144 PROCEEDINGS OF THE AMERICAN ACADEMY. 



known to transmit the black character in a certain proportion of cases, 

 hence this supposition seems plausible. Moreover, from the writer's 

 crosses of black and chocolate mice, as well as from the data given by 

 Bateson for similar crosses, it seems fairly clear that the chocolate under 

 these circumstances may behave as a simple recessive towards black. In 

 this case, the black heterozygotes, on being back-crossed with chocolate 

 mice, should give both black and chocolate individuals in the ratio, 1:1. 

 This, in fact, was approximately the result obtained. In six litters a 

 total of 10 black and 13 chocolate mice was obtained, each class being 

 thus within two of the expected relation, 11.5:11.5. 



A different result came from back-crossing with a chocolate mouse a 

 gray heterozygote whose parents were respectively a chocolate and 

 a white mouse. The ancestry of the gray animal follows : 



$ bl.-wh. 1 • 2 b. 9 wh. $ wh. 9 house mouse 



S gr. 82 9 gr. 86 $ gr. 51 9 wh. 



$ ch.-wh. 328 9 wh. 192 



$ gr. 407 



This gray mouse gave, in two litters by a chocolate mouse, 3 black, 3 

 gray, and 3 golden-agouti young. If the chocolate mouse produced only 

 gametes having the chocolate pigment-character, while those of the gray 

 mouse in half the cases underwent resolution and in the other half retained 

 the gray character unresolved, the gametic unions would be: 



gr. -f gr. = gametes of gray parent (unresolved), 

 ch. -f ch. = " chocolate parent, 



4 gr. := offspring. 



bl.-ch, -f yell.-ch. = gametes of gray parent (resolved), 

 ch. + ch. = " chocolate parent, 



2 bl. -H 2 golden-agouti = offspring. 



The result is that three classes of young, gray, black, and golden- 

 agouti, are produced, in the proportions 2:1:1. In tlie experiment these 

 three classes were present, but in the proportions 1:1:1. 



