252 On Inheritance of Coat-Colour in, the Grei/hound 
Further, four were from R. x F. crosses, and the remaining five introduced 
Be., Bk., Bd. and F. crosses of a miscellaneous character. 
It is obvious from these results that the breeding out of white would be at 
least a difficult task, and that when white has substantially disappeared for a 
couple of generations, it may suddenly reappear in full force. If it be suggested 
that Bk. in the nomenclature would not exclude a white toe, nor W. a small black 
spot, then the reply must necessarily be that no theory of heredity can be of 
service for the purposes of evolution which would make one class of a white dog 
with a black toe and a black dog with a few white hairs on the throat, and 
professes only to predict when such a class as a whole will occur. The protective 
value of the coat- colour in the two cases would be wholly and entirely different, 
and we should have to seek further for those features of the pedigree with which 
the difiFerentiation could be associated. 
(4) Colour Scales and Reduction Metliods. When we first started work on 
the greyhounds, the method of contingency had not yet been developed, and 
accordingly we made tables for the inheritance of melanism and of red pigment 
and proceeded to find the correlations by the fourfold division process*. In our 
classification we must admit having been influenced by the statement of breeders 
that brindle is the result of the crossing of red with black. A glance at our 
Tables I and II will show, however, that we cannot look upon red and black as 
pure forms giving a heterozygous brindle. We have in percentages based on 332 
E. Bd. AV. F. r. Bk. M. Bk. 
22% 8% 1% 12% 41 7„ 16% 
cases: a result which is capable of intcri^rctation when we pay attention to ancestry 
(or, again, on the basis of a mixture of alternative inheritance and reversion), 
but is scarcely reducible to any simple Mendelian proportions, still less is it 
compatible with R. x Bk. = Bd. 
The fact that R. x R. or R. x F. very rarely gives black of any kind (although 
F. X F. can give black), while Bd. x Bd. in the rarest cases gives Bk. of any kind, 
seems to denote more red than black in ordinary brindle and a closer relation of 
brindle to red than of fawn to red, although F. x F. in the bulk gives equal 
amounts of red and fawn. Accordingly we grouped our classes for red and black 
pigments as follows : 
Red Pigment Black Pigment 
(a) Red. {a) Pure Black. 
{h) Brindle. {h') Mixed Black, 
(c) Fawn. (c') Brindle. 
{d) No Red. {d') Fawn. 
(e) No Black. 
* Phil. Trans. Vol. 195, A, [ip. 1—47. 
