Duration of the Process of Selection. 91 



of independent types. "Every race of plants possesses 

 only a low degree of constancy," says one of our most 

 considerable agricultural authorities, Prof. Kurt von 

 RuMKER.^ Without a continuation of selection they 

 would soon lose their good qualities. In this respect 

 they behave quite differently from a true species or con- 

 stant variety. 



There is no need for me to go further into this matter 

 now, for I shall make it my business in the following 

 paragraphs to point out what the experience of breeders 

 can teach us on this point. 



Finally I should like to discuss a very instructive 

 example more thoroughly. I refer to the important ob- 

 servations of R. VON Wettstein on seasonal dimor])hism 

 as a point of departure for the origin of species in the 

 vegetable kingdom.^ He deals with the genera Gentiana, 

 Euphrasia, Alectorolophus (Rhinanthus). In the alpine 

 meadows there occur in the case of many species of these 

 genera two forms (varieties, subspecies, or elementary 

 species) of which one flowers early and the other late. 

 Moreover the early and late flowering forms of one pair 

 are usually distinguished from one another by a series 

 of further characters of the value of the differences be- 

 tween elementary species. 



The time for mowing the grass in Central Europe 

 falls between the flowering season of these two varieties. 

 The early kinds ripen their seeds before this time ; the 

 later only begin their main growth after it. 



The work of von Wettstein clearly shows that 

 these species are associated in pairs and has proved that 



^ Der wirthschaftliche Mehrwertli guter Culfiirvanetdten, 1898, 

 p. 136 of the offprint. 



^ Bcrichte d. d. hot. Gesellsch., 1895, Bd. XIII, p. 303, and Bot. 

 C entralblatt , 1900, No. i, p. 15. 



