226 Atkinson. 



should be judged to indicate patrocliny in one cross, the same hybrid 

 appears in the reciprocal cross where it would be a case of matrocliny 

 for the same hybrid. While the behavior of pycnocarpa and nutans in 

 reciprocal crosses is similar to that of isogamous species, in that the 

 hybrid production in one cross is the same in the reciprocal cross there 

 are two to three different hybrids instead of one as in the wild iso- 

 gamous species studied by de Vries. The behavior is distinctly diffe- 

 rent then from that of isoganious species like Oe. hooJceri, cocJcerelU and 

 strigosa. Nevertheless, since the same types of hybrids are produced 

 on each side in reciprocal crosses, it seems to indicate that the same 

 heritable characters are transmitted in the pollen and Q.gg cells. In 

 this respect the parents {pycnocarpa and nutans) behave like isogamous 

 species, but differ from such isogamous species as Oe. hooJceri, cocker elli 

 and strigosa in the production of several distinct hybrid types all of 

 which, in their vegetative characters are blends, or recombinations of 

 the parental characteristics. 



Constancy of the hybrids in the second generation. 

 De Vries has found that the majority of Oenothera hybrids are con- 

 stant in the second and succeeding generations, though there are a few 

 exceptions to this rule. Seed from two different plants of hyhrida pyc- 

 nella has been growm, and all the plants of the second generation 

 repeated the pycnella type both in the annual and biennial forms. 

 Under the usual circumstances seed from a larger number of plants 

 should be used in order to determine whether the second and following 

 generations are constant or not. But the fact that de Vries has found 

 the majority of Oenothera hybrids to be fixed from the first, and con- 

 stant in the succeeding generations, I feel reasonably sure that hybrida 

 pycnella is a constant hybrid. It represents, therefore, an interesting 

 type of hybrid in which there is produced in the first generation of a 

 cross, a new creation having the important characteristics of fixity of 

 type or constancy, a high degree of fertility, rapid germination and 

 growth and earlier maturity than either of its parents. 



Interpretations of splitting in the first generation. 



Segregation of characters in the zygote versus segre- 

 gation in the gonotokonts. The sorting and blending of the char- 

 acters which make up the constitution of the two parents (nutans and 

 pycnocarpa), into the two segregate hybrids {pycnella and tortuosa) and 



