T Mutation 287 



preference of the one special group of facts — the merlstic — 

 as being in their nature more fundamental and homogeneous, 

 and my object was rather to map out the ground than to 

 erect a definite proposition upon it. The book was to have 

 been followed by similar collections dealing with the other 

 manifestations of variation ; but with the development which 

 genetics almost immediately underwent, it became clear that 

 the method of miscellaneous collection was no longer the 

 most direct, and that by experimental investigation of special 

 cases progress of a far more valuable order was possible. 



Views somewhat similar to those that I had formed from 

 a general survey of the facts of variation were shortly 

 afterwards published by de Vries in his famous book Die 

 Mutationstkeorie, 1901-3. Having at command a mass of 

 evidence far larger and more coherent than mine he was 

 at last successful in bringing workers of many schools to 

 give these suggestions a serious consideration. For the 

 first time he pointed out the clear distinction between the 

 impermanent and non-transmissible variations which he 

 speaks oi^j^ fluctuations, and the permanent and transmissible 

 variations which he calls mutations. Of his proofs, the 

 most striking, to many the most convincing, is that provided 

 by his study of Oenothera^ in which he witnessed the actual 

 occurrence of sudden departures from type — not one but 

 several — by which at one step *in descent distinct and 

 frequently pure-breeding types were produced. Whatever 

 be the true interpretation of these particular observations, 

 they manifestly provide examples of something so like the 

 generation of new species that in any future discussion of 

 Evolution they cannot possibly be passed over. 



We may be doubtful of the validity of the superstructure 

 which de Vries has created, and yet in full agreement with 

 him in recognizing the fundamental truth, t hat there js_a_ 

 natural distinction between fiuctuational variations and actu al 

 genetic variations ; that the latter are those alone bv which 

 perma nent evolutionary change of type cq n b<^ pffprf-prl ; and 

 that commonly, though, as it seems to n ^e^ not a lways, the 

 ^teps~b y which such cha nj ^es ocm r are s o discontinuous as 

 to~merit th e name Mutations^ 



CIt is^^at this point that Mendelian discovery aids. 

 Whereas formerly, though the fact of Discontinuity was not 



