A. D. Darbishire 
27 
completely recessive in Mendel's sense ; the eye-colour of the hybrids is always 
dark ; the coat-colour is variable, generally a mixture of wild-grey and white, the 
character of the coat being distinctly correlated with characters transmitted both 
by the albino and by the coloured parent. There is thus no proper dominance in 
Mendel's sense, so far as eye-colour and coat-colour are concerned, the hybrids 
dififering always in eye-colour and generally in coat-colour from both their parents. 
2. When the hybrids, produced fi'om the cross described, are paired together, 
the resultant young exhibit a segregation into three groups, so far as eye-colour 
and coat-colour are concerned, into two so far as regards the waltzing habit. The 
phenomenon of segregation is closely similar to that described by Mendel, and in 
the colour, whether of eyes or of fur, the proportions are closely identical with 
those observed by him, a quarter of the young resembling their albino grand- 
parent, half representing their hybrid parents, and a quarter resembling their 
waltzing grandparents in so far that they have pink eyes and some coloured fur, 
but differing from any of their immediate ancestors in the range of coat-colours 
exhibited. The proportion of individuals which exhibit the waltzing habit is less 
than one-fifth of the whole number of young, and is not a Mendelian proportion. 
3. When hybrids are paired with albinos, half the young produced resemble 
their albino parent, half resemble their hybrid parent; this result is in accord 
with Mendel's theory. 
4. The correlations between coat-colour in tlie hybrids and descendants on 
the one hand, and the ancestry of their albino grandparents on the other, are not 
consistent with the view that albinism is a recessive Mendelian character, trans- 
mitted " gametically pure " by any homozygote which possesses it. 
5. The behaviour of " extracted albinos," " extracted hybrids," and " extracted 
dominants " is not consistent with any theory of gametic purity yet propounded. 
There is no evidence that any individuals which could be properly described as 
" pure dominants " or " pure recessives " exist among the whole series produced. 
6. The effect of varying ancestry is so great, in every case in which it has 
been examined, as to show that the phenomena observed cannot be adequately 
described except in terms of the ancestry of all the individuals used as parents. 
7. There is no evidence to show whether the ancestral correlations observed 
are consistent or not with the consequences which would follow from the applica- 
tion to this particular case of the Galton-Pearson law of inheritance ; the only 
consequence of that law which can at present be deduced, namely, the negative 
correlation between one parent and the offspring of the first cross, is in agreement 
with observation. 
8. In this, as in many other cases, the characters of the hybrids produced by 
crossing two races which have been long distinct strongly suggest reversion to 
ancestral characters. 
I cannot bring these observations to a close without expressing my gratitude 
to those to whom it is due. Professor Weldon has helped to bridge over a gap 
