fessor De Vries removed certain plants of this species to his experi- 

 mental garden and found that they contimied to exhibit this wide 

 range of variation, a number of new types arising from the parent 

 form and most of them breeding fairly true to their particular type. 

 When the seeds of 0. Lamarchiana were sown in quantity about 1 

 to 5 per cent of the plants developing from them belonged to one 

 of the new types, the remainder being like the parent form. 



This process of the sudden origin of a series of new types from 

 one parental form De Vries called "mutation," and chiefly on his 

 observations of phenomena of this sort in the Evening Primroses, 

 extending over many years, he founded his mutation theory of evolu- 

 tion. This theory may be very briefly stated as follows : This is 

 the general method of species origin in nature, new forms originating 

 suddenly and breeding true from the start, these jumps or saltations 

 replacing the slow and gradual Darwinian process of na-tura,! selec- 

 tion. Each of the new types presents a series of new unit characters 

 which were not present in the parent species. Some mutants are 

 said to be retrogressive, in which there is a loss of characters through 

 their becoming latent. Others are considered to be progressive 

 mutants in which additional units have appeared which the parent 

 form did not contain. 



In" order to explain the sudden origin of new unit characters 

 in the case of progressive mutants, De Vries made an assumption 

 which to many of us seems unjustified, namely, that at some time in 

 the previous history of the mutating species, representatives of these 

 new units had made their appearance in some mysterious maimer 

 in the germ plasm. The fact that this is purely an assumption 

 without any basis in observation needs to be emphasized. 



My own studies of these plants have led me to the opinion that 

 this assumption is wholly unnecessary and that the phenomena con- 

 cerned may be more reasonably explained in another manner. I 

 have gradually come to view this process of mutation as one of 

 analysis in which each of the new types or mutants is lacking in 

 some character or set of characters which the parent form possessed. 

 This view places quite a different evolutionary value on these phe- 

 nomena, but it would take far too long to discuss these facts in all 

 their bearings on present day views of evolutionary processes at this 

 time. 



According to this view all the mutants, with one exception,^ to 



