MUTATION THEORY OF DE VRIES 93 



forms. They seemed to arise by leaps and bounds, by organic 

 jerks ; they illustrated what De Vries has called " Mutation." 



Besides the smooth-leaved O. l&vifolia and the short-styled 

 O. brevistylis, both of which appeared in the potato-field, the cultiva- 

 tion of O. lamarckiana resulted in the emergence of seven constant 

 elementary species O. gigas (rare), O. rubrinervis, O. oblongata, 

 O. albida, O. leptocarpa, 0. lata, and a dwarf O. nanella. Besides these 

 there were a few inconstant variants and a few which were sterile. 



One form, O. scintillans, that only appeared eight times, was not 

 constant like the others. When self -fertilised it produced O. ob- 

 longata, O. lamarckiana, and others like itself. 



It is interesting to notice that some of the forms e.g. O. oblongata 

 were produced over and over again ; that five of the new forms 

 appeared afterwards in the field or from seeds collected in the field, 

 which shows that the cause of their origin was not to be found in the 

 cultivation. 



As De Vries says, the new elementary species arise suddenly 

 without transitional links ; for the most part they are quite con- 

 stant ; within the limits of their essential constancy they exhibit 

 similar minor fluctuations ; they are usually represented by nu- 

 merous individuals within the same period of time ; the observed 

 changes affect many organs and parts, and in no definite direction ; 

 and the mutability seems to be periodic, not continuous. 



If cases like that of 0. lamarckiana are indicative of what often 

 occurs and has occurred in nature, then our view of the evolution- 

 process must be in several respects modified. 



It will be necessary to distinguish more sharply between fluc- 

 tuating variations and discontinuous mutations. If a new ele- 

 mentary species may arise as it were ready-made, " at a single 

 advance," it is not necessary to hold to the formula that species 

 have arisen by the gradual accumulation (under selection) of 

 minute individual variations. As mutations occur in large 

 numbers and occur repeatedly and are very constant, the familiar 

 difficulties in regard to the swamping of novelties, the inappre- 

 ciable value of incipient stages, the apparent non-utilitarian 

 character of some specific differences, and so on, will be greatly 



