CAUSES OF THE EVOLUTION AND EXTINCTION OF THE TITANOTHERES 



879 



Number of extinct and surviving families of large and of small 

 mammals since Cretaceous lime 



The above figures would seem to indicate that the 

 chance of survival of small mammals is nearly twice 

 that of the survival of bulky mammals, but when 

 we consider that many extinct families of small mam- 

 mals remain to be discovered among the Marsupialia, 

 Insectivora, Cheiroptera, and especially the Rodentia, 

 the disparity indicated by the figures given is not so 

 great. 



Among the large surviving land mammals are the 

 families of elephants, hippopotami, rhinoceroses, 

 giraffes, elands, and elks, and among the sea mammals 

 are walruses and the various families of Cetacea, 

 which, so far as known, are the largest mammals of all 

 time. 



Further, we may recall the facts (1) that in lower 

 Eocene time the small Condylarthra became extinct 

 far earlier than the large Amblypoda; (2) that in 

 lower Oligocene time many families of small Artio- 

 dactyla and Perissodactyla became extinct at the 

 same period as the large titanotheres, lophiodonts, 

 and amynodonts; (3) that in Pleistocene time the 

 relatively small Mylodon disappeared as early as the 

 large Megatherium; (4) that the extinction of the 

 mammoth in North America during or after the glacial 

 epoch attracts attention because of the animal's large 

 size, but, as indicated above, many smaller quadrupeds 

 disappeared at the same time — for example, the horses 

 and the sloths. 



Bulk and slow breeding not inimical to rapid evolu- 

 tion. — The following arguments of Wallace and of 

 Andrews in regard to bulk and slow breeding are in 

 part fallacious and receive little support from paleon- 

 tology. Wallace remarks (1876.1, vol. 1, pp. 158-159) : 



There is, however, another cause for the extinction of large 

 rather than small animals whenever an important change of 

 conditions occurs, which has been suggested to me by a cor- 

 respondent but which has not, I believe, been adduced by Mr. 

 Darwin or by any other writer on the subject. It is dependent 

 on the fact that large animals as compared with small ones are 

 almost invariably slow breeders, and as they also necessarily 

 exist in much smaller numbers in a given area, they offer far 

 less materials for favorable variations than do smaller animals 

 In such an extreme case as that of the rabbit and elephant, the 

 young born each year in the world are probably as some mil- 

 lions to one; and it is very easily conceivable that in a thousand 

 years the former might, under pressure of rapidly changing 

 conditions, become modified into a distinct species, while the 

 latter, not offering enough favorable variations to effect a 

 suitable adaptation, would become extinct. 



The above argument, however, is not in accord with 

 the facts ; the slow-breeding elephant evolved with far 

 greater rapidity than the swift-breeding mouse. 



C. W. Andrews has recently (1903.1, p. 2) revived 

 this argument that the lengthening of the time taken 

 to attain sexual maturity may affect the rate of evolu- 

 tion and under changed conditions where a rapid rate 

 of evolution is essential may cause extinction. He 

 writes : 



In many ungulates this increased longevity is indicated by 

 various modifications of the teeth, tending to give them a 

 longer period of wear: generally this end is attained by the in- 

 creasing hypselodonty of the cheek-teeth. A necessary con- 

 sequence of the longer individual life will be that in a given 

 period fewer generations will succeed one another, and the rate 

 of evolution of the stock will therefore be lowered in the same 

 proportion. If now the conditions of life undergo change, the 

 question whether a given group of animals will survive or be- 

 come e.xtinct will depend upon whether it can undergo suf- 

 ficiently rapid variation to enable it to avoid getting so far out 

 of harmony with its surroundings that further existence becomes 

 impossible. It seems to follow, then, that the smaller animals, 

 in which the generations succeed one another rapidly, will have a 

 better chance of surviving than the larger and more slowly 

 breeding forms, which at the same time will be stiU further 

 handicapped if, as is usually the case, they are more highly 

 specialized than the smaller forms and therefore have a more 

 restricted range of possible variation. 



This argument is contradicted by all the facts of 

 paleontology: there is no relation between rapid 

 breeding and rapid evolution. 



As against these purely hypothetical considerations 

 paleontology shows that during Pliocene and Pleisto- 

 cene times the slow-breeding Proboscidea evolved 

 quite as rapidly if not more rapidly than the rapid- 

 breeding Rodentia. 



Relation oj bulJc to nutrition. — Stromer (1905.1, pp. 

 97-132) discusses in detail the relation of bulk to 

 nutrition. He observes that nourishment or lack of 

 it must not be overestimated as a factor in extinction, 

 for it has been shown among mammals that the nourish- 

 ing surfaces are only squared as the mammals increase 

 in size, while the bulk of the body is cubed, and that 

 small forms eat much more relatively than large forms. 

 The fact remains that the great terrestrial animals need 

 as a rule a large amount of vegetable food and also an 

 abundance of water. Aquatic mammals, on the other 

 hand, embrace more giant carnivores (cetaceans, 

 walruses) than herbivores (hippopotami, Rhytina). 

 Naturally, defenseless giant forms incapable of fight- 

 ing can maintain their existence only in the absence 

 of destructive Carnivora — for example, the giant rodent 

 {AmUyrliiza) of the Lesser Antilles. Such factors as 

 the diminution of food supply would weigh much more 

 heavily against ponderous animals like Glyptodon and 

 Megatherium than against agile ones such as giraffes, 

 elephants, and rhinoceroses. While the glacial epoch 

 may have destroyed the northern representatives of 

 Hippopotamus, it must be recalled that this genus at 



