zyxwvu zy zyxwvut zyx zyx zyx zyxwv AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 53285-295 (1980) Indonesian “Solo” (Ngandong) Endocranial Reconst ructions: Some Preliminary Observations and Comparisons With Neandertal and Homo erectus Groups RALPH L. HOLLOWAY Deparhnent ofAnhiv&y, Columbia Universi&,New York, New York 10027 Endocasts, Brains, Asymmetry, Hominids, Platycephaly, PaKEY WORDS leoanthropology, Petalias, Handedness ABSTRACT Latex rubber brain endocasts of Solo I, V, VI, X, and XI crania were prepared in order to determine accurate volumes by H,Odisplacement and morphological description. The new volumes are not markedly different from previous estimates that had been done by mensuration formulae. The mean volume is 1,151 ml. The endocast surfaces, while devoid of gyral and sulcal relief patterns, do show a preponderance of left-occipital-right-frontal petalia patterns, suggesting right-handedness. Metrical observations support the viewpoint that the Solo crania are more closely related to earlier Indonesia Homo erectus than to Neandertals. In 1971 and 1972, through the courtesy of Professor R. von Koenigswald, this author had the opportunity to endocast the interiors of all the Solo crania directly by liquid latex rubber. The purpose of that work was to gain as much endocranial detail as possible, to make complete reconstructions to enable accurate volume determinations, and to compare these with previously published estimates. The literature on the Solo crania, a t least with regard to endocranial volumes, is confusing. Weidenreich’s (’51) monograph on the subject was published posthumously and does not include discussion of the volumes or the methods used to calculate them. Instead, these appear in Weidenreich’s (‘43a) classic monograph on the skull of Sinanthropus, but the methods are not precisely described, except for the Sinanthropus crania, where endocasts were made and direct water displacement was used to obtain the volumes. The discussion of the Solo endocranial volumes on p. 116, Table XXII (cm 74, p. 1121, is in the context of external cranial dimensions, L, B, H, and the Lee-Pearson formula. Oppenoorth‘s (‘37) estimates were very definitely formula-based, as were those of Dubois (’37) in which Froriep’s (‘11) methods were applied. Weidenreich‘s values in Table XXII (see Table 1) differ from those of both Dubois and Oppenoorth, but the differences are not discussed in any detail. Morphological description has so far been lim- ited to Ariens-Kappers (’36) paper, which briefly described the endocranial portions of Solo IV and V. This paper will deal mostly with the volume determination and some of the major morphological characteristics and their mensuration and will offer some suggestions regarding taxonomic affinities. zyx MATERIALS AND METHODS Making the casts Liquid latex rubber (ADMOLD #3280)’ was used to endocast all of the Solo cranial pieces. Only crania I, V, VI, X, and XI are reported here, owing to incompleteness of numbers 11, 111, VII, and VIII and the excessive distortion and poor internal condition of specimen IX. Each of the specimens required somewhat different treatment but, by and large, the overall methods were similar, as described in detail elsewhere (Holloway, ’78). Basically, each cranium was checked for cracks, which were filled with plasticene. Six or seven coats of liquid latex were poured into each specimen, 1 z zyx Obtainable frum Adhesive pmduds Gorp., 1660 Bwne Ave., New York,NY 10060. Not for citation without permission of author. Received October 25,1979 accepted March 20, 1980. zyxwvuts 0002-948318015302-0285$02.300 1980 ALAN R. LISS, INC. 285 286 zyxwv zyxwvutsr zyxwvu zyxw zyxw zyx zyxw R.L. HOLLOWAY TABLE I . Comparisons of cranial mpacities for the solo endocash* ~ Solo crania solo I solo v Solo VI solo IX solo x solo XI ~~ Author Dubois (‘37) Oppenoorth (‘37) Weidenreich (‘43a) 1,143 cc (2.5%) 1,160 cc (-1.0%) 1.316 cc (5%) l;l60 cc (12.7%) 1,035 cc (13%) 1,225 cc (2.1%) 1,035 cc (-2.1%) 1,135 cc 1,055 cc (16.7%) 1,060 cc (2.6%) 1,087 cc (6.8%) *Avg. = 1,151 oc;S.D. = 99; Var.= 7,881.36C.V. N.A. = Not Available Present study 1,172 cc 1,251 cc 1,013 cc N.A. 1,231 cc 1,090 cc = 8.6%. each coating dried by means of a hair dryer (warm air) before the next was introduced. Next, the specimens were placed into an oven and dried a t 50°C for 4 hours. After these cooled down to room temperature the cured rubber endocast shells were detached from the endocranial surfaces, with the exception of crania VI and XI, and were immediately fitted back into the cranial portions. This was done to ensure easier separation of the endocasts from the crania after plaster retainers were built inside the crania. In the case of crania VI and XI, which are the most complete and which include basal cranial portions, both were filled with liquid plaster of paris under water, so as to stabilize dimensions. This process can only be done with complete shells, which maintain enough rigidity to avoid the problems of buckling and distortion that occur with incomplete (basal portions missing) shells. Crania I, V, IX,and X all have their basal portions missing; thus the stabilizing of the endocast shell had to be done by “piece-moulding” removeable sections within the crania. Once the pieces (about six or seven each crania) are dried, they are removed from the endocast shell, bonded together, and the endocast latex shell is fitted over them. In this way the original dimensions of the situ endocast are maintained with minimal distortion. This was checked, incidentally, by comparing landmark distances on the convex surface of the endocast with the corresponding measurements on the original endocranial surface. No distortions greater than 1%were found. The original cranial portions were not broken to extract the endocasts. Reconstructions None of the Solo crania are fully intact, and to achieve the most reliable volume estimates, reconstruction of the basal portions is necessary. This was done by using plasticene, by allowing the more complete basal portions of crania VI and XI to serve as shape templates for reconstructing the bases of I, V, and X (Figs. 1-51, and by following the curvature of the present portions to fill in the missing sections. (See Holloway, ’78, for further discussion of this approach.) Volume determinations Volume measurements were done by water displacement in two separate ways. First, each endocast was made water-tight by covering any gaps between plasticene and latex wjth a thin film of latex. Using a large beaker (about 3,500 cc) fitted with a narrow tube, the beaker was first filled to overflow and allowed to drain. Next, the endocast was slowly but fully immersed, and the resultant run-off was collected in another pre-weighed beaker and consequently weighed, giving a n immediate volume, since the specific gravity of water is 1. Five determinations were made for each endocast, and the average was calculated. In the second method, each endocast was weighed in air to the nearest 0.1 gram. Next, the endocast was weighed in water (nearest 0.1 gram) by suspending the endocast from the rim of the weighing pan into a large plastic tank of water. Unless Archimedes (N.D.) was mistaken, the loss in weight exactly equals the volume of the endocast. This method provides a narrower range of values and was the method finally adopted. zyxwvu zyxwvut zyxwvu SOLO ENDOCRANIAL RECONSTRUCTIONS zyxw z 287 zyx zy zyxwvuts zyxwv zy Fig.1. Photographs of the reconstructed endocast of Solo I. First row, left to right, frontal and occipital views; second row, lateral views; third row, dorsal and basal views. In each view, the endocast has been oriented in a plane passing through frontal and occipital poles. 288 zyxwvu zyxwvutsrq R.L. HOLLOWAY zyxwvuts zyxw zyxw Fig. 2. Recomtructixl endocast of Solo V. (Dem-iptionas in Fig. 1.) zyxwvutsr SOLO ENDOCRANIAL RECONSTRUCTIONS 289 zyxwvu zyxwv zyxw Fig. 3. Reconsh-uctedendocast of Solo VI. (Description as in Fig. 1.) 290 zyxwvut zyxwvutsrq R.L. HOLLOWAY zyxwvuts zyxw Fig. 4. Reconstructed endocast of Solo X. (Description as in Fig. 1.) SOLO ENDOCRANIAL RECONSTRUCTIONS 291 zyxwvuts zyxwvuts zyxwvu zyxwvu zyxwv zyxwv zyxw Fig. 5. Recastructed end& of Solo XI. (Description as in Fig. 1.) 292 zyxwvu zyx zyx zyxwvutsrqp zyxwvutsr R.L. HOLLOWAY TABLE 2. Some metrical observations on the Solo endocasts Max. length solo endocasts L (a) R Max. height (cm) Max. breadth (a) 16.4 Solo v Solo vl solo x solo XI Mean: S.D. 17.7 16.1 17.4 16.6 16.84 0.680 4.04 C.V. ( 2 ) Height Length Height Breadth ~ ~~ Solo I Breadth Length 16.3 13.5 10.4 0.826 0.636 0.770 17.7 16.2 17.4 16.6 16.84 0.680 4.04 or less 13.3 12.6 13.4 12.3 13.02 0.536 4.1 10.9 9.8 10.6 10.45 10.43 0.402 3.8 0.751 0.780 0.770 0.741 0.774 0.033 4.3 0.616 0.607 0.609 0.629 0.619 0.0127 2.05 0.819 0.778 0.791 0.849 0.801 0.032 3.99 zyxwvuts zyxwv zyxw Metrical studies There is, to this author’s mind, little point to a comprehensive list of linear measurements on these endocasts, given their distortion in various regions, as mentioned above. Table 2 provides the three basic measurements of anterior-posterior length for both right and left endocast halves, maximum breadth, height, and the respective indices. Maxjmum lengths were measured with spmxhng calipers to the nearest mm, using light ink marks first placed on the left and right frontal and occipital poles and found through visual inspection. Maximum breadth measurements are only approximate and are usually located on the superior and posterior portion of the temporal lobe, more or less where the middle meningeal vein is evident; maximal measures are never on the parietal surface of these endocasts. Height was measured by stretching a metal tape across the temporal poles and measuring to the mid-point of the tape from vertex, in the mid-sagittal plane. Vertex was marked on the endocast as the most superior projection of the dorsal surface once the endocast had been aligned in a horizontal plane through frontal-occipitalpoles. apparent, without undue speculation. There does exist a right-frontal-left-occipitalpetalia combination. Only the anterior division of the menigeal vessel is present. The sagittal sinus appears to lead into a more or less equal division of transverse sinuses, although I would favor a right-sided pattern on the basis of an apparent greater thickness of the initial flux of the right transverse. Solo XZ. As in Solo VI this endocast is complete with a basal portion, except for a somewhat larger inferior rostral bec portion than was seen in VI. This endocast is also more distorted than VI, with a deeper right temporal pole and a skewed left frontal area on the dorsal surface. Again no sulci or gyri are apparent except perhaps in the left frontal lobe, within the complex of “Broca’s area,” which is more pronounced on the left side. There is a right frontal-left occipital petalia combination. The meningeal vessles are somewhat more clearly retained on this endocast than on VI, showing both anterior and posterior branches, which divide immediately superior of the styloid foramen. The posterior region is not clear enough to determine with preciseness whether the sagittal sinus drained RESULTS to right or left transverse sinus, although I would be inclined to favor the right side. Morphological observations Both endocasts VI and XI have undamaged Solo VZ. This endocast is by far the most rostral portions, and these both served as the complete, missing only a small portion of the templates for reconstructing the basal porrostral bec, which according to Jacob (personal tions of the remaining Solo endocasts. communication) has been more recently Solo I. The basal portion is missing, from cleaned out and is quite deep. There is but slight distortion, which appears limited to the just inferior to the frontal poles and back to left temporal region, in its superior portion. A the mid-foramen magnum region, being more small addition of approximately 10 cc of plas- extensively present on the right side. The ticine would provide the requisite symmetry endocast is distorted in the region of the right to match the right side. No sulci or gyri are sylvian region and in the posterior temporal zyxwvutsr zyxw zyxwvu 293 SOLO ENDOCRANIAL RECONSTRUCTIONS (and inferior parietal) lobes on the left side. right occipital region. Sulcal and gyral patNo gyral or sulcal markings are safely inter- terns are not evident. The question of petalias pretable. Again, as in VI and XI, there is a is difficult, given the breakage of the cranium right frontal-left occipital petalia which can- in the right occipital portion and its subsenot be accounted for by distortion but as a quent distortion. There appears to be a clear part of its “natural” asymmetry. There is also left frontal petalia, but it cannot be stated a right transverse sinus lead from the superior with absolute certainty that there was a right sagittal sinus, although it is not distinctly occipital petalia. The left frontal petalia does visible. not appear to be any artifact of distortion but rather a “natural” asymmetry. The sagittal Solo V. In this largest of the endocasts the sinus drained toward the right transverse sinbase is missing from well inferior of the frontal us. poles to mid-cerebellar lobes, although both right and left temporal lobes are more extenVolumetric determinations sively present than in I. The distortion is Table 1provides the volume determinations mostly related to the right side, resulting in a collapsed lateral aspect of the temporal lobe from this study and compares them to previand initial portion of the sylvian fissure. There ously published values. The values reported are no fissurational patterns worth mention- here are by the second method, i.e., Aring. There is, as in each of the cases above, a chimedes’weight loss method, since it is more right frontal-left occipital petalia combina- accurate with less variation between meastion; but it is mainly demonstrable on the left uring trials. In addition, the parenthesized occipital region not by the posterior projection values indicate the percent of deviation of any of the occipital lobe, but by its thickness from results from those previously published. The midline to lateral extent. The right transverse average of the five solo endocranial volumes sinus appears to receive the major supply from is 1,151mm, with an S.D. (standard deviation) the sagittal sinus. The left Broca cap is par- of 99 ml and C.V. (coefficient of variation) ticularly pronounced; the right cap is some- equal to 8.6%. The percent difference figures indicate that what smaller, even though more pointed through damage to the original cranium in in those cases in which reconstruction of the basal portions is not substantial, the present that area. values are very close to those predicted by Solo X . This endocast is roughly as com- linear measurements and formula, i.e., Solo plete as V above, but the temporal lobes, VI and XI. Where basal reconstruction is more particularly on the right side, are more pres- substantial, as in I, V, and X, the formulaent than they are in V and show the same based methods probably underestimate the tucked-under aspect as those in the complete height dimension, and thus the percent differVI and XI endocasts.There appears to be some ences are positively higher using reconstructslight depression and thus distortion in the ed endocasts. In any event, the volume deterright and left temporal lobes, immediately minations presented herein do not differ posterior to the Sylvian fissure, and in the strikingly from those previously published. zyxw zyxwvuts zyxwvutsrq zyxwvut zyxwvutsrqp TABLE 3 . Comparing Endocranial Indices Solo endocasts (N = 5) Homo ereetus Indonesia (N = 6) Homo sapiens neandertnlensis (N = 5) Mean e h length S.D. C.V.% len& S.D. C.V. Mean height breadth S.D. C.V. 0.774 0.033 4.3 0.619 0.013 2.05 0.801 0.032 3.99 0.735 0.007 0.95 0.616 0.016 2.58 0.838 0.017 2.06 Mean M t zyxwvutsrqp 0.793 0.039 4.9 0.673’ 0.027 3.97 0.789l 0.072’ 9.12l ‘ The only available Neandertal endocasts were La Chapelle, La Ferrassie, La Quina,Neandortal,and Qebel Ihroud I. The Neundertnl specimen has no base, thus height was not available, and the relevant indices are based on N = 4. 294 zyxwvutsr zyxwvutsr R.L. HOLLOWAY As the section on morphological description indicated, the temporal poles are characteristically tucked under, as are those in other endocasts for Homo erectus (non-African), and the reconstructions of the basal portions for I, V, and X followed such a pattern, which should, I believe, add to the accuracy of the volume determinations. Tables 2 and 3 provide the endocranial lengths, breadths, and heights for the Solo endocasts, and compare #em with selected Homo erectus endocasts from Indonesia and European Neandertals. The results emphasize only two things in my opinion: 1)the endocasts are clearly platyencephalic, yielding height/ length indices most similar to the Indonesia specimens; 2) the endocasts show considerable homogeneity, given the low S.D.s and very low C.V.S. The figures are published herein, should they be useful in some other context to other authors. DISCUSSlON As a group, the endocasts are homogeneously devoid of sulcal and gyral relief, with minor exceptions in the frontal lobes. The endocasts show little, if any, significant shape differences, each of them being considerably platycephalic (Table 2). That 80% (four out of five) show a combination of right frontal-left occipital petalias is interesting and possibly significant when combined with the observation that the third inferior frontal region, “Broca’s area,” is well developed on all endocasts, more often on the left than right sides. The precise significance of these findings is unclear, although LeMay (’76, ’77) and Galaburda et al. (‘78) have suggested that the petalia combinations are possibly indicative of both handedness and hemispheric dominance with regard to language. From a study in progress, this writer is finding strong evidence for leftoccipital-right-frontal petalias in most of the hominid endocasts studied to date. However, preliminary observations show such a combination t o be very strong in Gorilla endocasts also. Until these results are clarified and the functional significance is made stronger, one cannot discuss this problem at any further length.’ The Solo or Ngandong crania have been recently reappraised by Santa Luca (’78) for their morphological affinities with both Indonesian Homo erectus and Neandertal (European and Near East) groups. His opinion that these crania were representative of a group of Homo erectus controverts the widely publicized views of Brace (‘67) and of Brose and Wolpoff (‘71), who have regarded the Solo crania as reflecting part of a worldwide Neandertal stage of human evolution. This view has had support from the past literature. Both Dubois (’37) and Oppenoorth (’37) considered Solo V as the most Neandertal-like of the Solo crania, but regarded the Solo crania a s more like the Sinanthropus material (see particularly DuBois, ’36), a conclusion shared by Ariens-Kappers (‘36). Weidenreich’s view (‘42, p. 65) was very explicit: “Homo soloensis differs decisively from the European Neanderthalians and resembles Pithecanthropus in general form and specific features so closely that he can be classified flatly as an enlarged Pithecanthropus type.” (See also Weidenreich, ’43b, pp. 44-45). Clark (’64) was very cautious in his book, retaining the possibility that the Solo crania fell within the Homo neanderthalensis (sic: p. 83) range of variation, but he did not discount entirely the view of an “independent evolutionary differentiation” (p. 85) of the Ngandong population with the ancestral Australoid peoples, a view revived by Coon (’62).While not wishing to enjoin in controversy here, examination of the morphological and metrical features of the Solo endocasts leads me to conclude that the Solo group is clearly more closely aligned with Homo erectus than Neandertals. Table 3 provides a comparison of the endocast measurements and indices for the Solo, Homo erectus (Indonesian), and Neandertal groups, all based solely on available endoca~ts.~ The platycephalic condition of the Solo endocasts is clearly much more like that of the Indoneasian materials and furthermore, the location for measuring maximum breadth is on the parietal z zyxwvu zyx zyxw zyx zyxwv * Indeed,the petalia question needs more investigation and extension into pongid endocast samples. In some cases, fmntoaxipital petalias are not discernible from the usual dorsal view, when the endocast is oriented in a plane passing h u g h frontal and occipital poles, but does bemme discernible when viewed fmm the basal aspect. In some cases,the A-P petalia is not apparent but is m k e d in its lateral aspect. Thee variations q u i r e more study than they have received thus far. zyxw As the S.Dsand C.V.s make apparent,all three p u p are fairly homogeneous in their metrical pmperties. In the cases of both the Solo and IndonesianHomo erectm etxlocasts, some small part of this homogeneitycan probably be attributedto reconstruction techniques, where the most complete endocasts act as templates to build up missing sections (basal) on those less complete. This latter observation should not, however,deemphasize their considerable homogene ity lobe in Neandertals (La Quina excepted) and is wholly temporal on both Solo and Indonesian Homo erectus endocasts. Further studies On these endocasts are in progress, which utilize a stereometric and comparative framework for measuring roughly 170+ points on the dorsal surface and, after each measurement is corrected for allomekY, or scaling, the residuals are examined by a variety of multivariate techniques, particularly discriminant analysis. So far the results (tentative) suggest much stronger relationships with both the Indonesian and Chinese Homo erectus endocasts and less strong than with European Neandertals.4 zy zyx zy zyxwv zyx zyxwv zy zyx zyx SOLO ENDOCRANIAL RECONSTRUCTIONS 295 Coon, C.S. (1962) Theorigin of Races. New York AE.XI A. a o p f . Dubois, E. (1936) Racial identity ofHomo soloensis Op penoorth (including Homo modjokertensis von Koenigswald) and Sinnnthropus pekemsis. Davidson Black. Konin. Akad. Van Wetemchappen, 39:l-6. Dubois, E. (1937) Early Man in Java and P z t b n t h m $~’!~~.~i~~~&,~~; $yLgFIYMan. Froriep, A. (1911) Ueber die Bestimmung der Schadelcapacitat durch Messung und Berechnung. 2. Morphol. Anthropol. Bd. XIII: 347-374 ~ ; g ~ science,199;85%856. Holloway, R.L. (1978) The relevance of endocasts for studying brain ev~htion.In Noba& CR. (.XI): Sensory Systems of Primates. New York: Plenum Press, pp. 181-200. LeMay, M. (1976) Morphological asymmetrics of modern ACKNOWLEDGMENTS man, fossil man, and non-human primates. In Harnad, H.D. SkMiS, and J. L~IUXW&~ (d): OriginS and 1 am very grateful to Dr. G.H.R. “on KO- S.R.7 evolution of language and speech. Ann. N.Y. Acad. Sci. enigswald, Senckenberg Institute, Frankfurt, 2&):34%360. Federal Republic of Germany, who gave me m y , M. (1977) Asymmetries of the skull and handd t J.N ~e ~~ lsci. . 32:243-253. the permission and encouragement to endosoloensis W3.F. (1937) “he Place of cast the original specimens, and for his has- %PenCJ”fi, among fossil man. In M a c C d y , G.G. (ed):Early Man. pitality and generous offer of facilities. I also Philadelphia: J.B. Gpphcotta., pp. 349-360. thank Dr. Teuku Jacob for his cooperation and santa LUB, A.P. (1978) A reexamination of p r e m e d Neandertal-like fossils. J . of Hum. Ewl. 7:619-636. permission to study the other Homo erectus F. (1942) Early man in Indonesia. Far materials in Indonesia, and for his interest Weidenreich, Eastern Qwuterly Nov. pp. 58-65. and advice. This research was supported in Weide-i&, F. (&3) heskull of,yimnthropus p h i nensis: A comparative study on a primitive hominid part by NSF grants GS 2931-1 and BNS skull. Palaeontologica Sinica, new ser. D, No. 10, whole 7911235 and in part by a Guggenheim Fellowseries No. 127, pp. 1-484. ship, for which I am most grateful. Weidenreich,F. (1943b) The “Neanderthal man” and the ancestors of “Homo sapiens”. Am. Anthrupol. 45:39-48. LITERACITED Weidenreich, F. (1951) Morphology of Solo Man. Vol. 43 Ariens-Kapprs, C.U. (1936) The endocranial casts of (Part 3), Anthropology Papers of the American Museum of Natural History, New York. the Ehringsdorf and Homo soloensis skulls. J. Anat. 71:61-76. Brace, L.C. (1967) The Stages of Human Evolution. En4 It would be premature to disclLss thee tentative results in any glewood Cliffs, New Jersey: F’rentice-Hall, Inc. detail, as they are O ~ supportive Y and are independent Of the B ~D.s.,~ad,M.H. wolp0ff (1971) ~~l~ upper morphological a d metrical aspeds d i x d herein. When discrimlithic man and late tools, Am. Anthropol. inant analyses are done, the small Solo group (N = 5 ) tends to be 73:115694. classified either in its own group or with either the Chinese or Clark, W.E. LeGm (1964) The Fossil Evidence for HU- Indonesian Homo erectus groups, rather than in the Neandertal man Evolution. Chicago: University of Chicago Press. group. G%:$(k$j. :dthFbzi’ ~ zyx we