THE EOCENE PERIOD. 231 



form, and the assumption of the light springy digitate habit, doubtless 

 through the need of a quick start and a swift flight to escape the car- 

 nivores that were also abandoning the palmate form for the digitate. 

 There was thus a sharp competition for increased speed, on the one hand 

 to escape, and on the other to overtake, and on both sides there was a 

 rising up on the toes with an increase of length of limb and a gain in 

 elasticity. The evolution of the hoofs and of the grinding teeth have 

 been thought to be intimately associated with an increased prevalence 

 of grassy plains. As we have seen, the grasses were present in some 

 abundance as early as the later Cretaceous 1 at least, and they had by 

 this time been given ample opportunity to spread widely, and to fasten 

 upon suitable ground and hold it with that remarkable virility and 

 tenacity which is characteristic of the grasses, and which has made them 

 so important a factor in modern food supplies. The firm turf which 

 the grasses give is quite in contrast with the soft soil of the forests and 

 ungrassed marshes. Because grasses are also much associated with dry 

 and even semiarid grounds, dessication intensifies the firmness of the 

 bottom, and gives additional occasion for the hoof. The tenacious fiber, 

 the siliceous stiffening, and the dryness of the grasses at certain seasons, 

 doubtless gave occasion for effective cropping incisors and grind- 

 ing molars. The foliage of the angiosperms, which was more available 

 for fodder than the needles and spines of the previous gymnosperms and 

 pteridophytes, gave occasion for similar cropping and grinding teeth, 

 and lent their influence to the transition, but served to retain in the 

 forests a notable section of the evolving order. 



Back of these influences lay the physical conditions that promoted 

 them. In the western American region, where the evolution is best 

 known, the great lakes and meandering rivers were characteristically 

 undergoing shiftings. If these followed the method of like modern agents, 

 they left behind them, as they shrank or shifted, a border of grassy or 

 sedgy ground which, on fuller drainage, often became prairie, though 

 this is not the sole explanation of prairies. Such changes were peculiarly 

 suited to the evolution of herbivorous prairie life, and this in turn must 

 have invited its appropriate contingent of predaceous animals. If these 

 considerations be valid, the prime factors in the evolution of the ungu- 

 lates were (1) an undifferentiated plastic animal group susceptible of 



1 Dawson, Plant Life, p. 195. 



