No. 490] ASPECTS OF THE SPECIES QUESTION 261 



sought are more or loss coordinate. In this process, it is 

 probable also that light will be thrown on the question 

 of their derivation from a common stock or from two 

 or more stocks. 



The four methods by which new forms originate have 

 been discussed before this society in a paper read last 

 year. These have also been given in some detail in print 2 

 and are so generally well known to botanists that de- 

 tailed consideration of them is unnecessary here. These 

 four methods are adaptation, mutation, 'variation and 

 hybridation. Variation alone has not yet been certainly 

 established by experiment as a method by which new 

 forms originate, but the evidence of it is so strong that 

 it amounts to presumptive proof. It seems certain that 

 for plants at least it is not the principal method, as 

 Darwin thought, and that it is probably much less im- 

 portant than adaptation, and probably also less impor- 

 tant than mutation. It is, however, a much more obscure 

 process, and it is impossible to tell in the absence of any 

 critical experiments just what it is, or what its impor- 

 tance may be. Darwin has said: "I have spoken of 

 variations sometimes as if they were due to chance. This 



acknowledge plainly our ignorance of the cause of each 

 1 »articular variation." The exact measurement of 

 habitats has made it clear that minute variation in the 

 factors of the habitat is a very important if not the ruling 

 cause in the production of minute variations in structure. 

 In the case of the minute initial variations shown by the 

 individuals of a group, it is impossible to tell at present 

 how much is due to heredity and how much to causal 

 variation of the habitat. It seems probable that the 

 former may control in some forms, and the latter in 

 others. Furthermore, while hybridation stands apart 

 sharply from the others as a process in which existing 



