LIMITATION BY BARRIERS. 183 



Such a distribution is not out of harmony with the 

 view that species are specially created for no dis- 

 tribution could be out of harmony with this idea, 

 but it will hardly be denied that the geographical 

 distribution of groups indicates a law of distribu- 

 tion : a law unexplained by special creation, but one 

 which must inevitably follow from the genetic de- 

 scent of animals and the origin of species from older 

 varieties. 



Do Barriers Limit the Areas of Specific Distri- 

 bution ? 



To this question the answer is that, as a rule, species 

 are not found on the opposite sides of impassable 

 barriers. But to this rule there are some remarkable 

 exceptions of species which are not thus limited by 

 seemingly natural barriers. And while the instances 

 of agreement with the rule outnumber the excep- 

 tions one thousand to one, it is the exceptions to the 

 rule which must occupy our attention. The most 

 important exceptions are the existence of the same 

 species of fish on either side of the Isthmus of 

 Panama ; the presence of the same species of fresh- , 

 water animals on the two continents ; and the v 

 presence of certain Arctic species of plants on the 

 mountains of Southern Europe and America. These 

 cases will be taken up, therefore, as illustrations of 

 the classes of difficulties and the method of meeting 

 them. 



Such cases of discontinuous distribution are not 

 to be explained by assuming a formerly wide dis- 

 persal, for how can we suppose that with any 

 amount of dispersal the barriers in question could 



