1888.] of Reptilian Character in Mammalian Teeth. 131 



is great and the wear considerable the crown of the tooth grows in 

 length, while the roots are relatively small ; but if the intermittent 

 strain on the tooth is great then the crown is relatively short and the 

 roots long. The latter condition is well seen in the molars of Carni- 

 vora ; the former in the molars of rodents and ungulates. The small 

 roots of ungulates and rodents illustrate a mode of development of 

 roots: for I have seen teeth of an aged fossil horse from the gravel 

 in which the crown was completely worn down, and then the roots 

 appeared to be relatively almost as well developed as in Rhinoceros.* 

 Perhaps no order is more instructive in regard to the classificational 

 value of roots of teeth than the Sirenia, because Manatus has tuber- 

 culate teeth and well-developed roots to the molars, while Halichore 

 has but one strong root to these teeth, indistinguishable from the 

 crown, with a hollow conical base, such as is often seen in Reptiles. 

 From these considerations I infer that the type of tooth — at least as 

 regards complexity — is to be correlated with the influences exercised 

 by food, and is not a distinctive inheritance. 



Secondly, the implantation of teeth in bony sockets is a mammalian 

 character which is not less well marked in the Crocodilia and some 

 extinct orders of Reptiles. The implantation in mammals with 

 single roots to the molars differs in no way from the conditions which 

 I have observed in Theriodont Reptilia. There are some exceptions 

 among mammals to the location of teeth in sockets, since in certain 

 Cetacea the teeth are in a groove at the posterior end of the series. 

 And the Ornithorhynchus may be regarded as another exception, since 

 it has three teeth on each side closely united together into one long 

 ovate mass which is contained in a groove. The teeth are closer 

 together than those of Ichthyosaurus, and there is no more definition 

 of the groove into separate sockets than in that genus ; but there is 

 nothing else in common, since the base of the dental plate of Ornitho- 

 rhynchus can scarcely be said to have roots. Frederick Cuvier 

 described these teeth as horny,f and many writers have been disposed 

 to regard them as horny plates rather than true teeth. Sir R. Owen 

 quotes a French analysis of the tooth substance as yielding 99 - 5 

 horny matter and 0*3 calcareous matter. J This may be true of the 

 long anterior horny plates on the jaws, but it can hardly apply to the 

 posterior teeth which are in a socket- groove. If the dental plate is 

 extracted from the jaw and examined against transmitted light, each of 

 the three teeth which form it will be seen to consist of a large opaque 

 subquadrate central portion, and an external translucent border of a 

 horny appearance. I regard the latter as representing the uncalcified 

 enamel of the tooth, while the central portion corresponds to the 



* The specimen was obtained by the Rev. N. Brady from near Cambridge, 

 f ' Des Dents des Mammiferes,' 1825, p. 203. 

 X ' Odontography,' p. 311. 



