210 Hints on the 



are often comprised within a single marine province. Tlius the 

 terrestrial faunse of Cuba, St, Domingo, Porto Rico, and 

 Jamaica, are distinct from each other ; and the same is true to a 

 great extent of the "West Indian Islands generally ; those of the 

 Bahamas and Bermudas are also distinct. But a single marine 

 province extends from the Bermudas to the southern part of 

 Brazil. 



14. The attempt to divide up the earth into one set of 

 zoological provinces for all animals, whether aquatic or terres- 

 trial, is therefore futile. 



15. The converse of proposition 13, in respect of bodies of 

 water surrounded by land, does not hold true. In any one 

 lake or river, the proportion of species peculiar to it is smalL 

 This may be said to be owing to the means of communication 

 in the case of rivers emptying into a lake or chain of lakes. 

 But the same extent of distribution is equally common in 

 respect of unconnected rivers and isolated ponds. It may,, 

 therefore, be inferred that the original plan of creation was 

 different from that of the insular species. 



16. Because change of climate is always nearly or quite in 

 one direction (N. and S.), and is always more or less gradual, 

 differences of climate do not cause several zoological provinces 

 of one sort to be comprised within zoological provinces of 

 another sort, but merely render their boundaries indistinct. 



17. Because zoological afiinities have some connexion with 

 distribution (§ 11), indeflniteness of boundaries will result from 

 the attempt to comprise any considerable portion of the animal 

 kingdom in one scheme of zoological provinces. Absolutely 

 distinct, that is, linear boundaries exist only between single 

 species, except that the boundaries between terrestrial and 

 insular faunae are linear, and those between indistinct species 

 are necessarily ill defined. 



The cause of indefiniteness of boundaries is to be found in 

 the fact that the areas of different species do not exactly 

 coincide; although sometimes several nearly coincide, as at 



