44 THE INHERITANCE OP COLOR IN MICE. 



From this table it will be seen that the spotted individuals of F2 have been 

 contaminated by the cross, having their dorsal pigmentation increased on the 

 average more than 50 per cent over the average of the pure Japanese race. 

 The range of the spotted forms in Fi is also very different from that in the 

 Japanese race. There are no F^ animals within 18 per cent of the average grade 

 of the Japanese mice (32 per cent) ; also there is only one F2 animal out of 21 

 that is as lightly pigmented as the darkest of the Japanese race. 



The back cross, between Fi animals and individuals from the pure Jap- 

 anese race, we should expect to give spotted and self animals in a 1 : 1 ratio. 

 The spotted animals, on a basis of contamination, should not be so extensively 

 modified as in the F2 animals; that is, their degree of dorsal pigmentation 

 should be less. There have been 96 animals produced in this back cross, of 

 which 49 are self and 47 spotted, showing that "spotting" and "self" are still 

 behaving as a pair of allelomorphic characters. The spotted animals from this 

 cross — see above table, (3) — show an average of 60.2 per cent of the dorsal 

 surface pigmented; their range is from 40 to 99 per cent. This coincides 

 qualitatively with the expectation. 



One more cross may be recorded. Certain of the 49 self s produced by the 

 back cross were bred inter se, thus producing a generation which may be desig- 

 nated as "back cross F2." Spotted animals resulting from this cross, on the 

 theory of contamination, should show a degree of dorsal pigmentation between 

 that of the first F2 (85.3 per cent) and of the back cross (60.2 per cent). As a 

 matter of fact the results show that their average pigmentation is 70.4 per cent. 



ASSOCIATION OF CHARACTERS. 



The question naturally arises as to whether the various characters are 

 completely independent of each other or whether coupling or gametic associ- 

 ation of any sort exists between some of them. Morgan (1911&) has observed 

 in DrosopMla certain results best explained on an hypothesis of "association" 

 of certain characters in the gamete. It remains to be seen whether any such 

 "association" of characters exists in mice. For this purpose the writer has 

 tabulated the results of four crosses: 



I. Cross 10a. Wild Black Agouti (Mus muscultjs) Crossed with Dilute Brown. 



The black-agouti parent possesses factors B =black, A = agouti, and D = 

 density. The dilute-brown parent lacks the factors B, A, and D. Fi resem- 

 bles the wild parent. In F2 we expect 8 classes, as follows : 



(1) Black agouti =BrBAD* (5) Brown ^BrD 



(2) Black =BrBD (6) Dilute brown agouti =BrAd 



(3) Brown agouti =BrAD (7) Dilute black =BrBd 



(4) Dilute black agouti =BrBAd (8) Dilute brown =Brd 



*A11 animals in this cross possess the general color factor Y, and the factor ior fullness 

 of black and brown pigmentation, P- They are therefore omitted from the formulae for 

 purpose of simplification. 



