ON BIOLOGY. 29 



various parts of the world, but the diversity in the physi- 

 cal geography of the earth's surface is another powerful 

 factor in determining faunal and floral limits. Plains 

 may encourage extended migrations, as do hills, moun- 

 tains, large rivers and seas check the same. The abun- 

 dance or absence and scarcity of certain foods over certain 

 areas also have an Influence. In many cases the faunal 

 limits can be but loosely defined, as any faunal district 

 may almost imperceptibly merge into any one of its ad- 

 joining ones, but certain surface contours may often, on 

 the other hand, sharply define some of the boundaries, 

 and animals of the contiguous areas not pass over the 

 line. Some species range through several faunal areas, 

 others are never found beyond, perhaps, some quite lim- 

 ited locality, these several restrictions and diffusions 

 nearly always being due to assignable natural causes. 

 Where certain species arrive at their maximum of 

 abundance it is generally supposed to indicate the area of 

 the origination of that species. 



Soils also play their part, and it will be clear that the 

 fauna of extensive marshy districts are sure to be quite 

 unlike the fauna of a great desert; indeed, in the case of 

 the latter, it has led an eminent authority to say that 

 "deserts may act much as inland seas to separate the ani- 

 mals of the adjoining more fertile tracts, and they afford 

 dwelling-places for animals which are incapable of living 

 elsewhere. Desert faunae have a general fades the world 

 over, though the original elements out of which the 

 faunae have been made up may radically differ." 



Although a great ocean, such as the Pacific or the 

 Atlantic, forms almost a positive barrier to the spread of 

 the land animals of its opposite coasis, which are in- 

 variably quite distinct, yet at the same time, within its 

 own self, the great ocean currents form means of dispersal 

 of prime importance for many land animals and the vast 

 majority of marine forms. Again, as we have already 

 pointed out, climate in most cases is a positive barrier to 

 the spread of animals in many directions, whereas the 

 alternation of the seasons, and even the severity or mild- 

 ness of the Summers and Winters powerfully urge many 

 animals to migrate, and thus for the time being, at least, 

 to pass beyond either their local or general faunal areas. 

 Migrations from such causes, in the main, however, take 

 place within the boundary of any faunal region where 

 they occur. A familiar example of this is our own 

 United States vernal and autumnal migration of birds 



