ALLEN. — THE HEREDITY OF COAT COLOR IN MICE. 107 



is interesting further to note the exact 1 : 1 ratio of tlie two classes, pig- 

 mented ami completely albino young, in this litter of six. A second 

 individual was tested in like manner. This time a gray mouse, showing 

 no trace of white in the pelage, was chosen, and bred to an albino of 

 spotted ancestry. Of the four young resulting, two were entirely gray 

 and two were spotted, the one being gray-white, the other black-white. 

 It is plain that in this case the gray parent was of the class D*DR. For 

 the possible unions of such an animal with the albino would result in two 

 animals of the character DR (here presumably represented by the two 

 all-gray young) and two of the character DR-R (the two spotted indi- 

 viduals). Moreover, these two sorts of j^oung are actually present in the 

 ratio 1 : 1 as expected. 



It is interesting to observe that of the four possible classes of young 

 by the heterozygous parents, viz. : D, DDR, D-R, and DR-R, only one, 

 the class DR'R, is expected to show a spotted coat. In the thirteen young 

 obtained representing these four classes, we therefore look for three, or 

 one-fourth the total number, to be spotted, and there are actually four. 

 It is unsafe, however, to assume that all these spotted young are of 

 the class DRR, for a mouse of the class D-DR may sometimes show 

 white flecks on the tail. The case appears to be an instructive one, how- 

 ever, and so far as worked out, conforms to the Mendelian laws. No 

 attempt has been made to interbreed at random the four sorts of indi- 

 viduals of Fi which have been considered, viz. : D, D*R, D'DR, and 

 DR'R. It may readily be calculated, however, that if all the possible 

 unions were to take place, the smallest total number of individuals repre- 

 senting all such combinations would be sixty-four. The same six classes 

 as previously worked out would be present, and the t^tal albinos 

 would still be one in sixteen. The relations of these classes to one an- 

 other would be: 16 D: 16 D-DR: IGDR: 4DR: 8 DR-R: 4 R. 

 It may be seen that an actual statistical test of a case of this sort is far 

 from being a simple matter. 



4. Albinos bred inter se. So far as certainly known, two albinos 

 when interbred produce only albinos, or, as is commonly said, albinos 

 breed true. It must not be assumed, however, as some recent writers 

 have done, that any given albino is the exact equivalent of any other 

 albino, in so far as concerns its influence in the heredity of characters 

 other than albinism. The writer has obtained a large number of young 

 both from interbreeding mice of albino parentage and from interbreeding 

 those which Batesou has termed " extracted recessives," that is, the 

 young of two pigmented heterozygotes which each contain the albino 



