302 



F. B. SUMNER AND H. H. COLLINS 



close with the generations just referred to. No careful compari- 

 sons were made between any of these mice and the 'a' and '6' 

 yellows, and only one of the skins was saved, so that it is now 

 impossible to make such a comparison. It is our recollection 

 that the 'c' mice resem'bled the 'b' rather than the 'a' strain, 

 though the single preserved skin is probably intermediate. 



It is worth noting, though perhaps not significant, that the 

 heterozygous (?) parents of five out of six of our yellow strains 

 gave numbers of yellows in excess of mendelian expectation. 

 Of the 68 mice thus produced, 44 were normal, 24 yellow, giving 

 a ratio of 1.8:1, instead of 3:1. Such a departure from the 



TABLE 4 



'expected' condition may well be accidental, however, particu- 

 larly in view of the fact that the 'heterozygous' parents of the 

 'a,' 'b/ and 'c' strains gave an excess of wild-color offspring when 

 mated with yellows (viz., 16:11).^" Whether these parents were 

 in reality heterozygous, rather than original producers of mutant 

 germ-cells, is not definitely shown by our records. The data 

 given seem compatible with either interpretation. 



Juvenile yellows 



As stated in earlier papers, the yellow variety is nearly or quite 

 as distinguishable in the juvenile pelage as in the mature (figs. 7 and 

 8). Table 4 gives the proportions of black, white, and color in 



1* This apparent excess of yellows may be due to another cause (see Summary 

 and Conclusions). 



