286 



THE All KMC AN NATURALIST [Vol. LII 



that it seems impossible to fix the condition, led Shu 11 and 

 the writer independently to the conclusion that certain 

 factors in addition to their functions as transmitters of 

 hereditary characters also had the faculty of carrying- 

 some sort of a developmental stimulus when in the hetero- 

 zygous condition. The recent work of Morgan on linked 

 characters, however, makes it possible to give another 

 interpretation, as Jones (1917) has demonstrated. If it 

 be assumed that several variations have occurred in each 

 of one or more chromosomes, then it can be shown that the 

 first-generation hybrid between such a variant and the 

 race from which it arose will bring together all dominant 

 or partially dominant characters. In the second hybrid 

 generation, on the other hand, Mendelian recombination 

 steps in and makes it improbable that many individuals 

 shall have such a zygotic composition. And only in the 

 rare cases where the proper breaks in linkage have oc- 

 curred can a homozygous individual of this type be 

 produced. 



The latter hypothesis holds the advantage that it 

 furnishes hope for a homozygous combination as valuable 

 as that of the first hybrid generation no matter how 

 rarely it may be assumed to occur, but whether it holds 

 for the majority of organisms or not may depend on a 

 future decision as to the frequency of side-by-side 

 synapsis as compared to end-to-end synapsis. Our knowl- 

 edge of linkage rests almost entirely on Morgan's work 

 on Drosophila where side-by-side synapsis occurs at the 

 maturation of the germ cells. If the break in linkage be- 

 tween groups of characters apparently carried by a single 

 chromosome, which Morgan finds to be so exact in Dro- 

 sophila, should actually depend on Jannsen's theory of 

 chromosome, twisting at synapsis, then some other type 

 of inheritance may be found in species having end-t<>-end 

 synapsis. Perhaps this is the reason why the Oenotheras 

 have such a peculiar heredity, for in them Davis ( 1009) 

 thinks end-to-end synapsis prevails. But, be this as it 

 may, the vigor of first generation hybrids is a fact and 

 not a theory, and the advantage it brings to the hetero- 



