Iran Journal of the British Institute of Persian Studies ISSN: 0578-6967 (Print) 2396-9202 (Online) Journal homepage: http://www.tandfonline.com/loi/rirn20 Cheshmeh Ali Ware: A Petrographic and Geochemical Study of a Transitional Chalcolithic Period Ceramic Industry on the Northern Central Plateau of Iran Edna H. Wong, Cameron A. Petrie & Hassan Fazeli To cite this article: Edna H. Wong, Cameron A. Petrie & Hassan Fazeli (2010) Cheshmeh Ali Ware: A Petrographic and Geochemical Study of a Transitional Chalcolithic Period Ceramic Industry on the Northern Central Plateau of Iran, Iran, 48:1, 11-26, DOI: 10.1080/05786967.2010.11864770 To link to this article: https://doi.org/10.1080/05786967.2010.11864770 Published online: 23 Mar 2017. Submit your article to this journal Article views: 2 View related articles Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=rirn20 CHESHMEH ALI WARE: A PETROGRAPHIC AND GEOCHEMICAL STUDY OF A TRANSITIONAL CHALCOLITHIC PERIOD CERAMIC INDUSTRY ON THE NORTHERN CENTRAL PLATEAU OF IRAN By Edna H. Wong,* Cameron A. Petrie** and Hassan Fazeli*** *University ofSydney, **University ofCambridge, ***University of Tehran Abstract A distinctive red ceramic known as Cheshmeh Ali Ware, which dates to the Transitional Chalcolithic period (c. 5200-4300 BC), has been found at sites on the Central Plateau in northern Iran, stretching from the Gorgan plain in the east to the Qazvin plain in the west. Geochemical and petrographic analyses were performed on samples collected from sites in the Qazvin and Tehran plains to investigate the mode of production and distribution. The results suggest local rather than centralised production ofCheshmeh Ali Ware ceramic vessels. Using discriminant analyses, geochemical groupings have been established that differentiate samples of vessels from the two plains, and also samples from different sites within the Qazvin plain. This has significance for our understanding of the mode and scale of ceramic production and distribution in this important formative period. Keywords Cheshmeh Ali Ware; Transitional Chalcolithic; north Central Plateau; ceramic production; ceramic distribution I. INTRODUCTION The fine black-on-red ceramics known as Cheshmeh Ali Ware or Cheshmeh Ali Fine Ware (Fig. I) were first reported by Eric Schmidt following his excavation at the site of Cheshmeh Ali in 1934-36.1 Although the original excavations at the site remain largely unpublished,2 Cheshmeh Ali Ware continues to be a key cultural and chronological marker for the interpretation of the late prehistoric chronology of northern Iran. In his synthesis of Iranian prehistory, Donald McCown used it as one of the markers for identifying the early cultures of northern Iran, which shared a tradition of painted pottery that can be clearly distinguished from the "buff-ware cultures" in west, south-west and south lran.3 Cheshmeh Ali Ware has since been found in numerous sites in northern Iran stretching from the Qazvin plain in the west to the Gorgan plain in the east,4 and this prompted Voigt and Schmidt 1935; 1936; Matney 1995. New excavations were carried out in 1997; see Fazeli et al. Dyson to suggest that the fine Cheshmeh Ali Ware can be used ''to establish a horizon style within western and northern Iran"S (Fig. 2). Between 1997 and 200 I, one of the authors (Fazeli) conducted new excavations at the sites of Cheshmeh Ali on the Tehran plain and Zagheh on the Qazvin plain. This work was expanded by systematic surveys of these two areas and excavations at the site of Tepe Pardis in the Tehran plain.6 A primary objective of this research has been the clarification of the relative and absolute chronology and to examine the development of social complexity during late prehistory in northern Iran. Fine Cheshmeh Ali Ware remains the hallmark ceramic tradition of the Transitional Chalcolithic period, which has traditionally been dated to c. 52004300 BC, and it is a key element in discussions of technological changes in ceramic production between the Late Neolithic and the Transitional Chalcolithic period in the northern Central Plateau. 7 Cheshmeh Ali Ware is a handmade pottery that is characteristically reddish brown, ranging in surface 2004. 3 4 McCown 1942: 3, 7. E.g. Burton Brown 1979; Voigt and Dyson 1992: 166; Fazeli et al. 2001; 2007; Coningham et al. 2004; 2006. Iran XLVIII2010, 11-26 0 2010 The British Institute of Persian Studies s 6 7 Voigt and Dyson 1992: 166. Coningham eta/. 2004; Fazeli et a/2001. Fazeli et al. 2004; 2005; 2006. 12 EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI Fig. I. CheshmehA/i Warefrom CheshmeA/i. colour from 2.5YR 5/6-5/8, 5YR 4/4, I OYR 5/6 to pink 7.5YR 7/4. It is typically decorated with blackpainted motifs that include simple or composite geometric designs including basketry/brickwork patterns, pendant loops, horizontally hatched diamonds in rows/ columns, hatched triangles in rows, crossed-hatched triangles, linked ovals, stylised birds in rows and columns, animals such as goats with notched horns and flowers and foliage (Fig. I). Forms consist of bowls with curved or flaring walls and concave or trumpet bases. Jars are rare. Thin-walled, fine grit-tempered vessels are evenly fired, while the thicker-wall vessels often show grey cores. II. THE PRODUCTION OF CHESHMEH ALI WARE There has been some debate over the nature and organisation of the production of Cheshmeh Ali Ware, and this has significance for our understanding of the socio-economics of ceramic production and distribution CHESHMEH ALI WARE 13 - ··.: ... ~->·_,. __Ch.atiw;~< Tehran Plain SeeFiguN3 . Te~ P1~; :-~ · - ., :. ~':· ~:·. .....';~ -:·:· ,.>. -~ :. ,. : · . · . . - .. ""· ·~ . ,.• -~f;. .: . ~ :i-~.; . .. .· \; ·: ,: : .. ~;. ~ ·, .. _ 0 ·~ 'c. ~ ·"',.. ~ ' . ,· ."·.. • ·-'o . '· ...... SIIIk . -.. • " • -- - ---- 0 25 50 100 150 200 Kilometers Fig. 2. Topographic map ofthe north Central Plateau showing the location of the Qazvin plain and Cheshmeh Ali. during the Transitional Chalcolithic period. In essence, there has been disagreement over whether Cheshmeh Ali Ware was produced locally at the village level or whether it was produced at a centralised site or sites and then distributed widely. Based on his initial excavations at Zagheh in the 1970s, Malek Shahmirzadi suggested that the type of clay used for making the Cheshmeh Ali Ware vessels at the site did not match the clay found in the vicinity of the settlement, and concluded that ..either the raw material or the finished product was brought to the site from elsewhere".s If this observation is accurate, it has important implications for the organisation of the Transitional Chalcolithic ceramic industry that produced these vessels, and also the development of socio-economic complexity in the Central Plateau, since it would assume centralisation of production and also possible site specialisation.9 According to Arnold, B 9 Malek Shahmirzadi 1977: 281. After Rice 1991. this type of production implies high volume of output with finished products being destined for a supraregional market.IO In order to study the organisation of production of the Transitional Chalcolithic period on the north Central Plateau, the ceramics of the Neolithic and Chalcolithic periods of the Tehran plain were examined using Inductively-Coupled Plasma-Atomic Emission Spectrometry.ll This study indicated that there was localised independent· production of'ceramics 'during the Transitional Chalcolithic period, with each site potentially producing for its own needs. The results of the analysis, conducted on samples from the Tehran plain, was clearly at odds with previous interpretations of ceramic production in the Qazvin plain, and at Zagheh in particular. However, in the new excavations at Zagheh in 200 I, artefacts and IO II Arnold 1991: 94. Fazeli eta/. 2001. 14 EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELl material discovered in trench K, which was located at the southern section of the settlement, suggest that this area might have been an industrial quarter.l2 No ovens, walls, floors or burials were found, but there was a high concentration of pottery sherds, including Cheshmeh Ali Ware, together with fragments of finished, unfinished and deformed figurines, spindle whorls, unfired lumps of clay, traces of pigment, and also the remains of kilns in several contexts.J3 These artefacts and materials were similar to those found by Malek Shahmirzadeh in what he referred to as the ''workshop area".l4 This raised the possibility that Zagheh might have been a regional production site for Cheshmeh Ali Ware. Fazeli and Djamali subsequently conducted a preliminary petrographic study on samples of Transitional Chalcolithic ceramics collected from Zagheh and Kamal Abad, a site that lies 20 km to the north, and concluded that the clay used for the Cheshmeh Ali Ware samples found at each site was likely to have been obtained close to each settlement.' s Furthermore, comparison of the ceramics from Zagheh and Kamal Abad revealed no significant differences in their petrofabrics. This provisional study suggested that although ceramic vessels appear to have been being made locally in different parts of the Qazvin plain, producers were using similar clay fabrics to do so. In order to clarify this issue, the present study was designed to use a combination of thin-section petrography and elemental analysis to establish the provenance of the Cheshmeh Ali ceramics in the Qazvin plain. Integrated methodologies, combining mineralogical and elemental techniques, are becoming more common in investigative studies of pottery production and distribution.16 The two main approaches used in this investigation are thin-section petrographic analysis and bulk chemical compositional analysis using Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES) and Inductively Couple Plasma Mass Spectrometry (ICP-MS) of the samples of Cheshmeh Ali Ware. Fazeli eta/. 2005: I 5. Fazeli eta/. 2005: I 5. 14 Fazeli eta/. 2005: I 5; Malek Shahmirzadeb 1977: 358-76. IS Fazeli and Djamali 2002. 16 Tite 1999. III. GEOLOGICAL CONSIDERATIONS Ceramic vessels are made from specific combinations of raw materials. The mineral and chemical content of raw clay and mineral inclusions, and the petrographic appearance of fired ceramic vessels are intimately related to the geology of the area from which the raw materials were obtained. Therefore, a brief discussion of the geographical and geological context of the north Central Plateau is appropriate. With the exception ofCheshmeh Ali, the sites from which samples were obtained for this study are situated in the Qazvin plain in Qazvin province. The Qazvin plain is bordered by the Alborz in the north and north-east, the Zagros in the west and low mountains from the Zagros range in the south. The southern Qazvin plain is adjacent to the Urumieh-Dokhtar Magmatic Assemblage (UDMA). This tectonic division is composed of various lithologic units including diorites, granodiorites, gabbros and granites as small and large plutonic bodies, as well as widely distributed basaltic lava flows, trachybasalts (locally shoshonitic), andesites, dacites, trachytes, ignimbrites and pyroclastics (mostly tuffs and agglomerates) that cover vast areas.t7 The youngest rocks are lava flows and pyroclastics that erupted from Quaternary craters.ts Zagheh is situated on a gravel plain that is currently covered by cultivated fields. To the north are areas of clay flats. To the south are older terraces and gravel fans, areas of coarse conglomerates, and rare patches of silty marl, siltstone and pockets of coarse grained calcareous sandstone.t9 The sites that lie in the north-eastern Qazvin plain are situated on younger terraces and silt and clay flats. To the north-east there are Eocenic formations composed of dark grey basalt, trachybasalt and andesitic basalt, grey to brown porphyritic dacite, trachyandesite, andesite and fold bearing volcanic lavas, green tuff and tuff breccia.2o Also present is a region of Miocenic formation of dark brown trachyte, dacitic andesite and fold bearing volcanic lavas. To the south of the sites are mud/salt flats.2t The rock types encountered in both the northern and southern parts of the Qazvin plain are therefore mainly igneous. The contrast lies 12 11 13 IS 19 2o 21 Alavi 1994:213. Ghaemi eta/. 2000. Ghaemi eta/. 2000. Jamshidi and Radfar n.d. Jamshidi and Radfar n.d. CHESHMEH ALI WARE in the fact that sedimentary rocks and the full range of basic to acidic igneous rocks are present in the southem area, whereas intermediate and basic igneous rocks predominant in the north-eastern region of the plain. The site of Cheshmeh Ali is located on the Rayy plain, situated at the southern edge of the Tehran alluvium, abutting the eocenic Karaj, Ziarat and Fajan formations of the ..Anti-Elburz". The extensive Karaj formation is made up of andesite/dacite lava and pyroclastics, shale, marl, sandstone and mudstone. The Ziarat formation consists mainly of limestone, while the Fajan formation consists of conglomerate and conglomeratic limestone. Large irregular outcrops of conglomerate are present in the northern and southern Rayy plain.22 In close proximity to the site of Cheshmeh Ali is the promontory of Mount Bibishaharbanu, which is composed ofTriassic dolomite and limestone. There are also small outcrops of Upper Cambrian dolomite/shale, dolomite limestone as well as Lalun sandstone (sandstone with quartzite) south-west of the promontory.23 On the whole the geology of the north Central Plateau is relatively homogeneous, and therefore presents considerable difficulties in using either geochemical or petrographic analyses to differentiate material made in any one area. In view of this, it was decided that thin-section petrographic and elemental compositional analysis would be used in conjunction, and that for the latter, the large number of trace elements and rare earth elements that are produced by ICP-MS analysis would be desirable. IV. SAMPLES AND METHODS The analysis presented in this paper focuses on samples taken from forty-two sherds of Cheshmeh Ali Ware collected from the excavation ofZagheh (twelve samples) and several sites located in different parts of the Qazvin plain, including Zagheh 2 (nine samples), Mahmoodian (five samples), Ebrahim Abad (seven samples), Kamal Abad (five samples) and Zahir Tepe (four samples) (Fig. 3). It also incorporates samples taken from twenty sherds collected from the surface of the site adjacent to an old trench at Cheshmeh Ali24 22 23 24 Geological Survey of Iran 1986. Geological Survey of Iran 1986. Cheshmeh Ali is situated in the outskin of the modem city ofTehran, 144 km south-west ofZagheh 15 by Maurizio Tosi in the 1970s (Table 1). This analysis was undertaken as part of a larger study of the ceramic material collected from archaeological sites in the Qazvin plain with dates ranging from the Late Neolithic to the Bronze age.2s For the thin-section petrographic study, the ceramic sections were cut perpendicular to the preserved wall surface. The cut samples were mounted on glass slides, sliced and then ground to a thickness of 0.03 mm. The slides were polished but not cover-slipped to facilitate future SEM analysis. All the samples from the Qazvin plain were processed in a private laboratory at Tehran. Six of the samples from Cheshmeh Ali were processed in the Geology Department at the University of New South Wales. All of the thin-section slides were examined under a polarising microscope with a rotating stage, using the system proposed by Whitbread,26 which includes the recording and description of Microstructure (in particular the voids and the arrangement of the coarse and fine non-plastic particles), Groundmass (including homogeneity, the nature of the fabric, the composition and size range of the inclusions) and Concentration features for each component. Such an approach allows for the study of the clay mixing, levigation, temper addition and firing, in addition to examination of mineral and lithic inclusions in the assessment of possible sources of raw materials. The chemical analyses were carried out at the Natural Environment Research Council (NERC) supported ICP facility in the Department of Geology, Royal Holloway, University of London. Each sample was cut into approximately 15 g slivers using a diamond saw, which was considered adequate because the inclusions in the fabric were very small. All original surfaces were taken off to avoid modem contamination and influence from slip and paint. Each sample was then put into a beaker of distilled water and cleaned ultrasonically for 10 minutes before being left to dry under a cover for 24 hours. Once dried, the sample was pulverised using an agate mortar and pestle. The beakers and instruments were washed thoroughly and rinsed with distilled water between each sample. The powderised samples were then put into labelled sterilised plastic tubes for analysis in the laboratory. For the ICP-AES analysis of AI, Fe, Mg, Ca, Na, K, Ti, P, Mn, Ba, Co, Cr, Cu, Li, Ni, Pb, Sc, 2s 26 Wong 2008. Whitbread 1989: 128-30. EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI 16 -:. : ~ •,'.' ;~: ? ' -~: ZahirTape ~ '-.~ ~.,/'v . ....__J) I \ I • ._- ) KamaiAbad J,/ ~,_ • Ebrahim Abad··- ' -•- --- Mahmoodian I / . ~ \ --------:-- (, ,- ---- __ .;.. . J~-;z:_./ Zaghe~- ~ ;:. ~ 1 >•. .:_·-~ ~\ ·L~- ' v r_;!, .,..., // :/~.- ,_ -- --- Zagheh .- • •Boein Zahra N ' -y,~: I ;'" .-.- f...• .~" - .J -·:~; -.. _,--.....__ ---, -· - ;~ J!'_ -~ I -~ ~ I~'-/ r ) ' . I---'·· -- - --- 0 5 10- 20 - 30 40 Kilometers Fig. 3. Topographic map ofthe Qazvin plain showing the location of Transitional Chalcolithic sites samples for this analysis. Sr, V and Zn, the following protocol was followed: I) 0.2 g ± 0.0010 g of powdered sample was dispensed into a 25 ml PTFE crucible; 2) Using an HF (hydrofluoric acid) dispenser 6 ml of a 1:2 mixture ofHCI04 (perchloric) and HF acids was added to each crucible; 3) The tray of crucibles was placed on the hotplate at setting 4 in a fume cupboard for evaporation, and once dried (usually takes 3-4 hours) it was removed from the hotplate and allowed to cool; 4) 2 ml ofHCI (hydrochloric) acid was added to each crucible using a hand-held repetitive pipette; 5) Each crucible was topped up with distilled water to about 3/4 full and warmed on the hotplate for 20-30 minutes, and the crucibles were removed from the hotplate and allowed to cool; 6) Once the crucibles were cool, each solution was transferred to a labelled tube and made up to 20.40 g ± 0.1 g on a top pan balance using distilled water (this equated to 20 ml, taking the density of the HCI into account), and the tube was capped and shaken by hand; 7) The solutions were then analysed using a Perkin Elmer Optima 3300RL ICP-AES system, and five standardisation solutions-Traces KCIO, Traces KCll, Traces KCI2, Traces KC14 and Traces RH21 were used as controls; and 8) The run commenced with measurement of the drift monitor solution, which was re-measured after CHESHMEH ALI WARE every I 0 samples and at the end of the run to correct for instrumental drift during the analysis. For the ICP-MS analysis ofZr, Cs, Nb, Rb, Ta, Th, Tl, U, Y and the rare earth elements, La, Ce, Pr, Nd, Sm, Eu, Gd, Dy, Ho, Er, Yb, Lu, the following protocol was followed: 1) 0.2 of powdered sample was dissolved in 6 ml. of HF and HC104 (2: I mixture), which was then evaporated to dryness, cooled and dissolved in 20 ml of 10% HN03; 2) a further 0.2 g of powdered sample was weighed into a graphite crucible and 1.0 g of Lithium metaborate (LiB02) added. The powders were carefully mixed and fused at 900°C for 20 minutes. The resulting mixture was dissolved in 200 ml of cold 5% nitric acid. Ga was added to the flux to act as an internal standard to improve instrumental precision.27 The instrument used for the analysis was a Perkin Elmer Elan 5000. The working detection limits for trace elements were below 5 ppm for the ICP-AES and 1 ppm for the ICP-MS, with a typical precision of 1%-1.5% RSD (relative standard deviation) when the concentration is 100 times greater than the detection limits. For the statistical analysis carried out on the elemental results, elements and compounds with an analytical error less than ±I 0% in three or more of the reference standards have been used. These included aluminium oxide (AI 20 3). iron oxide (Fe20 3), magnesium oxide (MgO), calcium oxide (CaO), sodium oxide (Na20), potassium oxide (K20), titanium oxide (Ti02), phosphorus oxide (P20 5). manganese oxide (MoO), cobalt (Co), nickel (Ni), scandium (Sc), strontium (Sr), vanadium (V), yttrium (Y), zinc (Zn) from the ICP-AES analysis, and cesium (Cs), hafnium (Hf), niobium (Nb), rubidium (Rb), zirconium (Zr) and the rare earth elements, uranium (U), thorium (Th), lanthanum (La), cerium (Ce), praseodymium (Pr), samarium (Sm), lutetium (Lu) from the ICP-MS analysis (Table I). Due to financial constraints, no repeat analyses were made on any of the samples. The statistical analyses were carried out using Statisti-XL v.l.7 and principal components analysis (PCA) and discriminant analysis (DA) using the find spot as classificatory information were used. A correlation matrix was used in the PCA and covariance matrix was used in the grouping function of DA, in 27 This step is used in addition to the dissolution by HF/ HCL04 by the Geochemical Laboratory in the Royal Holloway for a more accurate determination of Zr. 17 which elements with tolerance values less than 0.00 I were eliminated from the analyses. The groups used in all the discriminant analyses are displayed and coded for ease of reference at the right of the relevant figures. Two sets of PCA were performed, one using raw data and a second with loglO standardisation of the raw data. There was no difference in distribution of the samples from specific sites, and the sets and relative dimensions of significant elemental determinants are less than 2% in the cumulative variances in each analysis. Standardisation of the raw data was deemed unnecessary for the DA since "this [was] taken care of by the mathematics of the method".28 Therefore, the results presented here are those based on analyses with the raw data. V. RESULTS V.I. Petrographic examination Macroscopically, it is easy to distinguish the fine Cheshmeh Ali Ware with its distinctive red colour and relatively well levigated paste from the contemporary buff wares that are prevalent at contemporaneous sites in west and south-west Iran. With the exception of four samples, there are minimal differences in the ceramic fabrics collected from the various sites, including those from Cheshmeh Ali in the Tehran plain. The typical fabric has a low percentage of voids, good homogeneity and coarse:fine:void inclusion ratio of approximately 5:90:5.29 Samples from Zagheh, Mahmoodian, Ebrahim Abad and Cheshmeh Ali tend to have a slightly lower percentage of voids than those from Zagheh 2 and Zahir Tepe. It is interesting to note that the samples from Zagheh also have a more consistent coarse:fine:void ratio and generally lower density of coarse fraction than samples collected from other sites. This may be due to the fact that the samples from Zagheh were largely taken from a trench thought to be located in a production area while samples from other sites were randomly collected from the surface. The typical coarse fraction consists predominantly of quartz, plagioclase and rounded to sub-angular igneous rock fragments, with the very occasional fragment of chert and limestone. The fine fraction consists 28 29 Baxter 200 I: 690. Coarse inclusion is defined as 0.2 mm-1 mm and fine inclusion as 0.2 mm or less in maximal dimension. TABLE I. Caption. Slle Sample EA4 ] < B :a I! .D Ill 'a .a< 'i B ;a a .!1 00 v v 138 23 u Tb Rb 2.34 11.20 101 CaO 1.67 NazO 1.69 3.90 1.57 I.SO 3.55 0.82 0.17 0.069 33 83 18 232 146 27 107 194 2.31 10.73 3.82 2.77 1.58 3.29 0.77 0.17 0.124 31 77 16 324 13S 25 97 174 2.13 9.89 AlzOl 16.93 Fez03 MaO 6.91 3.41 EAS 16.91 7.65 EA6 15.98 7.70 MaO 0.06 KzO TIOz PzOs 3.34 0.86 0.13 Co 34 Nl 69 Sr 18 268 Sc Zn 100 Zr 180 5.36 La 27.S Ce 61.8 Pr 6.4 Sm Lu S.76 0.38 16.1 7.95 5.23 32.7 70.7 7.2 6.22 0.44 14.9 7.08 5.00 30.4 71.3 7.2 6.07 0.38 6.52 0.40 Nb 15.9 Cs 7.37 110 102 Hf 15.74 7.33 3.39 1.51 1.48 3.03 0.8 0.13 0.07 32 100 17 245 126 24 99 196 2.56 11.43 114 16.3 5.05 33.3 72.1 7.7 EAB 16.09 7.2 4.03 4.94 1.53 3.33 0.8 0.19 0.09 30 77 17 348 132 22 178 2.53 10.41 106 15.4 7.S8 4.89 32.2 67.0 7.3 6.08 0.38 17.1 8.20 5.12 33.0 7S.2 7.6 6.21 0.42 ::z: 71.3 7.6 6.55 0.37 ~ 0 EA9 16.58 7.48 4.01 1.39 1.39 3.58 0.83 0.06 0.16 38 83 18 299 140 24 102 109 2.33 190 11.39 116 EAIO 15.19 6.86 3.02 8.3 1.06 3.36 0.81 0.35 0.12 34 73 IS 411 119 22 88 173 2.S2 11.80 110 17.3 7.11 4.56 34.4 KA4 15.65 6.47 3.2S 9.26 1.99 3.24 0.74 0.23 0.143 30 so 16 766 Ill 21 76 147 2.28 8.22 80 13.2 4.69 4.1S 25.1 5S.O 5.3 4.74 0.30 KAS 15.9 7.28 3.9 1.8S 1.66 3.61 0.83 O.IS 0.08 3S 80 17 296 135 20 98 185 2.25 II.OS 118 16.7 7.79 4.75 30.8 1l.S 7.1 5.95 0.37 KA6 13.36 5.51 2.9S 13.11 1.8 3 0.63 0.23 0.11 19 45 13 773 106 19 15 140 2.18 7.82 76 11.6 4.0S 3.72 23.6 S0.6 5.6 4.74 KA7 15.79 7.26 3.59 2.11 1.51 3.32 0.74 0.19 O.IOS 26 73 16 259 130 24 199 2.08 9.79 96 IS.S 6.49 s.os 27.6 62.3 6.8 S.36 0.37 3.34 22.4 47.9 S.4 4.S3 0.25 66.7 6.7 S.11 0.33 KAB 13.78 5.36 2.81 11.74 1.78 2.97 0.6 0.1 0.33 18 45 13 746 lOS 100 19 70 2.04 118 7.59 72 11.4 4.46 0.31 MDI 13.88 6.58 3.23 7.83 1.56 3.19 0.81 0.2 0.15 36 63 14 488 114 21 81 168 1.99 9.52 87 16.S S.22 4.18 28.7 MDZ 16.5 7.15 3.95 2.51 1.64 3.45 0.82 0.28 0.06 36 83 17 314 146 25 Ill 169 2.S9 109 16.4 6.98 s.os 30.9 69.S 7.1 6.03 0.38 11.07 1B MDJ 16.S2 6.96 3.28 2.12 1.6S 3.3 0.8 O.IS 0.06 21 62 17 330 161 21 94 170 2.2S 10.50 101 17.4 6.82 4.S9 27.S 63.3 5.34 MD4 16.91 7.84 2.24 1.42 1.2 3.25 1.02 0.21 0.08 49 77 18 231 140 23 109 232 2.18 13.63 116 26.3 6.46 6.13 39.9 82.4 8.2 8.03 0.39 ::E MDS 16.47 6.84 4.04 7.11 1.61 3.62 0.67 0.24 0.2 25 S4 17 630 108 25 103 ISS 1.95 9.70 96 13.0 6.85 4.04 27.2 62.2 6.0 5.62 0.35 63.1 6.8 S.99 0.3S ~ rn c z > 7.06 0.36 .... . I EA7 6.4 .• j ZAII 15.40 8.95 3.37 1.83 1.44 3.40 0.76 0.41 0.076 48 87 IS 352 159 25 98 185 2.04 9.94 95 14.5 6.SO 4.59 29.8 ZAIZ 17.42 8.16 2.73 1.85 1.01 3.16 1.03 0.18 0.13 46 88 18 303 152 21 118 213 2.09 12.20 110 23.8 6.87 5.11 35.0 79.7 8.2 7.41 0.42 ZAIJ 16.06 7.54 3.9 2 1.66 3.4 0.79 0.22 0.38 28 87 17 313 132 23 112 180 2.27 10.25 99 IS.O 6.31 4.78 26.4 59.7 6.6 6.05 0.39 ZAI4 IS.77 7.45 4.10 2.89 1.41 3.6S 0.72 0.21 0.116 32 75 IS 337 127 23 97 IS2 2.00 9.37 98 14.7 6.84 4.04 26.4 S6.S 6.2 S.49 0.34 ZAIS 16.42 7.16 2.92 2.29 1.39 3.2S 0.82 0.09 0.04S 41 53 17 320 137 20 88 205 3.S2 10.70 106 15.5 7.64 4.88 27.1 SS.3 6.0 4.74 0.36 ZAI6 IS.SI 6.34 3.11 2.9S 1.94 3.33 0.64 0.17 0.076 17 63 14 391 125 22 106 164 2.01 9.08 86 13.6 5.84 4.32 24.2 47.4 S.8 4.95 0.38 ZAI7 15.84 6.5 3.1 1.63 1.78 3.2 0.9 0.1 0.06 34 66 17 31S 138 22 92 204 2.27 11.35 109 17.8 7.36 5.18 31.4 74.8 6.9 5.98 0.44 ZAII IS.39 6.72 3.93 S.SI 1.48 3.47 0.81 0.2 0.09 26 73 16 426 131 22 94 179 2.36 10.71 Ill 16.6 7.28 4.86 36.3 76.9 7.3 S.9S 0.40 ZAZO 15.37 6.18 3.64 3.73 1.69 3.27 0.82 0.2 0.07 22 64 16 390 133 19 93 201 2.38 10.81 102 16.0 7.14 5.16 30.8 73.4 6.9 6.03 0.36 L_ -- - - - z 0 (') > 3:: rn ~ 0 z > 'tl rn o-j ~ -rn > z c ::z: > Cll Cll > z 'Tl > N m r MaO Co Nl Se Sr v y Zn Zr u Th Rh Nb Cs ur La Ce Pr Sm Lu 0.132 30 40 14 450 102 23 117 139 2.07 8.47 97 13.3 7.29 3.93 26.1 54.1 5.6 4.93 0.34 0.19 0.06 29 61 14 306 liS 20 102 168 2.21 8.90 79 12.9 4.87 4.03 24.8 60.3 5.6 5.00 0.30 0.24 0.18 Ill 46 16 487 112 23 99 143 1.95 8.05 91 11.6 6.42 3.47 21.5 48.7 4.8 4.89 0.34 0.17 56 so 15 398 87 30 137 196 2.14 9.76 91 12.9 6.39 5.17 29.1 68.1 6.6 5.59 0.50 0.22 0.068 52 58 14 278 109 23 97 156 2.06 9.39 87 15.1 5.64 4.50 28.0 60.5 6.3 5.67 0.37 0.28 0.16 34 41 13 417 76 28 172 182 2.02 8.79 79 13.2 5.14 4.96 26.5 60.7 6.4 5.68 0.44 0.15 0.08 82 77 18 24S 139 22 107 202 2.28 11.40 113 17.8 7.71 5.09 33.9 69.3 7.1 6.06 0.42 0.54 0.20 0.141 23 42 12 360 77 27 149 169 2.10 9.18 94 14.7 6.56 4.73 27.9 62.4 6.3 6.53 0.43 3.52 0.85 0.18 0.15 54 89 18 244 142 23 109 187 2.26 11.08 118 16.7 7.94 4.70 31.9 69.3 7.4 6.53 0.40 1.59 3.41 0.65 0.27 0.122 24 41 IS 483 106 21 91 140 1.96 8.46 94 12.7 7.42 4.04 25.2 48.5 5.4 4.51 0.3S 6.00 1.72 3.19 0.75 0.22 0.073 64 69 IS 325 lOS 23 97 183 2.04 10.05 94 15.7 6.78 5.04 29.0 61.3 6.8 6.15 0.41 3.36 1.47 1.5 3.23 0.82 0.12 0.06 32 93 17 216 126 23 103 191 2.24 11.22 liS 18.8 7.79 5.07 32.5 67.4 7.1 6.19 0.39 7.75 4.01 2.37 1.52 3.SS 0.86 0.16 0.07 138 83 18 240 138 26 Ill 186 3.11 11.35 119 17.6 7.69 4.56 31.7 67.0 7.2 6.14 0.40 AlzOJ 15.91 Fe2o, MgO CaO Na20 K 20 TI02 ZHI6 6.20 3.01 10.44 1.62 3.25 0.60 ZHII 14.12 5.87 3.7 3.88 1.73 2.93 0.7 ZHI9 16.4 6.42 2.78 10.25 1.66 3.48 0.68 Zll20 16.16 5.98 3.15 7.69 1.89 3.73 0.61 0.21 ZHZI 15.07 6.00 3.93 4.43 1.75 2.93 0.66 ZH22 14.79 5.23 3.21 6.06 2.09 3.61 0.54 ZH23 16.86 6.84 3.64 1.67 1.88 3.51 0.86 ZH24 15.34 5.46 3.28 8.27 1.79 3.37 ZH25 17.15 7.61 4.28 1.79 1.6 .... ZH26 16.32 6.36 3.04 9.10 ZH27 15.67 6.60 4.05 ~ ZH28 15.95 7.72 ZR2 16.93 Site . .... S.mple PzOs 0.42 ! ZR3 IS.SS 7.71 3.46 I.SS 1.85 3.08 0.8 0.16 0.08 29 75 IS 262 125 19 98 189 2.11 10.80 104 17.0 7.30 4.94 29.2 61.0 6.2 5.33 0.36 ZR4 17.78 8.74 2.66 2.58 0.9 2.98 0.96 0.32 0.22 104 86 19 326 160 22 117 180 2.41 12.4S 121 22.4 7.84 5.22 36.4 76.7 8.0 6.61 0.44 ~ ZR6 16 6.91 2.61 1.29 1.72 2.97 0.86 0.09 0.03 65 58 17 237 146 21 78 187 2.45 11.27 99 lS.S 6.87 4.89 27.7 58.4 6.4 5.52 0.41 CAl 14.60 5.36 2.31 2.48 2.16 2.94 0.67 0.13 0.059 14 48 14 299 178 2S 89 185 2.44 10.03 101 15.4 S.69 5.02 27.7 62.3 6.5 5.72 0.39 ...J: ~ 1i ....e ....(.III CA2 13.81 6.03 3.39 9.70 1.88 2.91 0.61 0.29 0.248 IS 63 14 459 125 22 96 149 2.00 8.70 88 13.7 S.60 3.85 26.9 55.4 5.8 4.86 0.39 CAl IS.41 6.50 3.28 3.91 1.96 3.00 0.65 0.47 0.110 16 63 IS 352 169 27 108 177 2.49 10.22 lOS 14.4 6.80 4.41 30.0 61.8 6.7 5.39 0.41 CA4 15.83 6.78 3.53 4.43 1.66 3.48 0.73 0.16 0.088 IS 66 16 292 123 24 95 169 2.37 11.56 Ill 14.6 7.84 4.65 27.9 68.6 6.6 5.36 0.39 CAS 15.39 6.13 3.27 2.00 2.06 3.25 0.69 0.25 0.136 17 67 IS 315 182 26 113 164 2.66 10.66 104 13.7 6.21 4.62 26.5 61.1 6.3 4.92 0.36 CA6 16.48 6.34 3.32 3.06 1.49 3.12 0.67 0.22 0.074 16 64 16 287 ISS 24 92 159 2.38 IO.S3 107 13.1 7.90 4.49 25.8 56.9 6.3 5.19 0.42 CA7 16.16 6.69 3.12 2.22 2.03 3.01 0.70 0.25 0.081 16 61 IS 280 148 24 93 182 2.35 10.44 97 14.1 6.26 4.72 26.0 57.8 6.6 5.33 0.36 CAS 15.99 6.87 3.16 2.04 2.16 3.12 0.70 0.26 0.066 16 57 16 307 164 26 98 180 2.34 10.95 108 14.4 7.77 4.35 29.8 64.9 6.5 5.59 0.39 CA9 15.50 6.39 2.69 3.15 2.05 2.98 0.71 0.14 0.072 16 51 14 298 191 24 73 196 2.10 9.49 95 16.1 5.98 4.81 28.6 64.0 6.3 5.22 0.39 CAIO lS.S6 6.30 3.69 3.87 1.84 3.19 0.63 0.24 0.138 16 70 16 308 liS 24 94 147 2.17 9.50 100 12.7 7.98 4.06 27.9 62.1 5.9 5.12 0.38 CAll 15.96 6.34 3.23 3.62 1.71 3.28 0.69 0.17 0.084 16 69 16 270 139 26 92 ISS 2.60 9.99 104 14.1 7.45 4.32 28.9 62.1 6.1 s.oo 0.42 CAIZ 16.58 6.64 3.48 1.7S 1.33 3.34 0.70 0.17 0.085 18 71 17 240 138 25 Ill ISS 2.33 10.68 116 13.8 8.47 4.32 27.5 62.4 6.2 5.12 0.35 CAll 15.66 6.39 3.88 3.78 1.70 3.36 0.66 1.09 0.134 16 65 16 513 135 24 125 139 2.33 9.81 109 13.3 7.45 4.21 24.4 51.5 S.8 4.87 0.35 CAI4 15.92 6.22 3.52 2.70 1.92 3.27 0.68 0.69 0.115 16 65 16 392 144 25 116 lSI 2.54 10.16 104 13.9 7.12 4.16 28.0 61.9 6.4 5.26 0.39 CAIS 15.77 6.58 3.63 3.55 2.2S 3.03 0.63 0.25 0.112 16 64 IS 367 139 24 107 165 2.30 10.11 103 14.1 7.62 4.29 27.S 58.2 6.3 5.13 0.40 CAI6 16.20 6.79 3.48 2.05 1.44 3.22 0.65 0.25 0.263 16 69 16 330 137 23 Ill ISS 2.61 10.57 112 13.8 8.86 4.45 27.2 58.4 6.3 5.23 0.41 CAI7 15.68 S.S8 3.01 2.03 2.01 3.22 0.70 0.21 O.OS3 14 46 14 381 125 20 87 179 2.16 9.64 99 13.7 6.79 4.75 23.8 55.4 5.7 4.98 0.31 CAll 15.91 6.67 3.29 1.77 1.51 3.12 0.66 0.17 0.101 16 66 IS 253 120 24 102 166 2.07 10.14 108 14.2 7.60 4.69 27.7 57.4 6.3 S.44 0.36 CAI9 14.42 S.99 3.37 7.13 1.84 3.22 0.61 0.23 0.099 14 58 14 385 131 22 89 153 2.26 8.90 86 11.8 6.01 3.92 25.2 54.4 5.8 4.72 0.35 CAZO 15.28 6.39 3.86 6.81 1.86 3.30 0.63 0.62 0.116 16 66 IS 379 114 25 112 ISO 2.28 9.29 95 13.2 7.42 4.13 28.2 56.9 5.9 5.47 0.37 (') :t tTl en :t 3:: tTl :t > r ~ > ::0 tTl 10 20 EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI 7 6 6 6 5 6 4 3 ~ N .. 2 :c. 6 6 6 0 6 0 -2 6 0 a a • .. i • o. a 66 • 6g •• .. • 0 -3 .. a • -1 a • • 0 -4 -10 0 -2 2 4 6 8 PC 1 (48.4'1.) Fig. 4. PCA ofCheshmeh Ali Ware samples from sites on the Qazvin plain. The site abbreviations are as follows; EA = Ebrahim Abad (n = 7), KA =Kamal A bad (n = 5), MD = Mahmoodian (n = 5), ZA = Zagheh 2 (n = 9), ZH = Zagheh (n = 12), ZR = Zahir Tepe (n = 4). predominantly of quartz and plagioclase with the rare occurrence of clinopyroxene, hornblende and haematite. A minor amount of secondary calcite is present in some of the samples, indicative of post-depositional changes. The four samples whose fabrics are significantly different from the main group (MD4, KA6, KA8 and EA I 0), have a higher coarse fraction, ranging from I 0-20%, and larger and more angular coarse fragments. MD4 from Mahmoodian also has a predominance of granitic elements in the igneous rock fragments when compared to the other samples from the same site and a higher degree of crushing and fissuring in the quartz mineral fragments. This is unexpected as basic and intermediate igneous rocks are more common in this region. More samples from an excavated context at Mahmoodian would undoubtedly help to ascertain the relative frequency of this fabric group. There is the possibility of it being an import from the southern plain. KA6 and KA8 from Kamal Abad have the highest proportion of coarse fragments (15-20%) and highly angular mineral and rock fragments. EA I 0 from Ebrahim Abad is distinguished by a predominance of angular to sub-angular larger fragments of intermediate igneous rocks and sub-round limestone. This is consistent with the geology of the northern part of the Qazvin region, and the higher proportion of coarse fraction is likely to be related to choices made during the production process. By examining a larger sample from these sites it would be possible to further evaluate the significance of these differences. V.2. Geochemical analysis The result of the first two principal components of a PCA carried out on all the samples from the Qazvin plain is presented in Figure 4. The cumulative variance for the first two PCs is 60.5%, which is slightly below average for ceramic analysis.JO However, the plot shows no definite site clustering and there is considerable overlapping, as might be expected from JO PC I is determined by Th (0.956), Rb (0.914), Pr (0.895), La (0.880), Ti02 (0.860), Nb (0.851), Ce (0.841), Hf (0.824), Zr (0.829), Sc (0.822), Ni (0.810), Sm (0.810), Fe20 3 (0.771), V (0.761), Cs (0.742), Al 20 3 (0.723), CaO (-{).804), Sr (-{).766), Na20 (-{).606). PC 2 is determined by Y (0.859), Zn (0.859), K 20 (0.655), Lu (0.672), V (-{).488), Ti02 (-{).416). CHESHMEH ALI WARE 21 6 • 5 • 4 3 ;q • • • • • 2 • 9.1 N 0: ... ~ ._ - 0 -1 D 6 6 6 6 • -2 D 6 6 6 D 6 6 6 D • • D D 6 0 0 0 D • • 6 -3 0 • 0 -4 -10 -5 0 5 10 Fn 1 144.2'11.1 Fig. 5. Discriminant analysis ofCheshmeh Ali Ware samples from sites in the Qazvin plain. For site abbreviations, see caption for Fig. 4. samples from a small area with similar surface geology. The samples from Kamal Abad and Mahmoodian are widely dispersed along the x axis, as a result of the petrographic "outliers" from these sites. Discriminant analysis performed on the same data revealed discrete groupings, with some overlapping of samples from Zagheh and Ebrahim Abad (Fig. 5). The proximity of the samples from Mahmoodian and Kamal Abad is likely to be a function of the high ratio of coarse fraction in the three "outliers" from these two sites. Zahir Tepe is the furthest from the drainage system at the north-eastern plain and might be expected to form a distinct group. The cumulative variance is 66.2%. Most of the discriminatory variables are trace elements in Fn I and the primary major element discriminator in Fn 2 is CaO.Jl In order to put the Cheshmeh Ali Ware samples from sites in the Qazvin plain into a broader interregional perspective, elemental values from the samples collected from the type site ofCheshmehAii 31 Fn I is dominated by Rb (3.291), Nb (2.251), K20 (1.902), Th (-3.791), La (-3.631), Cs (-2.466), Zn (-2.356). Fn 2 is detennined by Sr (2.894), Ti02 (2.797), Cs (2.448), Ni (2.247), La(2.091), Lu (2.050), CaO (-5.158), Sc (-4.029), Pr (-3.911), Th (-3.110), Fe20 3 (-2.789). in the Tehran plain were then added to both PCA and DA statistical analyses. The resultant PCA shows no obvious segregation of samples by collection sites, as may be expected from the similar geology across the northern Central Plateau (Fig. 6). The cumulative variance of the first two PCs (54.4%) is relatively low,32 A second PC analysis using PC 2 and PC 3, however, shows better segregation of the Cheshmeh Ali samples from those collected in the Qazvin plain (Fig. 7). The cumulative variance of these two PCs is 18.6%,33 When the same data is analysed using discriminant analysis, it can be seen that the samples from Cheshmeh Ali are effectively separated from those collected from sites in the Qazvin plain by Fn I (Fig. 8). Samples from Zagheh and Zahir Tepe are in close proximity, while those from Ebrahim Abad, Kamal Abad, Mahmoodia and Zagheh 2 overlap. 32 PC 1 is dominated byTh (0.935), Pr(0.901), La(0.87l),Ce (0.850), Rb (0.844), Ti02 (0.837), Nb (0.842), Hf (0.808), Sm (0.813), Sc (0.808), Zr (0.781), Fe20 3 (0.768), Ah0 3 (0.723), CaO (-0.721), Sr (-0.701). Majority of these are rare earth and trace elements. PC 2 is detennined by Zn (0.780), Y (0.713), K20 (0.641), MgO (0.496). 33 PC 3 is detennined by V (0.565), U (0.477), Cs (0.432), Sr (0.423), CaO (-0.520), and Sm (-0.437). EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI 22 5 6 6 4 6 3 • .. • 2 iii -.; 1 6 ~ . 6 N u -1 • 6 -3 " D " oKA a MD 6 • •ZA 0 D ..• • 6ZH • 6 D • •ZR • • • •• 0 aEA D D 0 • D 6 • •CA D 6 .0. • 0 -2 • ·-6 • • 0 • • • • • ... •• 6 " -4 -10 -4 -8 0 -2 6 4 2 10 8 PC1 (4UI'JI.) Fig. 6. PCA ofCheshmeh Ali Ware samples from sites in the Qazvin plain and also Cheshmeh Ali. Site abbreviations are as follows; CA = Cheshmeh Ali (n = 20). EA =Ebrahim A bad (n = 7), KA =Kamal A bad (n = 5). MD = Mahmoodian (n = 5), ZA = Zagheh 2 (n = 9), ZH = Zagheh (n = 12). ZR = Zahir Tepe (n =4). 3 • •• • 2 • • • • • • • .. ~ " • " 0 6 u • 0 6 .. -1 D • 6 6a .. • • 6 a • • 6 . • 6 0 -2 •• • • • D " -:.. • " 0 6 6 D 0 -3 " -3 • 6 • " -2 -1 0 1 2 3 4 PC 2 (10.3"4) Fig. 7. PCA ofCheshmeh Ali Ware samples from sites in the Qazvin plain and also Cheshmeh Ali. For site abbreviations, see caption for Fig. 6. 5 23 CHESHMEH ALI WARE 6 II. 4 • II. II. II. II. II. 2 ..... • • • • • • -2 • • • • • • • II. ~ II. • II. -• .. II. .. 0 •a • • a a a •CA aEA oKA II. •• • a 0 A AMD •lA • a .a.ZH Aa •ZR 0 .. 0 0 -6 -6 0 -2 2 4 6 Fn 1(42.0%) Fig. 8. Discriminant analysis ofCheshmeh Ali Ware in the Qazvin plain and Cheshmeh Ali. For site abbreviations, see caption for Fig. 6. The cumulative variance of the first two functions is 70.7%,34 The detennining elements that are common in the second PC analysis and the discriminant analysis are Sr, Cs, Y and Zn, in particular Cs, which is a main negative detenninant in Fn I in the discriminant analysis and positive detenninant in PC 3 in the PC analysis. VI. DISCUSSION AND CONCLUSION There is a range of archaeological evidence indicating that the Transitional Chalcolithic on the north Central Plateau was a period of significant socio-economic development, particularly in tenns of the organisation of production. Zooarchaeological evidence suggests that there was an intensification of agriculture, particularly in the exploitation of domesticated cattle, pig, sheep and goat,JS and there are also indications of increased complexity in tenns of social ranking, long distance trade and specialisation in craft pro34 3S Fn I is detennined by Fe20 3 (1.361), Ti02 (0.946), Th (-1.127), Cs (-1.092), Y (-1.044). Fn 2 is detennined by Co (1.098), La (1.406), Zn (0.934), Ti02 (0.931), Rb (0.909), Sc(-1.538), Nb(-1.329), Pr(-1.246), Sr(-0.918). Young and Fazeli 2008. duction.J6 A number of major technological changes in ceramic production appear to have occurred with the commencement of the Transitional Chalcolithic, including a shift toward the use of standardised raw materials, and there is evidence from several sites for the use of the slow wheel,37 and the use of ceramic kilns in workshop areas.JS As noted above, the 2001 excavations at Zagheh, revealed evidence for ceramic production (trench K), including kiln remains, which paralleled material recovered at the same site by Malek Shahmirzadeh in the "workshop area". More definitive evidence for Transitional Chalcolithic ceramic production was subsequently revealed at Tepe Pardis, where a pivoted wheel and the remains of at least six large kilns with dimensions of up to 12m2 each, indicating that they could have been used to fire very large vessels (Fazeli eta/. 2007). The study ofCheshmeh Ali Ware from sites located in different parts of north Central Iran that is presented here makes an important contribution to our understanding of the operation of the Transitional Chalco36 37 38 Fazeli 2001; Fazeli eta/. 2005. Chesmeh-Aii, Dipilato and Laneri 1998; Tepe Pardis, Fazeli eta/. 2007, Fazeli eta/. in press. Fazeli eta/. 2005; Fazeli eta/. 2007. 24 EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI lithic ceramic production and distribution systems. The petrographic analysis confirms that similar approaches to fabric preparation were being used at the various sites from which samples were analysed, including samples from Cheshmeh Ali and various sites in the Qazvin plain. Apart from four samples, there was no significant difference in the fabric of the ceramics collected from the different sites. The clay was likely to have been Ievigated to achieve a fairly consistent ratio of coarse and fine inclusions, but it is also possible that naturally fine clay was deliberately selected by the potters. In either case, it appears that specific types of raw materials were chosen. When this observation is taken together with the similarity in forms, surface treatment and painted motifs that are evident at sites spread across the north Central Plateau, it can be concluded that there was a high degree of standardisation in the production of Cheshmeh Ali Ware in this region.J9 The discovery of workshop areas at Zagheh and the more consistent coarse:fine:void ratio in the petrography of the samples from this settlement supports this hypothesis, since it indicates the emergence of specialist pottery production. The presence of "outlier" samples from Kamal A bad, Mahmoodian and Ebrahim A bad does, however, suggest that not all Cheshmeh Ali Ware vessels recovered from some sites were manufactured using raw materials from the same or similar sources. The significance of this finding is difficult to assess with small numbers of samples, but it is possible that potters at some settlements made use of more than one distinct source of raw materials. There is also the possibility that these outlier vessels were produced at sites from which we lack other samples, potentially indicating that there was some exchange of vessels. However, the evidence in hand is insufficient for building robust conclusions, and the provenance of the outliers will only be resolved from more comprehensive analyses incorporating samples from more sites. It is notable that principle component analysis was not successful in differentiating the samples into site specific groups which might be expected when ceramics are locally produced at sites with very similar J9 The tenn "standardisation" is defined here as the relative degree of homogeneity or reduction in variability, or the process in achieving that homogeneity as defined by Rice (1991: 268). This concept applies to fonns as well as manufacturing technology and chemical compositions (Blackman eta/. 1993:61). geology, as is the case in the Qazvin and neighbouring Tehran plain. However, discriminant analysis, using the find site as the grouping assumption, was able to demonstrate discreet clusters within the Qazvin plain, and distinguishes those clusters from the samples collected at the type site of Cheshmeh Ali, which lies in the Tehran plain. Thus, even though the samples are petrographically similar, the material from some sites are compositionally distinct. Although the analysis incorporated a relatively small number of samples from any one site, the patterning evidence in the discriminant analysis indicates that there were multiple production sites rather than one central production site. The likelihood that there were multiple production locales for Cheshmeh Ali Ware across the north Central Plateau is particularly significant, as the similarity in stylistic and petrographic features is so readily apparent. Such a scenario invites speculation about the mechanisms that would have made the overt sharing of approaches to fabric preparation and vessel decoration possible across such a wide area. It is possible that the apparent standardisation in production is a function of technological transfer that operated as a result of inter-regional interaction that had its roots in the late Neolithic period. This may also have been occurring alongside a diachronic reorganisation of production in the ceramic industry where a household-based production operating during the Neolithic period developed into a workshop-based production system during the Transitional Chalcolithic period. Nevertheless, these workshops do not appear to have operated as centralised pottery production centres manufacturing material for the surrounding region. This indicates that while the pottery production economy involved elements of specialisation, it was not integrated into a broader regional redistribution economy. The evidence for the use of similar fabric recipes, vessel forms and decorative schemes, however, indicates that at the least, potters and the consumers of the pots being produced desired pottery that was similar to that being used at neighbouring sites and in neighbouring regions. This analysis has provided important insight into the specific types of interaction and communication between the Transitional Chalcolithic populations of the north Central Plateau, as there is clear evidence for interregional socio-cultural integration, but little evidence of direct and widespread economic interaction, at least in terms of pottery production and distribution. CHESHMEH ALI WARE Acknowledgements The authors are grateful to the NERC, Carlyle Greenwell Bequest and the Near Eastern Archaeology Foundation for supporting the analytical work, to Dr Nick Walsh and Ms Jacqui Duffel at Royal Holloway for their assistance in carrying out the chemical analyses, to Mr Chris Doherty in the Research Laboratory of Archaeology and Art History, Oxford for his patient guidance in petrographic examination and analysis, to Professor A.M. Pollard and Professor Richard Wright for their very helpful discussions on the statistical approaches and to Dr Maurizio Tosi for generously allowing access to his collection of sherds from Cheshmeh Ali. This work was also supported by a grant from the British Institute of Persian Studies. The authors also acknowledge the kind assistance and support of the members in the Department of Geology in the University of Tehran and the Institute of Archaeology, Tehran. Edna Wong Department ofArchaeology A 14 University of Sydney Sydney2006 Australia hye_ wong@hotmail.com Cameron Petrie Department ofArchaeology University of Cambridge Cambridge CB23DZ UK cap59@cam.ac.uk Hassan Fazeli Institute ofArchaeology University of Tehran Tehran Iran hfazelin@ut.ac. ir Bibliography Alavi, M. 1994. ''Tectonics ofthe Zagros orogenic belt oflran: new data and interpretations", Tectonophysics 229: 211-38. Arnold, P.J. 1991. Domestic Ceramic Production and Social 25 Organization: A Mexican Case Study in Ethnoarchaeology, Cambridge University Press, Cambridge. Baxter, M. 2001. "Multivariate Analysis in Archaeology", in D.R. Brothwell and A.M. Pollard (eds.), Handbook of Archaeological Sciences, John Wiley and Sons Ltd., Chichester: 685-94. Blackman, M.J., Stein, G.J. and Vandiver, P.M. 1993. "The Standardization Hypothesis and Ceramic Mass Production; Technological, Compositional and Metric Indexes of Craft Specialization at Tell Leilan, Syria", American Antiquity 58 (1): 60-80. Burton Brown, T. 1979. Excavations at Kara Tepe 1957, Wooton, Woodstock. [Private Publication] Coningham, R.A.E., Fazeli, H., Young, R.L. and Donahue, R.E. 2004. "Location, Location, Location: a Pilot Survey of the Tehran Plain in 2003", Iran 42: 1-12. Coningham, R.A.E., Fazeli, H., Young, R.L., Gillmore, G.K., Karimian, H., Maghsoudi, M., Donahue, R.E. and Batt, C.M. 2006. "Socio-economic Transformations: Settlement Survey in the Tehran Plain and Excavations at Tepe Pardis", Iran 44: 33--62. Dipilato, S. and Laneri, N. 1998. "Sequential Slab Construction and Other Problems Concerning HandBuilding Techniques in Chalcolithic Iran: Experimenting with Mammographic X-Ray Images", in S. Milliken and M. Vidale (eds.), Craft Specialization: Operational Sequences and Beyond: Papers from the European Association of Archaeologists Third Annual Meeting at Ravenna 1997, Volume IV, BAR International Series 720, Archaeopress, Oxford: 59--68. Fazeli, H. 2001. Social Complexity and Craft Specialization in the Late Neolithic and Early Chalcolithic Period in the Central Plateau of Iran. Unpublished PhD dissertation, University of Bradford. - 2006. The Archaeology of Qazvin. From the Sixth to the First Millennium BC, University of Tehran Press, Tehran. [In Persian] - , Coningham, R.A.E. and Batt. C.M. 2004. "CheshmehAii Revisited: Towards an Absolute Dating of the Late Neolithic and Chalcolithic oflran's Tehran Plain", Iran 42: 13-23. - , Coningham, R.A.E. and Pollard, A.M. 2001. "Chemical Characterization of Late Neolithic and Chalcolithic pottery from the Tehran Plain, Iran", Iran 39: 55-71. - , Coningham, R.A.E., Young, R.L., Gillmore, G.K., Maghsoudi, M. and Raza, H. 2007. "Socio-economic Transformations: Final Season of Settlement Survey and Excavations at Tepe Pardis", Iran 45: 267-85. -and Djamali, M. 2002. ''The Process of Specialized Ceramic Production at Zagheh Prehistoric Site Based on the Archaeological Evidence and Petrography Result", in M. Azarnoush (ed.), The First Symposium of Archaeometry in Iran: The Role of Sciences in Archaeology, Cultural Heritage Organization Press, Tehran: 203-24. - , Vidale, M., Bianchetti, P. and Giuseppe, G. in press. "The Evolution of Ceramic Manufacturing Technology During the Late Neolithic and Transitional Chalcolithic Periods at Tepe Pardis, Iran", AMI. 26 EDNA H. WONG, CAMERON A. PETRIE AND HASSAN FAZELI - , Wong, E.H. and Potts, D.T. 2005. "The Qazvin Plain Revisited: A Reappraisal of the Chronology ofNonhwestem Central Plateau, Iran, in the 6th to the 4th Millennium BC", Ancien/ Near Easlern Sludies 42: 3-82. Geological Survey of Iran. 1986. Geological Map of the Rayy plain. Geological quadrangle map oflran No F5. Ghaemi, J., Zohrab, E., Haddadan, M., Eghlimi, B. and Mosavvari, F. 2000. Geological Map of Danesfahan (Khiaraj), Geological Map of Iran Series No. 5961, Geological Survey of Iran, Tehran. Jamshidi, K. H. and Radfar, J. n.d. Geological Map ofQazvin, Geological map oflran Series No. 6062, Geological Survey oflran, Tehran. Malek Shahmirzadi, S. 1977. Tepe Zagheh: A Sixth Millennium BC Village in the Qazvin Plain of the Central Iranian Plateau, PhD dissenation, University of Pennsylvania. McCown, D. 1942. Comparalive S/raligraphy of Early Iran, Studies in Ancient Oriental Civilization No. 23, University of Chicago Press, Chicago. Matney, T. 1995. "Re-excavating Cheshmeh-Ali", Expedilion 37(2): 26-32. Rice, P.M. 1991. "Specialization, Standardization and Diversity: a Retrospective", in L.R. Bishop and F.W. Lang (eds.), The Ceramic Legacy of Anna 0. Shepard, University Press of Colorado, Boulder: 257-79. Schmidt, E. 1935. "The Persian Expedition", Universily Museum Bulle/in 5(5): 41-49. - 1936. "Rayy Research", Universily Museum Bulle/in 6(3): 79-87. Tite, M. 1999. "Pottery Production, Distribution, and Consumption-The Contribution of the Physical Sciences", Journal ofArchaeological Me/hods and Theory, Vol. 6, No. 3: 181-233. "The Chronology Voigt, M. and Dyson Jr., R.H. 1992. of Iran, ca. 8000-2000 BC", in R.W. Ehrich (ed.), Chronologies in Old World Archaeology, Vol. I, University of Chicago Press, Chicago and London: 122-78. Whitbread, I.K. 1989. "A Proposal for the Systematic Description of Thin Sections Towards the Study of Ancient Ceramic Technology", in Y. Maniatis (ed.), Archaeomelry: Proceedings of/he 25/h fnlernalional Symposium, Elsevier, Amsterdam: 127-38. Wong, E.H.Y. 2008. Ceramic Characterization and Inter-site Relationships in the Northwestern Central Plateau, Iran, in the Late Neolithic to the Bronze Age, Unpublished PhD dissertation, University of Sydney. Young, R. and Fazeli, H. 2008. "Interpreting Animal Bones in Iran: Considering New Animal Bone Assemblages from Three Sites in the Qazvin Plain within a Broader Geographical and Chronological Perspective", Pa/eorien/34.2: 153-72.