EVIDENCE FROM DISTRIBUTION 147 



sea. This may be directly proved for young oceanic 

 islands. Now, since we have all stages from the 

 youngest to the oldest islands and see how the 

 degree of specialization of the animals increases in 

 just such a graduated way, and how their number 

 and kind are different in correspondence with the 

 distance of the island and the animals' power of 

 dispersal and how they everywhere display a rela- 

 tionship with those of the nearest lands, the conclu- 

 sion is cogent that even on the most ancient islands 

 the animals were not created, but immigrated after 

 the islands had been formed. If this be admitted, 

 the further conclusion cannot be escaped, that the 

 greater or less difference of the insular forms from 

 those of their former home must be ascribed to a 

 mutability and capacity of change on the part of 

 the animals." * 



As in the case of all the topics dealt with in the 

 preceding lectures, I cannot but lament the inade- 

 quacy of treatment which is made necessary by the 

 demands of brevity. Nobody would pretend for a 

 moment that we understand all the phenomena of 

 distribution and can account for every seemingly 

 anomalous fact, or that all difficulties have been 

 removed. We are attempting to reconstruct a long 

 history of past changes, and under such circum- 

 stances, partial success is all that can be hoped for. 

 But if, in looking over the vast array of data that 

 have been gathered illustrating distribution and 



1 A. Brauer. in Die Abstammungslehre, pp. 161-2. 



