92 MR. W. BATESOX ON COLOUR-HEREDITY [May 26, 



while the dark-eyed blacks should give only blacks, pei'haps blues, 

 and albinos. The dai-k-eyed greys will probably give both l^lacks 

 and yellows, though in view of von Guaita's evidence that grey 

 may exist as a new homozygote, this is perhaps doubtfiil. As far 

 as the published evidence goes, the most probable constitution of 

 the sevei'al foi-ms in F., is as follows : — 



Albinos albino x albino. 



Pink-eyed yellows yellow x yellow. 



Dark-eyed yellow x albinos. 



Pink-eyed fawn-yellows ... fawn-yellow x fawn-yellow. 



Dark-eyed blacks black x albino (?) 



Dark-eyed greys fawn x albinos (and perhaps 



some other combinations). 



Pink-eyed lilacs lilac x lilac. 



The absence of blacks with pink eyes is noticeable, and i-aises 

 the question whethei- there is not a permanent synthesis in these 

 blacks. 



Finally, we have two impoi-tant problems, the nature of the 

 cZar^-eyed fawn-yellows and of the dwk-eyed " lilacs." As they 

 are dark-eyed they presumably both contain albino. But as 

 regards the first, it is difiicult to see what the other gamete 

 can be in that case. For from F^ we learn that fawn x albino give 

 grei/, not fawn-yellow. On the other hand, as there are black and 

 yellow gametes, we ought to find their heterozgyote, which will 

 presumably he f man. But if this combination follows the lules 

 of the others, the heterozygote should be pink-eyed, not dark-eyed. 

 The number of dark-eyed fawn-yellows, three, is too few to make 

 it likely that these are the black x yellows, which we expect to 

 appear as a fairly frequent combination ; and the general indica- 

 tions are quite unfavourable to the view that any considerable 

 number of heterozygotes can be dark-eyed without the presence of 

 the albino, though it is not impossible that such I'oal synthesis 

 may take place. 



Next the " lilacs " raise certain questions. We must suppose that 

 the dark-eyed " lilacs " contain albino ; but in the ofispring of F, x 

 albino there are no " lilacs." As F^ is giving ofi" gametes capable 

 of forming " lilacs," we see that the lilac x pure albino gives some 

 other colour. Next, which of the groups can be supposed to 

 represent the lilac-bearing gametes in their other combinations ? 

 This also is a question we cannot answer. 



A similar difficulty is created by the scarcity of yellows in the 

 offspring of F, x albino. There were only 3 in 88. We might 

 have expected the numbers of yellows and blacks to be equal, but 

 there were 15 blacks. Moreover, all the yellows were in one 

 family. So far this is quite inexplicable. It probably indicates 

 that some of the albinos possessed powers of resolution different 

 from those of others, or conversely that some of the original 

 " fawns " were more easily resoluble than others, [of. Cu^not's 

 new results (27), where blacks were resolved out, but apparently 

 no yellows.] 



