Evolution of Evolution-Theory. 231 



There is not any obvious way of proving or disproving 

 an ingenious hypothesis of this sort, but it is in line 

 with the central idea of Darwinism. If a process of 

 germinal selection can be admitted as aiding and abet- 

 ting the processes of selection at higher levels (intra- 

 selection and individual selection), a new strength is 

 given to the general selectionist position. 



(b) Mr. Bateson's great work, entitled Materials for 

 the Study of Variation, is an endeavour to get out of the 

 speculative mire in which, to the physicist's contempt, 

 the biologist still flounders. It is an attempt to get 

 beyond the vagueness of the assumption that "varia- 

 bility exists " to a sure knowledge of what variations do 

 actually occur. Life is so abundant and so protean 

 that we draw cheques upon nature almost ad libitum, 

 and in our impetuosity scarce wait to see whether they 

 are honoured. 



By an examination of specimens in many collections 

 and museums, by detailed investigations in regard to 

 particular cases of importance, and by careful sifting 

 of recorded instances of variation, Mr. Bateson has 

 given us a sound foundation upon which to build. It 

 must be noted, however, that he has as yet confined 

 himself almost entirely to one kind of variation, which 

 he terms meristic, i.e. variations in the number, sym- 

 metry, and arrangement of parts. He leaves to a future 

 volume almost all discussion of substantive variations, 

 that is to say, changes in quality and substance, which 

 to most biologists are probably of greater interest. 

 Many of the variations with which he deals, such as 

 branched legs in insects, are not of the sort which we 

 suppose to have furnished the raw material of evolu- 

 tionary progress. In fact, they are too " monstrous ". 



As is well known, the ordinary, though not universal, 

 conception of the process of organic evolution is that 

 from an ancestral form by minute and, at first, almost 

 insensible differences a new form arises. The minute 

 variations may be indefinite and indeterminate, as most 

 Darwinians follow their master in believing; or they 

 may be definite and determinate, along particular lines, 

 as is suggested from many sides, by Lloyd Morgan with 



