I9I5-] PURE SPECIES OF OENOTHERA. 11 



hastening their germination. A description of a metliod of seed 

 germination which will, I think, prove to be satisfactory in gen- 

 etical work on CEnothera may be found in the Proceedings of the 

 National Academy of Sciences, Vol. I., p. 360, 1915. 



The first investigator to make use of the facts of seed sterility 

 in suggesting Mendelian interpretations of the behavior of La- 

 marckiana and certain CEnothera crosses has been Renner ('14) 

 and his line of investigation has opened a field of research and spec- 

 ulation that must be reckoned with in the future. Renner has 

 studied the seed structure in Lamarckiana, biennis and muricata, 

 and in certain crosses among these forms. His conclusion on the 

 genotype of Lamarckiana will illustrate the principles underlying 

 the method of attack. Since Lamarckiana when crossed with bien- 

 nis and certain other species gives in the Fj^ hybrid generation the 

 twin hybrids Iceta and velutina it may be assumed to develop two 

 classes of gametes which function. These may be spoken of as the 

 IcBta and velutina gametes and are produced in about equal numbers. 

 When Lamarckiana is self-pollinated the Iceta and velutina gametes 

 may combine in proportions to give i pure lata: 2 lata^velutina : i 

 pure velutina. It is a fact that more than one half of the seeds 

 of Lamarckiana fail to develop normal embryos and Renner con- 

 cludes that these sterile seeds represent zygotes homozygous re- 

 spectively for the Iceta and velutina factors. The fertile seeds de- 

 velop from the heterozygotes with both Iceta and velutina factors 

 combined and this combination gives the characters of Lamarcki- 

 ana. CEnothera Lamarckiana may thus be an impure or heiterozygous 

 species breeding true because of the death of such zygotes as carry 

 the factors for Iceta and velutina in homozygous conditions. This 

 simple Mendelian explanation of the behavior of Lamarckiana 

 points a line of interpretation and study certain to be fruitful in 

 CEnothera research. 



Among hybrids of CEnothera the seed sterility sometimes runs 

 extraordinarily high. The most remarkable illustrations of this fact 

 so far known appear in the second generations of crosses involving 

 the Dutch biennis and the Dutch muricata which exhibit certain 

 remarkable morphological peculiarities discovered and described by 

 De Vries ('13). First generation hybrids of reciprocal crosses 



