Endocranial form of gorilla skulls. With special reference to the existence of dolichocephaly as a normal feature of certain primatesкод для вставкиСкачать
ENDOCRANIAL FORM OF GORILLA SKULLS WITHSPECIAL REFERENCE TO THE EXISTENCE OF DOLICHOCEPHALY AS A NORMAL FEATURE OF CERTAIN PRIMATES H. A. HARRIS Instgtute of Anatomy, University College, London; and Washington University, Saint Louis INTRODUCTORY REMARKS The recent discovery of a fossilized skull at Taungs in South Africa was an event of great importance. Professor Dart of the Witwatersrand University (Johannesburg)-our erstwhile colleague at Saint Louis and a t London-described the specimen as a member of an ancient group, intermediate between living anthropoids and man, and gave it the name of Australopithecus africanus. British anatomists, prominent among them Sir Arthur Keith, Professor Elliot Smith and Professor Robinson, heralded the importance of the discovery and differed considerably as regards the true morphological significance thereof. Dart (1),in describing Australopithecus, says that “the whole cranium displays humanoid rather than anthropoid lineaments. It is markedly doZichocephaZic and Zeptoprosopic.” Thus the longheadedness, together with the long narrow face is regarded as a human attribute. Keith (2) lays great stress on this character and says: “Even if it be admitted that the Australopithecus is an anthropoid ape, it is a very remarkable one. It is a true longheaded or dolichocephalic anthropoid-the first so f a r known.” Robinson (3), obviously impressed by the dolichocephalic character of the specimen is reported as follows : “The Taungs specimen was the distorted skull of a chimpanzee just over four years old, probably a female.” Elliot Smith (4) in a lecture at University College said “Even if it be admitted, which I am not prepared to do, that some flattening of the brain case occurred after death, the absence of salient eyebrow ridges, the smallness of the jaws, and especially the reduction of the prominence of the snout, confirm the claim that, although Australopithecus is an ape nearly akin to the gorilla and chimpanzee, it does reveal a definite though slight indication of that refinement of the features which represents the early stage in the process of the assumption of distinctively human characters.”] With characteristic relevance, Bolk ( 5 ) immediately published a description of a strange gorilla skull from the collection in the Anatomical ’For latest and original study of the specimen, with new measurements and conclusions, see HrdliEka (A.), The Taungs Ape. Am. J . Phys. Anthrop., 1925, VIII, 0ct.-Dec. No., 379-392. AM. J. PHYS.ANTHROP..1926,Val. IX,No.2. 157 H. A. HARRIS 158 Museum of the University of Amsterdam. Out of fifty skulls, this one alone displayed marked dolichocephaly and leptoprosopy. The description of this single specimen of a longheaded anthropoid was particularly suggestive, in view of the previous controversy as to the value of dolichocephaly in classifying the Taungs skull. At this time, in company with Mr. J. Thornton Carter, I had the privilege of visiting the Natural History Museum of Lord Rothschild at Tring Park. The whole problem of the value of external and craniometric characters in determining the range of variation on the one hand, and in determining the classification of species on the other, has been discussed by Lord Rothschild (6). In particular the variations in craniometric characters were so lucidly and convincingly put forward by him that, a t his instigation, facilities were made for the staff of the Institute of Anatomy to make a survey of the anthropoids in the Rothschild collection. The collection of gorilla skulls was moved to the Institute of Anatomy, and under the guidance of Professor Elliot Smith, a survey was commenced. MATERIAL AND METHODS The material consists of 45 gorilla skeletons from the Rothschild Collection, numbered in series from A. D. 1 to A. D. 45, of which 12 are female and 33 male; with these are considered four skulls of the Anatomical Museum of University College, of which C. A. l , 2 and 4 are male, and C. A. 5 is female. Thus the total represents 49 skulls, 36 male and 13 female. The only skull which is not completely adult is that of a young male, A. D. 2, in which the third molars have not erupted and the occipital and sagittal crests are not well developed. (A) CRANIAL CAPACITY Wingate Todd (7) says that "there is no method of obtaining cranial capacity with absolute accuracy and there is no method of approximation which is not open to serious objection." His learned analyses of the various methods described by Martin, Pearson, Macdonnell and HrdliEka and the severe checking of his own methods show that the error may amount to 3%. In view of this fact the white mustard seed method (HrdliEka) was employed, filling the skull by Flower's method and emptying the seed into a 1000 ccs. cylinder by means of a large funnel the stem of which was cut short, the bore at the exit being 2 cms. Each reading was repeated five times and the average was regarded as the cranial capacity. (B) FACIAL INDEX The facial index is usually measured in terms of the ratio of the nasoalveolar distance to the maximum bizygomatic width. This method ENDOCRANIAL FORM OF GORILLA SKULLS 159 was extensively employed as the “upper facial index of Kollmann.” It is unsuitable for anthropoids as the nasion is not easily determined owing to the variable development of the supraorbital crest and the early union of the nasals and frontal a t the fronto-nasal suture. Accordingly, the method of Bolk was pursued. The greatest breadth is measured between the right and left points at which the zygomatic arch bends itself round into the lateral orbital margin. The greatest length is measured in profile from the highest point in the mid-line of the crista supraorbitalis to the anterior margin of the premaxilla between the central incisors. The ratio of this breadth to the length, multiplied by one hundred, will be referred to as the Facial Index of Bolk. Where the Facial Index of Bolk is 100 or more, the face is classed as chamaeprosopic or broad and where the index is less than 100, the face is classed as leptoprosopic or long. The method of taking the measurements is depicted by the arrows in Fig. I (A) to (D), which is after Bolk (op. cit.). (C) CEPHALIC INDEX. THE RADIOGRAPHIC METHOD No satisfactory craniometric method for the skulls of anthropoids has yet been obtained. The only method which would permit of true comparison with man would necessitate the comparison of endocranial dimensions. This necessitates section of the skull in the mid-line, an act which destroys a band of bone from 1.5 to 2 mms. thick as a result of the spread of the teeth of the saw. All external measurements of length in the anthropoid skull are vitiated by the strongly developed crista occipitalis and, in the gorilla, the more strongly developed crista supraorbitalis. A fact which has not been so clearly recognized is the still greater error in determining the breadth of the brain case. The parietal eminences of the anthropoids are not well developed : they do not overlie the point of maximum breadth: the mastoid air sinuses in the male gorilla extend into the squama of the temporal bone superior and even anterior to the external auditory meatus and may invade the lower margin of the parietal bone by transgressing the squamo-parietal suture and so reach even to the parietal eminence. The point of maximum breadth sometimes lies as much as 4 cms. anterior to the external auditory meatus. Keith (8) says: “The present manner of description by angles and indices is a method that leads only to the accumulation of a mass of most useless, cumbersome material. The describers seem to have lost all sight of the skull as a functional organ, with its form adapted for its two main uses, as a brain cover and a tooth carrier.” Bolk (loc. cit.) FIG.1.-After B D Bolk. The arrows indicate the method of measuring the facial length and breadth in the short-faced (A) and (c) and long-faced skulls (B) and (D). P P ENDOCRANIAL FORM OF GORILLA SKULLS 161 says that in studying the gorilla, “the terms dolichocephalic and brachycephalic are not truly applicable in that the index cephalicus is not taken in the same way from the full grown Gorilla as from man. Keith (Q), in his determination of the average length of 10 gorilla skulls, excluded from the length the prominences due to the frontal air sinuses and the external occipital protuberance. But he measured the breadth from one parietal eminence to the other. Such a measurement of breadth is open to question, as the parietal eminence may be obscured in the adult male gorilla by the great development of the air sinuses, and the point of maximum endocranial breadth may lie as much as 4 cns. anterior to the external auditory meatus a t a considerable distance from the poorly developed parietal eminence. Bolk accordingly obtained what he calls an Index Encephalicus for the length-breadth relation of the cranial cavity. He made a sagittal section of the skull and obtained the length and breadth therefrom. The Fronton of Bolk is the point where the frontal wall and the base of the skull meet anteriorlyinthe middle-line. The Occipiton is the point on the occipital surface of the endocranial wall furthest from the Fronton. This is the greatest length of the cranial cavity. The greatest breadth is found by adding together the greatest depth of the cranial cavity of the right and left halves. This is a troublesome method, as in the first place a band of bone is destroyed by the act of sawing, and secondly the maximum breadth of each half has to be determined by a laborious method of trial over a considerable area by means of a depth gauge or other comparable arrangement. Accordingly a new method was devised for the breadthlength relation of the skull. The largest gorilla skull is mounted upside down in the Frankfort plane, with cylindrical metal pointers in the right and left external auditory meatus and a third pointer supporting a t its tip the left infraorbital margin. The third pointer is not quite in the same horizontal plane as the other two but lies a t a distance of 2 mms. below, so as to have its sharp tip a t the same level as the lower margin of the two cylindrical pointers which are thrust in to the right and left external auditory meatus. By this manoeuvre it is possible to maintain the true superior margin of the left and right external auditory meatus a t the same level as the left infraorbital margin, so guaranteeing that the skull is in the Frankfurt Plane. The level of these pointers is maintained constant, but the stands supporting them can be moved to and fro on the horizontal table. The largest male skull is mounted in this manner so that the sagittal crest just clears the table. 162 H. A. HARRIS A radiographic filmis placed on the horizontal table beneath the skull; another film is mounted in a vertical plane parallel to the sagittal plane of the skull at such a distance that the distance from the Frankfurt plane to the horizontal film on the table is equal to the distance from the sagittal axis of the skull to the vertical film. Two X-ray tubes are mounted so that the upper tube is centered over the basion which is marked by a small arrowhead of lead foil, and the lateral one is centred over the external auditory meatus. The tube distance from the film in each case is constant so that the radiograms of the norma verticalis and norma lateralis are of equal magnification. Lateral and vertical radiograms of all the skulls were thus taken, so that two series of radiograms were obtained of constant and equal magnification. Two of the skulls, C. A. 1 and C. A. 4 were now sectioned in the sagittal plane by a thin band saw. The actual maximum length and breadth was determined after the method of Bolk. Furthermore, lateral radiograms of the half skull were taken by placing the half skull with its sagittally sectioned surface in contact with the film,so that the magnification of the image was unity. Thus three standards are obtained for determining the magnification of distances in the Frankfurt plane and in the sagittal plane on the two series of radiograms. The magnification can be calculated : (1) By actual measurement in terms of the distance of the Frankfurt plane and sagittal plane respectively from the x-ray tube and from the photographic film. (2) By actual measurement of the two sectioned skulls. (3) By actual measurement on the radiograms of the half skulls of C. A. 1 and C. A. 4 in which the sagittal surface is in contact with the photographic film. All three methods showed the magnification of the radiograms to be approximately 1.1. Thus distances in the two standard planes can be reduced to actual distances by dividing such distances on the radiogram by the common magnification factor of 1.1. It is important to stress the fact that this magnification factor can only be employed for distances which lie in the Frankfurt plane or in the sagittal plane in the radiograms of the nonna verticalis and norma lateralis respectively. First of all consider the problem of the maximum breadth of the skull. In Fig. 2 the Frankfurt plane is F P and the plane of the image on the photographic film is I M. The magnification of all distances in the plane F P on the image I M is 1.1. But the actual maximum breadth of the skull lies in some position such as B H and the FIG,2. Gorilla skull in position to show magnification of maximum breadth BH at BIH' in the plane of the film irnaEe IM. FP is the Frankfurt plane. FIG. 3. Brachycephalic Gorilla skull in posiiton to show magnificationof maximum length LH at L W in the alane of the film image IM. FP is the Frankfurt H. A. HARRIS 16# magnification of this line on the plane of the image I M will be less than 1.1, in proportion as B H recedes from F P and approaches I M. The actual apparent maximum endocranial breadth on the radiogram was measured and divided by 1.1. This gives a reading for the actual breadth which is too small, as the magnification of B H on an average is in the neighbourhood of 1.07 to 1.08. The apparent maximum breadth was measured on the series of radiograms of the norma verticalis and in each case this dimension was divided by 1.1, so giving a series of values for the breadth of the skull which were uniformly high. As regards the maximum length of the skull, in addition to the factor considered above for breadth, there is another factor due to the obliquity between the actual line of maximum length and the Frankfurt plane. In Fig. 3, F Pis the Frankfurt plane, I M is the plane of the image and L H is the actual maximum length in the sagittal plane. The magnification of all distances in the plane F P on the image I M is 1.1. The magnification of L H in the plane I M is less than 1.1, according as L H recedes from F P and ranges between 1.08 and 1.09. Moreover the projection of L H on the plane I M is decreased proportionally to the angle of obliquity of L H to F P. In the case of a longheaded skull (Fig. 4) this obliquity is usually less than 10" and in the case of a broad headed skull the obliquity is almost nil. The obliquity of L H to F P and to I M will thus cause the length to be shortened on the image by a factor ranging from cosine 10" to cosine 0" i.e. from 0.98 to 1.00. Accordingly, in view of the relative insignificance of this small obliquity, the apparent maximum length as measured on the series of radiograms of the norma verticalis has been divided in each case by the standard factor of magnification 1.1, in order to obtain the endocranial length of the skulls. Both length and breadth have thus been calculated by dividing the maximum length and breadth on the series of radiograms of the norma verticalis by 1.1 and the index is called the "Radiographic Encephalic Index." The errors of the method due to the position of the lines of maximum length and breadth have been shown to be within 27,. It is hoped by further development of this technique to obtain an absolute measurement for the actual maximum endocranial length and breadth. (D) EXOCRANIAL CEPHALIC INDEX For the purposes of comparison a series of measurements were made on the external aspect of the skulls, the length being measured ad maximum from the external occipital protuberance to the supraorbital crest, and the width from the upper margin of one external 166 ENDOCRANIAL FORM OF GORILLA SKULLS auditory meatus to that of the other. This is called the “Exocranial Cephalic Index.” Such an index wiII show a very wide range because of the marked variation in the development of the crests and air sinuses of the skull. GORILLA SKULLS OF ROTHSCHILD COLLECTION No. Sex 13 2 4 A. D. 2 A. D. 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 C.A. 44 45 C. A. 5 9 A. D. 1 3 4 5 6 7 8 9 10 11 12 13 Cranial capacity In ccs. 560 652 510 490 550 560 547 460 500 495 460 470 ? 490 ? 510 ? ? ? ? ? 550 ? 580 440 465 480 ? 520 ? 560 455 ? ? 480 ? 475 460 415 450 425 430 ? 420 450 442 457 ? 440 Facial index of Bolk 110.9 129.2 111.0 100.7 113.7 97.5 111.6 111.3 100.0 111.8 112.5 109.1 113.6 113.5 120.8 108.9 110.3 113.6 101.4 110.4 97.7 96.3 114.4 113.6 111.1 108.1 104.1 106.1 97.9 92.9 125.3 105.0 97.5 96.8 110.5 ? 108.7 96.2 110.6 114.0 110.4 110.9 106.5 102.4 106.0 98.5 110.9 107.5 105.8 “Radiographic” “Exocranial” Encephalic index, cephalic index endocranial index 84.7 80.5 84.7 85.8 74.4 78.2 76.9 80.3 80.3 80.8 76.8 77.2 75.2 73-2 76.6 72.1 77.2 81.1 77.2 79.7 79.2 78.7 74.0 74.4 76.6 74.2 79.5 80.1 76.0 76.0 72.7 82.5 81.4 77.2 74.2 75.0 82.6 85.2 80.0 78.3 79.1 84.2 84.8 86.8 82.8 84.9 82.8 79.7 80.7 87.2 70.9 84.7 78.8 67.6 74.6 78.5 73.9 72.6 66.5 77.3 77.5 78.6 70.1 78.0 72.9 83.0 76.4 75.1 70.4 65.6 66.7 69.3 78.2 68.4 77.8 68.6 84.7 75.6 67.0 72.4 64.9 76.0 76.9 60.4 82.9 79.4 77.1 75.5 75.5 76.4 77.1 75.1 79.5 76.8 76.5 ? 81.0 Locality Gaboon Gaboon French Congo French Congo Cameroons Gaboon S. Cameroons S. Cameroons Nguni River Nguni River Cameroons Gaboon French Congo Cameroons Cameroons Cameroons S . Cameroons Gaboon Cameroons Bonila R. Gaboon Nguni River Gaboon Cameioons Gaboon Bonila R. French Congo Gaboon Gaboon Cameroons H. A. HARRIS 166 (A) CRANIAL CAPACITY Collection Rothschild (H. A. H.) Total No. of skulls 34 23 11 29 5 27 7 Bolk(5) Martin(l0) Hr&Eka( 11) Keith(l2) 39 27 12 ? ? 7 ? 13 7 T* M. & F. M. F. Chamaeprosopic M. & F. Leptoprosopic M. & F. Brachycephalic (Radio. endocranial) M. & F. Dolichocephalic (Radio. endocranial) M. M. & F. M. F. M. F. M. F. M. F. AVWW Max. Min. 490ccs. 513 442 487 652ccs. 652 475 652 415ccs. 440 415 415 506 560 442 482 652 415 519 580 465 528 550 478 508 435 ? ? 497 450 655 655 595 585 550 390 450 390 623 580 573 496 420 370 420 370 425 393 Notes:-The greater the number of gorillas examined, the greater does the average cranial capacity tend to become. Whereas Keith gives the average capacity as 470 ccs. with a maximum of 620 ccs. Hagedoorn (13) says that this average is far too small as it is already surpassed by that of certain female skulls. The Rothschild skulls give an average of 490 ccs. for both sexes; 512 ccs. for the male and 442 ccs. for the female. This is not as large as the average given by Bolk i.e. 528 ccs. for both sexes, 550 ccs. for males and 478 ccs. for females. HrdliEka discusses the ratio of the female cranial capacity to the male capacity as an index in terms of 100, and quotes Oppenheim (14) to the effect that the value is 85. Bolk’s figures give a value of 86.9, Keith’s 90.5, and the present collection 86.3, a value which is very close to that of Bolk. The average cranial capacity of the nine leptoprosopic skulls is 506, with a range from 442 to 560 ccs. The average of the eight dolichocephalic (radiographically) skulls is 519, ranging from 465 to 580 ccs. Bolk’s single specimen falls within these limits with a capacity of 550 ccs. ENDOCRANIAL FORM OF GORILLA SKULLS 167 (B). FACIAL INDEX OF BOLK. Collection Rothschild (H. A. H.) Total No. of skull6 48 35 13 39 9 28 7 11 Max. Mis. 107.9 108.3 106.8 110.4 129.2 129.2 114.0 129.2 92.9 92.99 96.2 100.0 96.8 98.5 92.9 Chamaepros M. - -. Leptopros. 111.1 129.2 100.0 96.7 97.9 92.9 Chamaepros. 108.5 114.0 102.4 97.3 98.5 96.2 111.5 122.6 103.4 Type Average M.& F. M. F. Chamaepros. M. & F. Leptopros. M. & F. M. F. 2 Bolk 49 1 Leptopros. F. Chamaepros. M. & F. Leptopros. 89.5 M. I H' FIG.4. Dolichocephalicgorilla skull in position to show magnification of maximum length LH at LlH' in the plane of the film image IM. The maximum length LH makes an angle with the FP and IM. H. A. HARRIS 168 Notes:-Leptoprosopy characterizes nine of the gorillas examined. The average value of the facial index in the nine leptoprosopic skulls is 96.8, ranging from 98.5 to 92.9. Not one of the skulls was so markedly leptoprosopic as Bob’s sole specimen which had a value of 89.5. Of the chamaeprosopic skulls 15 range from 100 to 1U9.9, 21 range from 110 to 119.9 and three have values above 120. The most markedly chamaeprosopic skull, (C. A. 2 ) is also that male skull which has the largest cranial capacity, with marked development of the crests and air sinuses. Two of the female skulls are leptoprosopic. (C). RADIOGRAPHIC ENDOCRANIAL INDEX, OR ENCEPHALIC INDEX. Collection Rothschild (H. A. H.) Total No. of skulls 49 36 13 41 8 40 9 Bolk Keith Keith (2) HrdliEka(l6) 10 ? 1 1 M.& F. M. F. Brachyceph. Dolichoceph. Chamaepros. Leptopros. Brachyceph. Brachyceph. Taungs Skull Taungs Skull Average Max. Min. 79.1 77.9 82.4 80.2 73.6 79.0 79.6 86.8 85.8 86.8 86.8 74.4 86.8 85.2 85.9 80.0 71.0 74.0 ? 72.I 72.1 78.3 75.0 72.1 72.1 76.0 80.6 ? ? Discussion:-The radiographic method of estimating the endocranial index shows that eight of the skulls are dolichocephalic. On the average, the male skulls display a much wider range of variation than the female. The female skulls range from 78.3 to 86.8 with an average of 82.4, whereas the male range from 72.1 to 85.8 with an average of 77.9. Thus the dolichocephalic character is more marked in the male, in terms of endocranial measurement. All eight dolichocephalic skulls were male. The thirteen female skulls were consistently brachycephalic. This brachycephalic character of the female gorilla skull is comparable to what is found in the female human slcull. Yet, two of the female skulls are leptoprosopic. The average endocranial index of the chamaeprosopic and leptoprosopic groups are very close together, 79 and 79.6 respectively. Although Bolk’s specimen was both leptoprosopic and dolichocephalic, it should be noted that there is no close correlation between the leptoprosopic and dolichocephalic characters in members of the Rothschild series. Not one of the nine leptoprosopic skulls is dolichocephalic as they have endocranial indices which range from 76 to 85.2, With an average of 79.6. Conversely, not one of the eight dolichocephalic skulls is leptoprosopic, the eight dolichocephalic skulls having facial indices which range from 108.1 to 125.3 with an average of 113.5. Hrdli&a ENDOCRANIAL FORM OF GORILLA SKULLS 169 (15), in a comparison of the facial with the cephalic index in a series of orang skulls, pointed out the lack of correspondence between these two indices, and suggested that “the facial growth is apparently controlled, unlike in man, much more by the development of the teeth and facial muscles than by that of the cranial vault.” The absence of close correspondence between the facial indices and endocranial indices in the gorilla demonstrates that the development of the skull and the development of the face cannot be too closely related in terms of “brain cover” and “tooth carrier.’’ The preliminary estimates of the cephalic index of Australopithecus have varied from 71 by Keith (2) to 74 by HrdliEka (16). These figures lie between the limits of the dolichocephalic gorillas of the Rothschild Collection and between the h i t s of dolichocephalic man. Too much importance should not be attached to the cephalic index in interpreting a given anthropoid or human specimen. Bunak (17), in a recent study of the lambdoid and sagittal crests of anthropoids has shown that there is a close, but not a full correspondence between the extent of the temporal muscles, the size of the canines and the development of the lower jaw. There is, apart from these factors, what he calls a “crest factor” which is a systemic character of certain species and sub-species of Primates. It is this same “crest factor” which Lord Rothschild has employed in the separation of the species and sub-species of gorilla. Similarly with regard to supra-orbital crests, the facts have been clearly stated by Elliot Smith (18) : “The development of the eye-brow ridges is not of much importance as an index of race. It is neither an exclusively primitive character nor a distinction of higher race. It is found developed in Pithecanthropus, Rhodesian Man, Neanderthal Man, and in the Australian and Alpine Races, whereas a defective development of the ridge is characteristic of Eoanthropus, the Negro, the Mongol and the Mediterranean Races, while the Nordic Race occupies a position between the two.” As for the skull crests so for endocranial skull form; there is not only a “muscle” factor, a “dentition” factor a “face” factor and a “brain” factor, but there is also a “skull” factor which is at present too complex to be analysed in terms of race alone. In Fig (5) are shown the radiograms of the norma verticalis of seven of the gorilla skulls selected a t random on the basis of the differences in skull form pointed out by Lord Rothschild. The range of the endocranial or encephalic index obtained by the radiographic method is from 72.1 t o 86.8, the upper row being brachycephalic; and the lower row dolichocephalic. TOPROW. A. D. 40=72.7. A. D. I4=74.4. A. D.25=72.1. Botlom Row. A. D.2=80.5 A. D. I =85.2. A. D. 8=86.8. A. D. 26=76.6. ences in skull form described by Lord Rothschild. The endocranial cephalic indices are:- FIG.5. Radiograms of the norma verticalis of seven gorilla skulls selected at random by sight to show differ- ENDOCRANIAL FORM OF GORILLA SKULLS 171 (D). EXOCBANIAL CEPHALIC INDEX. Collection Rothschild (H.A. H.) Total No. of skulls 48 36 12 30 18 Average Max. Min. M.& F. M. F. 74.9 73.9 77.7 87.2 87.2 82.9 60.4 60.4 75.1 Brachyceph. (exocranial) Dolichocmh. (exocraHa1) 78.3 87.2 60.4 69.2 74.6 60.4 Type The wide variation in the exocranial cephalic index is due to the great variation in the development of the crests and air sinuses in the male skull. Here again the female shows but a small degree of variation, with a range from 75.1 to 82.9 and an average of 77.7 whereas the male ranges from 60.4 to 87.2 with an average of 74.9. SUMMARY 1. A method is devised for rapidly obtaining by radiographic means a breadth-length ratio for the endocranial aspect of the intact skull. 2. Dolichocephaly is demonstrated in eight gorilla skulls in a series of 49 and Ieptoprosopy is also demonstrated in nine of the skulls. In no case was a given skull both leptoprosopic and dolichocephalic, so that the correlation between skull form and face form is not very marked within the series. 3. Dolichocephaly, per se, cannot be employed as a character of any importance in deciding whether a given fossil skull possesses humanoid characters. 4. The range of cephalic index in this series is comparable to the range shown by man, from 72 to 87, both in the New World and the Old. CONCLUDING REMARKS Further attempts are being made to devise a speedy method of obtaining the maximum endocranial length and breadth of the skulls of anthropoids by means of the use of the fluoroscopic screen and internal callipers. Our thanks are due to Lord Rothschild for placing this unique collection of gorilla skulls at our disposal and it is proposed t o study the question of geographical distribution of species and sub-species when an opportunity is found to examine the records concerning location of the specimens in detail. Mr. Melville and Mr. Farey, technical assistants a t University College, have devoted much time and thought to the perfecting of the method. My colleagues at Washington University have also been most ready to assist. 1’7e H. A. HARRIS LITERATURE Dart, R. A. Australopithecus Africanus: The Man Ape of South Africa. Nature, Feb. 7 , 1925, 195-199. Keith, Sir Arthur. The Fossil Anthropoid Ape from Taungs. Nature, Feb. 14, 1925, 234. Robinson, A. The Taungs Skull. Br. M d . J., March 25, 1925, 622. Elliot Smith, G. Lecture delivered at University College, London, May 22, 1925, and I l l w t r a k d London News, Feb. 14, 1925. Bolk, L. “On the existence of a dolichocephalic race of Gorilla.” Kon. AR. Wet. Amsterdam, 1925, XXVIII, 204-213. Rothschild, Lord. Notes on Anthropoid Apes. Proc. ZooE. SOC.LOW‘.,1904, 413440. Further Notes on Anthropoid Apes. Proc. ZooE. SOG.Lond., 1906, 465468. Todd, Wingate & Kuenzel, W. The Estimation of Cranial Capacity. A m . J . Phys. A t ~ t l r ~ p1925, . , VIII, 251-260. Keith, A. An Introduction to the Study of the Anthropoid Apes. Natural Science Reprints, London, 1897, 5. Keith. A. On the ChimDanzees and their relationship to the Gorilla. Proc. Zoob. Soc. London, Makh 7, 1899, 311. Martin, R. Lehrbuch der Anthropologie. Jena. 1914, 652. HrdliEka, A. Weight of the Brain and of the Internal Organs in American Monkeys, with data on Brain Weight in other Apes. Am. J . Phys. Anthrop. 1925, VIII, 201-226. Keith, A. The Growth of Brain in Men and Monkeys. J. A m t . Er Phys., 1895, XXIX, 282-301. Hagedoorn, A. Schadelcapaciteit van Anthropomorphen. Ned. Tijdschr. v. Geneesek., 1923. Oppenheim. Zur Typologie der Primaten Craniums. 2.Morph. 6 Anthrop., 1911, xrv. HrdliEka, A. Anatomical Observations on a Collection of Orang Skulls from Western Borneo: with a Bibliography. Proc. U. S, Nat. Mas., 1906, XXXI, 539-568. HrdliEka, A. The Taung’s Ape. Am. J. Phys. Anthrop., 1925, VIII, 379-392. Bunak, V. The Crest on the Skulls of Primates. Russ. Anthrop. J. MOSCOW, 1923, XII; also Anthroropologie (Prague), 1924, 11, 131-134. Elliot Smith, G. Essays on the Evolution of Man. So, Oxford Univ. Press, 1924, 11.