No. 524] CHROMOSOMES AND HEREDITY 469 



suggested by Sliull and by Spillman. It is a quantita- 

 tive conception, and I shall try to point out some of its 

 applications to the chromosomes. 



Let me recall once again the familiar fact that in ani- 

 mals and plants two homologous chromosomes of each 

 kind are present in every cell. This gives the diploid 

 number. One of each kind suffices to produce the char- 

 acter in some cases, but each is nevertheless present in 

 double. We might think of the doublcness as a sort of 

 reserve and the double group be conceived as a "mechan- 

 ism of safety." That the double number is not always 

 necessary for the formation of the characters is shown 

 in embryos that develop from non-nucleated fragments 

 of eggs. These embryos have all the characters peculiar 

 to their species. The importance of the double set is 

 illustrated, however, in certain hybrids. The best case 

 is that of the hybrid between horned and hornless races 

 of sheep. The male hybrids from this union have horns, 

 the female hybrids lack horns, irrespective of the way in 

 which the first cross was made, i. e., the results are the 

 same whether the mother was horned and the father 

 hornless, or the mother hornless and the male horned. 



in the male hybrids suffices to call forth horns; but one 

 dose in the female hybrids is insufficient to call forth 

 horns. In terms of chromosomes this may mean that 

 one horn-bearim_r chromosome suffices in the male to call 

 forth horns, but in the females one chromosome is not 

 enough. 



When these hybrids are inbred they produce in the 

 second generation four kinds of individuals, horned 

 males and females; hornless males and females. The 

 numerical results appear to coincide with the assump- 



ehromosomes necessary to call forth horns in the two 



certain number of combinations in which females will 

 contain two doses of horns and these females should be 

 horned. 



