponikarov, V.P., Kazmin, V.G., Mikhailov, I.A., Razvaliayev, A.V., Krasheninnikov, V.A., Kozlov, V.V., SulidiKondratev, E.D., Mikhailov, K.Y., Kulakov, V.V., Faradzhev, V.A. and Mirzayev, K.. 1966. The Geology of
Syria, Explanatory Notes on the Geological Map of Syria, Scale 1:500,000. Vol. 1: Stratigraphy, Igneous Rocks
& Tectonics. (Syrian Arab Republic, Ministry of Industry, Department of Geological and Mineral Research)
Porat, N. 1984. Petrography and Mineralogy of Pottery from Archaeological Sites from Southern Israel.
(Unpublished M.Sc. thesis, the Hebrew University) Jerusalem. (Hebrew)
Porat, N., 1987. Local industry of Egyptian pottery in southern Palestine during the Early Bronze I period. Bulletin
of the Egyptological Seminar 8:109-129.
Porat, N. 1989a. Petrography of pottery from southern Israel and Sinai. In: de Miroschedji, P., ed. Lurbanisation de
la Palestine lge du Bronze ancien, bilan et perspectives des recherches actuelles I. (British Archaeological
Reports International Series 527) Oxford. pp. 169-188.
Porat, N. 1989b. Composition of pottery - application to the study of the interrelations between Canaan and Egypt
during the 3rd millennium B.C.. (Unpublished Ph.D. dissertation, the Hebrew University) Jerusalem.
Porat, N. 1998. Pottery analysis from Site 12. In: Epstein, C. The Chalcolithic Culture of the Golan. Jerusalem. pp.
344-345.
Porat, N. and Goren, Y. 2002. Petrography of the Naqada IIIa Canaanite pottery from Tomb U-j in Abydos. In: van
den Brink, E.C.M. and Levy, T.E., eds. Egypt and the Levant, Interrelations from the 4th Through the Early 3rd
Millennium B.C.E. London and New York. pp. 252-270.
Porat, N., Yellin, J., Heller-Kallai, L. and Halicz, L. 1991. Correlation between petrography, NAA, and ICP
analyses: application to Early Bronze Egyptian pottery from Canaan. Geoarchaeology 6:133-149.
Porath, Y., Dar, S. and Applebaum, S. 1985. The History and Archaeology of Emeq-Hefer. Tel Aviv. (Hebrew)
Porath, Y., Yannai, E. and Kasher, A. 1999. Archaeological remains at Jatt. >Atiqot 37:1-78. (Hebrew)
Portugali, Y. 1982. A eld methodology for regional archaeology (the Jezreel Valley survey, 1981). Tel Aviv 9:170188.
Prausnitz, M.W. 1975. The planning of the Middle Bronze Age town at Achzib and its defences. Israel Exploration
Journal 25:202-210.
Pythian-Adams, W.J. 1923. Second report on soundings at Gaza. Palestine Exploration Fund Quarterly Statement 18-30.
Raban, A. 1982. Archaeological Survey of Israel Nahalal Map (28) 16-23. Jerusalem.
Raban, A. 1999. Archaeological Survey of Israel Map of Mishmar Ha->emeq (32). Jerusalem.
Rabikovitz, S. 1981. The Soils of Israel: Formation, Nature, and Properties. Tel Aviv. (Hebrew with English summary).
Rainey, A.F. 1968. Gath-padalla. Israel Exploration Journal 18:1-14.
Rainey, A.F. 1975a. The identication of Philistine Gath - A problem in source analysis for historical geography.
Eretz-Israel 12:63*-76*.
Rainey, A.F. 1975b. Toponymic problems: >Ain-Shasu. Tel Aviv 2:13-16.
Rainey, A.F. 1976a. Toponymic problems: Shim>on-Shimron. Tel Aviv 3:57-69.
Rainey, A.F. 1976b. A tri-lingual cuneiform fragment from Tel Aphek. Tel Aviv 3:137-140.
Rainey, A. F. 1978. El Amarna Tablets 359-379. Supplement to J.A. Knudtzon, Die El-Amarna-Tafeln. 2nd revised
ed. (AOAT 8). Kevelaer and Neukirchen-Vluyn.
Rainey, A.F. 1979. Toponymic problems (contd.): Rium = roiu? Tel Aviv 6:158-162.
Rainey, A.F. 1989. Review of W.L. Moran, Les lettres del-Amarna, Correspondance diplomatique du pharaon,
Paris 1987. Biblica 70:566-572.
Rainey, A.F. 1993. Sharn/Sharuhen - The problem of identication. Eretz-Israel 24:178*-187*.
Rainey, A.F. 1995. Unruly elements in Late Bronze Canaanite society. In: Wright, D.P., Freedman, D.N. and
Hurvitz, A., eds. Pomegranates and Golden Bells: Studies in Biblical, Jewish, and Near Eastern Ritual, Law,
and Literature in Honor of Jacob Milgrom. Winona Lake. pp. 481-496.
376
Rainey, A.F. 1995/1996. A new English translation of the Amarna letters. Archiv fr Orientforschung 42/43:
109-121.
Rainey, A.F. 1998. Syntax, hermeneutics and history. Israel Exploration Journal 48:239-251.
Rainey, A.F. 1999. Ta>anach Letters. Eretz-Israel 26:153*-162*.
Rainey, A.F. 2003. Some Amarna collations. Eretz-Israel 27:192*-202*.
Redford, D.B. 1965. The coregency of Tuthmosis III and Amenophis II. Journal of Egyptian Archaeology 51:107122.
Redford, D.B. 1982. A Bronze Age itinerary in Transjordan (Nos. 89-101 of Thutmose IIIs list of Asiatic
toponyms). Journal for the Study of Egyptian Antiquities 12:55-74.
Reviv, H. 1963. Regarding the history of the territory of Shekhem in the el-Amarna period. Tarbitz 33(1):1-7.
(Hebrew)
Rice, P.M. 1977. Whiteware pottery production in the Valley of Guatemala: specialization and resource utilization.
Journal of Field Archaeology 4:221-233.
Rice, P.M. 1981. Evolution of specialized pottery production: a trial model. Current Anthropology 22:219-240.
Rice, P.M. 1987. Pottery Analysis: A Sourcebook. Chicago.
Riedel, W. 1920. Untersuchungen zu den Tell-elAmarna-Briefen. Tbingen.
Robertson, A.H.F. 1986. The Hatay ophiolite (southern Turkey) in its eastern Mediterranean tectonic context: A
report on some aspects of the eld excursion. Ooliti 11:105-119.
Rohrlich, V. and Metzer, A. 1980. Potential iron ores in the Lower Cretaceous of Israel and their origin. Israel
Rognon, P., Coude-Gaussen, G., Fedoroff, N. and Goldberg, P. 1987. Micromorphology of loess in the Northern
Negev (Israel). Soil Micromorphology: 631-638.
Ronen, A. and Olami, Y. 1983. Archaeological Survey of Israel Map of HaifaEast (23) 15-24. Jerusalem.
Ross, J.F. 1967. Gezer in the Tell el-Amarna letters. Biblical Archaeologist 30:62-70.
Saidah, R. 1969. Archaeology in the Lebanon 1968-1969. Berytus 18:119-142.
Salam-Sarkis, H. 1972. Ardata-Ard dans le Liban-Nord: Une nouvelle cit cananenne identie. Mlanges de
lUniversit Saint-Joseph 47:123-145.
Salame-Sarkis, H. 1973. Chronique archologique du Liban-Nord. Bulletin du Muse de Beyrouth 26:91-102.
Sandler, A. and Herut, B. 2000. Composition of clays along the continental shelf off Israel: Contribution of the Nile
versus local sources. Marine Geology 167:339-354.
Sanlaville, P. 1977. tude gomorphologique de la rgion littorale du Liban. Beirut.
Sanlaville, P. 1979. Quaternaire et prhistoire du Nahr el Kbir septentrional. (Travaux de la RCP 438, CNRSditions) Paris.
Sanlaville, P., Besanon, J., Copland, L. and Muhesen, S. 1993. Le Palolithique de la Valle moyenne de lOronte
(Syrie). (British Archaeological Reports International Series 587) Oxford.
Sartorio D. and Venturini S. 1988. Southern Tethys Biofacies. (Agip Stratigraphic Dpt., AGIP S.p.A.)
Sass, B. 1991. Studia Alphabetica. On the Origin and Early History of the Northwestern Semitic, South Semitic and
Greek Alphabets. (Orbis Biblicum Orientalis 102) Freibourg.
Sass, E. 1957. The Volcanic Occurences in Mt. Carmel. (Unpublished M.Sc. thesis, the Hebrew University) Jerusalem.
(Hebrew)
Sass, E. 1968. Geology of the Umm el Fahm area, northern Israel. Israel Journal of Earth-Sciences 17:115-130.
Sass, E. 1980. Late Cretaceous volcanism in Mount Carmel, Israel. Israel Journal of Earth-Sciences 29:8-24.
Sayre, E.V. 1975. Brookhaven Procedures for Statistical Analyses of Multivariate Archaeometric Data.
(Unpublished manuscript)
Schmitt, G. 1987. Geba, Getta und Gintikirmil. Zeitschrift des Deutschen Palstina-Vereins 103:22-48.
377
Schreier, C.M. 1988. Geochemie und Mineralogie oberkretazischer und alttertirer Pelite der stlichen Wste
gyptens und ihre geologische Interpretation. (Berliner Geowissenschaftliche Abhandlungen 97) Berlin.
Schubert, P. 1986. Petrographic modal analysis - a necessary complement to chemical analysis of ceramic coarse
ware. Archaeometry 28:2:163-178.
Schulman, N. 1962. The Geology of the Central Jordan Valley. (Unpublished Ph.D. dissertation, the Hebrew
University) Jerusalem. (Hebrew, with English abstract)
Schroeder, O. 1915. Vorderasiatische Schriftdenkmler der Kniglichen Museen zu Berlin, XI-XII. Berlin.
Seeden, H. 1983. Busra 1983: An Umayyad farmhouse and Bronze Age occupation levels. Annales Archologiques
Arabes Syriennes 33(2):161-173.
Seeden, H. 1986. Bronze Age village occupation at Busra, AUB excavations on the northwest tell 1983-84. Berytus
34:11-81.
Seeden, H. 1988. Busra 1983-84: Second archaeological report. Damaszener Mitteilungen 3:387-411.
Seeligmann, A. 1984. The identication of Gaba Hippeon. Palestine Exploration Quarterly 116:89-93.
Serpico, M., Bourriau, J., Smith, L., Goren, Y., Stern, B. and Heron, C. 2003. Commodities and containers:
A project to study Canaanite amphorae imported into Egypt during the New Kingdom. In: Bietak, M.,
ed. The Synchronisation of Civilisations in the Eastern Mediterranean in the Second Millennium B.C.
II. Wien. pp. 365-375.
Shaliv, G. 1972. The Geology of the Wadi Far>ah Area (Part I: Geology). (Unpublished M.Sc. thesis, the Hebrew
Shaliv, G., Mimran, Y. and Hatzor, Y. 1992. The sedimentology and structural history of the Bet She>an area. Israel
Shatsky, V.N., Kazima, V.G. and Kulakov, V.V. 1966. The Geological Map of Syria, Scale 1:200,000, Sheets I-37XIX, I-36,XXIV, Explanatory Notes. Damascus.
Shavit, A. 1992. The Ayalon Valley and its Vicinity During the Bronze and Iron Ages. (Unpublished M.A thesis, Tel
Aviv University) Tel Aviv. (Hebrew, with English abstract).
Shaw, S.W. 1947. Geological Map of Southern Palestine, 1:250,000 Scale, with Explanatory Notes. Jerusalem.
Shenhav, H. 1964. Petrographic Analyses of Sherds from the Hazor Excavations (Unpublished M.Sc. thesis, the
Hebrew University). Jerusalem.
Shepard, A. 1965. Ceramics for the Archaeologist. Washington.
Shiloh, Y. 1984. Excavations at the City of David I: 1978-1982 Interim Report of the First Five Seasons. (Qedem
19) Jerusalem.
Shoval, S. 1996. Microscopy of Minerals and Magmatic Rocks, Geology of the Carmel. Guide for the Geological
Excursion to the Volcanoes of Karnei Hittin and Kerem Maharal. (The Open University) Tel-Aviv. (Hebrew)
Shoval, S., Beck, P., Kirsh, Y., Eisenberg-Levy, D. and Yadin, E. 1992. Using calcareous and non-calcareous clays
as raw materials for Iron Age vessels. In: Polishook, B., ed. Israel Geological Society Annual Meeting, 1992,
Ashqelon 30 March-1 April. Jerusalem. pp. 136
Siegelman, A. 1984. The identication of Gaba Hippeon. Palestine Exploration Quarterly 116:89-93.
Singer, I. 1983. Western Anatolia in the thirteenth century B.C. according to the Hittite sources. Anatolian Studies
33:205-217.
Singer, I. 1991. Appendix: The history of Amurru. In: Izreel, S. A Linguistic Study Volume I. (Harvard Semitic
Studies 40) Atlanta. pp. 134-195.
Singer, I. 1999. A political history of Ugarit. In: Watson, W.G.E. and Wyatt, N., eds. Handbook of Ugaritic Studies.
Leiden. pp. 666-673.
Singer-Avitz, L. and Levy, Y. 1992. An MBIIa kiln at the Naal Soreq Site. >Atiqot 21: 9*-14*. (Hebrew)
378
Sivan, D. 1996. Paleogeography of the Galilee Coastal Plain During the Quarternary (Geological Survey of Israel
GSI/18/96) Jerusalem. (Hebrew with English summary).
Sivan, D., Gvirtzman, G. and Sass, E. 1999. Quaternary stratigraphy and paleogeography of the Galilee Coastal
Plain, Israel. Quaternary Research 51:280-294
Slatkine, A. and Pomerancblum, M. 1958. Unstable Heavy Minerals as Criteria of Depositional Environment.
(Geological Survey of Israel Bulletin 19) Jerusalem.
Smith, G.A. 1901. Notes of a journey through Hauran, with inscriptions found by the way. Palestine Exploration
Fund Quarterly Statement 33:340-361.
Smith, R.H. and Potts, T. 1992. The Middle and Late Bronze Ages. In: McNicoll, A.W. et al., Pella in Jordan 2.
Sydney. pp. 35-81.
Sneh, A., Sass, E., Bein, A., Arad, A. and Rosensaft, M. 1996. Geological Map of Israel 1:50,000, Sheet 5-I,
Hadera. Jerusalem.
Sneh, A., Bartov, Y. and Rosensaft, M. 1998. Geological Map of Israel 1:200,000. Ministry of Infrastructures,
Geological Survey of Israel. Jerusalem.
South, A., Russell, P.A. and Keswani, P.S. 1989. Vasilikos Valley Project 3: Kalavasos-Ayios Dhimitrios II Ceramics,
Objects, Tombs, Specialist Studies (Studies in Mediterranean Archaeology LXXXI:3). Gteborg.
South, A. 1996. Kalavasos-Ayios Dhimitrios and the organisation of Late Bronze Age Cyprus. In: strm, P. and
Herscher, E., eds. Late Bronze Age Settlement in Cyprus: Function and Relationship. (Studies in Mediterranean
Archaeology and Literature, Pocket-Book 126) Jonsered. pp. 39-49.
South, A. 1997. Kalavasos-Ayios Dhimitrios 1992-1996. Report of the Department of Antiquities of Cyprus 1997.
pp. 151-175.
Stager, L.E. 1995. The impact of the Sea Peoples in Canaan (1185-1050 BCE). In: Levy, T.E., ed. The Archaeology
of Society in the Holy Land. London. pp. 332-348.
Stepansky, Y. 1999. The Periphery of Hazor during the Bronze Age, the Iron Age and the Persian Period: A
Regional Archaeological Study. (M.A. thesis, Tel Aviv University) Tel Aviv. (Hebrew)
Stern, E. 1984. Excavations at Tel Mevorakh (1973-1976) Part Two: The Bronze Age. (Qedem 18) Jerusalem.
Stern, E. and Beit-Arieh, I. 1979. Excavations at Tel Kedesh (Tell Abu Qudeis). Tel Aviv 6:1-25.
Stieglitz, R.R. 1974. Ugaritic Md the harbor of Yabne-Yam? Journal of the American Oriental Society 94:137-138.
Stienstra, P. 1986. Systematic macroscopic description of the texture and composition of ancient pottery - some
basic methods. Newsletter of the Department of Pottery Technology IV:29-48.
Strange, J. 2000. The Late Bronze Age in northern Jordan in the light of the nds at Tell el-Fukhar. In: Stager, L.E.,
Greene, J.A. and Coogan, M.D., eds. The Archaeology of Jordan and Beyond, Essays in Honor of James A.
Sauer. Winona Lake. pp. 476-481.
Strange, J.F., Groh, D.E. and Longstaff, Th.R. 1995. Excavations at Sepphoris: The location and identication of
Shikhin, Part II. Israel Exploration Journal 45:171-187.
Stubbings, F.H. 1951. Mycenaean Pottery from the Levant. Cambridge.
Tadmor, M. and Prausnitz, M. 1959. Excavations at Rosh Hanniqra. >Atiqot 2:72-88.
Taitel-Goldman, N., Heller-Kallai, L. and Sass, E. 1992. Clay minerals and feldspars in argillaceous strata of the
Judea Group in the Jerusalem Hills. Israel Journal of Earth-Sciences 44:71-79.
Taitel-Goldman, N. 1993. Mineralogy of the Argillaceous Strata of the Judean Group in the Jerusalem Hills.
(Unpublished M.Sc. thesis, the Hebrew University) Jerusalem.
Taraqji, A. 1993. Tell Saqqa. Syria 70:453-456.
Tejirian, H.G. 1970. Preliminary Study of Submarine Topography and Sediments of Junieh Bay, Central Lebanon.
(Unpublished M.Sc. thesis, University of Beirouth) Beirut.
379
Tekeli, O. and Erendil, M. 1986. Geology and petrology of the Kizilda Ophiolite (Hatay). Bulletin of the Mineral
Resources and Exploration Institute of Turkey 107:21-37.
Thalmann, J.-P. 1991. Lge du Bronze Tell >Arqa: Bilan et perspectives (1981-1991). Berytus 39:21-38.
Thalmann, J.-P. 1993. Lge du Bronze Tell >Arqa (Liban-Nord). (Ph.D. dissertation, Universit de Paris) Paris.
Thalmann, J.-P. 2000. Tell >Arqa. BAAL 4:5-74.
Thierry, J. 1980. Ophiolites of Turkey. Ooliti 2:199-237.
Thureau-Dangin, F. 1933. La stle dAsharn. Revue dAssyriologie et dArchologie Orientale 30:53-56.
Tite, M.S., Maniatis, Y., Meeks, N.D., Bimson, M., Hughes, M.J. and Leppard, S.C. 1982. Technological studies of
ancient ceramics from the Near East, Aegean, and southeast Europe. In: Wertime, T.E. and Wertime, S.F. Early
Pyrotechnology, the Evolution of the First Fire-Using Industries. Washington D.C. pp. 61-72.
Tite, M.S., Huges, M.J., Freestone, I.C., Meeks, N.D. and Bimson, M. 1990. Technological characterisation of
refractory ceramics from Timna. In: Rothenberg, B., ed. The Ancient Metallurgy of Copper: ArchaeologyExperiment-Theory. London. pp. 158-175.
Tobia, S.K. and Syre, E.V. 1974. An analytical comparison of various Egyptian soils, clays, shales and some ancient
pottery by Neutron Activation Analysis. In: Bishay, A., ed. Recent Advances in Science and Technology of
Materials. New York. pp. 99-128.
Tolun, N. and Pamir, H.N. 1975. 1:500,000 lekli A Trkiye Jeoloji Haritalsi, Hatay (with explanatory notes).
Ankara (Turkish and English).
Tufnell, O. 1958. Lachish IV: The Bronze Age. London.
Tzaferis, V. and Hess, O. 1992. A Late Bronze Age tomb at Tell Eitun. >Atiqot 21:9-21.
Ulu, . 1996a. 1:100,000 lekli Ainsama Nitelikli Trkiye Jeoloji Haritalari Serisi No 43, Suru L 25 Paftasi.
Ankara (Turkish with English abstract).
Ulu, . 1996b. 1:100,000 lekli Ainsama Nitelikli Trkiye Jeoloji Haritalari Serisi No 42, Urfa K 25 Paftasi.
Ankara (Turkish with English abstract).
Ussishkin, D. 1985. Levels VII and VI at Tel Lachish and the end of the Late Bronze age in Canaan. In: Tubb, J.N.,
ed. Palestine in the Bronze and Iron Ages, Papers in Honour of Olga Tufnell. London. pp. 213-230.
Ussishkin, D. 1995. The destruction of Megiddo at the end of the Late Bronze Age and its historical signicance.
Tel Aviv 22:240-267.
Ussishkin, D. and Woodhead, J. 1994. Excavations at Tel Jezreel 1992-1993: Second preliminary report. Levant
26:1-48.
Van Soldt, W. 2002. Additions to studies on the skinu-ofcial (1): UF 33 (2001). Nouvelles Assyriologiques Brves
et Utilitaires 2002: No. 74.
Vaughan, S. 1987. A Fabric Analysis of Late Cypriote Base Ring Ware: Studies in Ceramic Technology, Petrology,
Geochemistry and Mineralogy. (Unpublished Ph.D. dissertation, London University College) London.
Vaughan, S. 1989. Petrographic and microprobe analyses of Base Ring Ware. In: South, A., Russell, P.A. and
Keswani, P.S., eds. Vasilikos Valley Project 3: Kalavasos-Ayios Dhimitrios II Ceramics, Objects, Tombs,
Specialist Studies. (Studies in Mediterranean Archaeology LXXXI:3) Gteborg. pp. 78-81.
Vaughan, S. 1991. Late Cypriote Base Ring wares: studies in raw materials and technology. In: Middleton, A. and
Freestone, I., eds. Recent Developments in Ceramic Petrology (British Museum Occasional Paper No. 81)
Vita, J.-P. 2000. The Gezer-Corpus von el-Amarna: Umfang und Schreiber. Zeitschrift fr Assyriologie 90:70-77.
Vita-Finzi, C. 1978. Archaeological Sites and their Setting. London.
Wachsmann, S. 2000. To the Sea of the Philistines. In: Oren, E.D., ed. The Sea Peoples and Their World: A
Reassessment. Philadelphia. pp. 103-143.
380
Wainwright, G.A. 1914-15. Alashia = Alasa; and Asy. Klio 14:1-36.

Walker, C.B.F. 1979. Another fragment from El-Amarna. Journal of Cuneiform Studies 31:249.
Walley, C. 1997. The lithostratigraphy of Lebanon, a review. Lebanese Science Bulletin 10:81-108.
Weber, O. 1907. Die Literatur der babylonier u. Assyurer. Leipzig.
Weber, O. 1915. In: Knudtzon, J.A. Die El-Amarna-Tafeln II. Leipsig.
Weinstein, J.M. 1981. The Egyptian empire in Palestine: A reassessment. Bulletin of the American Schools of
Weippert, M. 1970. Die Nomadenquelle. In: Kuschke, A. and Kutsch, E., eds. Archologie und altes Testamnent.
Festschrift fr Kurt Galling. Tbingen. pp. 259-272.
Weippert, M. 1974. Semitische Nomaden des zweiten Jahrtausends. ber die 3w der gyptischen Quellen.
Biblica 55:265-280, 427-433.
Weippert, M. 1997. Isralites, Aramens et Assyriens dans la Transjordanie septentrionale. Zeitschrift des
Deutschen Palstina-Vereins 113:19-38.
Weisman, R.M. 1996. Appendix 7: Petrographic analysis of pottery from Alambra. In: Coleman, J.E., Barlow,
J.A., Mogelonsky, M.K. and Schaar, K.W., eds. Alambra: A Middle Bronze Age Settlement in Cyprus:
Archaeological Investigations by Cornell University. Gteborg. pp. 447-473.
Whitbread, I. 1995. Greek Transport Amphorae, a Petrological and Archaeological Study. (The British School at
Athens, Fitch Laboratory Occasional Paper 4) Athens.
Whitechurch, H. and Parrot, J.F. 1974. Les cailles mtamorphiques infra-pridotitiques du Bar-Bassit (nord-ouest
de la Syrie). Cahier ORSTOM, Service Gologique 6:173-184.
Whitechurch, H., Juteau, T. and Montigny, R. 1984. Role of Eastern Mediterranean ophiolites (Turkey, Syria, Cyprus)
in the history of the Neo-Tethys. In: Dixon, J.E. and Robertson, A.H.F., eds. The Geological Evolution of the
Eastern Mediterranean. Oxford. pp. 301-317.
Wieder, M. and Gvirtzman, G. 1999. Micromorphological indications on the nature of the Late-Quaternary
paleosols in the southern coastal plain of Israel. Catena 35:219-237.
Wilhelm, G. 1989. The Hurrians. Warminster.
Wilson, J.A. 1969a. Egyptian historical texts. In: Pritchard, J.B., ed. Ancient Near Eastern Texts Relating to the Old
Testament. Princeton. pp. 227-264.
Wilson, J.A. 1969b. An Egyptian letter. In: Pritchard, J.B. ed. Ancient Near Eastern Texts Relating to the Old
Testament. Princeton. pp. 475-479.
Wilson, R.A.M. and Ingham, F.T. 1959. The Geology of the Xeros Troodos Area with an Account of the Mineral
Resources. (Cyprus Geological Survey Department Memoir No. 1) Nicosia.
Wimmer, D.H. 1987a. The excavations at Tell Safut. Studies in the History and Archaeology of Jordan 3:279-282.
Wimmer, D.H. 1987b. Tell Safut excavations, 1982-1985: A preliminary report. Annual of the Department of
Antiquities of Jordan 31:159-174.
Wright, G.E. 1971. A problem of ancient topography: Lachish and Eglon. Biblical Archaeologist 34:76-86.
Wood, B.G. 1990. The Sociology of Pottery in Ancient Palestine: The Ceramic Industry and the Diffusion of
Ceramic Style in the Bronze and Iron Ages. Shefeld.
Worschech, U. and Ninow, F. 1999. Preliminary report on the excavation at al-Balu> and a rst sounding at al-Misna
in 1997. Annual of the Department of Antiquities of Jordan 43:169-173.
Xella, P. 1995. Ugarit et les Phniciens. In: Dietrich, M. and Loretz, O., eds. Ugarit: Ein ostmediterranes
Kulturzentrum im Alten Orient. Mnster. pp. 239-266.
Xenophontos, C. 1991. Petrographic analysis of ceramics. Report of the Department of Antiquities of Cyprus
1991:62-63.
381
Ya>alon, D.H. 1971. Soil forming processes in time and space. In: Ya>alon, D.H., ed. Paleopedology - Origin,
Nature and Dating of Paleosols. Jerusalem. pp. 29-39.
Ya>alon, D.H. and Dan, J. 1974. Accumulation and distribution of loess-derived deposits in the semi-desert and
desert fringe areas of Israel. Zeitschrift fr Geomorphologie Supplementbnde 20:91-105.
Ya>alon, D. H. and Ganor, E., 1973. The inuence of dust during the Quatarnary. Soil Science 116:146-155.
Yamada, M. 1992. Reconsidering the letters from the King in the Ugarit texts: Royal correspondence of
Carchemish? Ugarit-Forschungen 24:431-446.
Yannai, E. 1993. Settlement and burial remains at Kh. Majdal, in the Sharon Plain. >Atiqot 22:1-8. (Hebrew)
Yadin, Y. 1972. Hazor. (The Schweich Lectures of the British Academy) London.
Yassine, K., Ibrahim, M. and Sauer, J. 1988. The east Jordan Valley survey, 1976 (second part). In: Yassine, K.
Archaeology of Jordan: Essays and Reports. Amman. pp. 189-207.
Yassine, K., Kerestes, T., Wood, B.G. and Lundquist, J.M. 1988. An archaeological survey of the three reservoir areas in
northern Jordan, 1978. In: Yassine, K. Archaeology of Jordan: Essays and Reports. Amman. pp. 209-255.
Yellin, J., Broshi, M. and Eshel, H. 1999 Revelations on the pottery from Qumran from Instrumental Neutron
Activation Analysis. (Paper given at the 10th International Conference on Modern Trends in Activation
Analysis, 19-23 April, 1999) Bethesda, Maryland.
Yon, M. 1999. Chypre et Ougarit la n du Bronze Rcent. Report of the Department of Antiquities of Cyprus
1999:113-118.
Youngblood, R. 1962. Amorite inuence in a Canaanite Amarna letter. Bulletin of the American Schools of Oriental
Research 168:24-27.
Younker, R.W. and Merling, D. 1999. Tall Jalul. ACOR Newsletter 11.1:6-7.
Zaccagnini, C. 1987. Aspects of ceremonial exchange in the Near East during the second millennium BC. In:
Rowlands, M., Larsen, M., and Kristiansen, K., eds. Center and Periphery in the Ancient World. Cambridge. pp.
57-65.
Zadok, R. 1982. Remarks on the inscription of HDYS>Y from Tall Fakhariya. Tel Aviv 9:117-129.
Zadok, R. 1986. Bcherbesprechungen: Eph>al, Israel, ed. The History of Eretz Israel, I. Introductions. The Early
Period. Jerusalem. 1982. Zeitschrift des Deutschen Palstina-Vereins 102:179-180.
Zertal, A. 1988. The Israelite Settlement in the Hill Country of Manasseh. Haifa. (Hebrew)
Zertal, A. 1992. The Manasseh Hill Country Survey: The Shechem Syncline. Haifa. (Hebrew)
Zertal, A. 1996. The Manasseh Hill Country Survey: The Eastern Valleys and the Fringes of the Desert. Haifa.
(Hebrew)
Zertal, A. and Mirkam, N. 2000. The Manasseh Hill Country Survey: From Naal >Iron to Naal Shechem. Haifa.
(Hebrew)
Zori, N. 1977. The Land of Issachar: Archaeological Survey. Jerusalem. (Hebrew)
Zori, N. 1962. An archaeological survey of the Beth-shean Valley. In: The Beth-shean Valley. (The 17th
Archaeological Convention) Jerusalem. pp. 135-198. (Hebrew)
Zorn, J.R. 1994. Estimating the population size of ancient settlements: Methods, problems, solutions and a case
study. Bulletin of the American Schools of Oriental Research 295:31-48.
Zwickel, W. 1994. Gat-Karmel. Biblische Notizen 71: 20-23.
382
INDEX OF THE PETROGRAPHIC FABRICS
Group / Fabric Type
Region
Chapter
Brown clay with fresh basaltic and limestone

sand
Dolerite-derived reworked clay
Galilee
12.V
241
Cyprus
3.IV
60
Esna Shales
Egypt
3.I
24
3.III
34
Euphrates sediments
Page
Ghareb marl
Israel
12.VI
246
Habur marly and clayey fabrics
Mitanni
3.IV
44
Hamra soil and calcareous/quartz coastal sand
Galilee coast
12.II
232
Hamra soil and coastal quartz sand
Central coastal plain of Israel
14.V
292
3.II
31
Hattusha fabric
Loess soil with coastal sand
Southern coast of Israel
7.I
112
Lower Cretaceous clay
Lebanon
7.I
103
Lower Cretaceous oolithic marl
Lebanon
8.I
126
Lower Cretaceous shales
Lebanon
9.IV
168
Moni Mlange fabric
Cyprus
3.IV
60
Moza clay and dolomitic sand
Central Hill Country, Israel
13.I
262
Neogene clay with Amphiroa fossils
Syro-Lebanese coast
7.I
108
Neogene coastal clay
Lebanese coast
9.I
143
Neogene coastal marl
Lebanese coast
9.I
134
Neogene marine marl
Lebanon
7.I
105
Neogene marl and basaltic minerals
Syria
10.I
170
Neogene marl with basalt and dolerite
>Akkar plain, Syria-Lebanon
7.I
114
5.I
88
Neogene marl with radiolarian chert

Nile Silt
Egypt
3.I
26
Pakhna marl
Cyprus
3.IV
60
Paleocene marl
Syria
5.2
91
Paleocene marl and basaltic sand
Syria
11.Concl.
223
Paleogene glauconitic marl
Syria
11.III
118
Paleogene marl and Jordan sand
Jordan Valley
12.III
234
Rendzina/loess with chalk and quartz sand
Southern Shephelah, Israel
14.II
280
12.VIII
252
Rendzina soil and Cretaceous tuff and basalt
383
Senonian Paleocene marl and basalt/chalk
Galilee
12.I
228
Taqiye marl
Israel
12.IX
256
Taqiye marl and coastal quartz sand
Lower Shephelah, Israel
14.I
271
Terra rossa soil with limestone
Lebanon
8.II
127
Terra rossa with quartz and chalk sand
Upper Shephelah, Israel
14.II
284
384
EMERY AND CLAIRE YASS PUBLICATIONS IN ARCHAEOLOGY

MONOGRAPH SERIES OF THE SONIA AND MARCO NADLER INSTITUTE OF ARCHAEOLOGY
1.
EXCAVATIONS AND STUDIES: Essays in Honour of Prof. Shmuel Yeivin. Y. Aharoni (ed.). 1973.
(Hebrew)
2.
BEER-SHEBA I: Excavations at Tel Beer-sheba 1969-1971 Seasons. Y. Aharoni (ed.). 1973.
3.
TIRAT-CARMEL: A Mousterian Open-air Site in Israel. A. Ronen and M. Davis. 1974. (Hebrew)
4.
INVESTIGATIONS AT LACHISH: The Sanctuary and the Residency (Lachish V). Y. Aharoni. 1975.
5.
CANAANITE GODS IN METAL: An Archaeological Study of Ancient Syro-Palestinian Figurines.

O. Negbi. 1976.
6.
THE CONQUEST OF LACHISH BY SENNACHERIB. D. Ussishkin. 1983.
7.
BEER-SHEBA II: The Early Iron Age Settlements. Z. Herzog. 1984.
8.
EXCAVATIONS AT TEL MICHAL, ISRAEL. Z. Herzog, G. Rapp Jr. and O. Negbi (eds.). 1989.
9.
PREHISTORIC ANATOLIA: The Neolithic Transformation and the Early Chalcolithic Period.
J. Yakar. 1991.
9a.
PREHISTORIC ANATOLIA: SUPPLEMENT No. 1. J. Yakar. 1994.
10.
SHILOH: The Archaeology of a Biblical Site. I. Finkelstein (ed.). 1993.
11.
ORVAT QITMIT: An Edomite Shrine in the Biblical Negev. I. Beit-Arieh (ed.). 1995.
12.
THE NAAL QANAH CAVE: Earliest Gold in the Southern Levant. A. Gopher (ed.). 1996.
13.
ARCHAEOLOGY OF THE CITY: Urban Planning in Ancient Israel and its Social Implications.
Z. Herzog. 1997.
14.
HIGHLANDS OF MANY CULTURES: The Southern Samaria Survey: The Sites.

I. Finkelstein (ed.). 1997.
15.
TEL >IRA: A Stronghold in the Biblical Negev. I. Beit-Arieh (ed.). 1999.
16.
APOLLONIA-ARSUF: Final Report of the Excavations. Volume I. I. Roll and O. Tal (eds.). 1999.
385
17.
ETHNOARCHAEOLOGY OF ANATOLIA: Rural Socio-economy in the Bronze and Iron Ages.

J. Yakar. 2000.
18.
MEGIDDO III: The 1992-1996 Seasons. I. Finkelstein, B. Halpern and D. Ussishkin (eds.). 2000.
19.
APHEK-ANTIPATRIS I: Excavation of Areas A and B. The 1972-1976 Seasons.

M. Kochavi, P. Beck and E. Yadin (eds.). 2000.
20.
TEL KABRI: The 1986-1993 Excavation Seasons. A. Kempinski. 2002.
21.
ARCHAEOLOGY IN SINAI: The Ophir Expedition. I. Beit-Arieh. 2003.
22.
THE RENEWED ARCHAEOLOGICAL EXCAVATIONS AT LACHISH 1973-1994. D. Ussishkin.

(Forthcoming)
23.
INSCRIBED IN CLAY: Provenance Study of the Amarna Letters and Other Ancient Near Eastern Texts.
Y. Goren, I Finkelstein and N. Naaman. 2004.
386

Inscribed in Clay Provenance Study of TH

Uploaded by

Copyright:

Available Formats

Inscribed in Clay Provenance Study of TH

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Inscribed in Clay Provenance Study of TH

Uploaded by

Copyright:

Available Formats

INSCRIBED IN CLAY

PROVENANCE STUDY OF THE AMARNA TABLETS

TEL AVIV UNIVERSITY

YUVAL GOREN, ISRAEL FINKELSTEIN AND NADAV NAAMAN

EMERY AND CLAIRE YASS PUBLICATIONS IN ARCHAEOLOGY

Under the auspices of the

Published by the Emery and Claire Yass Publications in Archaeology

of the Institute of Archaeology, Tel Aviv University

Printed in Israel by Top

METHOD AND THEORY

LETTERS OF THE GREAT POWERS

Ceramic ecology and petrographic references

Ceramic ecology and petrographic references

Interpretation of the petrographic data

NORTH SYRIAN KINGDOMS

III. NII (?)

SYRIAN KINGDOMS IN THE MIDDLE ORONTES AREA

THE KINGDOM OF AMURRU AND NEIGHBOURING POLITIES

THE LEBANESE BEQA>

THE LEBANESE LITTORAL

EA 102 (BM 29806), from Rib-Hadda to an Egyptian ofcial

EA 147 (BM 29812), from Abi-Milku to the King of Egypt

Chapter 10 SOUTHERN SYRIA AND THE NEIGHBOURING AREAS

Chapter 11 THE BASHAN AREA

Chapter 12 THE GALILEE, THE COASTAL PLAIN OF ACCO AND THE

Chapter 13 THE CENTRAL HILL COUNTRY

Chapter 14 THE SHEPHELAH AND THE SOUTHERN COASTAL PLAIN

EA 282 (BM 29851), from Shuwardatu to the King of Egypt

Chapter 15 UNIDENTIFIED CITIES IN CANAAN

EA 230 (BM 37646), from Yama to the King of Egypt

RESULTS OF THE CHEMICAL ANALYSES

SUPPLEMENT: CATALOGUE OF LATE BRONZE AGE II SITES IN THE EGYPTIAN

INDEX OF THE PETROGRAPHIC FABRICS

The Near East during the Amarna period.

LIST OF COLOUR PLATES OF THIN SECTIONS

Letters of the Great Powers. I. Egypt.

METHOD AND THEORY

TERMINOLOGY AND DESCRIPTION OF THE PETROGRAPHIC DATA

QUANTITATIVE ANALYSIS OF THE ELEMENTAL DATA

SOURCES FOR PROVENANCE INTERPRETATION OF THE TABLETS

COMPARATIVE EXAMINATION OF RESOURCES

LETTERS OF THE GREAT POWERS

Fig. 3.1: The Near East during the Amarna period.

CERAMIC ECOLOGY AND PETROGRAPHIC REFERENCES

EA 339 (VAT 1887), Canaanite letter written in Egypt

CHEMICAL ANALYSES OF SELECTED EGYPTIAN LETTERS

CONCLUSION: THE EGYPTIAN TEXTS

CERAMIC ECOLOGY AND PETROGRAPHIC REFERENCES

CHEMICAL ANALYSES OF THE HITTITE TABLETS

CONCLUSION: THE HITTITE CORRESPONDENCE

CERAMIC ECOLOGY AND PETROGRAPHIC REFERENCES

CHEMICAL ANALYSIS OF SELECTED BABYLONIAN TABLETS

CONCLUSION: THE BABYLONIAN CORRESPONDENCE

Fig. 3.2: Principal component

CERAMIC ECOLOGY AND PETROGRAPHIC REFERENCES

EA 21 (VAT 190), from the King of Mitanni to the King of Egypt

EA 28 (BM 37645), from the King of Mitanni to the King of Egypt

CHEMICAL ANALYSES OF TWO MITANNIAN TABLETS

CONCLUSION: THE MITANNIAN TABLETS AND THE LOCATION OF WAUKANNI