424 The Living Plant 



line, where issues are doubtful. It must suffice to say that our 

 knowledge of these subjects is in process of active extension at 

 this moment. 



Fourth, exact study devoted to determining whether the selec- 

 tion of variations, in the Darwinian sense, can actually produce 

 a new species have given very largely a negative result, — much 

 evidence tending to show that selection simply isolates the bio- 

 types, but cannot in any way alter them. If, however, biotypes 

 originate from other biotypes, as it seems that surely they must, 

 then the method of evolution would be substantially that im- 

 agined by de Vries for his mutants, and that represented in our 

 comparative diagram (figure 172). Thus, selection would still 

 rank as the great decisive, though not as an originating, factor 

 in evolution. As to adaptation, that still stands as a corollary of 

 any kind of evolution by selection, for selection imposes a step- 

 by-step development in touch with the environment. The con- 

 ception of biotypes is wholly consistent therewith, and indeed 

 helps to explain some of the peculiarities of adaptation, — es- 

 pecially the somewhat loose, clumsy, or generic character that 

 most adaptation displays in conjunction with the occasional 

 existence of highly exact fitness. In general in Nature, structure 

 fits function about as well as a man's physique fits his trade, — 

 that is, always in a general way, and sometimes very exactly. 

 We cannot expect rigid biotypes to fit intergrading environments 

 any more than we can expect polygons to match circles, — though 

 with some many-sided kinds, the correspondence can be appre- 

 ciably close. But it is perfectly clear that the first great problem 

 of present-day experimental evolution is the determination of the 

 origin of biotypes, or, to be exact, of the variations or characters 

 which constitute biotypes. I should not be surprised if it were to 

 turn out that the origination of new characters or biotypes is a 

 normal function of organisms, adaptively acquired by them pre- 

 cisely as any other physiological function has been, and represents 

 their method of securing survival in changing environments. It 



