120 PROCEEDINGS OF THE AMERICAN ACADEMY. 



relative degree of in-bredness or pure-bredness of the albino stocks. 

 Moreover, the cliaracters total and partial pigmentation are inherited quite 

 apart from the character complete albinism, so that in the Fg generation 

 it is possible to obtain both spotted and self-colored animals having the 

 character complete albinism recessive. An albino having tlie spotted 

 character may be regarded as a spotted mouse in which the spots do not 

 show, perhaps because of the absence of chromogenic substance, if we 

 adopt the theory of Cuenot (:03). 



Darbishire's mice offer another point of interest in this connection. 

 The spotted Japanese individuals were of a peculiar strain, having pink 

 eyes. All the Fi heterozygotes had dark eyes, so that we may suppose 

 the pigmented parents of the albinos to have had dark eyes. From his 

 second generation, obtained by interbreeding the dark-eyed heterozy- 

 gotes, Darbishire had 17 pink-eyed spotted mice, 36 dark-eyed individ- 

 uals, and 13 albinos. These last doubtless represent both the pink- and 

 the dark-eyed condition potentially, but until tested they may be disre- 

 garded. It is expected, therefore, that 25 per cent of the pigmented 

 individuals will show also the pink-eyed character. This, in fact, is 

 practically the outcome, for 17 young are pink-eyed, 13+ being expected. 

 The character of the albinos could be tested by back-crossing them with 

 the pink-eyed pigmented stock, and some would probably give pigmented 

 mice, all of which would have pink eyes, while others would give young 

 some with pink, some with dark eyes. 



To test further the character partial pigmentation, the writer has 

 back-crossed with spotted mice both albino and pigmented individuals 

 which might be supposed to be heterozygous with respect to the two 

 characters. The derivation of the animals used is, 

 spotted X non-spotted or self 



self (spotted recessive), 

 and the expectation is that by the cross with a spotted mouse, self-colored 

 and spotted offspring will result in the ratio, 1 : 1. Tiiis was practically 

 the result obtained. Table P contains a list of crosses of this nature. 



The expectation is that the two classes will each consist of 14.5 individ- 

 uals, so that the deviation amounts to but 0.5 for each class, — as close an 

 approximation as is possible in an odd number of offspring. The writer 

 purposely selected as the heterozygotes for these tests albinos coming of 

 a cross between spotted and self-colored mice, in order to throw light on 

 the suggestion of Bateson (:03'') that a segregation of pigment characters 



