Problems and Concepts of Evolution 15 



a later chapter data will be found dealing with this problem, showing that the 

 statement is true or not true, depending entirely upon the definitions employed 

 and the point of attack from which the investigation is made. 



With regard to the mutations of De Vries, I am of the opinion that De Vries 

 has described a specific operation, which is not the same as the formation of 

 sports and saltations described by other writers, and I would distinguish, there- 

 fore, a process of mutating as distinguished by De Vries and shown by him to be 

 present in Oenothera lamarckiana, and a saltation process in which there is a 

 single sporadic production of one or more divergent individuals as the result of 

 the single application of the productive force to the parental strain. Both of 

 these types of the production of new conditions in the qualities of organisms 

 are experimentally confirmed in these reports, in addition to those already 

 existing in the literature elsewhere, and should be recognized as distinct as to 

 causation. To these a third process must be added, which also produces dis- 

 continuous changes either repeatedly in the same strain or rarely as an apparent 

 sport. Morgan's experience with germinal disintegration in Drosophila and my 

 experience with this same process in my materials leaves no doubt of the exist- 

 ence of this third category as a general type of operation, the products of which 

 are all now collectively characterized as " mutations." 



The three processes are alike only in the sense that they produce at one opera- 

 tion new aspects in the organism which are discontinuous with the parent 

 stock. In mutation there is a process operating in which there is repeatedly 

 produced in succeeding generations one or more new types which appear again 

 and again through a series of generations. I have shown in some experiments 

 how such a race can be produced experimentally and give products which are in 

 every way in animals the duplicate of those described by De Vries in Oenothera 

 lamarcTciana, and hence I think it necessary to recognize a " mutating process " 

 as distinct from the others which may also give products of quite diverse natures. 

 The sporting process seems uniformly to give but a single result, following the 

 application of the productive cause, and there is thus far nowhere in any of my 

 experiments, or in the modifications produced by MacDougal, Kammerer, or 

 Woltereck, any indication of repeated or of subsequent production from the race 

 of the new types. In many of my examples the change is distinctly that of 

 adding a character to the organism, in others it is the displacement of the 

 character, and so no doubt the sporting process works both ways, either to add 

 or subtract characters to or from the race. 



In germinal disintegration there is present an operation in which characters 

 are removed from the race one after the other, but which can be put back into 

 the race by proper crossing methods, thus restoring the original ; and differing 

 in many respects from the products of mutation or the products of sporting. 

 This germinal disintegration is possibly the process which is largely responsible 

 for the production of large numbers of domesticated varieties, which can be 

 shifted back to the original condition merely by adding the requisite character. 



In these three types of change there is concerned solely the action of genetic 

 and environic factors, either through the combination of agents hitherto sepa- 

 rated, and hence with the resulting appearance of a new character, or the actual 

 production of new factors, and the removal of agents, all of which show that the 

 entire series of operations resolves itself into the study of the factorial composi- 

 tion and operation of the gametic substance. 



