AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 59157-174 (1982) Basicranial Anatomy of PlioPleistocene Hominids From East and South Africa M.C. DEAN AND B.A. WOOD Department of Anatomy, The Middlesex Hospital Medical School, Cleuehnd Street, London WlP 6DB KEY WORDS Basicranium, Hominids, Systematics ABSTRACT The results of a metrical analysis of the basicraniurn of 19 PlioPleistocene fossil hominid crania are presented. The sample includes crania attributed to Australopithecus africanus, Australopithecus boisei, and robustus, and Homo erectus as well as crania whose attribution is still under discussion. These results confirm significant differences between the cranial base patterns of the “gracile”and “robust”australopithecines and the three crania attributed to Homo erectus have a pattern which resembles that of modern humans. None of the crania examined from East Africa sites have base patterns which resemble that of the “gracile” australopithecines. The crania KNM-ER 407 and 732 have patterns which are compatible with them being smaller-bodiedfemalesof Australopithecus boisei; KNM-ER 1470 and 1813 have base patterns which most closely resemble that of Homo erectus. The cranial base pattern of KNM-ER 1805 is compatible with its inclusion in either Australopithecus boisei or Homo. When account is taken of the immaturity of Taung, the evidence of its cranial base pattern suggests that if it had reached adulthood it would have resembled the “gracile”australopithecine crania from Sterkfontein and Makapansgat. In an earlier paper we established a comparative metrical framework for assessing the significance of variations in the basicranial morphology of early hominid crania (Deanand Wood, 1981a). Linear and angular measurements were made with the basicranium viewed in normal basilaris, and special emphasis was given to establishing the relative positions of important bilateral landmarks. In this way we hoped to be able to integrate what was already known about cranial base flexion in the sagittal plane, with changes in the position and orientation of bilateral structures. The results of the comparativestudy showed a remarkably consistent pattern of cranial base morphology in the three pongid taxa. The basicrania of Gorilla, Pongo, and Pan are relatively long and narrow with sagittally orient&ed petrous temporal bones and a posteriorly situated foramen magnum. In contrast, the cranial base in Homo sapiens is broad but compressed anteroposteriorly. The foramen magnum is more anteriorly situated and the petrous temporal bones are shorter and orientated with their long axes more aligned to the cor- 0002-WW6902-0157 $&.OO 0 1982 ALAN R. LISS, INC. onal plane (Fig. 1).Four hominid crania were examined in the earlier study, all belonging to the genus Australopithecus. Sts 5 and MLD 37/38 represented Australopithecus africanus, the “gracile”australopithecine, and KNM-ER 406 and OH 5 sampled one of the species of “robust” australopithecine, Australopithecus boisei. All four hominid crania retain a relatively long cranial base and the overall morphology of the two “gracile” crania is remarkably pongidlike. However, the anteriorly positioned foramen magnum and the more coronally orientated petrous bones of KNM-ER 406 and OH 5 are reminiscent of the basicranial morphology of Homo sapiens. In this second part of our investigation we have enlarged the fossil sample from 4 to 19. This has not only allowed us to make modest, but important, increases in the sample sizes of the two australopithecine taxa, but the larger sample also includes three crania attributed to Homo erectus. Most of the remaining fossil W i v e d July 6,1981; accepted April 26,1982. 158 M.C. DEAN AND B.A. WOOD A B Fig. 1. Skull base diagrams of A. Homo sapiens and B. Gorilla with the landmarks used to define measurements employed in this study. A full list of the landmarks together with the measurement definitions are given in Dean and Wood (1981a). hominid crania have either been attributed to taxa for which there are no other good examples of basicranial morphology, e.g., OH 24; or they have been proposed as examples of smaller-bodied females of the “robust” australopithecines, e.g., KNM-ER 407 and 732 (Leakey, 1974; Wood, 1978); or they are crania whose affinities are sufficiently enigmatic for their taxonomic attribution to have been the subject of debate, e.g., KNM-ER 1470 and 1813 (Leakey, 1974; Leakey, et al., 1978; Walker, 1976 and Wood, 1976) and SK 847 (Clarke et al., 1970; Clarke and Howell, 1972; Wolpoff, 1971). Two other specimens from South African sites have been treated separately: TM 1517 from Kromdraai because it is fragmentary, and the skull from Taung because of its immaturity. The aims of this new study are to test whether, with a larger sample, it is still possible to discern two patterns of basicranial morphology in Austrulopithecus; to establish whether Homo erectus crania show a consistent pattern of cranial base morphology, and to attempt to clarify the taxonomic affinities of the individual fossil crania by comparing their basicranial shape with that of established fossil groups. A comparative metrical framework for this analysis was established in an earlier paper (Dean and Wood, 1981a). This paper should be consulted for the detailed results of the measurements taken on the samples of extant taxa, but they are presented in a summarized form in Table 3. The problems of simultaneously comparing a series of variables are often solved by resorting, with little or no preliminary analysis, to sophisticated multivariate analytical techniques. However, although such an analysis is in progress, we are conscious that it cannot replace, and should always be preceded by, careful inspection and consideration of the original data. It is, of course, possible to employ simpler bivariate statistical tests of affinity, but we considered that the small sample sizes of the fossil groups in this study seriously limited the usefulness of such tests. We decided therefore, to present the data for each specimen separately, and to 159 BASICRANIUM OF PLIO-PLEISTOCENEHOMINIDS base our interpretations on observed patterns of overall cranial shape, and on variables which were shown to be useful discriminators in our earlier comparative study. We believe that metrical data like these, from a region such as the cranial base, and collected from extant taxa as well as fossils could be used as part of a cladistic analysis of fossil hominids, and such a study, based on a series of cranial morphological features, is in progress. MATERIALS AND METHODS Nineteen fossil crania were examined and measured, and except for two specimens, OH 5 and OH 9, all the observations were made on the original fossils. The reference numbers of the specimens, their site of origin, and the location of the original specimens are given in Table 1. All specimens are judged to be adult, except for SK 47 which is sub adult, and Taung which is juvenile. Details of the comparative extant samples of Homo, Gorilla, Pongo, and Pun are given in Dean and Wood (1981a). The fossils are not equally well preserved. Five of the specimens used in the study (KNMER 406, KNM-ER 3733, KNM-ER 3883, OH 5, and Sts 5) are sufficiently well preserved to identify the landmarks, and it was possible to make reliable estimates of all the measurements on these crania. Three crania (Sts 19, MLD 37/38, and SK 47) are damaged in the left or right infratemporal region and the position of the damaged infratemporal crest was derived by doubling the distance from the preserved side to the midline. Five specimens (KNM-ER407,1470,1805,1813, and OH 9)are damaged in the region of the basioccipital and the foramen magnum; interpretation of the base of KNM-ER 1470 is further complicated by distortion. Where landmarks are preserved on one side of the cranium, measurements were derived by doubling the distance to the midline. The cranial base is particularly poorly preserved in KNM-ER 1470; nonetheless, by using the preserved margins of the both foramina ovale and the right carotid canal, and part of the right petrous temporal bone, it is possible to estimate the distances FOIFO, CCICC, and the angle a.The positions of the infratemporal crest (preserved on the right) and the lateral extremities of the tympanic plate can also be estimated. The cranium OH 24 was found in a badly distorted state embedded in a mass of calcareous matrix, but many of the pieces of the crushed skull have been separated, cleaned, and reassembled (Leakey et al., 1971). Despite this extensive and skilled attempt at restoration the cranium remains slightly distorted and the right petrous and tympanic plate are still in their crushed position. There has also been some reconstruction of the lateral end of the left tympanic plate. TABLE 1. Details o f the Plio-Pleistocene hominid fossils used in the studv Site of origin Koobi Fora (East Rudolfl, Kenya Olduvai Gorge, Tanzania Sterkfontein South Africa Swartkrans, South Africa Makapansgat, South Africa Kromdraai South Africa Taung South Africa Museum or Collection registration number Institution where original is kept KNM-ER 406 KNM-ER 407 KNM-ER 732 KNM-ER 1470 KNM-ER 1805 KNM-ER 1813 KNM-ER 3733 KNM-ER 3883 OH 5 (Cast) OH 9 (Cast) OH 24 Sts 5 Sts 19 Sts 25 SK 47 SK 847 MLD 37/38 National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Nairobi National Museum, Dar es Salaam National Museum, Nairobi National Museum, Nairobi Transvaal Museum, Pretoria Transvaal Museum, Pretoria Transvaal Museum, Pretoria Transvaal Museum, Pretoria Transvaal Museum, Pretoria Department of Anatomy, University of Witwatersrand, Johannesburg. Transvaal Museum, Pretoria TM 1517 Taung 1 Department of Anatomy, University of Witwatersrand, Johannesburg. 160 M.C. DEAN AND B.A.WOOD Five crania (KNM-ER 732, Sts 25, SK 847 TM 1517, and Taung) are well enough preserved on one side only. Where any two or more of the following midline landmarks were preserved (prosthion, staphylion, hormion, sphenobasion, basion, opisthion, or opisthocranion), or where there were other indications of the sagittal plane, these were used to construct a midline, and measurements and the positions of landmarks were derived by doubling the distance from the midline to the intact side. In some cases the position of a landmark was inferred from the topography of nearby preserved bone. All landmarks whose positions have been derived by reconstruction, or by mirror imaging, are indicated by broken lines in Figures 2-6, and measurements involving these landmarks are put in parenthesis in Table 2. I t must be emphasized that measurements taken on crania reconstructed in the ways outlined above are inherently less reliable than those made on intact specimens. However, it is our belief that in most cases the measurements made using reconstructed landmarks are accurate to within a few millimetres, and it is on this basis that these specimens were included in the study. Nevertheless, the results should be interpreted with this important caveat in mind. The landmarks and the measurements are those that were listed and defined in Dean and Wood (1981a),and the landmarks have been located in a diagram (Fig. 1)of two crania seen in normal basilaris, one Homo sapiens and the other a Gorilla. Measurements were made on three occasions and the mean values were recorded. In cases of doubt the measurements were checked by a second observer. Definitions of the landmarks and details of the measurement technique are given in Dean and Wood ( 1981a). surements of Homo sapiens and Gorilla have been included at the top of each figure for comparison. The results will considered in the four groups used in Table 2. 1. Adult fossils attributed to Australopithecus africanus (Sts5, Sts 19, Sts 25, and MLD 37/38) (Figure 2.) The skull base diagrams of all these fossils more closely resemble the pattern of the pongid samples than they do the modern human sample. Overall they are small and the bilateral structures all lie relatively close to the midline, except for the carotid canals which are relatively further apart than those of the pongid taxa. Compared to the pongids the length of the cranial base as a whole is reduced in Sts 19,25, and MLD 37/38, but rather less so in Sts 5, and the foramen magnum is closer to the bitympanic line. The bitympanic width is also reduced in comparison with the pongid taxa, but not to the extent it is in the modern Homo sapiens sample; this may be related to a reduction in the size of the masticatory apparatus (Dean and Wood, 1981a).The petrous axes are aligned more sagittally than in the modern Homo sapiens sample, but slightly less so than in the pongid taxa. The range for a in the “gracile”australopithecines is 59“ to 72”. This is comparable to the distribution of values in Pongo ( a = 60”-71”), but it is at the lower end of the ranges of the Pan and Gorilla samples. The orientation of the tympanic axis is very similar to that in the pongid taxa but the range within the “gracile” group also overlaps the range of the modern Homo sapiens sample. The more sagittal orientation of the petrous axis is the most notable contrast between the cranial base patterns of the “gracile”australopithecines and modern Homo sapiens, and the shortened tympanic plates and reduced skull base length of the “gracile”australopithecines are features which contrast with the pattern typical of the pongids. RESULTS 2. Fossils attributed to Australopithecus The results of the 15 linear measurements boisei and robustus (KNM-ER 406, OH 5, and and the two angles for the fossil sample are SK 47) (Figure 3). The cranial base diagrams of given in Table 2. The crania are grouped in the this group of fossils are distinct from the patTable according to accepted taxonomic attri- tern of both the comparative pongid sample bution, or treated as individual specimens, as and the “gracile”australopithecines, and they outlined in the introduction. The parameters share some features of the cranial base pattern for the comparative samples are given in Dean of the modern Homo sapiens sample. Overall and Wood (1981a. Table 2); the measurement the cranial base of this group is large and parorder has been revised in the present paper to ticularly wide across the lateral extremities of separate those which relate to basicranial the tympanic plates. The cranial base is width from length measurements. The results shorter than that of the pongids and the anterare also presented in a series of two-dimension- ior margin of the foramen magnum is characal plots (Fig. 2-6). In each of these plots the teristically positioned well in front of the mean values for the linear and angular mea- bitympanic line, unlike either the modern 48 44 44 47 58 57 43 61 54 57 40 (41) (45) (51) 44 49 (46) (41) (36) 48 50 42 47 65 59 51 58 57 67 53 136) (60) 55 48 48 (44) (50) (38) 60 160) 70 (65) 60 61 (62) (61) (44) 50 (49) 50 50 61 64 (54) 66 69 76 72 78 72 (63) 67 73 69 85 83 78 89 84 92 69 68 28 114) (25) (20) (20) 23 (21) (18) (15) 26 25 (26) 25 29 29 22 35 31 30 29 (26) 25 29 28 29 27 27 35 31 29 27 (30) 29 28 25 32 26 20 32 26 24 22 25 33 (30) 19 32 30 23 (35) 9 23 22 19 26 34 36 25 29 29 25 (25) 24 (20) 20 20 (21) (14) 19 23 24 30 24 20 26 25 28 19 24 20 'All linear measurements in mm. 'All angles to the nearest degree. 'Measurements in parentheses are either estimates, or, in the case of widths, were derived by doubling measurements taken on one side only. Sts 5 92 66 Sts 19 89 63 Sts 25 (90P MLD37138 97 62 Australopithecus KNM-ER406 132 79 (Paranthropus) OH5 122 72 boisei and robustus SK47 100 68 Homo erectus KNM-ER3733 120 79 KNM-ER3883 115 78 OH9 123 84 Individual fossils KNM-ER407 102 64 KNM-ER732 (100) KNM-ER1470 131 KNM-ER1813 103 66 KNM-ER1805 118 72 OH24 (103) 67 SK847 (98) (60) TM1517 (118) 66 TAUNG 54 Fossil Groups Australopithecus africanus Lengths (16) (16) (25) (22) 21 22 29 (28) 12 27 24 30 25 20 47 50 (56) 44 60 45 49 63 29 53 41 (43) 47 57 60 45 55 55 60 (27) 24 42 28 31 35 18 24 28 22 37 24 30 31 23 30 28 30 49 45 (56) (45) 45 46 (58) 46 38 54 41 50 (57) 42 32 50 40 95 112 108 110 115 90 112 105 (105) 50 65 103 39 59 98 (40) 72 104 48 60 93 53 44 100 55 45 102 37 45 107 57 48 104 52 55 105 60 50 108 Angles' IT/IT- FOIFO- IT/IT TP-TP SP-SP CC-CC FO-FO IT-IT SM-SM AP-AP FM-FM BS-OP TP-CC CC-PA SB-BS TPITP TPlTP -BS a p Widths Table 2. Cranial base data for hominid fossils' rn 8 5 0 5: 5 M c, F 1 0 * G 0 "1 cd k d 3 : *m 162 M.C. DEAN AND B.A. WOOD TABLE 3. Cranial base data for comparative groups Widths Comparative groups Homo sapiens X n = 30 S.D. cv Range Pan troglodytes n = 30 Gorilla gorilla n = 30 - X S.D. cv Range X S.D. cv Range Pongo pygmaeus i7 n = 30 S.D. cv Range TP-TP SP-SP CC-CC FO-FO IT-IT SM-SM PA-PA FM-FM 99.2 3.8 3.8 91-108 78.0 3.8 4.9 71-86 56.8 3.4 5.9 50-63 49.4 2.8 5.7 45-54 65.5 4.7 7.3 57-77 81.6 4.2 5.2 72-90 32.2 3.4 10.4 27-39 29.9 2.1 7.1 25-33 101.8 6.3 6.2 92-118 59.1 3.6 6.1 54-66 42.5 2.8 6.4 38-48 42.9 2.8 6.4 37-48 51.0 3.3 6.5 42-59 63.6 3.6 5.6 58-69 24.0 2.4 10.6 19-28 23.4 1.9 8.3 20-28 125.9 9.0 7.1 112-143 65.4 3.8 5.6 60-72 47.5 4.5 9.5 40-57 51.3 3.2 6.2 42-57 59.1 4.1 6.9 51-67 79.8 5.2 6.5 72-89 30.4 2.8 9.4 23-34 28.6 2.2 7.8 24-35 114.7 11.2 9.8 93-136 68.2 5.5 8.1 55-77 49.5 3.8 7.6 41-57 45.3 3.6 7.8 41-54 51.5 5.1 9.8 41-63 76.1 5.4 7.1 65-86 28.8 2.7 9.3 23-35 24.4 2.1 8.6 21-31 Angles - Lengths BS-OP TP-CC CC-PA SB-BS ITIITTPlTP 35.0 2.7 7.8 29-40 21.9 2.1 9.7 18-25 18.6 2.3 12.0 16-23 25.5 2.8 11.2 20-31 47.3 4.1 8.8 41-55 25.9 2.2 8.3 22-32 48.0 4.7 9.7 40-58 45.7 4.8 12.7 31-55 107.0 3.3 3.1 102-114 28.2 2.9 10.1 22-34 31.8 2.9 9.2 26-36 24.0 2.5 9.3 20-29 26.3 2.4 9.1 20-31 48.8 4.2 8.6 42-56 24.6 3.3 13.2 19-29 51.8 3.8 7.4 41-58 68.9 5.3 7.7 60-78 96.0 4.9 5.1 86-105 31.8 2.5 7.9 28-37 40.5 3.5 8.6 34-46 29.8 3.1 10.2 20-37 31.4 3.1 9.9 25-36 61.6 6.6 10.6 52-75 31.3 3.8 12.1 25-38 68.3 7.8 11.4 52-85 71.9 4.9 6.8 60-81 95.0 3.0 3.2 88-102 30.7 3.0 9.8 24-35 34.7 3.8 10.9 29-41 28.4 2.9 10.3 23-35 30.2 4.4 14.5 21-37 57.9 6.7 11.6 43-70 28.0 3.7 13.1 21-38 61.5 7.1 11.6 45-75 68.1 3.0 4.4 60-71 101.0 4.5 4.4 93-108 humans or the “gracile”australopithecines and the pongids. The orientation of the petrous axis closely resembles that of the modern human sample and values for this angle in the “robust”group of fossils fall within the middle of the range, a! = 31” - 55”, for modern Homo supiens. The orientation of the tympanic axis also resembles that of the modern human group, but values for 0 for the “robust” group fall towards the lower end of the modern human range. In this particular fossil group this is almost certainly due to the long tym- FOIFOTPITP ITIlT -BS a P panic plates reducing the value of /3. The bicarotid canal width (CC-CC)in KNM-ER 406 and OH 5 exceeds the modern human mean value, but whereas in the Homo sapiens sample the bi-infratemporal fossa width (IT-IT)is disproportionately wider than the other bilateral landmarks this is not the case in the “robust” group. There appears to have been little reduction in the length of the body of the sphenoid bone in the “robust”crania (if spenoid length is taken as equivalent to the distance ITIIT-FOIFO). 163 BASICRANIUM OF PLIO-PLEISTOCENEHOMINIDS x X X X GoriIla Homo X X x x Sts 19 Sts 5 X x Sts 25 + X ,1( MU) 37138 I All diagrams are to scale and are reduced to 40% of life size II Landmarks in Homo and correspond to their mean value I I I Where the position of a landmark can only be estimated, or is derived by mirror imaging the preserved side, it is marked by a broken line Fig. 2. Skull base diagrams of fossils attributed to Australopithecus africanus, with Homo and Gorilla for comparison. 164 M.C. DEAN AND B.A.WOOD X X X X Homo - Gorilla X X X x OH 5 KNM-ER 406 X x SK 47 I All diagrams are to scale and are reduced to 40% of life size II Landmarks in Homo and G M correspond to their mean value I I1 Where the position of a landmark can only be estimated. or is derived by mirror imaging the preserved side, it is marked by a broken line Fig. 3. Skull base diagramsof fossils attributed to Austmlopithecus boisei and robustus, withHomo and Gorilla for comparison. 165 BASICRANIUM OF PLIO-PLEISTOCENE HOMINIDS X X X X Homo - Gorilla X X X KNM-ER 3733 X KNM-ER 3883 X X OH 9 I All diagrams are to scale and are reduced to 40% of life size II Landmarks in Homo and Gorilla correspond to their mean value I II Where the position of a landmark can only be estimated, or is derived by mirror imaging the presented side, it is marked by a broken line Fig. 4. Skull base diagrams of fossils attributed to Homo erectus, with Homo and Gorilla for comparison. 166 M.C. DEAN AND B.A. WOOD X X X KNM-ER 407 KNM-ER 732 x X 0 I II x 0 All diagrams are to scale and are reduced to 40% of life size Landmarks in w a n d Q&Jg correspond to their mean value i I I Where the position of a landmark can only be estimated. or is derived by mirror imaging the preserved side,it is marked by a broken line Fig. 5. Skull base diagrams of individual Plio-Pleistocenehominid crania from East Africa, with Homo and Gorilla for comparison. 167 BASICRANIUM OF PLIO-PLEISTOCENEHOMINIDS X X X X Gorilla - Homo X 0 X 8 , ’.-__-’ ‘ SK 847 OH 24 \ i X .4 r: U 0 x 0 TM 1517 s , .-. L.’ Taung I All diagrams are to scale and are reduced to 40% of life size II Landmarks in Homo and Gorilla correspond to their mean value II I Where the position of a landmark can only be estimated. or i s derived by mirror imaging the preserved side,it is marked by a broken line Fig. 6. Skull base diagrams of individual Plio-Pleistocenehominid crania from South Africa, with Homo and Gorilla for comparison. 168 M.C. DEAN AND B.A. WOOD KNM-ER 1470 and KNM-ER 1813 3. Fossils attributed to Homo erectus (KNMER 3733,3883, and OH 9) (Figure 4).The craniThe cranial base patterns of these two fossils al base pattern of this group of fossils resem- resemble those of the Homo erectus group. The bles that of modern Homo sapiens. They differ foramen magnum lies on the bitympanic line, from the “robust” group and resemble the Ho- the angle p is low and the angle a high. The mo sapiens sample in two ways. The first is the sphenoid bone is foreshortened, but wide position of the foramen magnum which lies ap- across the infratemporal fossa at the base of proximately in the bitympanic line, and not the greater wings. Two obvious differences bewell in front of it as it does in the “robust” tween these two crania are the relatively short australopithecine fossils. The second is the sphenoid, and the smaller overall size of KNMbroadening of the sphenoid so that the bi-infra- ER 1813. temporal fossa width exceeds the biforamen ovale and bicarotid canal widths; this trend is, KNM-ER 1805 and TM 151 7 however, not as marked in Homo erectus as it These fossils combine a relatively long craniis in Homo sapiens. The petrous axis in Homo erectus is more sagittally orientated than in al base and a foramen magnum which is posiHomo sapiens and the “robust”australopithe- tioned well in front of the bitympanic line. The cines, with the values of a falling at the upper values for angle 0 are high, with the tympanic end of the Homo sapiens range; in contrast the axis being orientated further forwards than it values for /3 fall towards the bottom of the is in modern Homo sapiens, but the angles a range for Homo sapiens. This is partly due to are comparable to the mean values for both the the relatively long tympanic plates in Homo “robust” australopithecines and the modern erectus which have the effect of reducing the Homo sapiens sample. The presumed position value of /3 and accentuating the difference be- of the spheno-occipitalsynchondrosis in KNMER 1805 is some way behind the line connecttween the tympanic and petrous axes. 4. Individual Plio-Pleistocene hominid cra- ing the posterior margins of the foramen ovale, nia (KNM-ER 407,732,1470,1805, and 1813, thus suggesting that the sphenoid body may OH 24, SK 847, TM 1517, and Taung) (Figures have been longer than indicated by the ITIIT5 and 6). For convenience these fossils have FOIFO distance. Apart from the relatively been divided into four groups on the basis of si- high value for angle (3, and a marked narrowing milarities in cranial base patterns; no other im- of the biforamen ovale width (a feature which plications are intended at this stage and the they share with KNM-ER 732), the basigroupings are not intended to be taxonomic crania of these two specimens are similar in shape to those attributed to the “robust” units. australopithecines. KNM-ER 407, KNM-ER 732, OH 24, and SK Taung 847 Developmentally, this juvenile specimen is This group of fossils shows a combination of features, some of which are characteristic of not directly comparable with the other fossil the “robust” australopithecines, and others of hominids used in this study, and for this reawhich are seen in Homo erectus and in the son it has been considered separately. It is posmodern Homo sapiens sample. The position of sible that gingival eruption of the permanent the foramen magnum relative to the bitympan- first molars has occurred in this skull but it is ic line resembles that in the “robust” sample, unlikely that the first permanent molars were but the relatively wide and foreshortened sphe- in functional occlusion. This corresponds to a noid bone are features of Homo. The value of a developmental age of about 6 years in modern for the orientation of the petrous axis is 49“ in man and to about 3 years 4 months in the three KNM-ER 407,45” in KNM-ER 732,54” in OH great apes (Dean and Wood, 1981b).A separ24, and 50” in SK 847. With the exception of ate study on the growth of the hominoid craniKNM-ER 732, these values are slightly higher al base in norma basilaris (Dean, in preparathan those for the “robust” sample, and within tion) suggests that at the time of eruption of the range for Homo erectus. There is little to the first molars the angles a and p would not distinguish between the cranial base patterns yet have reached their adult values. Of all the of these four fossils except that the petrous measurements used in this study, only the angle and the sphenoid lengthlwidth ratio of length and width of the foramen magnum of OH 24 and SK 847 are more indicative of the the Taung skull can be expected to approach the adult values (Ashton and Spence, 1958). pattern seen in Homo erectus. BASICRANIUM OF PLIO-PLEISTOCENE HOMINIDS The value of 50” estimated for angle a is already greater than the mean value for the adult Homo sapiens sample, and it also exceeds the values for the adult and subadult “robust” australopithecine fossils used in this study. During the growth of the cranial base from the time of eruption of the first permanent molars to adulthood an increase of about 10O occurs in both the pongid taxa and Homo sapiens (Dean, in preparation). If this is added to the existing value for a,50°, then the inference is that the petrous axis in the Taung adult wouId have been approximately 60”. This would place it out of the observed range of the “robust”crania and well within the “gracile”australopithecine range. DISCUSSION Differences in the degree of cranial base flexion in the skulls of hominoids have been linked with a variety of morphological and functional correlates. Huxley (1863)suggested that there was a relationship between facial prognathism and cranial base flexion, and since then Huxley (1867),Weidenreich (1947),Bjork (1950),Scott (1958), and Cramer (1977)have presented evidence to support such an association. Other workers have concentrated on the effects of the expansion of the neurocranium on cranial base flexion during both human phylogeny and ontogeny (Moss, 1958 and Biegert, 1963). The type of habitual posture, and the relative proportions of neurocranium and viscerocranium have also been related to the degree of cranial base flexion (Schultz, 1955; Clark, 1959; Du Brul and Laskin, 1961, and Du Brul, 1977). More recently Laitman (1977)and Laitman et al. (1978,1979)have presented evidence to suggest that the degree of exocranial base flexion may indicate the position of structures in the upper part of the respiratory tree in primates. In an earlier paper (Dean and Wood, 1981a) we emphasized the importance of recording variation in the shape and form of the cranial base when it is viewed in norma basilaris as well as in the sagittal plane. Among the problems of sagittal studies, not touched on in our earlier discussion, is the difficulty of describing a complicated anatomical region across a range of primate comparative groups without a reliable set of homologous landmarks (Lestrel and Moore, 1978). For example, while nasion is a reliable indicator of the plane of the foramen caecum (the anterior end of the cranial base) in modern human crania, because of an upward remodelling which occurs during growth in many nonhuman primates it is a 169 much less reliable indicator of the position of the foramen caecum in these groups (Ashton, 1957; Scott, 1958, 1963; Cramer, 1977). Likewise intrataxonomic variations in the degree of upward remodelling have been shown to occur in the region of the sella turcica (Latham, 1972; Gould, 1978), and there is also evidence that growth changes occurring at the spheno-occipita1 synchrondrosis may differ among primate taxa (Sirianni and Van Ness, 1978). In discussing the significance of the results of this study of hominid basicrania we are, of course, conscious that they refer to a restricted set of metrical criteria from only one area of the skull. However, the results of the preliminary study of the variation of this region in a series of samples of extant primates, established the existence of consistent morphological patterns which could, moreover, be plausibly related to known patterns of cranial base flexion in the sagittal plane. Thus it is against this comparative background, and the knowledge that the cranial base must reflect, or respond to, a whole series of important functional demands, that the results should be assessed. The additional of Sts 19 and 25 to the sample of adult crania attributed to Aus tralopithecus africanus has reinforced the conclusions based on the initial observations on Sts 5 and MLD 37138. In general Sts 19 and 25 follow the pattern of MLD 37/38 in having a more foreshortened sphenoid than Sts 5. Sts 19 has both a shortened sphenoid and the least sagittally orientated petrous axis of the group which makes it the most “Homo-like”of the sample. The suggeston that Sts 19 may not be a typical “gracile”australopithecine is not a new one. When Sts 19 was first described in detail, and referred to as skull No. 8 (Broom and Robinson, 1950).the authors referred to the “human” proportions of the posterior cranial fossa. In the same monograph Schepers (1950) commented that “the manner in which the cerebellum has come to be shifted forward and below the cerebral occiput is most striking. The general arrangement is that found for the human brain, especially in Homo sapiens.” Clarke (1977)considered the morphology of the cranial base of Sts 19 in detail and suggested there were several features of the morphology of the temporal and sphenoid bones that were Homolike. He considered the possibility that Sts 19 should be excluded from Australopithecus africanus, but also noted the alternative, that the “moreHomo-like characteristics of Sts 19 compared to the more pongidlike characteristics of Sts 5 might be indicative only of a wide range 170 M.C. DEAN AND B.A. WOOD of variation among the Sterkfontein Australopithecus population.” The results of this metrical analysis suggest that the Clarke’ssecond interpretation is the more probable one. The subadult “robust”australopithecine cranium SK47 has been included with the adult crania KNM-ER 406 and OH 5 in this study because it is the only “robust” australopithecine specimen from South Africa with a more or less complete and undistorted cranial base. Despite its immaturity Ashton and Zuckerman (1952) noted that in SK 47 “themuscular markings of the occipital region of the fossil are far more pronounced than in either man, or the chimpanzee, and are as well developed as in gorillas of corresponding age.” In addition, they comment that the inion and external occipital crest in SK 47 are better marked than those of adolescent gorillas and chimpanzees. The results of our metrical study suggest that, if it had reached full adulthood, SK 47 would have come to resemble the two adult “robust” australopithecine crania. Olson (1978) has previously drawn attention to cranial and dental features of SK 47 which he considers are hominine, but these cranial base metrical data on their own are not sufficiently discriminatory to make any significant contribution to deciding the relative strengths of the hominine and “robust” australopithecine affinities of SK 47. Nonetheless some features, such as the anteriorly situated foramen magnum, suggest a closer association with the “robust” taxon. Howell (1978)has maintained a separation at the species level between specimens recovered from Kromdraai and Swartkrans, but the incomplete cranial basc of TM 1517 resembles the cranial base pattern of the “robust” sample used in this study much more closely than any of them resemble the pattern in the “gracile” australopithecines. However, it differs from other “robust” crania in having a more elongated base (IT-ITITP-TP)and a larger angle p. The cranial base pattern of TM 1517 also resembles those of KNM-ER 732 (a probable female of the “robust”taxon) and KNM-ER 1805. The calvarium KNM-ER 407 and the cranium KNM-ER 732 were recovered from the lower part of the Upper Member at Koobi Fora in North Kenya (Leakeyet al., 1978).Both specimens show marked post-orbital constriction, and have expanded and heavily pneumatized mastoid regions. The mandibular fossae are wide and extend laterally well beyond the sides of the cranial vault. Although the specimens are incomplete they have sufficient anatomical areas in common to suggest that they should be considered as belonging to the same taxon. A preliminary assessment of KNM-ER 407 lead to the conclusion that it “probably is either a gracile species of Australopithecus or else a very early representative of Homo” (Leakey, 1970).Since then Wolpoff (1978a)has espoused the former proposition, but the latter has received no support. From the outset, KNM-ER 732 was considered as a possible smaller-bodied female of Australopithecus (Leakey, 1971), and although these reports made no specific attribution, the clear inference was that the material should be referred to Australopithecus boisei (Leakey, 1970 and 1976). Since these preliminary pronouncements two substantial reviews of the fossil hominid evidence from Koobi Fora concurred with the judgement that KNM-ER 407 and 732 are probable females of Australopithecus boisei (Howell, 1978; Wood 1978). I t is clear from the cranial base diagrams (Fig. 5) and the results in Table 2, that neither KNM-407 or 732 have a basicranium which resembles that of the “gracile” australopithecines. The forward position of the foramen magnum in both crania is a feature which resembles the “robust” australopithecine group, and in view of the features of the frontal and malar region which KNM-ER 732 shares with specimens attributed to Australopithecus boisei (Wood,1978, in press), the most parsimonius taxonomic placement of these two specimens is within Australopithecus boisei. The cranium KNM-ER 1470 was recovered from the Lower Member of the Koobi Fora Formation and its age is between about 1.8 Myr. and 2.8-3.0 Myr. (Drake et al., 1980; Gleadow, 1980; McDougall et al., 1980), and palaeomagnetic evidence suggests that it may be close to 2.4 Myr old (Hillhouse et al., 1977). Initial assessment suggested that its affinities were with Homo (Leakey, 1973; Wood, 1976) but Walker argued for caution, and quite properly stressed the features, particularly of the face, which are shared between KNM-ER 1470 and Australopithecus (Walker, 1976). Subsequent reviews have favoured its inclusion in Homo habilis (Howell, 1978) or treated it as a conspecific of the “gracile” australopithecine material from Sterkfontein (Olson, 1978). Evidence derived from the cranial base of KNMER 1470 has to be interpreted in the light of the distortion which has been assessed by stereographic projection (Walker, 1981). This indicates that the right temporal has been pushed forwards, but not laterally. If allowance is made for these changes it gives no cause to modify the conclusion that the hominid group KNM-ER 1470 most closely resem- BASICRANIUM OF PLIO-PLEISTOCENE HOMINIDS 171 bles is Homo erectus. There are clear contrasts to Homo habilis, the pattern of basicranial between the cranial base of KNM-ER 1470 and morphology in KNM-ER 1813 is strongly in that of the “gracile”australopithecines, and the favour of its inclusion in Homo, and for it not breadth across the infratemporal crests to be regarded as a “gracile”australopithecine. distinguishes it from “robust” australopithe- The foreshortening of the skull base, widening cine crania. Although the overall morphology of the sphenoid, and the angulation of tymof the cranial vault and the face of KNM-ER panic and petrous parts of the temporal bone 1470 set it apart from crania and calottes at- are all clear and unambiguous evidence against tributed to Homo erectus, the form of its KNM-ER 1813 being placed within Australocranial base suggests that if it is assigned to pithecus africanus. Homo habilis then the cranial base of this taxThe cranium OH 24 was found at site DK in on had already developed some of the probable Bed I of Olduvai Gorge. The cranium was derived features we presently associate with found in a crushed state and was skillfully reHomo erectus and Homo sapiens. stored by Dr. R.J. Clarke. The initial descripThe skull KNM-ER 1805 was recovered from tion pointed to the similarities between OH 24 just below the Okote Tuff in the Upper Mem- and the Homo habilis skull, OH 13, from Bed ber at Koobi Fora. Radiometric and magneto- I1 (M.D. Leakey et al., 1971). Although some stratigraphic evidence all point to a date differences were noted between OH 24, and around 1.6 Myr (Fitch and Miller, 1976; Brock OH 7 and 16, the report concluded that “beand Isaac, 1976). Its relatively small cranial yond doubt, the new specimen represents the capacity, 582 m13, (Holloway, 1978), and the genus Homo as defined by Leakey, Tobias, and combination of a compound nuchal crest and Napier and differs fundamentally from the parasagittal crests have perhaps been features australopithecines.” Although its inclusion in which have led some workers to attribute it to Homo habilis has been supported by some Australopithecus boisei (Tobias, 1980). Pre- (Holloway, 1976, 1978; Tobias, 1980, 1981). sumbably different features prompted Howell many workers have been impressed by the sim(1978)and Wolpoff (1978b)to include ER 1805 ilarities between OH 24 and Australopithecus in Homo erectus, and other workers to place it africanus (Leakey, 1974; Howell, 1978; Olson, in a separate species of Australopithecus, 1978; Leakey and Walker, 1980). The cranial either named (Olson, 1978), or unnamed base of OH 24 is, however, unlike that of the (Holloway, 1976; Leakey 1976). The evidence “gracile”australopithecines. The petrous axes from the cranial base is compatible with the in- are more coronally aligned and the greater clusion of KNM-ER 1805 in either Australo- width across the infratemporal crests suggests pithecus boisei or a species within the genus that the sphenoid in OH 24 was broader than in Homo, but is strong evidence against its at- the Australopithecus africanus specimens. We tribution to Australopithecus africanus. are confident that the differences in cranial The precise stratigraphic relationships of base length between OH 24 and AustralopitheKNM-ER 1813 have yet to be worked out, but cus afn’canus are not simply the result of it is likely that it was buried in sediments in the distortion. Since the discovery that the parts of the cralower part of the Upper Member, or the upper part of the Lower Member, at Koobi Fora. Its nium SK 847 from Swartkrans fitted with a small brain size, 509 m13(Holloway, 1978) and left temporal (SK 846b) and a maxillary fragthe overall shape of the cranial vault and face ment (SK 80),the affinities of this “composite” have prompted many workers to suggest affin- cranium have been in dispute. The initial asessment of its affinities considered that it differed ities with, or its inclusion in, Australopithecus africanus (Holloway, 1978; Olson, 1978),while significantly from the “robust” australopithemore hominine aspects of relative tooth size, cines, and it was regarded as Homo sp. indet. dental morphology, and tooth wear have led (Clarke et al., 1970), and a similar taxonomic others to include it in Homo habilis (Howell, conclusion was reached after a much more de1978; Tobias, 1980).Many features of the mor- tailed analysis (Clarke, 1977). Howell (1978) phology of the cranial vault, face, and denti- has assigned it to Homo habilis, but Wolpoff has argued consistently against any taxonomtion convincingly preclude the inclusion of KNM-ER 1813 in Australopithecus boisei ic distinction between SK 847 and the “robust” (Wood,1978).Thus, with the taxonomic claims australopithecines. (Wolpoff, 1970, 1971, 1974). In his survey of nasal and mastoid morlimited to its inclusion in either Australopithecus africanus or Homo habilis the evidence phology Olson (1978)regarded the “composite” from the cranial base is crucial. If KNM-ER cranium as a conspecific of the “gracile” 1470 and OH 24 (uide infra) do prove to belong australopithecines from Sterkfontein. In this 172 M.C. DEAN AND B.A. WOOD present study the shape of the basicranium of SK 847 has been shown to be unlike that of the “gracile”australopithecines, and thus must be counted as evidence against them being regarded as conspecific. Although the overall pattern of basicranial morphology of SK 847 shows similarities to both Homo and the “robust” australopithecines, the relatively wide bi-infratemporal fossa breadth suggests interesting affinities with specimens which have been attributed to Homo. Although the cranial base of Taung is incomplete, any information it does contain is potentially important contributory evidence for any assessment of its affinities. In the wake of new evidence about the dating of Taung, Tobias suggested that serious consideration be given to the proposal that the Taung child may be a “late surviving member of A. robustus or A. cf robustus” (Tobias, 1973).Tobias (1981)has recently restated his views on Taung and has reemphasized his belief that Taung has closer affinities with the “robust” than with the “gracile”australopithecines. The assessment of the significance of differences between adult specimens is notoriously difficult. One of us (Dean, in preparation) has investigated growth changes in the basicrania of extant higher primates which have been aged using dental criteria (Dean and Wood, 1981b).These results suggest that in the adult equivalent of the Taung skull the petrous axes would have been aligned at an angle of approximately 60”. This is a much more sagittal orientation than is found in the “robust” australopithecines and is inferential morphological evidence in favour of regarding the Taung child as a juvenile representative of Australopithecus africanus as known from Sterkfontein and Makapansgat. A more general feature of these results is that, because of the similarities between the cranial base patterns of “robust”australopithecines and Homo, it would be unwise to try to distinguish between the two taxa solely on the evidence of the cranial base. We have discussed elsewhere the possibility of parallel evolution in Homo and the “robust” australopithecines (Dean and Wood, 1981a),and while many apparently derived features have been adduced in support of the concept of a ‘robust’ australopithecine clade (White, et al., 1981; Rak, 1981)we submit that these similarities in cranial base morphology between Homo and the “robust” taxa, together with evidence of shared patterns of tooth eruption (Clements and Zuckerman, 1953) and jaw growth (Skin- ner, 1978)should be given due consideration in any analysis of hominid phylogeny. The implications of the differences between the cranial base patterns of the “robust” and “gracile” australopithecines have been discussed elsewhere (Dean and Wood, 1981a)and confirm the preliminary observations of Tobias (1967) and Du Brul (1977). There are profound differences in basicranial morphology between Homo habilis and Homo erectus on the one hand, and Australopithecus africanus on the other. The reorientation and rearrangement of the cranial base which has taken place in these early Homo taxa suggests that the relationship between Homo habitis and Australopithecus africanus may not be as close as some workers have claimed (Robinson, 1972; Leakey, 1974; Olson, 1978; Leakey and Walker, 1980). The taxonomic and phylogenetic relationships of the small brained “gracile” crania from East Africa (e.g., KNM-ER 1813 and OH 24) remain an enigma, but the results presented here suggest that their inclusion in Australopithecus africanus may be premature. The basicranial morphology of hominids recovered from Laetoli and Hadar will be crucial evidence in any phylogenetic analysis, and more details of well-preserved cranial specimens are awaited with interest (Olson, 1981). We are conscious that the taxonomic and phylogenetic relationship of fossil hominids can be neither confirmed nor refuted on the basis of observations made on one region of the cranium. However, it is clear that one profitable way of analyzing the affinities of fossil groups is to accumulate evidence from a series of investigations dealing with a comprehensive set of anatomical regions or morphological variables, and we submit that this study is a contribution to such an analytical framework. ACKNOWLEDGMENTS M.C. Dean was in receipt of an NERC Research studentship and B.A. Wood was supported by a project grant from the NERC. We are grateful to the Trustees of the National Museum, Nairobi, the Transvaal Museum, Pretoria, the Tanzanian Government, and the Head of the Department of Anatomy, University of the Witwatersrand for allowing us access to material in their care. Our thanks go to C.K. Brain, Alan Hughes, M.D. Leakey, R.E. Leakey, P.V. Tobias, and Elizabeth Vrba for providing hospitality and for stimulating and productive discussion of many ideas included in this paper. BASICRANIUM OF PLIO-PLEISTOCENE HOMINIDS We thank Elizabeth Marshall for typing the manuscript, and we are indebted to several referees for their careful assessment of this paper. Some of their suggestions have been incorporated into the revised version of the manuscript. LITERATURE CITED Ashton, E H (1957)Agechanges in the basicranial axis of the Anthropoidea. Proc. 2001. SOC.Lond. 129:61-74. Ashton, E H and Spence,TF (1958)Age changes in the cranial capacity and foramen magnum of hominoids. Proc. Zool. SOC.Lond. 130:169-181. Ashton, E H and Zuckerman, S (1952)The base of the skull in immature hominoids. Am. J. Phys. Anthropol. 14:611-624. Biegert, J (1963) The evaluation of characteristics of the skull, hands, and feet for primate taxonomy. In SL Washburn (ed): Classification and human evolution. Chicago: Aldine. pp. 116-145. Bjork. A (1950)Some biological aspects of prognathism and occlusion of the teeth. Acta Odontol. Scand. 9:l-40. Brock, A and Isaac, GLI (1976)Reversal stratigraphy and its application at East Rudolf. In Y Coppens, F Clark Howell, GL1 Isaac, and REF Leakey (eds): Earliest man and environments in the Lake Rudolf Basin, Chicago: University of Chicago Press, pp. 148-162. Broom, R and Robinson. J T (1950)Further evidence of the structure of the Sterkfontein ape-man Plesianthropus. Part I. Trans. Mus. Mem. No. 4:ll-85. Clark, WE Le Gros (1959)The Antecedents of Man. An introduction to the evolution of primates. Edinburgh: Edinburgh University Press. Clarke, RJ (1977)The cranium of the Swartkrans hominid SK 847 and its relevance to human origins. Ph.D. Thesis Johannesburg: University of the Witwatersrand. Clarke, FkJ and Howell, FC (1972) Affinities of the Swartkrans 847 hominid cranium. Am. J. Phys. Anthropol. 37;319-336. Clarke, RJ, Howell, FC. and Brain, CK (1970) More evidence of an advanced hominid at Swartkrans. Nature 225: 12 19- 122 2. Clements, EMB and Zuckerman, S (1953)The order of eruption of the permanent teeth in the Hominoidea. Am. J. Phys. Anthropol. 11:313-332. Cramer, DL (19771 Craniofacial morphology of Pun paniscus. A morphometric andevolutionary appraisal. Basel: S Karger. Du Brul. EL (1977)Early hominid feeding mechanisms. Am. J. Phys. Anthropol. 47:305-320. Du Brul, EL and Laskin. DM (1961) Preadaptive potentialities of the mammalian skull: an experiment in growth and form. Am. J. Anat. 109:117-132. Dean, MC and Wood, BA (1981a) Metrical analysis of the basicranium of extant hominoids and Australopithecus. Am. J. Phys. Anthropol. 5453-71. Dean. MC and Wood, BA (1981h)Developing pongid dentition and its use for ageing individual crania in comparative cross-sectional growth studies. Folia primatol. 36: 111- 127. Drake, RE, Curtis, GH, Cerling. TE, Cerling. BW, and Hampel, J (1980)KBS Tuff dating and geochronology of tuffaceous sediments in the Koobi Fora and Shungura Formations, East Africa. Nature 283:368-372. Fitch, FJ and Miller, JA (19761 Conventional PotassiumArgon and Argon-40iArgon-39 dating of volcanic rocks from East Rudolf. In Y Coppens, F Clark Howell, GL1 Isaac, and REF Leakey (eds):Earliest man and environ- 173 ments in the Lake Rudolf Basin, Chicago: University of Chicago Press, pp. 123-147. Gleadow, AJW (1980)Fission track age of the KBS Tuff and associated hominid remains in northern Kenya. Nature 284:225-230. Gould, SJ (1978) Ontogeny and phylogeny. Cambridge: Belknap. Hillhouse, JW. Ndombi, JWM, Cox, A, and Brock, A (1977) Additional results on palaeomagnetic stratigraphy of the Koobi Fora Formation, east of Lake Turkana (Lake Rudolf), Kenya. Nature 265:411-415. Holloway, RL (1976)Some problems of hominid brain endocast reconstruction, allometry and neural organization. In PV Tobias and Y Coppens (eds):Les Plus Anciens Hominidks, Paris: CNRS, pp. 69-119. Holloway, RL (1978)Problems of brain endocast interpretation and African hominid evolution. In CJ Jolly (ed):Early hominids of Africa, London: Duckworth, pp. 379-401. Howell, FC (1978)Hominidae. In VJ Maglio and HBS Cooke (eds): Evolution of African Mammals, Cambridge: Harvard University Press, pp. 154-248. Huxley, TH (1863) Evidence a s to man’s place in nature. London: Williams and Norgate. Huxley, TH (1867) On two widely contrasted forms of the human cranium. 3. Anat. Physiol. 1:60-77. Laitman, J T (1977)The ontogenetic and phylogenetic development of the upper respiratory system and basicranium in man. Ph.D. Dissertation, Yale University. Laitman, JT, Heimbuch RC, and Crelin, ES (1978)Developmental change in a basicranial line and its relationship to the upper respiratory system in living primates. Am. J. Anat. 152:467-483. Laitman, J T , Heimbuch, RC, and Crelin, ES (1979)The basicranium of fossil hominids as an indicator of their upper respiratory systems. Am. J. Phys. Anthropol. 5115-33. Latham, RA (1972)The sella point and postnatal growth of the human cranial base. Am. J. Orthod. 61:156-162. Leakey, MD. Clarke, RJ, and Leakey, LSB (1971)New hominid skull from Bed I, Olduvai Gorge, Tanzania. Nature 232308-312. Leakey, REF (1970)Fauna and artefacts from a new PlioPleistocene locality near Lake Rudolf in Kenya. Nature 226:223-224. Leakey, REF (1971) Further evidence of Lower Pleistocene hominids from East Rudolf, North Kenya. Nature 231:241-245. Leakey, REF (1973)Evidence for an Advanced Plio-Pleistocene hominid from East Rudolf, Kenya. Nature 242447-450. Leakey, REF (1974) Further evidence of Lower Pleistocene hominids from East Rudolf, North Kenya, 1973. Nature 2483353-656. Leakey, REF (1976) An overview of the Hominidae from East Rudolf, Kenya. In Y Coppens, F Clark Howell, GLI Isaac and REF Leakey (eds): Earliest man and environments in the Lake Rudolf Basin, Chicago: University of Chicago Press, pp. 490-506. Leakey. RE, Leakey, MG, and Behrensmeyer. AK (1978) The Hominid Catalogue. I n MG Leakey and RE Leakey (eds):Koohi Fora Research Project, Vol. 1: The fossil hominids and an introduction to their context, 1968-1974. Oxford: Clarendon Press, pp, 86-182. Leakey, REF and Walker A (1980)On the status ofAustrulopithecus afarensis. Science 207;1103. Lestrel. PE and Moore, RN (1978) The cranial base in fetal Macaca nemestrina: A quantitative analysis of size and shape. J. Dent. Res. 57:395-401. McDougall, I. Maier. R, Sutherland-Hawkes. P, and Gleadow, AJW (1980) K-Ar age estimate for the KBS Tuff, East Turkana, Kenya. Nature 284: 230-234. Moss, ML (1958)The pathogenesis of artificial cranial defor- 174 M.C. DEAN AND B.A. WOOD mation. Am. J. Phys. Anthropol. 16:269-286. Olson, TR (1978) Hominid phylogenetics and the existence of Homo in Member 1 of the Swartkrans Formation, South Africa. J. Hum. Evol. 7:159-178. Olson, TR (1981)Basicranial morphology of the extant hominoids and Pliocene hominids: the new material from the Hadar Formation, Ethiopia, and its significance in early human evolution and taxonomy. In CB Stringer (ed):Aspects of human evolution. London: Taylor and Francis, pp. 99-128. Rak. R (1981)The morphology and architecture of the australopithecine face. Ph.D. Dissertation. University of California, Berkeley. Robinson, J T (1972)The bearing of East Rudolf fossils on early hominid systematics. Nature 240:239-240. Schepers, GWH (1950) The brain casts of the recently discovered Plesianthropus skulls. Part 11. Trans. Mus. Mem. No. 4: 87-117. Schultz, AH (1955) The position of the occipital condyles and of the face relative to the skull base in primates. Am. J. Phys. Anthropol. 13:97-120. Scott, J H (1958)The cranial base. Am. J. Phys. Anthropol. 16:319-348. Scott, J H (1963) Factors determining skull form in primates. In J.R. Napier and N.A. Barnicot (eds)Symp. Zool. Sac. Land. 10:127-134. Sirianni. J E and Van Ness, AL (1978) Postnatal growth of the cranial base in Macaca nemestrina. Am. J. Phys. Anthropol. 49:329-340. Skinner, MA (1978)Dental maturation, dental attrition, and g o w t h of the skull in fossil hominids. Ph.D. Thesis Cambridge: University of Cambridge. Tobias, PV (1967) Olduvai Gorge, Volume 2. The cranium and maxillary dentition of Austrulopithecus (Zinjanthropusl boisei. Cambridge: Cambridge University Press. Tobias, PV (1973) Implications of the new age estimates of the early South African hominids. Nature 246:79-83. Tobias, PV (1980) The natural history of the helicoidal occlusal plane and its evolution in early Homo. Am. J. Phys. Anthropol. 53:173-187. Tobias, PV (1981) A survey and synthesis of the African hominids of the late Tertiary and early Quaternary periods. In L-K Konigsson (ed):Current Argument on Early Man. Oxford Pergamon. pp. 86-113. Walker, A (1976) Remains attributable to Austrulopithecus in the East Rudolf succession. In Y Coppens, F Clark Howell, GL1 Isaac and REF Leakey (eds): Earliest man and environments in the Lake Rudolf Basin, Chicago: University of Chicago Press, pp. 484-489. Walker A (1981)The Koobi Fora hominids and their bearing on the origins of the genus Homo. In BA Sigmon and J S Cyhulski (eds): Homo erectus. Papers in Honor of Davidson Black. Toronto: University of Toronto Press, pp. 193-215. Weidenreich, F (1947)Some particulars of skull and brain of early hominids and their bearing on the problem of the relationships between man and anthropoids. Am. J. Phys. Anthropol. 5:387-428. White, TD, Johanson, DC, and Kimbel, WH (1981)Austrulopithecus ufricanus: its phyletic position reconsidered. S. Afr. J. Sci. 77:445-470. Wolpoff, MH (1970)The evidence for multiple hominid taxa at Swartkrans. Am. Anthrop. 72:576-607. Wolpoff, MH (1971) Is the new composite cranium from Swartkrans a small robust australopithecine? Nature 230: 398-40 1. Wolpoff. MH (1974)The evidence for two australopithecine lineages in South Africa. Yb. Phys. Anthropol. 17: 113-139. Wolpoff. MH (1978a) Analogies and interpretation in palaeoanthropology. In CJ Jolly (ed): Early hominids of Africa. London: Duckworth, pp. 461-503. Wolpoff, MH (1978h)Some aspects of canine size in the australopithecines. J. Hum. Evol. 7:115-126. Wood, BA (1976)Remains attributable to Homo in the East Rudolf succession. In Y Coppens, F Clark Howell, GLI Isaac and REF Leakey (eds): Earliest man and environments in the Lake Rudolf Basin, Chicago: University of Chicago Press, pp. 490-506. Wood, BA (1978) Classification and phylogeny of East African hominids. In DJ Chivers and KA Joysey (eds): Recent Advances in Primatology, Val. 3, London. Academi, Press, pp. 351-372.