GEOGRAPHIC DISTRIBUTION OF ANIMALS: CONTINENTS 267 



from its northern neighbor. During most of the Tertiary, South America 

 was cut oflf in this manner. We recall that this period saw the evolution of 

 the placental mammals (Chap. 10). The fact that the evolution of South 

 American mammals followed its own course, largely independently of 

 that of the rest of the world, is doubtless attributable to the fact that during 

 long periods South American forms had no contact with those on other 

 continents. Thus in isolation interrupted only occasionally the peculiar 

 guinea-pig-like rodents (cavy, agouti, capybara, paca, and their kin), the 

 distinctive South American monkeys, the porcupines, the armadillos, the 

 sloths, the anteaters, the opossums, and many other unique animals were 

 free to undergo adaptive radiation only slightly less striking than that of 

 Australian marsupials. 



In somewhat similar manner the animals in the regions of Africa south of 

 the Sahara and adjoining deserts have undergone independent evolution 

 in at least partial isolation. The animals of northern Africa more closely re- 

 semble those of Europe than they do those of central and southern portions 

 of the continent. This again is understandable upon a basis of accessibility, 

 since at various times the barrier presented by the Mediterranean Sea has 

 been bridged — at the narrow Strait of Gibraltar, for example. On the other 

 hand, the deserts form an effective barrier to dispersal of mammals 

 adapted for life in forests, or for life on open plains which are not deserts. 



We have been developing the idea that accessibility and inaccessibility 

 play major roles in the distribution of animals. If an animal is to live in a 

 certain region, ( 1 ) it must be able to reach that region, and (2) the region 

 must be suitable for the existence of that animal. The second point is so 

 self-evident as to need little elaboration. Obviously, for example, animals 

 like frogs which have no adequate means of preventing loss of water from 

 the body cannot live in deserts. Or again, since frogs and toads burrow into 

 the soil to hibernate through the winter, they are not found in regions so 

 far north that the subsoil remains frozen throughout the year. Examples of 

 limitation of distribution by unsuitability of environments might be multi- 

 plied almost endlessly. But from the standpoint of the present discussion 

 chief interest lies in the observed fact that animals do not inhabit all re- 

 gions suitable to them. We cannot conclude that because an animal is not 

 found in a given region the latter is necessarily unsuitable for it. If ani- 

 mals were separately and specially created, failure to find an animal in 

 every region suited to it would be mysterious, to say the least. If, on the 

 other hand, animals have evolved from predecessors which differed in 

 structure and, frequently, in place of origin, failure to find an animal in a 

 region suited to it but inaccessible to it is exactly what we should expect. 



