THE COEFFICIENT OF MUTATION IN OENOTHERA BIENNIS L. 7 



Gates assumes that crosses between species or between elemen- 

 tary species often occur in nature among allogamous or openflowered 

 forms. 1 ) But, according to my own experience, even in such cases 

 hybrids are rare in the wild state, and hybrid races must be much 

 rarer still. The slightest degree of weakening of the individual 

 vigor will doom such hybrids to extermination, even as most of 

 the occasional whire flower mutations in nature disappear sooner 

 or later, without starting a permanent variety. 



In order to save the hypothesis of hybridism as a cause of the 

 mutable condition of the evening primroses, different authors have 

 proposed different auxiliary suppositions. And since the possi- 

 bility is acknowledged that mutability may be far more widely 

 spread within this group than we now know, such suppositions 

 must not be of a limited nature, but applicable to large divisions of 

 the vegetable kingdom. Kearney, in studying the mutations of 

 the Egyptian cotton, comes to the conclusion that these and other 

 mutations might be the result of crosses between remote ancestors, 

 but that these crosses have left no other traces in their descendants 

 than "the disturbance of germinal equilibrium, which manifests 

 itself in the production of mutants." 2 ) It is not very clear how 

 this supposition is to bring the problem nearer to its solution. 



In a recent article in the Botanical Gazette 3 ) Jeffrey takes an op- 

 posite position. He assumes that the ancestral crosses have left 

 another visible trace in their descendants, which is the partial sterility 

 of their sexual cells. It is a well known fact that many hybrids have 

 partially sterile pollen, while acknowledged species have, as a rule, 

 only fertile pollen grains. Jeffrey assumes this rule to be without 

 exceptions, but does not adduce any arguments in favor of this 

 hypothesis. It is difficult to judge the value of an argument so 

 long as the facts upon which it rests have not been submitted to 

 criticism. But I might suggest that it seems rather hard to recon- 

 cile this view with the fact that in angiosperms three of the four 

 megaspores are usually sterile, while only one produces an embryo 

 sac. Are we to deduce from this fact, in connection with Jeffrey's 



i) Gates, R. R., Mutation in Oenothera. Amer. Nat. 45:577 — 606, 

 191 1 ; see pp. 578 — 579- 



2) Kearney, T. H., Mutation in Egyptian cotton. Jour. Agric. Research 

 2:287—302, 1914. 



3) Jeffrey, E. C., Spore conditions in hybrids and the mutation hypo- 

 thesis of De Vries. Bot. Gaz. 58:322 — 336, 1914. 



