Mat 11, 1906.] 



SCIENCE. 



747 



tain that the mutations have bred true. They 

 surely did not do this in the beginning of the 

 experiments, since they were throwing oS, in 

 each generation, additional mutants, and it 

 was only after some time that de Vries suc- 

 ceeded in obtaining a relatively pure strain. 



Consequently, de Vries's contention that 

 mutations are species is not supported at all 

 by his experiments; whatever they are, they 

 are not species, since they do not show the 

 characteristic features of such. 



However, if de Vries had claimed that spe- 

 cies might be made out of mutations, nothing 

 could be objected to this view; but this is no 

 new idea. Similar experiments have been 

 undertaken by animal and plant breeders, and 

 a large number are on record. In fact, the 

 breeding of domestic races has always been 

 regarded as a process analogous to the one in 

 nature by which new species are produced. 

 But the main features of this process in 

 nature as well as under domestication are 

 selection and segregation. This is exactly 

 what de Vries has done with his mutations : 

 he selected and segregated them (preventing 

 crossing), and thus he imitated nature's way, 

 and finally obtained more or less pure strains, 

 which are analogous to natural species. But 

 before he began this process of selecting and 

 segregating, the mutations were by no means, 

 species, but only varieties. 



Aside from the above claim, de Vries further 

 maintains that it is the mutations, and not 

 the variations, that give origin to new species, 

 and he thinks that there is a fundamental 

 difference between them. However, I have 

 been unable to see where he draws the line 

 between variations, constituting small steps, 

 and mutations representing sudden leaps, and 

 I do not think that he has solved the old 

 sophistic problem of how much must be added 

 to a small thing in order to make it a large 

 one. His discussion of unit-characters does 

 not offer any help in this respect, since in 

 many cases he confesses himself that he does 

 not know what should be regarded as a unit- 

 character. 



Mutations are by no means as frequent as 

 de Vries would fain make us believe. He 

 concedes himself that he had considerable 



trouble in finding a fit object for his experi- 

 ments, and, indeed, among living animals and 

 plants in the wild state, mutations in de 

 Vries's sense are extremely rare, and in this 

 respect I agree entirely with Merriam's con- 

 tention, not only with reference to animals, 

 but also to plants. True mutations, that is to 

 say, variations which represent sudden leaps, 

 are found chiefly among domestic forms, and 

 this fact, I think, is well established; and the 

 form that finally furnished the material for 

 de Vries's experiments, (Enothera lama/rcki- 

 ana, is a domesticated, a garden form, and 

 not a native species of Europe. It is true, 

 it lately has become a habit with some biolo- 

 gists to hunt for mutations in nature, but the 

 search has been quite unsuccessful, for the 

 so-called mutations in part do not at all rep- 

 resent sudden leaps ; in part it was not consid- 

 ered worth while to investigate whether the 

 sudden leaps discovered were connected with 

 the original form by transitions or not. 



Paleontological evidence for the former ex- 

 istence of mutations should be excluded from 

 the beginning, since it is in the very nature 

 of paleontological facts to be fragmentary, 

 and, in this connection, it is well to call atten- 

 tion to the former use of the term mutation 

 by paleontologists (Waagen, Neumayr, W. B. 

 Scott) ; it means just the opposite of de Vries's 

 mutation, namely, a change during phylo- 

 genetic development, which is characterized 

 by slow, small, almost insensible steps. For 

 this we possess positive proof. 



Thus we arrive at the conclusion that de 

 Vries has not made good his claim that muta- 

 tions are species, since his conception of spe- 

 cies is defective. If he should change his 

 view, and claim that species could be made 

 out of mutations, he would be right, but then 

 it is the selection, and chiefly the segregation, 

 that has this effect; and further, this would 

 be no new theory. If he claims that it is mu- 

 tation as distinguished from variation that 

 starts the species-forming process, we must 

 point out that mutations are rare in nature, 

 that there is no sharp line to be drawn between 

 mutation and variation, and that mutation 

 has always been regarded as a special form 

 of variation (sporting, halmatogenesis). Con- 



