162 MUTATION. 



species of Oenothera used, the pollen of a true-breeding form 

 can differ geno- typically from the ovules. The breeding true of 

 hybrids is demonstrated and partially explained, but at the 

 same time it is shown that such apparently pure species as 

 Oenothera Muricata and O. Biennis can be heterozygous for 

 very many genes without having for that reason a hetero- 

 geneous descendance. It is clear that we may expect any true- 

 breeding Oenothera, even if it belongs to a recognized species, 

 to be in this peculiar state. The fact that reciprocal hybrids 

 between such species are far from identical, and at the same 

 time breed true, shows that in every such an instance at least 

 one of the species crossed must have pollen which differs geno- 

 typically from the ovules. From the experiments of de Vries it 

 is clear, that Oenothera biennis, Oenothera Lamarckiana, and 

 Oenothera rubrinervis are all three in this condition. The pollen 

 of a pant of Oenothera Lamarckiana need not contain the same 

 genes as the ovules of the same plant. A true-breeding plant in 

 this group may even be heterozygous for a certain gene if we 

 regard it as a male, and heterozygous for other genes if re- 

 garded as a female. 



The interesting facts discovered by de Vries in his exten- 

 sive work, show very conclusively, that here there is a peculiar 

 mechanism, as yet unexplained, which makes that plants, 

 heterozygous for several genes, may yet breed true. 



And it is these facts, in my opinion, which furnish the key 

 to the very peculiar phenomena de Vries had observed long 

 before, in some of these species of Oenothera, notably in O. La- 

 marckiana. The sudden production of new forms in this plant 

 looks very much like mutation, and we know that de Vries has 

 inteipreted them as such, even to the point of generalizing 

 this sudden production of novel forms into a mutation-theory 

 of evolution. 



Several authors, including myself, have tried in vain to in- 

 terpret this "mutability" of Oenothera lamarckiana as a case of 

 complex heterozygosis, with the resulting production of reces- 

 sive forms, given in the genotype of the individuals. The dif- 





