Genetical Stuilies on Oenothera. V. 203 



meal, about 1 Idlonrani to the s(|uare meter, but even with tliis special 

 care the cultures proved absolutely sterile when selfed (105 trials) and 

 my experience with respect to the sterility of this cross was the same 

 as that of de Vries ('13, p. 181). From gigas pollinated by tmiricata 

 I obtained very small, shrunken seeds (?) obviously abortive. Some 

 1300 of these structures were sown with the result that only 1 plant, 

 13'31a, was obtained from seed pans kept 6 — 7 weeks. The reciprocal 

 cross, muricata X gigas, yielded seeds of healthy appearance and from 

 224 of these, in seed pans for 5 — 7 weeks, culture 13"32 was obtained 

 of 42 plants. 



The single plant 13'31a, gigas X muricata H, and the 42 plants 

 in culture ]3'32, muricata H X gigas, proved to be similar to one 

 another. They were also essentially indistinguishable from the broader- 

 leaved hybrids previously described in which the muricata parent was 

 of the race from Woods Hole. There were no narrow-leaved forms 

 among the 42 plants of culture 13"32, muricata H X gigas, such as 

 have been noted by de Vries ('13, p. 181) for the same cross and 

 such as appeared in my earlier studies in which the muricata from 

 Woods Hole was involved. Perhaps the small number of plants in the 

 culture did not happen to include any representatives of the narrow- 

 leaved tj7)es. 



General Considerations. 



The reader of this account will have noted that in the descriptions 

 of the reciprocal crosses the terms patroclinous and matroclinons have 

 throughout been used in a relative sense. I have observed no certain 

 evidence that a morphological character of either species in a cross is 

 passed on to the Fi hybrids exactly as it is represented in one or the 

 other of the parents. Even where the resemblance to one of the parents 

 has been strongest, I have always found some trace of the influence 

 of the other species. Therefore when the terms patroclinous and matro- 

 clinons have been employed to designate a condition in the hybrids the 

 meaning has always been that of strong resemblance but not for a perfect 

 duplication of the characters concerned. The resemblance of a character 

 in a hyl)rid to that of one of the parents may be very striking, but I 

 am not convinced that the exact duplication of a character is ever 

 attained in the material under consideration. 



With respect to the characters of morjihology it has usually not 

 been difficult to find evidence of the influence of both parents however 



14* 



