ALLEN. — THE HEREDITY OF COAT COLOR IN MICE. 147 



characters, but not the black, then the goklen-agouti, 225, may well liave 

 retained the chocolate character recessive. Goklen-agouti 9 224, from 

 two gray parents, may be assumed to contain no chocolate recessive. 

 The mating of 224 and 225 will then result in golden-agoutis of two 

 sorts, the one pure, the other with chocolate recessive, and 295, of 

 course, must be taken as of the latter class. This explanation is put 

 forward tentatively, pending test by further experiment, but another 

 possibility is also not to be overlooked, namely, that the color characters 

 of a golden-agouti may undergo resolution. If the yellow and the choco- 

 late pigment characters become segregated into separate gametes, then 

 it is clear that golden-agouti as well as chocolate young would result, but 

 in addition there should be some pure yellow offspring as well. The 

 fact that none have yet occuiTed, speaks against a complete resolution of 

 the yellow-chocolate combination. Possibly, however, the resolution of 

 the yellow-chocolate takes place, only in such a way that gametes are 

 formed which are respectively yellow-chocolate and chocolate, as regards 

 their pigment characters. 



The second pair of golden-agoutis that produced chocolate offspring as 

 well as those of their own color, were of the \followiug rather compli- 

 cated descent : 



$ wh. 9 house mouse 



<? ?!'• o 1 S bl.-wh. „ , 



51 ^^^- l-2b. ^^^• 



c?bl.-wh. ^ J S bl.-wh. p j^ 

 1 • 2 b. ^ ^ • 1.2b. ^ ^ • 



$ wh. 9 wh. S gr. 9 gr. 



142 141 82 81 



$ bl. 9 gr. S hi. 9 gr. I [ 



92 91, 93 92 91, 93 (? wh. 9 wh. 



I II I 202 204 



S gr- 9 gr. 



$ gold.-ag. 9 ch. j' gr.-wh. 9 gr.-wh. $ wh. 9 gold.-ag. 



245 244 230 231 248 224 



$ gold.-ag. 307 9 gold.-ag.-wh. 353 $ gr. 372 



I I I ., I 



^ gold.-ag. 451 9 gold.-ag. 477 



