﻿8 DARBISHIRE, Mendelian and Biometric Theories. 



heterozygote nature of the hybrids was made, as suggested,, 

 by mating them with albinos. In all cases but two I 

 found as I had expected that albinos were produced : 

 and I ascribed the absence of albinos from the other two 

 litters to chance, and had no doubt that I had only to- 

 mate them with albinos again to obtain the required 

 proof of their hybridity. But the next litters from the 

 two mice, mated thus, contained no albinos. So that it 

 began to look very much as if these apparent hybrids 

 really were dominants. This fact in itself pointed to the 

 truth of the suggestion that mice with coloured coats and 

 eyes were of two kinds — hybrids and dominants ; but 

 coupled with the result of mating the gametically tested 

 hybrids inter se it afforded fairly complete confirmation 

 of that hypothesis, for of the 92 young resulting from 

 such union, 14 had a pink eye and coloured coat, 58 

 a dark eye with coloured coat, while 20 were albinos. 



I wish particularly to remind the reader that I do not 

 think that these numbers are large enough to draw any 

 numerical conclusions from ; they are introduced solely as 

 part of an argument which is intended to shew how I came 

 to see that the facts summarized in the Table on p. 6, were 

 equally in accord with Mendelian and Galtonian theory r 

 and their value in this respect will not be in the least 

 impaired if it ultimately turns out that the proportion in 

 which the three categories of coat and eye colour occur are 

 "in every respect discordant with" Mendelian prediction. 



That the two mice which gave no albinos in the two 

 matings referred to really are dominant is, I think, placed 

 beyond doubt by the fact that in a third mating with 

 albinos they have failed to produce anything but coloured 

 mice with black eyes.* That the remaining mice are 



* Besides these two mice which have been tested thrice there are three 

 which have been tested twice, two bucks and a doe, with the same result. 



