1910.] Prwcijilcs of Mammalian PJiylogoiy. HI 



The shape of the teeth in a given race is of course determined primarily 

 by germinal, hereditary, perhaps "orthogenetic," factors. But even if it 

 is true that changes in the soma are in general not transmitted to the germ, 

 yet the form and conditions of one part of the soma must in some wav, in 

 the long run, act and react on the evolution of other parts of the soma; 

 because the correlation (mutual adaptation) of organs is a universally ob- 

 served fact. The shape of any given tooth must then be determined to some 

 degree by somatic conditions: such as the interaction between upper and 

 lower teeth (p. 100), the indirect influence of each tooth upon adjacent teeth 

 (p. 194), the size and arrangement of the muscles of mastication, and the like. 



The shape of the teeth is also profoundly influenced, in the long run, bv a 

 wholly external, "environmental," factor, namely the nature of the food, 

 and this is true even although it is not known how the result is efifected. 



In the case of the carpals and tarsals there is also a set of somatic factors : 

 such as the weight of the animal, the pull of the muscles, the interaction 

 between adjoining facets, etc. Each of these somatic factors must be in 

 part hereditary, in part adaptive i. e., depending on the nature of the environ- 

 ment. But assuming that in some way environmental conditions do affect 

 evolution, then the environment can not operate on the evolution of the 

 carpals and tarsals directly, but only through the intervention of the parts 

 that are in contact with them. 



Primitive characters often retained in sheltered parts of the organism. — 

 In general, primitive or largely hereditary (palseotelic) characters seem to be 

 retained longer in parts which come less directly into relation with the en- 

 vironment. This law is well illustrated in the mammalian reproductive 

 system. Amid the innumerable modifications of the external form, teeth, 

 limbs, skeleton and digestive apparatus in the Placental orders, the uterus 

 varies only from the "duplex" to the immediately derived bicornuate and 

 "simplex" types. In many instances (e. g., OrnithorJnjnchus, Ori/cteropus, 

 Castor) a very primitive type of uterus may coexist with a very highly spe- 

 cialized skull and dentition. 



G7rat phi/Iogenetic value of so called "non-adaptive" "morphological" 

 or palceotelic ^ characters. — We are thus brought to one of the guidino- 

 principles of the following investigation, a prihci])le which also seems to have 

 led de Blainville, Gill (cf. p. 91) and others to the best features of their 

 classifications: namely, that the parts which come more directly and simply 

 into relation with special food-habits and special environments (such as the 

 teeth, claws or hoofs, digestive system, etc.) are more plastic, and frequently 



1 The words palwotrlic and ccenotelic may be invented, as being more expressive than "mor- 

 phological" vs. "adaptive," to contrast ancient, with newer, characters. A palaeoteHc char- 

 acter becomes csenotelic by a change of function. 



