188 



SCIENCE 



[N. S. Vol. XXV. No. 631 



2. The essence of de Vries's experiments, 

 pedigree-culture, consists of ' selection ' and 

 ' segregation.' This becomes most evident in 

 MacDougal's description of his methods as 

 given in the paper under discussion (p. 214 

 f.), and anybody may see at a glance that 

 MacDougal, as well as de Vries, did nothing 

 that has not the purpose of selecting certain 

 variations and their seeds, and of segregating 

 (separating) them from disturbing influences. 



Granting that these fundamental views are 

 correct, we may now look at the bare facts 

 represented in de Vries's experiments, without 

 any attempt at explanation or theoretical 

 speculation. The following two stand out 

 prominently : 



1. By pedigree-culture de Yries succeeded 

 in making certain variations hreed true. 



2. In other cases of variations he did not 

 succeed. 



The first sentence tells an old story. The 

 same has been done since times immemorial, 

 and scientific investigation has taken notice 

 of this fact since the time of Darwin. The 

 process is now rather well understood, that is 

 to say, with reference to the essential features 

 of the action required of man : they are selec- 

 tion and segregation. De Vries did not 

 change the old method in the slightest degree, 

 he only introduced additional precaution and 

 refinement in detail, taking particular pains 

 to insure the full efiiciency of these two fac- 

 tors by carefully excluding all possible inter- 

 ference with them. In addition, he was the 

 first to keep proper scientific records of what 

 he was doing. 



The second fact, on the contrary, is new, 

 and it is the point in de Vries's experiments 

 which needs explanation and a theory. Why 

 is it that certain variations did not breed true 

 under de Vries's hands, although they were 

 treated exactly like those belonging to the first 

 group? The general belief, up to this time, 

 was that any variation might be transformed 

 into a true breeding form by proper treatment. 



De Vries's explanation of this fact is given 

 in his mutation theory. Believing that his 

 experiments are conclusive, and that, since he 

 himself did not succeed in cases of the second 

 group, noiody would he ahle to do so, he pro- 



pounds the theory that there are actually cer- 

 tain variations, in which selection and segre- 

 gation (pedigree-culture) are impotent to pro- 

 duce true breeding, and, consequently, that 

 there are two classes of variations, the one of 

 which he calls ' mutation,' which produces 

 forms which respond to the effort of the breed- 

 er, the other in which the art of the breeder 

 has no effect, and which he calls ' fluctuating 

 variation.' Then the first is, of course, all 

 important for the species-forming process, 

 while the other is of no consequence. There 

 is no saying, with respect to any particular 

 variation, whether it may belong to the one 

 or to the other class, before the actual test 

 (pedigree-culture) has been made, although 

 it seems that mutations often or generally 

 differ from fluctuating variations in the de- 

 gree of deviation from the original form. 

 This is the essence of the mutation theory. 



The above conclusion and theory would be 

 perfectly correct, if the proposition was cor- 

 rect that it is actually impossible to make 

 certain variations breed true. But just in 

 this point, I believe, de Vries is wrong, since 

 his experiments were not conducted in such a 

 way as to absolutely preclude the possibility 

 that even so-called ' fluctuating variations ' 

 may be successfully transformed into true 

 breeding forms. We always are to bear in 

 mind that it is at least thinkable that a par- 

 ticular form ' may be bred true only under 

 particular conditions, under conditions which 

 are congenial or essential to its very existence. 

 To present an imaginary example: a plant 

 species may possess a peculiar variation, which 

 is due to lack of direct sunlight (shade-form). 

 Suppose this shade-form is cultivated accord- 

 ing to de Vries's method in the botanical 

 garden, in beds where it gets its full share of 

 sunlight. I never believe, in such a case, that 

 pedigree-culture will succeed in making this 

 shade-form breed true, since always the con- 

 ditions of environment will have the tendency 

 to paralyze the effort of the breeder. If, how- 

 ever, this particular shade-form is bred in the 

 shade, under proper environment, the attempt 

 possibly may not be in vain. 



This is only an example to illustrate what 

 I thinlc may be correct. Every plant breeder 



