This is not the case with the lower jaws of Naulette, Aurignac, and Arcy, which are undoubtedly genuine and of great antiquity. The Naulette jaw is, indeed, very imperfect, yet we can trace the construction of the symphysis of the chin, which provokes comparison with the lower jaws of many anthropoids, especially those of the gorilla and chimpanzee (Fig. 37). The resemblance consists chiefly in the uprightness of the anterior surface, and especially of the body of the maxillary bone. In anthropoids this surface of the bone retreats from the row of teeth backwards and downwards to the lower edge of the body of the maxillary bone (Fig. 38); and in the Naulette specimen, as well as in the lower jaws of some modern Papuan skulls (of New Hebrides and elsewhere), there is a certain approximation to the simian type. A fossil ape (Dryopithecus Fontanii) has been found in the Middle Miocene of Saint-Gaudens, assumed to be one of the higher anthropoids, and in this case the jaw is only slightly retreating. Gaudry considers that the Dryopithecus was about the size of a man. The incisor teeth were small. The cusps of the back molar teeth were less rounded than in Europeans, and more like those of Australians. It has been surmised, although the fact cannot be established, that the last molar teeth were only cut after the canine teeth, as is the case with the human wisdom teeth. Gaudry gives the illustration of the lower jaw of a Tasmanian, from eleven to twelve years old, together with that of Dryopithecus Fontanii. In the human jaw the first molar tooth is larger than in the Dryopithecus, while the canine tooth and the pre-molars are much weaker. This distinction is important, since the smaller size of the front teeth is connected with the slight projection of the face, which is always a sign of human superiority. Although the canine tooth of the Dryopithecus is broken, we can see that it must have been considerably higher than the other teeth, and indeed the canine teeth of the male animal must have been very powerful. There is also a slight prominence in the teeth of this ape, which is absent in those of men. Mesopithecus, from the Miocene of Pikermi, Attica, was an ape less closely resembling the anthropoids. In the structure of the head it resembles the slender ape (Semnopithecus), and in the structure of the limbs it is like a macaca (Macacus). Gaudry believes that Sansan’s Pliopithecus was related to the gibbon. An ape of the size of the orang-utan, which belongs to the slender apes (Semnopithecus sub-himalayanus),45 was found by Baker and Durand in the Miocene of the Sewalik mountains.

In the comparative study of the human organization, and that of anthropoid apes, it is important to examine sections, and especially longitudinal sections, of characteristic skulls.46 Virchow has caused drawings to be made, from specimens in the Berlin Museum, of a gorilla, a chimpanzee, an orang-utan, and an Australian woman. The gorilla’s skull, when compared with the Australian’s, is so narrow that it looks as if compression had been applied to it; and yet the Australian skull is extremely small in comparison with that of men in general, since its cubic space is only 1150 ccm. In the gorilla47—at least in the old male, from which the drawing is taken—the immense size of the frontal sinuses, and the swellings which cover them, together with the strongly developed jaw, increase the impression of size. But, as Virchow observes, “all which adds to the size of the skull is bestial, and not human.” It is much the same in the orang-utan. Only in the chimpanzee the cubic space of the skull may be somewhat more favourably compared with that of the human skull. It approaches in size to that of a microcephalic native of the Rhein-Pfalz (of which an illustration is also given), which ranks a good deal below the Australian skull, and approximates more closely to the simian type. The internal space of the skulls of an adult female gorilla or orang may also be more favourably compared with those of men.

We have already mentioned the presence of extensive sinuses and cells in the skulls of anthropoids, exceeding those of human skulls, and this is apparent in the accompanying illustration of a longitudinal section of the skull of a chimpanzee carried through its centre (Fig. 39). The length of this skull between the nasal partition and the most prominent part of the occipital bone is 128 mm.; that of the internal space is 108 mm. 10 mm. of this difference is due to the depth of the frontal sinuses, and the rest is owing to the thickness of the bony part of the skull. In an aged male gorilla, the first measurement is 153 mm., the second 115 mm. In another aged gorilla the measurements were respectively 183 mm. and 117 mm. In a still more aged male orang they were respectively 140 mm. and 114 mm. The comparative thinness of the centre of the squamous occipital portion is to be noted in the aged gorilla male. In the adult chimpanzee the large cells of the squamous portion of the temporal bone extend into this bone, and indeed without interruption into the parietal bone adjoining it. For such investigations the thin and light bones of individuals which have lived a wild life are more suitable than the heavy and fat specimens which have died after prolonged confinement.

Zuckerkandl has observed that among Europeans the orbital part of the nose, or that part which is between the orbits, is longer than the infra-orbital or lower part. In anthropoids the infra-orbital portion is considerably the longest, although only in adult animals. There are stages in the period of development in which these animals display the characteristics of an adult European, or indeed of a child. The proportions of the skulls of Malays take a middle place between those of Europeans and of apes. The growth of the infra-orbital part of the nose in the Malay does not equal that of apes, but in many cases it differs essentially from that of Europeans. Zuckerkandl makes a skilful attempt to establish this statement by statistics.

The same inquirer makes some interesting remarks on the comparative height and width of the orbits. He observes that the skulls of adult apes and men differ more in these respects than the young specimens of these organisms. The orbits both of a child and an adult, especially in the case of a European, are much more like those of a young ape than of an aged animal of the same species. In the chimpanzee and the orang-utan the proportions are the same as in men; that is, the width of the orbit exceeds its height. In man, this seems to arise from the exceptionally strong development of the supra-orbital ridge. It is most probable that in very young anthropoids the width of the orbit exceeds its height.48 Zuckerkandl goes on to say that in anthropoids the height of the orbits is greater than their width, and that this difference increases with age. But this is not absolutely correct, for even in aged animals the proportions vary, and the height and width of the orbits sometimes, although rarely, remains the same.

In comparing the vertebral column in men and anthropoids, Rosenberg has sought to show in the embryo, that the first sacral vertebra assumes the form of a lumbar vertebra, and that in a later stage of development it is enclosed by the ilia, and anchylosed with the sacrum. The same author has proposed a theory of the homologous or genetic equivalents of the vertebræ, which we must now consider. According to this theory, as Welcker has observed,49 the twentieth vertebra of an animal A is homologous to the twentieth vertebra of an animal B, the thirtieth vertebra of one animal to the thirtieth of another, although in one case it may be a lumbar vertebra, in another a pelvic vertebra, and in a third a coccygeal vertebra. The dorso-lumbar vertebræ of the lower apes have, in the case of men, their descendants, undergone a threefold metamorphosis, and, after their modification into sacral vertebræ, have assumed their fourth form as coccygeal vertebræ.

Froriep, a follower of Rosenberg, remarks that the lumbo-sacral vertebræ, i.e. those constituents of the vertebral column which form the transition from the lumbar to the sacral vertebræ, are invested with fresh interest by Rosenberg’s hypothesis. According to their position in the vertebral column, they are to be regarded as lumbar vertebræ, introduced too early or too late into the structure of the sacrum. If the twenty-fourth vertebra is assimilated with the sacrum, so as to form an upper promontory or outwork, this variety offers a point of transition to a future formation (?) in which this vertebra normally becomes the first sacral vertebra, and the column will now display twenty-three free vertebræ. If, again, this transition occurs in the twenty-fifth vertebra of the series, which thus becomes the chief sacral vertebra, this is, in Rosenberg’s opinion, a characteristic survival of the racial development, an atavism.50

According to Welcker’s theory, the chief sacral vertebra in one animal corresponds to the same sacral vertebra in another animal, whatever their number may be. The cervical vertebræ of one animal, which may be five, seven, or even eleven in number, correspond to the cervical vertebræ of another animal. The vertebral column of one animal corresponds to the vertebral column of another, taken as a whole, but not to two-thirds or three-fourths of that column. In accordance with the requirements of a given animal, that part of the bone which belongs to the sections of the breast and loins is more or less abundant, and the vertebræ are homologous in accordance with their region, and not with their number.

Holl has asserted that one vertebra is in close connection with the ilium, joined with it throughout its extent, and that this vertebra at the same time always appears to support the pelvis. This vertebra is, in normal cases, the first sacral vertebra, and the twenty-fifth of the series. It may be termed, as Welcker suggests, vertebra fulcralis. Such a main support is found, according to Holl, in every vertebral column, however anomalous its other conditions may be, and the only irregularity consists in its number in the series. This bone serves as a natural starting-point in our division of the vertebral column. The vertebra fulcralis must always be regarded as the first sacral vertebra. It begins the series of sacral vertebra, and, on account of its subsequently important position, it must be regarded as primary. Holl finds that it is followed by four lower vertebræ, which are afterwards included with it in the sacrum. When in its primary condition the vertebra fulcralis is twenty-fifth in the series, the twenty-fifth to the twenty-ninth vertebræ are included in the sacrum. When the fulcralis is the twenty-sixth vertebra, the sacrum includes the thirtieth. Hence it follows that the sacrum is, from the first stages of its development, a formation which begins with the twenty-fifth or twenty-sixth vertebra, and includes four other vertebræ. Holl considers that the lumbo-sacral form of the last lumbar vertebra, which stands between the lumbar and sacral vertebræ, does not indicate a gradual transition into a sacral vertebra, but rather an arrest in its development.51

When we examine a human sacrum we see that its first vertebra, the twenty-fifth of the series, is formed like the lumbar vertebræ in its upper part, setting aside those portions of it which form part of the lateral masses of the sacrum. These lateral masses, which serve as a support to the ilia, owe most of their substance to the first sacral vertebra. Thus, since it has to support the whole weight of the pre-sacral vertebræ, it is in fact a true vertebra fulcralis.

Holl justly says that there are few instances in which the human os sacrum consists of less than five vertebræ, and in no case are there less than four. In such a case the first sacral vertebra defines the pre- and post-sacral segment of the vertebral column.

In anthropoids the lower segment of the lumbar vertebral column is deeply sunk between the high, wide, and flattened ilia, which converge closely towards the vertebral column. In man these bones are not so much higher than the base of the sacrum, and their crests diverge more widely from the vertebral column. In the large apes the lateral masses of the sacrum are comparatively deeply set below their anchylosis with the pelvic bones. In an aged male gorilla, for instance, the transverse processes of the two lower lumbar vertebræ often extend to the hinder borders of the ilia, although the second of the lower lumbar vertebræ is somewhat higher than the top of the crest of the ilium. This is still more remarkably the case in an old male chimpanzee, in which the lowest lumbar vertebra seems to be wedged in between the two ilia. In a young male chimpanzee, and in the adult female, both the lower lumbar vertebræ are almost compressed between the upper segments of the ilia. In the orang the lowest lumbar vertebra is placed between the ilia. Out of the five sacral vertebræ the first and second are articulated with these bones.

In the gorilla the highest sacral vertebra, the twenty-fifth of the series, is the fulcralis. In this animal the first to the third sacral vertebræ form part of the connection with the crests of the ilia. In the chimpanzee the twenty-fifth is also the vertebra fulcralis, and from the first to the third are likewise connected with the ilia, but the third only to a limited extent; and in young males and in old females the connection is generally confined to the first and second sacral vertebræ. In the orang-utan the twenty-fourth vertebra is generally the fulcralis.

In the gibbon the twenty-fifth vertebra is usually the fulcralis. In the siamang I found that the fifth of the five lumbar vertebræ was between the ilia. Out of the five sacral vertebræ the first and second were articulated with the said pelvic bones. In Hylobates agilis the fifth and sixth of the six lumbar vertebræ were between the ilia, and the first and second of the five sacral bones were articulated with these.

In the vertebral columns of the gorilla, the chimpanzee, and the orang we may observe an inconsiderable forward projection between the penultimate cervical and the second and third dorsal vertebræ. In the region below the second lumbar vertebra a similar forward projection may sometimes be observed. The so-called promontory at the entrance of the pelvis, that is, in the region developed between the lumbar and sacral vertebræ, which is remarkable in man, is only faintly apparent in anthropoids. The vertebral column is arched behind, since there is a dorsal curvature (see Figs. 17 and 23).

Aeby observes that the bodies of the vertebræ are tapering in the gorilla, and this is, in fact, the case. In climbing, or when he goes on all fours, the dorsal curvature of an anthropoid maintains its position. This curvature is still more apparent when the animal, in climbing, withdraws his body from the tree, mast, or whatever it may be, and bends forward his head. A similar dorsal curvature of the vertebral column may be observed in men who stiffen their hands and feet to climb up a tree or mast. If an anthropoid holds himself so erect as to be able to place his hands behind his head, the dorsal curvature of his spine is necessarily straightened, and indeed it becomes rather a ventral curvature.

The bony pelvis of anthropoids, with its high, narrow, and projecting ilia, and the lowest lumbar vertebræ deeply embedded between them, together with the sacral and coccygeal vertebræ, which directly remind us of the vertebræ of a rudimentary tail, present the points of unlikeness with the human skeleton in this part of the skeleton of these animals in the strongest light (comp. Figs. 40 and 41).

The bony thorax of anthropoids is distinguished from the human thorax in normal cases by the abrupt way in which it widens outwards. The thorax of the gorilla, and the widely diverging pelvic bones, which enclose the belly and give it a tun-shaped form, contrast with the graceful moulding of the corresponding parts of the human form.

Certain peculiarities in the structure of the bones of the shoulder-girdle and of the extremities of anthropoids, in which they differ from corresponding parts in the human structure, have been already mentioned.

With reference to the humerus of the gorilla, Aeby asserts that the head of the bone forms a cycloid, placed transversely, while in man its shape is that of the segment of a sphere. But I have pointed out in my treatise on the gorilla that there is a not inconsiderable variation in the form of the head of the humerus in these animals, and it is sometimes cycloidal or vertically-cycloidal, sometimes a segment of a true sphere. In the chimpanzee, orang, and gibbon this part of the humerus is always a segment of a sphere, while in man its form is not equally invariable. Aeby further observes that the transverse-cycloidal form of the head of the humerus in the gorilla justifies the inference that this animal, in the use of its fore-limbs, is accustomed to turn them transversely on their axis. But the direct observation of a living anthropoid, as well as the examination of its dead body, make it clear that the action of the ball and socket is remarkably free, and this theoretical surmise is contradicted by the perfection of the natural mechanism.

The excessive curvature of the forearm which we notice in the gorilla and the chimpanzee in their natural condition is rare in man, and when it does occur it must be regarded as an abnormal and pathological phenomenon.

The orang-utan always displays a ninth carpal bone, corresponding to de Blainville’s os intermedium and Gegenbaur’s os centrale carpi. In a very young animal I found that this small bone was furnished with a peculiar point of ossification. The bony structure of the wrist is developed in the following succession:—First, the os magnum and unciform bones; second, the scaphoid bone; third, the trapezium; fourth, the semi-lunar bone; fifth, the cuneiform bone; sixth, os centrale carpi; seventh, the trapezoid bone. The pisiform bone and the sesamoid bone, between the trapezium and the scaphoid bone, of which we shall speak presently in their relation to the muscular system, are at first simply cartilaginous.

Up to this time my search for this ninth carpal bone in the gorilla and the chimpanzee has been fruitless, since its occurrence is only exceptional. In the gibbon it is plainly inserted between the scaphoid, semi-lunar, trapezoid, and os magnum. Gegenbaur considers the os centrale to be a true constituent of the wrist, dating from an earlier condition, but he has nothing to suggest as to its subsequent survival. Rosenberg has lately given an incontestable proof of the presence of this bone in the human embryo. It is generally absorbed again, but sometimes it persists, and may be found in an adult as a well-formed ninth carpal bone. Cases of the persistence of the os centrale in man have been chiefly collected and published by the diligence of the Russian anatomist, Gruber. It is now suggested that there may also be indications of os centrale in the carpus of embryos of the gorilla and chimpanzee, but up to this time materials for such researches have been wanting.

I cannot accept the theory that os centrale carpi is merely a detached portion of the scaphoid bone. In a very young chimpanzee this bone is undoubtedly superficially indented with two transverse furrows, but the three segments display only one uniform development of bone. The distinct formation of os centrale, and its occasional appearance in man, testify that it has an independent existence. Rosenberg holds that this bone is not merely the os centrale of mammals, but that it is homologous with the two ossa centralia of the fossil Enaliosauria. It has become abortive in proportion to the reduction in size which has taken place.52 There would be no great difficulty in tracing back this bone to remote types of vertebrate animals, even as far as the Urodela (Wiedersheim) of Eastern Asia.53 The persistence of this bone in man must be regarded as a reversion, not as an arrest, of development.

On the femur of several mammals, especially in the horse, ass, rhinoceros, and tapir, and more slightly indicated in the carnivora and other families, there is, in addition to the two great and small trochanters, a third, termed by Waldeyer trochanter tertius.54 Such a formation, low, blunt, and generally placed at the top of the outer ridge of the superior bifurcation of the linea aspera, may be observed in human skeletons of all races, but is either absent in anthropoids or only faintly indicated. Virchow justly regards its presence as theromorphic, but not as a characteristic of savage or lower races.55

The human tibia displays in some instances a compression or lateral flattening of its shaft or centre-piece, so that its transverse diameter is quite out of proportion to its depth. Such a tibia is termed sword-bladed, or platycnemic. Bones of this form have been chiefly discovered in ancient deposits, as, for instance, at Gibraltar, at Perthi-Chwareu, in Wiltshire, in Lozère, at Clichy, at Saint-Suzanne (Sarthe), and especially at Cro-Magnon (Fig. 43), Janischwek, etc.

A similar formation has also been observed among men belonging to cultured races, both of ancient and modern times. Virchow, for example, discovered such bones in Transcaucasia (of the third and fourth century of the Christian era) and at Hanai-Tepe in Troas. All the large schools of anatomy in Europe contain specimens of tibiæ, which are to some extent platycnemic. These are also observed in the skeletons of primitive peoples of our time, as for example in the Negritos, Kanakas, and other African races. While some scientific men regard these bones as the result of an unhealthy condition, and the effect of rachitis, others more justly ascribe them to a vigorous exercise of the muscles in a one-sided direction. The idea expressed by Busk and others, that the platycnemic tibiæ discovered in ancient sites of Europe have belonged to a degraded race diffused over the whole continent, is contradicted by the wide diffusion of this characteristic, even in modern times. And it is doubtful whether platycnemy is absolutely restricted to the lower races. At Janischewek, Virchow found an extremely platycnemic tibia, exhumed from a kujawish grave of the Stone Age, which belonged to a skull remarkable for its unusual beauty and size, so that, taken by itself, the impression which it gave to an anatomist was that of a highly organized race.56

It is important to remark that platycnemy has been regarded as a pithecoid structure, and for this reason the attempt has been made to establish the degraded position of those peoples which are most remarkable for platycnemy. But, as Boyd-Dawkins has already observed, although the tibiæ of the gorilla and the chimpanzee are to some extent platycnemic, they are much less so than the platycnemic bones of the human skeleton. The tibia of a male gorilla in the College of Surgeons Museum has an index width of 68·1, that of a female of 65·0, while the index of the chimpanzee’s tibia is 61·1, which is about the average of the tibias of Perthichwareu. It is unnecessary to indicate the other marked distinctions between the tibiæ of men and apes; if platycnemy is to be regarded as genetic, it must be admitted that man has in this particular far exceeded apes.57 Neither the gorilla, the chimpanzee, the orang-utan, nor even the baboon possesses a tibia which is flattened in its upper or middle part. In all these apes the middle of the bone is more or less rounded, almost as if it had been rounded by a turning-lathe. According to my experience, the degree of platycnemy in anthropoids is subject to certain variations. It appears to me to be least marked in the aged male gorilla (Fig. 41), and in the gibbon (Hylobates agilis, syndactylus), in which latter animal the transverse section of the tibia represents an almost equilateral triangle. The platycnemy was more marked in an almost adult female gorilla, still more decided in an aged male chimpanzee, which came from the river Kiulu, and again in an aged female chimpanzee. On the other hand, the centre of the shaft of the tibia in another aged male chimpanzee which came from Loango, was rounded, and not platycnemic. In the tibia of an adult orang-utan which I examined, the platycnemy was very marked. But I agree with Boyd-Dawkins in never having met with an anthropoid in which the platycnemy is so considerable as it is, for instance, in the Cro-Magnon tibia, and in another found at Troy.

If we give a cursory glance at the lower limbs of apes, we see that all the same characteristics are present in their tarsus that we find in the human tarsus. In each case there is an astragalus, an os calcis, a scaphoid bone, three cuneiform bones, and a cuboid bone. There are undoubtedly several peculiarities in which the tarsus differs from the corresponding part of the human foot. The first metatarsal bone is joined to the first cuneiform bone by an articular facet which extends from the back to the sole of the foot. This joint plays a part resembling that of the thumb of the human hand (see Figs. 20 and 46).

In Huxley’s opinion, the hinder limbs of the gorilla terminate in a true foot, with a very movable great toe. It is undoubtedly a prehensile foot, but in no sense a hand. It is a foot which does not differ from the human foot in any essential characteristics, but only in relative circumstances, in the degree of flexibility, and in the subordinate arrangements of its parts. Huxley adds that it must not be supposed that he wishes to undervalue differences which, however, he does not regard as fundamental. They are important enough of their kind, since in any case the structure of the foot is in close correlation with the other parts of the organism. Although it cannot be doubted that the increased division of labour in man, which relegates the function of support entirely to the legs and feet, is a significant advance in structure; yet, regarded as a whole from the anatomical point of view, the points of agreement between the human foot and that of the gorilla are much more striking and significant than their differences.

The differences in the foot of the orang are still greater; in the very long toes and short tarsus, the short great toe and the removal of the heel from the ground, in the great obliquity of the joints which connect the foot with the shank-bones, and in the absence of a long flexor muscle to move the great toe, the orang’s foot differs still more from that of the gorilla than the latter differs from the human foot. In some of the lower apes the hands and feet are still further removed from those of the gorilla than in the case of the orang. In the American apes the thumb can no longer be opposed; in the ateles it is reduced to a mere rudiment, covered with skin; in the sahius it is bent forwards and provided with a curved claw like the other fingers. In all these cases there is no doubt that the hand differs more from that of the gorilla than the gorilla’s hand differs from that of man.58

Flower remarks that the chief distinction between the foot of a man and an ape consists in the fact that the latter is transformed into a prehensile organ. The tarsal and metatarsal bones, and the phalanges are of the same number in both orders, and in the same relative position, only in the foot of the ape the facet for articulation of the first cuneiform bone with the great toe is saddle-shaped, and obliquely directed towards the inner or tibial side of the foot. Thus, the great toe is separated from the others, and so placed, that when it is bent, it is directed downwards towards the sole, and is opposed to the other toes, much more opposed to them than is the case with the thumb of the human hand.59 Owen also speaks of the characteristic transformation of the great toe of an ape’s foot into a thumb, opposed to the other toes, and adapted for grasping.60

K. E. von Bär does not agree with Huxley in considering that there is less difference between man and the gorilla than that which exists between different species of apes. “There are,” Von Bär remarks, “differences of various kinds among apes. In some the thumb is only a stump; in others, as in the orang-utan, the fingers of the hinder extremities are so long and curved that they cannot be extended on flat ground; in many of the smaller apes this member is still more like a hand than in the larger species, and the fingers can be easily spread out on the ground. In this case the foot is of a much blunter form, and is more flexible, so that the sole, which is properly turned inward, can lie flat on the ground. The heavier the body of the animal, the more sharply cut the structure of the foot must be, so that it does not admit of the free movements which are possible in the hand. But all these are only modifications of a climbing foot, or prehensile member—that is, of a hand, not modifications of a foot resting firmly on the ground and supporting the whole weight of the body.

“It must not be forgotten that the structure of the skeleton is subject to mechanical laws, which may be traced through the whole series of the animal world. This is readily apparent when we turn to the human structure.

“The human foot rests for the greater part of its length on the ground, that is to say, with the heel and centre of the foot, which form together a firm arch. The tarsus consists of the astragalus, and also of the os calcis, which in man form a very prominent part, taking a backward and downward direction, and of five other bones. The metatarsus consists of five bones, on which the five toes are inserted. In man these metatarsal bones are considerably longer than the separate phalanges. Thus, the arch on which man is supported in an erect position extends from the heel to the extremities of the metatarsal bones. The several bones are slightly movable, but they are so firmly connected that they can diverge but little from each other, unless muscular power is exerted. In order to press the toes upon the ground, it is again necessary to exert the muscles. The arched instep has this advantage, that the foot can take a better hold of the slight inequalities of the ground. In a profile view of the skeleton of a human foot, the shortness of the toes, in comparison with the length of the arched instep, is very apparent. In any natural position, even when man is not walking or standing, the sole of the foot is not turned inwards, but downwards.... The toes of the gorilla take the form of a hand, since the great toe stands separate like a thumb, while the other toes are turned outwards. In the gorilla the tarsus is short, and the heel is bent inwards. The several bones of the human foot are undoubtedly present in the hind hand of a gorilla, but the organ is changed into a prehensile organ or hand. The conditions are the same as in the parts of the mouth in insects which in some cases form movable mandibles, while in others they are attenuated into a proboscis. When it is asserted that apes are not quadrumanous, it is as if we were to say that flies have no proboscis, but attenuated mandibles.”61

All apes, including anthropoids, occasionally make use of their hinder extremities in order to snatch at objects. They also grasp with them in climbing. On such occasions, when they wish to secure the fruit they have seized from the voracity of their fellows, they take it between the toes of one hinder extremity, in order to be able to get away more quickly by means of the other, and by the use of both hands.

From what we have said, it will be seen how difficult it is to reconcile the views of different observers with respect to the fitting term to be given to the hinder extremities of apes. Against those who uphold the designation of hind hands we must oppose the anatomical structure, and also the fact that a true hand ought to possess the power of rotation in a degree which exists in the fore, but not in the hind, extremities of apes. On this account I have already adopted, as more suitable and equally distinctive, the term of prehensile foot for this member.62 I agree with Haeckel in rejecting the common designation of apes as four-handed or quadrumanous.

The bands or ligaments which connect the different parts of the anthropoid skeleton together, and convert the detached elements into a movable machinery, do not on the whole differ much from the same structure in man. A detailed account of these ligaments would, for several reasons, be out of place in this work, and I shall only mention a few special and more interesting distinctions. Such, for example, is the uncommon strength of the ligamentum nuchæ in the gorilla, which is quite in harmony with the great development of the spinous processes of the upper cervical vertebræ, and with the flattening of the squamous occipital portion. Since the sacral vertebræ are deeply inserted between the high ilia, the ilio-lumbar ligaments (ligamenta iliolumbalia) and the sacro-iliac ligaments (ligamenta iliosacralia) are of considerable size. In agreement with the projection in a downward direction of the high, narrow ischial bones, the sacro-sciatic ligaments which extend between these and the sacrum are very long in the chimpanzee. Although in this case the ischial spine is only represented by a roughness of the bone, yet there is on either side between this and the sacrum a powerful lesser sacro-sciatic ligament (ligamentum spinoso-sacrum).

The well-known anatomist, J. F. Meckel, has asserted that the depression in the head of the femur (fovea capitis), which serves for the insertion of the round ligament (ligamentum teres), is absent in the chimpanzee and orang, and he adds that it is also absent in the gibbon. In a skeleton of a young chimpanzee which had not shed its milk-teeth, and of which the ligaments were also preserved, Welcker found a fully developed round ligament inserted almost in the centre of the head of the femur. This agrees in every particular with the same formation in man. On the other hand, no trace of a round ligament was to be found in the hip-joint of a young orang-utan. The cartilaginous envelope of the head of the femur was smooth throughout, without any indication of a place for inserting the ligament. Welcker again found no such depression in the femur of an aged male orang-utan, nor was there any trace of it in another aged male orang, designated as Simia Morio. Welcker believes that he has established the fact that the round ligament is wanting in the orang-utan, but that it is present in the gorilla, chimpanzee, and gibbon. The same naturalist remarks that, although we may certainly assume that the round ligament is absent wherever there is no depression in the head of the femur, yet the existence of such a depression in the acetabulum (fovea acetabuli) is not enough to prove that a round ligament was inserted in it. The innominate bones of an adult orang-utan were examined by Welcker, and displayed a small, but well-defined depression, as if destined for a receptacle for this ligament,63 running from the cotyloid notch down to the bottom of the acetabulum, between the two horns of the semilunar-shaped articular cartilage.

In a subsequent paper, Welcker states that the absence of the round ligament in the orang-utan, and its presence in the chimpanzee, had been previously established by Camper and Owen.64 In three specimens of orangs which he had obtained immediately after death, Owen found that the round ligament was imperfectly developed on both sides. The chimpanzee differs from the orang in possessing a depression on the head of the femur. In the gorilla, as Owen observes, this depression has almost the same depth and relative position as in man. At Welcker’s request, Professor Dippel ascertained the presence of the depression in the femur of a gorilla skeleton which is preserved in the natural history collection at Darmstadt. St. George Mivart saw the skeleton of an orang in which the femur was marked with a slight but plainly indicated depression, just where the round ligament is usually attached. Welcker thinks it probable that in some specimens of the gorilla the round ligament is only slightly developed, and that in others it is altogether wanting. On several femurs of gorillas, this naturalist observed only doubtful traces of the depression in question. Duvernoy found the round ligament fully developed in the gorilla and chimpanzee. Vrolik failed to find it in the orang-utan, but ascertained its presence in the chimpanzee. Gratiolet and Alix saw that it was fully developed in Troglodytes Aubryi.

In addition to these somewhat conflicting assertions, I have myself observed, in the gorilla innominate and femur bones examined by me, more or less distinct indications of the depression which receives the round ligament. The ligament itself has been preserved with the body of a gorilla. The same remark applies to the skeletons and bodies of chimpanzees. In the case of the skeleton of an orang, slight indications of a depression were observed on the head of the left femur, and these indications were absent in the femurs of other specimens. In a large orang-utan which died in the Berlin Aquarium, only short, filamentous tufts of streaky fibres were apparent in the right acetabulum, and these were intermingled singly or in groups with the cartilaginous cells, somewhat resembling the cartilaginous corpuscles of the synovial membrane. From these facts we may conclude that the round ligament is generally but not invariably present in the gorilla and chimpanzee, and that it is altogether absent in the orang-utan. In the gibbon it is present in the majority of cases. I have myself observed it in Hylobates agilis, leuciscus, and syndactylus. Owen asserts that the unsteady gait of the orang is partly due to the absence of this ligament, but the truth of this surmise is rendered doubtful by the fact that the ligament is not unfrequently absent in other anthropoids. Moreover, the gait of all these arboreal and climbing animals is extremely ungainly.

The muscular system of anthropoid apes is very interesting. I must necessarily refrain from giving a detailed account of it, and will only mention some points in connection with this organic system, and their relation to corresponding points in the muscular system of man. I rely partly on the researches of others, and partly on my own. The amount of material which has been collected up to this time is, unfortunately, too scanty to enable us to draw satisfactory conclusions in all cases. We are often unable to decide whether the conditions presented to us in the case of anthropoids are normal or exceptional. Nor are the statistics of muscular variations in the human subject by any means firmly established. My own labours in this direction are not yet concluded. The assertions on the subject which have been published to the world and accepted as authoritative have already been shown to be to some extent untrustworthy. Even the little which I am now able to produce may not altogether stand the test of subsequent research. Brühl justly remarks that in no department of anatomy more than in that which treats of the muscles, is it more essential that we should not decide whether a form is normal or exceptional until it has been repeatedly examined.65