BISEXUAL REPRODUCTION 221 



the case of Oenothera to what extent variation in single 

 genes is occurring, because those who, following De Vries, 

 have studied "mutation" in this multifarious genus of 

 plants, have directed their attention, almost without excep- 

 tion, to the major variations which they have called "mu- 

 tations," and have neglected or denied the existence of 

 minor genetic variations, such as have been studied in much 

 detail in Drosophila. In some cases, such as the "gigas" and 

 the "lata" types of mutation, irregularities of cell-division 

 seem to have resulted in duplication of entire chromosomes, 

 or of the entire set of normal chromosomes. As a comple- 

 mentary phenomenon we should expect entire chromosomes 

 to be lost from the germ-cell in other cases, and this may 

 possibly be the explanation of some classes of Oenothera 

 mutants whose associated cytological conditions have not 

 been determined. In the presence of such striking genetic 

 changes it is not surprising that variations in single genes 

 have scarcely been detected, although the "nanella" and 

 "rubri-calyx" mutants may be mentioned as manifesting 

 simple uni-factorial Mendelian inheritance. It seems prob- 

 able that when the minor variations of Oenothera are studied 

 as intensively as its peculiar "mutations" have been studied, 

 they will be found to be not less frequent in occurrence. 



In domesticated mammals and birds, where asexual repro- 

 duction, self-fertilization, and parthenogenesis are unknown, 

 and where so much racial or family hybridization is con- 

 stantly being carried on with a view to increasing vigor or 

 variability, it Is difficult to say how much of the genetic vari- 

 ability is of contemporaneous origin and how much of it has 

 been handed down in the stock from previous generations. 

 Theoretically, it should be possible to make any stock of 

 animals homozygous for practically every gene by inbreed- 

 ing continued for twenty or more generations, mating brother 

 with sister, parent with offspring, cousin with cousin, or 

 uncle with niece (Jennings). If this is done and genetic varia- 

 tion is subsequently observed to occur, this must have orig- 

 inated after the stock had been purified. By this means we 



