176 THE PROBLEMS OF EVOLUTION 



diflferent from the method of past evolution, but 

 we cannot expect to find it easily nor to disclose 

 it as a simple process. 



After admitting the efficacy of selection, of pre- 

 adaptation, and of isolation in the development of 

 taxonomic units of specific degree or less through 

 the differential assortment of existing characters, 

 and the possible occurrence of new species through 

 mutations of large degree, the remainder of the 

 problem narrows down again to the association of 

 individual adaptations with the species. Of the 

 details of the association I can say no more than I 

 have already written. In the end it is still a mat- 

 ter of speculation, of probabilities, and the most 

 valuable step now to be taken is the search for an 

 effective method of investigation in the light of the 

 successes and failures of the past. 



