ALLEN. — THE HEREDITY OF COAT COLOR IN MICE. 151 



of three color types in the proportions: 1 black, 11 gray, 4 golden- 

 agouti. The writer actually obtained, in five litters, 1 black, 14 gray, 4 

 golden-agouti young, as well as 5 albinos, which were not tested. The 

 close agi'eement of the observed with the calculated total seems to 

 corroborate the conclusion previously stated, that the writer's original 

 albino stock produced gametes some with the gray, some with the black 

 pigment character. It also is good evidence of the segregation of the 

 several pigment characters of the heterozygote in such a manner as 

 simply to unloose them from their combination practically unchanged. 

 Five males and four females contributed to the result just given, but only 

 one of the females was proven to have the black character among her 

 gametes, while the male by which she had the black mouse subsequently 

 died before further tests could be made. The same female by three other 

 of her brothers gave only albino, gray, and golden-agouti young. 



To test the matter still further, however, three of the same lot of gray 

 males were back-crossed with golden-agouti females. Assuming as before 

 that some of the gray animals produce gametes with the Mack and the 

 golden-agouti characters respectively, and others those with the gray and 

 the golden-agouti characters respectively, the possible combinations are 

 as follows : 



gr. -|- yell.-ch. = gametes of gray parent, 



yell.-ch. + yell.-ch. = gametes of golden-agouti parent, 



2 gr. -f- 2 gold.-ag. = offspring. 



bl.-ch. + yell.-ch. = gametes of gray parent, 

 yell.-ch. + yell.-ch. = gametes of golden-agouti parent, 



2 gr. + 2 gold.-ag. =: offspring. 



The expected result is that only gray and golden-agouti young will 

 be produced, and that these two classes will be equally numerous. This 

 result was realized qualitatively, but not (juantitatively, for there were 

 actually produced 19 gray and 7 golden-agouti young. In other words, 

 there are nearly one-quarter fewer golden-agoutis than expected. If 

 this discrepancy be simply the result of chance, the explanation suggested 

 may still be considered valid, yet the excess of gray animals is so great 

 that it seems hardly proper to ignore it entirely. 



Having thus followed throuffh the results of crossing fjolden-agoutis 

 with albinos of mixed extraction, it remains to mention two other crosses 

 in which the albinos used were of a yet different nature as regards pig- 



