Revival of the Piltdown Controversy

New Discoveries Relating to the Antiquity of Man 1931

Arthur Keith

[31] The London Skull

Readers may be surprised to find that I have discussed the London skull at great length. Two chapters are devoted to it–Chapters XXIX and XXX. I look upon the London skull as the most important addition which England has made to our knowledge of prehistoric man since the first edition of Antiquity of Man appeared. A full consideration of the evidence has led me to transfer this skull to an older geological horizon than the discoverers have given to it, and to assign it, not as they have done, [32] to an unknown type of modern humanity, but to a recognized type of ancient man, viz. man of the Piltdown type.

Revival of the Piltdown Controversy

In coming to this conclusion I fear I am throwing sparks into the smouldering fire of the Piltdown controversy. It is not the revival of controversy which disturbs me, but the fact that by coming to such a conclusion concerning the racial nature of the London skull I have had to reconsider certain major problems concerning the history of man in ancient Europe. The story of early man in Europe was becoming simple and straightforward. The Piltdown type, we had concluded, had died out leaving no issue. The remains of modern man which had been found in early pleistocene deposits of the valleys of the Thames and Seine we had cast aside as falsely dated documents. Having removed these two human types from the arena of early pleistocene Europe, the way was cleared for giving men of the Neanderthal type sole sway in the western part of the Old World from early pleistocene times until the extinction of the type in the last ice age. It was then that Neanthropic man made his first appearance in Europe. Thus the story of early man in Europe was becoming simple and straightforward. If I am right in regarding the London woman as a lineal descendant of Piltdown man, then our story of Western man becomes complicated once more. Indeed, I think we have again to consider the status of Piltdown man. We had excluded him from the ancestry of modern man because of ape-like features in his jaw and teeth. With the evidence of parallel evolution before us and the presence of similar ape-like features in the lower jaw of Sinanthropus, some of our objections to the ancestral position of Piltdown man disappear.

We must take into account the fact that in his cranial features he was essentially of the modern type. In the light of these facts, it seems to me that the ancient man of Sussex comes very near to being the ancestor we have been in search of – the early pleistocene ancestor of modern races of mankind. . . .

[119] If we regard the Piltdown mandible as that of a chimpanzee with human leanings, then we have to explain how it came to be embedded in the same pocket of gravel as [120] a skull which is marked by so many early and peculiar characters. In the case of Pithecanthropus we have to regard skull, teeth and thigh bone as parts of the same body or suppose that by some rare chance three beings, all in different stages of human evolution, became mingled together in the gravel bed of an ancient stream. If the theory of evolution is true, and it is the only theory which explains the diversity and history of living forms, then the discoveries at Piltdown in Sussex, at Trinil in Java, and in the Broken Hill mine, Northern Rhodesia, should not surprise us. Fossil forms in which humanoid and anthropoid characters mingle are just what we ought to find if the theory of man's simian origin is well-founded. . . .

[273] In concluding this chapter I will attempt to sum up the gains to our knowledge of ancient man which have accrued from the discoveries made in this cave, situated among the hills to the south-west of Peking, gains which we owe to the united enterprise of the Geological Survey of China and to the Peking Union Medical College. Until this discovery was made, we had obtained only three glimpses of early pleistocene man in widely separated parts of the earth–in Java (Pithecanthropus ), in England (Eoanthropus ) and in Germany (Palaeoanthropus ). Pithecanthropus we may exclude from the direct ancestry of modern man; when we find that there were big-brained human forms early in the pleistocene period we must regard the small-brained Pithecanthropus as a [274] survival from some earlier phase of evolution. Piltdown man, save for the simian features of his tooth and jaw, probably does represent the stage of evolution reached by the ancestry of modern races early in the pleistocene. . . .

[290] The most surprising feature of Sinanthropus is the smallness of his brain. Does he represent in this respect the stage of evolution reached by humanity at the end of the pliocene or beginning of the pleistocene period? In size and conformation of brain, Sinanthropus of China rose little above his contemporary in Java–Pithecan[291]thropus. So small and lowly was the brain of the latter, that we have been in the habit of regarding him, not as a representative of contemporary humanity, but as a non-progressive survival which had retained an ancestral type of brain. * But now, with the Peking skull before us we have to reconsider the problem. Were, then, the Peking and Java men representative of early pleistocene humanity in their development of brain?


Fig. 95.–The skull of Sinanthropus superimposed on that of Eoanthropus (Piltdown man). The outline of the Piltdown skull is reproduced from fig. 252, p. 591, vol. ii, Antiquity of Man..

* See discussion, Antiquity of Man, vol. 2, p. 436.

[446] Let us look first at the part of the skull which has been preserved; from it we have to interpret the nature of a human being who lived in the valley of the Thames before even the ground on which London is built had come into existence. In fig. 148 an exact drawing of the skull is given in true profile, 1 and oriented on the subcerebral plane. 2

Fig. 148.–The London skull seen in true profile and oriented on the subcerebral plane. The parts actually found are shaded; the missing parts are indicated by stippled lines.

The part actually recovered is shaded and consists of two bones, the occipital which encloses the hinder part of the brain-the cerebrum above, cerebellum below-and the parietal bone. At one point the latter bone reaches forward almost to the coronal suture (fig. 148). We know that this is so because the lower anterior part is marked by the groove for an artery [447] which ascends just behind the coronal suture....



Fig. 149.–London skull oriented on the subcerebral plane and viewed from above. The parts actually preserved are shaded and the restored parts stipple. The outline of the more intact left side is reversed and indicated on the right side.


Fig. 150A.–Occipital view of the London Fig. 150B.–Similar view of a woman's

skull. The missing parts are stippled. The skull–one dredged from the bed of the

skull was oriented on the subcerebral plane. Thames.

Thus in our preliminary survey of the London skull the features we have met with are those which we are familiar with in modern specimens. There have, however, been three important exceptions: (1) the lowness of the vault; (2) the gradual way in which the curvature of the vault passes into the side of the skull, as seen in an occipital view (fig.150A); (3) equally remarkable is the curvature by which the occipital bone, as seen in a profile view of the skull (fig. 148) passes into the curvature of [451] of the vault. All these three features we must take into account in our search for the true racial affinities of this ancient and British specimen of humanity. Certainly they have not yet been met with in any human skull of late palaeolithic date. The Langwith skull, 3 Cheddar skull, 4 Baker's Hole skull 5 and Halling skull 6 have none of these features, but are n every respect comparable to modern European skulls....

Fig. 151.–The London fossil skull seen in profile superimposed on the corresponding parts of the Gibraltar skull. The comparison has been made so that the upper (sagittal) borders of the parietal bone are made to correspond.

[452] The neck or nuchal part of the occipital bone of the Gibraltar skull is bent sharply forwards under the upper or supra-inial part of the bone (fig. 151), whereas the same part of the London skull continues downwards in the same gentle curvature as the upper part of the bone (fig. 151). Now this sharp nuchal bend which is seen in the Gibraltar skull is characteristic of many primitive and early human types: Pithecanthropus, Sinanthropus, Rhodesian man, and Neanderthal man–indeed, of all primitive skulls, save one–the Piltdown skull. At an early stage of my inquiries into the Piltdown skull my attention was arrested by its strange open nuchal angle and also the long gentle contour of the supra-inial part of its occipital bone-features which we again meet with in this ancient British skull found under the foundations of Lloyd's buildings. To whatever type or species we may assign the London woman, it cannot be to the Neanderthal species or type.

To elucidate this point 1 reproduce an old diagram, published in the first edition of Antiquity of Man (1915).7 A series of sections shows the varying degrees of the nuchal [453] angle in the occipital bone in various skulls. The first of the series illustrates the inclination of the occipital bone in the orang; it is almost vertical. The second shows the sharp flexure or nuchal angle in a Neanderthal skull–the Gibraltar specimen.

Fig. 152.– A series of sections showing the angle which the lower or nuchal part of the occipital bone makes with the upper part and also the relationship of the occipital bone to the base of the skull and to the car passages.

Now, in his original reconstruction, Sir Arthur Smith Woodward sought to give the occipital bone of the Piltdown skull the forward slope seen in other primitive skulls, with the result–as shown [454] in the lowest section (fig. 152)-that the base of the skull had to be unduly curtailed. On the other hand, I gave it the vertical position, shown as "Piltdown I' (fig. 152), and thus obtained room for a normal base and full-sized foramen magnum.

Fig. 153–-A profile of the London skull applied to one dredged from the bed of the Thames. The "river-bed" skull is 178 min. long, 135 min. wide, its vault rising 93 min. above the subcerebral plane.

1 was moved to place the Piltdown occipital thus because of several other considerations: (1) the forehead, so far as it was known to us, was modelled on orang lines; 8 (2) articulation of the occipital with neighbouring bones could be obtained only when the occipital bone was thus placed. Further, it was necessary to presume that the upper or supra-inial part of the Piltdown occipital was relatively long and only gently curved. Now these features we meet with again in the London skull; the length of its supra-inial part, measured along the middle line to the [455] lambda, is 82 mm.; in the Piltdown skull I presumed it to have been 75 mm. ...

Before we proceed to compare the London and Piltdown skulls, let us take the nearest approach to the London skull which can be found amongst modern specimens and make a comparison. In the Museum of the Royal College of Surgeons there are over 100 skulls–-more or less fragmentary–which have been dredged from the gravel bed of the modern river. The "London" skull is deceptively modern to the first glance of the eye; it is not until one has searched for its counterpart that one realizes the peculiarity of its features. The nearest approach to it 1 could find in the "river-bed" series was the skull of a woman shown in fig. 153. The height of the vault in the specimen chosen for comparison is 93 mm., nearly the same as in the fossil skull. In height, and also in the curvature of the occipital and parietal part of the vault, the two skulls correspond closely (fig. 153), but at three other regions the modern and ancient skulls differ from each other: (1) the nuchal part of the occipital bone descends more steeply in the fossil skull; (2) the lower or squamous border of the parietal of the London skull rises more rapidly as it passes forward than in the river-bed skull; (3) the bregma of the London skull was placed at a higher level and at a more forward point on that level than in the river-bed skull. Now in all these three points the London skull, while differing from specimens of the modern type, resembles the fossil skull from Piltdown.

In fig. 154 the London skull has been superimposed upon my reconstruction of the Piltdown skull.' When we allow for the greater massiveness of the Piltdown skull and the greater thickness of its walls, the agreement in type is very close. The bones of the vault of the Piltdown skull vary In thickness from 10 to 12 mm.; in the London skull the thickness varies from 5 mm. to 7 mm.–the usual thickness of modern skulls. It must be remembered [456] that if we had found only the cranial parts of the Piltdown man we should never have hesitated in regarding him as the direct ancestral type of modern man; the simian features of his lower jaw and of his teeth led us to exclude him from this position. The markings of the Piltdown temporal bone, of the occipital and parietal bones, are essentially as in modern skulls and therein differ from the corresponding bones of Neanderthal man and of Sinanthropus. In the points wherein the Piltdown bones differ from those of modern skulls they agree with those of the London skull–except in thickness. The point wherein the London and Piltdown reconstructions differ lies in the position which has been given to the lambdold suture in my reconstruction of the Piltdown skull. In this reconstruction the lambdoid suture has not been placed far enough back. 9

Fig. 154–Profile of the London cranial fragment (stippled) superimposed on the author's reconstruction of the Piltdown skull (shaded).


[457] In fig. 155 is reproduced an occipital view of my reconstruction of the Piltdown skull; 10 in fig. 156 is given a corresponding view of the London skull, set within the same standard of lines, so that a comparison of their respective dimensions and shapes may be made. The Piltdown skull is represented as 150 mm. in width; that of the London skull as 140 mm., but the actual brain cavity is approximately of the same width in both, namely 136 mm..–the greater width of the Piltdown skull being due to the thickness of its bony walls. Even when we make allowance for the thickness of the vault, the skull height and brain height of the Piltdown individual are the greater. As regards the shape of the various bones and the muscular and other impressions which are to be seen on them, the correspondence is close, closer than when we compare the London skull to any other type.

Fig. 155–Occipital view of the Fig. 156.– Occipital view of

author's reconstruction of Pilt- London skull set on subcerebral

down skull (shown in fig. 154) plane, for comparison with

for comparison with corres- Piltdown skull.

ponding view of London skull.

Thus when we compare the known parts of the Piltdown and of the London skull we find a remarkable series of agreements, which certainly suggest that there is a closely evolutionary connection between them. The one point in which they differ is in the thickness of their [458] cranial walls. Thickness of vault, such as we find in the Piltdown skull is certainly primitive, but in fragments of other Piltdown skulls, 11 the thickness is not so great as in the original. We have also to consider the influence of sex; in the La Quina (Neanderthal) woman the skull is scarcely so thick as in the London skull. In former editions 1 have expressed some hesitation as to the sex of the Piltdown individual, inclining rather to the point of view that it was female, 12 but with a very distinct female of the same type before me I am strongly inclined to regard the original Piltdown skull as representing the male sex of its species and that some of the differences which distinguish the Piltdown from the London types are sexual in nature.

Fig. 157.–Occipital view of the endocranial (brain) cast of the Piltdown skull – the restored parts being stippled.

[459] Before we discuss the relationship of the London type (Homo londonensis) to the Piltdown type (Homo pilidownensis) and the place of each in the phylogenetic tree, let us look at certain other features which are revealed on their endocranial casts–casts taken from their brain cavities. In fig. 157 is reproduced an occipital view of the author's reconstruction of the Piltdown brain cast. 13 There are most definite and clear impressions on the occipital fragment of the skull which tell us that the occipital lobes of the Piltdown brain were highly asymmetrical in their development.

Fig. 158.–Occipital view of the endocranial (brain) cast of the London skull. It is shown oriented on the subcerebral plane.

As in modern man, the hinder end of the left hemisphere of the brain was much the greater (fig. 157, L., O), pushing over towards and invading the territory of its counterpart on the right side (fig. 157, R, 0). As Professor Elliot Smith has proved, this preponderance of the left occipital pole and cortex is symptomatic of right-handedness–of specialization of the right and left hemispheres of the brain for different offices. Asymmetry is not a low or primitive feature–-it is the opposite; it indicates an evolutionary advance [460] towards the modern state. With this asymmetry of the occipital poles of the brain goes a crossed asymmetry of the lobes of the cerebellum-the right lobe predominating.

Fig. 159.–Lateral aspect of the endocranial cast taken from the London cranial fragment. It is represented in true profile and on the supercerebral plane.


In fig. 158 is reproduced a corresponding view of the London endocranial cast. Professor Elliot Smith has described the asymmetry, of its occipital lobes, 14 in the London woman there is a reversal of the usual arrangement-the right occipital pole preponderating. The London woman, one may infer, was "left-handed". In her the longitudinal blood sinus descended almost in the middle line of the occipital bone, whereas in the Piltdown individual this sinus was pushed far to the right of the middle line. On the outer aspect of the occipital pole of the London cast can be seen the impression of the lunate (simian) sulcus which marks the limits of that area of the cortex concerned with vision. The right and left cerebellar lobes are separated by a deep and narrow fissure, whereas in the Piltdown brain the separation is shallow and wide. This difference has only an individual significance. If we examine a series of modern skulls we will find that the crest or falx of bone which lies between the two lobes of the cerebellum may be deep and sharp as in the London skull, or low, wide and blunt as in the [461] Piltdown occipital bone. When the occipital outline of the Piltdown brain is examined (fig. 157), a depression will be observed on the lateral aspect of the brain which marks the fissure of Sylvius, the same depression being indicated on the London cast (fig. 158). Below the fissure of Sylvius are the elevations caused by convolutions on the temporal lobe of the brain, which are more developed on the surface of the "London" than on that of the Piltdown brain. On the other hand, the eminence of the parietal lobe, which appears just above the fissure of Sylvius, is more emphasized in the Piltdown than in the "London" brain cast (figs. 157, 158). Still higher up, on each side of the middle line, separated by a depression from the parietal eminence, are the parasagittal eminences–the significance of which will be alluded to in the next chapter. These eminences are more prominent in the Piltdown than in the "London" brain cast.

Fig. 160.–The endocranial cast taken from the river-bed skull shown in fig. 131.


Lastly, to bring our comparison of the London skull with the modern and Piltdown types to an end, let us look at the features revealed on the lateral aspect of the London brain cast (fig. 159). Below the lateral sinus we see the vertical deposition of the cerebellum; above this sinus, but below the fissure of Sylvius (fig. 159), are well marked elevations caused by the convolutions of the [462] temporal lobe-parts of the brain which we have a right to suppose are connected with the interpretations of sounds-and are therefore concerned in speech and the correlation of speech with vision. These parts are well developed in the London cast. Above the fissure of Sylvius come parietal convolutions, certainly subserving higher functions of the brain, but here less well marked

than in the Piltdown brain. Higher up comes the parasagittal depression, then the parasagittal eminence.

Fig. 161.–Profile drawing of the brain cast taken from the reconstruction of the Piltdown skull by the author. It is represented half natural size, and set within a standard frame of lines. The positions of sutures between cranial bones are indicated. The missing parts are stippled.

The reader will remember that I found a river-bed skull (fig. 153) with a close resemblance to the London specimen; the endocranial cast, depicted in fig. 160, is taken from this skull. Here we again meet with the fissure of Sylvius; below the hinder part of that fissure and above the lateral sinus are the same temporal elevations [463 as are seen on the London skull–only in the modern instances these elevations are more extensive. Above the fissure of Sylvius the parietal convolutions are also more extensive and the parasagittal eminence less marked. Thus the London woman, ancient as she is, has an ample development of those parts of the temporal lobe which we suspect are connected with memory and with speech.

When we turn to the endocranial cast of the Piltdown brain cast which is represented in fig. 161, 15 we see the same vertical position of the cerebellar lobes as in the London cast (fig. 159), but the temporal elevations, between the lateral sinus and fissure of Sylvius, are less emphasized and less extensive. On the other hand, the parietal convolutions above the fissure of Sylvius are more prominent and more extensive than in the brain of the London woman. In the next chapter we shall return to the interpretation of endocranial casts, both ancient and modern, but in the meantime we may state most definitely that there is no feature in the cast of the Piltdown brain cavity, nor in that of the London specimen, which marks the one as being decisively higher or lower than the other, nor is there any feature in either which clearly distinguishes them from brain casts taken from many modern skulls.

Thus by a process of anatomical analysis I have been led to the conclusion that the human skull found in the heart of the city of London, under the foundations of Lloyd's buildings, represents a modification of the human stock first revealed to us at Piltdown in Sussex. The Piltdown and London individuals are separated from each other by a long period of time–which readers will best understand by consulting the provisional calendar of the pleistocene period constantly referred to in this work (fig. 162). Piltdown man was living in what is now the south-eastern part of England when pleistocene deposits began to be laid down. The oldest of the forest-beds of Cromer were then being formed; so were the deepest beds of the 100-foot terrace of the Thames


Fig. 162.–Time chart. (A) diagrammatic representation of the pleistocene and pliocene cultural periods and of their duration. The scheme, purely provisional, is that which is adopted in this work. Three glacial periods are depicted. (B) is a magnification of the later cultures shown in (A).


[465] Valley and the high terrace of the Sussex Ouse in which the Piltdown remains were embedded. The climate of Europe was temperate–more so than now–and the earliest forms of Chellean stone tools and weapons were being invented. The London woman and her people lived in the Thames Valley when glacial conditions were setting in. The contorted drifts and deep boulder clay were then being laid down in East Anglia over the Cromerian formations, and also the great gravel bed, which represents the oldest element of the 50-foot terrace of the Thames Valley, was being deposited. The period of Chellean culture was approaching its end and about to pass into another, the Acheulean. 16 Thus the Piltdown race lived at the beginning of the first long temperate interglacial of the pleistocene period. To this interglacial a duration of 120,000 years has been assigned in the time chart reproduced here (fig. 162). The London woman represents English humanity at the end of this long interglacial period.

Is it possible for any type of humanity to exist in one part of the world throughout such a long period of time and remain unchanged? The evidence which is now available leads us to conclude that such a stability is incredible. Mr. A. C. Hinton 17 has made a close study of the fauna which existed in England during the earlier and temperate half of the pleistocene period. He finds no evidence of the arrival of new species by migration during this long period, but ample evidence of the local production of new species by evolutionary change. If this is true of the smaller mammals, may it not also be true of man? May not the crude type which existed at Piltdown in [466] Sussex at the beginning of the pleistocene period have become refined into the modified type represented by the London type in mid-pleistocene times? This is what 1 suppose to have happened; it is the most likely explanation of the facts known to us. Such a supposition raises issues of the first magnitude relating to the evolution of modern man, issues which are discussed elsewhere. 17

Fig. 163–The phylogenetic tree of man's evolution showing the position formerly attributed to the Piltdown type (Piltdown 1) and the position the author would now give to that type (Piltdown 2) and to the newly discovered London type (London 1).


In conclusion let us ascertain the place which is to be assigned–in the light of the anatomical and geological evidence just discussed–to this strange and ancient London type in the evolutionary tree of humanity. In fig. 163 is reproduced part of the evolutionary tree used in former editions of Antiquity of Man. Piltdown man is there [467] represented as springing from the main human stem in the pliocene period and becoming extinct early in the pleistocene period. The objections to placing the Piltdown type in the direct line of the ancestry of modern man are the simian characters of his mandible and of his teeth. These objections are less weighty to-day than they were in 1915 when the first edition of Antiquity of Man appeared. We did not know then that in other very early pleistocene types of humanity the canine teeth had assumed–apparently independently–a human shape and size, and that their chin regions were coming by manlike characters. Evidence is accumulating which leads us to believe that similar structural changes may occur in groups of humanity long after their separation from a common ancestor–that, in brief, we must accept "parallelism" as having been a factor in the evolution of human races. It is therefore possible that Piltdown man does represent the early pleistocene ancestor of the modern type of man. He may well be the ancestor we have been in search of during all these past years. I am therefore inclined to make the Piltdown type spring from the main ancestral stem of modern humanity (fig. 163) and to represent this stock as becoming modified in the early pleistocene period so as to culminate later in that period as the London type. If Mr. Hinton is right as to the late age of the London gravels, then the Piltdown line has to be carried up almost to the end of the pleistocene period (fig. 163, London 2). Certain it is that the London skull cannot be assigned to the Neanderthal stock or stem; its anatomical characters are those seen in the skulls of Piltdown and modern types of humanity, but its nearest affinity is to the Piltdown type.



1 The drawings have been made from an excellent and accurate cast made by Mr. F. 0. BarIow. Thanks to the courtesy of Professor C. Elliot Smith, 1 have had opportunities of comparing the cast with the original and of studying the latter.

2 For an explanation of this plane–which I introduced for the reconstruction of fragmentary human skulls–see Antiquity of Man, vol. ii, p. 582.

3 See Antiquity of Man, p. 133.

4 Ibid., p. 137.

5 Ibid., p. 162.

6 Ibid., p. 115.

7 Ibid., fig. 262, p. 707, and the test which that figure illustrates.

8 Professor F. Frassetto (Man, 1927, vol. 27, p. 12) has demonstrated resemblances in the Piltdown mandible to that of the orang and concludes that Eoanthropus" represents a primitive race belonging to a genus of the orang type".

9 See Antiquity of Man, vol. ii, p. 547.

10 See Antiquity of Man, vol. ii, p. 517, fig. 174

11 See Antiquity of Man, vol. ii, pp. 592, 594.

12 Ibid., pp. 594, 604.

13 See Antiquity of Man, vol. ii, p. 628, fig. 228.

14 Evolution of Man, 2nd ed., p. 176.

15 See Antiquity of Man, vol. ii, p. 614, fig. 220.

16 In my time chart (fig. 162) the first pleistocene (Mindel) glaciation is represented as having taken place between early Chellean and Chellean periods of culture; as already mentioned (p. 443), there is a growing body of evidence in favour of assigning this glaciation to the end of the period of Chellean culture. In the deposits at Hoxne Mr. Reid Moir (Proc. Prehist. Soc. East Anglia, 1927, vol. 5, p. 152) finds evidence of three pleistocene glaciations–corresponding to the Mindel, Riss and Wurm of Penck, but as Chellean implements occur under the Mindel and Acheulean over the Riss, it is impossible to regard these two as parts of the same glaciation.

17 See reference, p. 444.


































 Back to the MAP

 Back to the