234 DAVIS— THE TEST OF A [April 23, 



abortive pollen suffices to stamp a form as hybrid in character. This 

 represents an extreme view which in consideration of our ignorance 

 of possible physiological reasons for pollen sterility can at present 

 scarcely be claimed as more than an hypothesis. For the Oenotheras 

 we are greatly in need of cytological and physiological studies on 

 pollen sterility more detailed than the incidental observations that 

 have so far been published. 



With respect to the abortion of ovules among the Oenotheras our 

 information is practically confined to the observations of Geerts 

 ('09), mentioned above. It appears that in O. Lamar ckiana and a 

 number of other species only about 50 per cent, of the ovules de- 

 velop embryo sacs. Other species also show varying degrees of 

 ovule abortion. The ovules that fail to mature are represented in 

 the capsules by a fine light brown powder known to all who work 

 with Oenotheras. Such powder is very common in the capsules of 

 various species and their hybrids, and it seems probable that ovule 

 sterility is as widespread in this group of plants as is the degen- 

 eration of the pollen. As in the case of pollen sterility we do not 

 know to what extent physiological conditions may also be respon- 

 sible for the abortion of ovules. 



Pollen and ovule sterility involve of course the elimination from 

 the life history of immense numbers of gametes and raise the fol- 

 lowing questions. Can it be that this elimination throws out of 

 the life cycle types of gametes with germinal constitutions differ- 

 ent from the gametes that matured and that function? It is pos- 

 sible that some of the CEnotheras species, in hybrid condition, reg- 

 ularly mature for the most part particular classes of gametes 

 which in conjugation will perpetuate the genetic line of the parent 

 plant? Gametes even when normally developed may still not func- 

 tion as when pollen grains fail to germinate upon the stigma be- 

 cause its secretions are not suitable. It must also be borne in mind 

 that there are yet other phases of the life history when gametes 

 may become inefifective as through failure to conjugate or because 

 of a high mortality among zygotes, embryos, or young plants ; such 

 forms of infertility are expressed in sterile seeds or in weak ofif- 

 spring which never mature. Possibly the so-called " mutants " arise 

 when unusual gametes from hybrids, occasionally surviving the ex- 



