336 VERTEBRATE ZOOLOGY 



have dealt with as the group of reptiles showing the most pronounced 

 mammalian tendencies, were not Permian animals at all, but lived in 

 Triassic times, a million or so years later. How then could they have 

 given rise to the mammals? The answer is that they themselves 

 probably did not produce the mammals, but that they and the mam- 

 mals were both derived from a common ancestral stock that lived in 

 the Permian. The mammals represented an offshoot of this ancestral 

 stock that went the entire course in developing mammalian charac- 

 ters, while the cynodonts represent a number of partially successful 

 experiments that fell short in various respects of full mammalian de- 

 velopment. Some day it is hoped that the true ancestral mammals 

 will be found in rocks deposited not later than the Middle Triassic 

 and not earUer than the Lower Permian. 



MESOZOIC MAMMALS 



The first actual relics of mammals proper appear in the Triassic 

 contemporaneously with the cynodont reptiles. It is believed that 

 the beginning of mammalian evolution took place about ten million 

 years ago and that the first mammals were very small creatures about 

 the size of rats or mice. Osborn believes them to have been arboreal 

 forms, probably insectivorous, and obliged to lead a furtive nocturnal 

 life. In the daytime they hid among the trees and thickets and at 

 night ventured forth in search of prey. It may well have been crea- 

 tures of this sort that were partially or largely responsible for the 

 slaughter of eggs and young of the great Mesozoic reptiles, for thej' 

 must have lived together during this period. Osborn thinks that 

 these small furry creatures probably resembled the modern tree- 

 shrews, such as Tupaia (Fig. 186, B), a species which he believes to 

 be the most nearly prototypic of the modern mammals. 



In the study of mammalian evolution particular attention must be 

 paid to the two mechanisms whose contact with the environment is 

 the most intimate; the teeth and the feet. For the evolution of the 

 mammaUan orders and families is primarily one of foot and tooth 

 specialization; hence these two characters are of fundamental im- 

 portance in the classification of the Mammalia. 



The teeth of reptiles, except the cynodonts, are simple conical 

 bodies, with little or no regional differentiation. The cynodonts, as 

 we have seen, had incisors, canines and primitive molars; the mam- 

 mals have carried out this differentiation much further. The molars 



