954 



SCIENCE. 



[N. S. Vol. VIII. No. 209. 



The different kinds of teeth seen in the 

 reptiles regarded as mammalian in type I con- 

 sider of comparatively small importance, for 

 the reason that the same general forms of teeth 

 are to be found, not merely in the reptiles sup- 

 posed to be nearest to mammals, but also in 

 other groups widely different. In the croco- 

 diles, for example, the extinct genus Notosuchus, 

 recently discovered in Patagonia, has all three 

 kinds of teeth well distinguished. Again, some 

 of the Dinosaurs, especially the genus Tricer- 

 atops, have teeth with two rows, a supposed 

 mammalian character. In some fishes, also 

 (Anarrhichas), three kinds of teeth may be 

 seen. It is more than probable, therefore, that 

 the peculiar resemblance between the teeth of 

 mammals and those of the lower vertebrates is 

 merely one of parallel development, the adap- 

 tation being along similar lines, and in no sense 

 an indication of genetic affinity. 



The great development of the squamosal bone 

 in Theriodonts is not seen in them alone, for 

 somewhat similar proportions are found in 

 other reptiles, for example, in the Plesiosaurs, 

 where the squamosal is very large, and wrapped 

 around the quadrate. In some of the Dinosaurs, 

 also (Torosaurus), the squamosal has an enor- 

 mous development, while the quadrate remains 

 of very moderate size. 



The quadrate bone, always present in birds, 

 reptiles and other lower vertebrates, is well 

 known as the suspensorium of the lower jaw, 

 which meets it with a concave articular face. 

 The quadrate, however, appears to be want- 

 ing in mammals, or at least has not yet been 

 identified with certainty. 



What represents the quadrate bone in 

 mammals is a vexed question in itself, and 

 some of the best anatomists in the past, 

 Cuvier, Owen, Peters, Huxley and others, 

 have endeavored to solve the problem. The 

 tympanic bone, the incus and the malleus have 

 each in turn been regarded as the remnant of 

 the quadrate, but up to the present time the 

 question has not been settled. It is not im- 

 probable that the quadrate may have coalesced 

 with the squamosal. 



The occipital condyles of mammals, as well 

 known, are two in number, and separated from 

 each other. This is not the case with any true 



reptile, although the contrary has been asserted. 

 The nearest approach appears to be where there 

 is a single bifid condyle, cordate in shape, with 

 the two lobes meeting below, as in some rep- 

 tiles and a few birds, but not separate as in 

 mammals and amphibians. 



Finally, in all known mammals, recent and 

 extinct, the lower jaw is composed of a single 

 piece, and has a convex condyle meeting the 

 skull by a distinct articulation. All reptiles, 

 even those supposed to be nearest the mam- 

 mals, have the lower jaw composed of several 

 pieces, and these show distinct sutures between 

 them, a profound difference that must be ex- 

 plained away before an approach can be made 

 between the two classes. 



It may fairly be said that the separate ele- 

 ments of the lower jaw, if present, would nat- 

 urally be looked for in the Mesozoic mammals, 

 and this point I have long had in mind. I may 

 safely say that I have seen nearly every species 

 of Mesozoic mammals hitherto described, and 

 have searched for evidence on this point with- 

 out success. I have also sought for the sepa- 

 rate elements in the young of recent forms, but 

 without finding any indications of them. 



Beside the crucial points I have mentioned in 

 the skull, there are others of equal importance 

 in the skeleton, which I must not take time to 

 discuss, but will venture to allude to one of 

 them in passing. I refer to the ankle joint, 

 which, when present, is at the end of the tibia 

 in mammals, but in reptiles between the first 

 and second series of tarsals. When we really 

 find an approach between these two classes 

 the ankle joint will probably show evidence 

 of it. 



Having thus shown, as I believe, that we can- 

 not, with our present knowledge, expect to find 

 the origin of mammals among the known ex- 

 tinct reptiles, and that in attempting this we 

 are probably off the true line of descent, it re- 

 mains to indicate another direction in which the 

 quest seems more promising. 



Since 1876, when Huxley visited me at New 

 Haven, and we discussed the probable origin of 

 both birds and mammals, I have been greatly 

 impressed by his .suggestion that the mammals 

 were derived from ancestors with two occipital 

 condyles, and these were doubtless primitive 



