12 



HEREDITY IN RABBITS, RATS, AND MICE. 



if we take the whitest 48 individuals to be extracted "whites" it is 

 evident that they are considerably modified from the condition of the 

 uncrossed race, since their mode Ues at 12 to 14, not at 16 to 17, and 

 the highest grade (17) of the uncrossed white race does not reappear 

 at all. The back-cross of Fi with white must be investigated before 

 one can interpret the F2 result with confidence. But the back-cross 

 (table 29 and text-figure 2, Fi X W) makes it very clear that segrega- 

 tion occurs on a 1 : 1 basis. The 116 young thus produced fall into 

 two groups which do not overlap and each of which is monomodal. 

 Each contains 58 individuals. But the ex- 

 tracted groups show mutual modification. The 

 mode of the lower group is not at grade 1, as 

 in Fi, but at grades 4 to 6, while the mode of 

 the upper group is not at 16 to 17, as in the 

 uncrossed white race, but at 15. 



The back-cross of Fi with self (table 29 and 

 text-figure 2, Fi X S) gives a variation nearly 

 covering the combined range of self and Fi, as 

 expected, but the two expected groups (if they 

 are distinct) lie so close together that it is 

 impossible to separate them. 



The several facts developed in crosses of 

 white with self indicate that white is a simple 

 allelomorph of self. If so, and if dark is also 

 an allelomorph of white, as indicated by text- 

 figure 1, then self and dark should be allelo- 

 morphs of each other. Such is probably the 

 case, but crosses of self with dark are incon- 

 clusive because the two conditions are so 

 close to each other on the grading scale that 

 it is difficult to demonstrate segregation. See 

 tables 27 and 29 and text-figure 3. This figure 

 shows (at the top) the variation of uncrossed dark in relation to self. 

 Below is shown the variation of Fi, self being usually dominant. A 

 back-cross with dark (Fi X D) produces less than the expected propor- 

 tion of self (nearly 50 per cent as indicated by Fi) and produces extracted 

 darks of lower mean grade than the uncrossed darks. This is evidence 

 of mutual modification of self and dark in the heterozygote, so that 

 they emerge in the gametes modified. So far, then, the evidence indi- 

 cates that self, dark, and white are all allelomorphs, but that they 

 fluctuate quantitatively and mutually modify each other when asso- 

 ciated in heteroz3'gotes. 



We may now consider the relation of these three to tan Dutch. If 

 tan is an allelomorph of either of the other forms of Dutch, it should 

 be an allelomorph of all three conditions, in fact sl fourth allelomorph. 



Grade 1 2 3 4 5 6 7 



Text-figure 3. 



