2o2 PRESIDENTIAL ADDRESS. 



At all events, however, the occurrence of variations is 

 abundantly proved. Next as to the nature of the variations. 

 Mr. Wallace says that the " variability extends, so far as 

 we know, to every part and organ, whether external or 

 internal." But I do not think the evidence brouo-ht forward 

 justifies either the assertion or the denial that all possible 

 variations occur — possible, that is to say, within the limits 

 of the organization of the species. It may be so. But it 

 may be that the variations occur in a number of definite 

 and determinate lines. If this be denied, — if it be said that 

 the evidence shows that variations of all kinds occur in 

 equal proportions, — the opponent of Darwinism will perhaps 

 be tempted to answer : What then is your Natural Selection 

 doing? If variations of all kinds, within the limits you 

 assign, are there in equal proportions, the presumption is 

 that no varieties have been eliminated, and that natural 

 selection, so far as these species are concerned, is inoperative. 

 And the candid biologist must, I think, admit that the 

 evidence in Mr. Wallace's chapter — I refer, of course, to 

 the chapter on the variability of species under nature, not 

 to the work as a whole — while conclusive as to the occur- 

 rence of variations, gives on analysis little or no evidence 

 of any selective agenc}^ at work. I do not regard this 

 absence of evidence as damaging to the Darwinian position. 

 It is due to the incompleteness of the evidence — an in- 

 completeness which also precludes our saying whether the 

 variations are indefinite or along certain determined lines. 



Evidence on this head is indeed particularly hard to 

 obtain. If we look to established varieties and species, 

 and, noting their determinate characteristics, are tempted 

 to regard this as evkleuce of determinate variations, we 

 are met at once hy the Darwinian assertion that these 

 characteristics are determinate through selection froPxi 



