466 GENETICS IN RELATION TO AGRICULTURE 
tabulated by Anderson only sixteen of other colors were recorded. An- 
derson is confident that these were all mistakes, and since herdbook records 
are in error to the extent of about 2 per cent. this explanation would 
appear to be acceptable. Suffolk Punch horses are always chestnut, 
but even in other breeds when chestnuts are produced from matings of 
horses of other colors, they, when mated inter se, produce only chestnut 
progeny. Of the shades of chestnut, sorrel appears to be recessive to 
chestnut proper and distinct from it, according to data collected by Mc- 
Cann from studbook records. According to Wentworth, lighter mane 
and tail, often met with in sorrels, is recessive to the normal darker 
coloration. 
Black is a simple dominant to chestnut, but the data in this case are 
not so clear cut as for chestnut. For this analysis, we represent the 
black factor by C. Black varies in shade from a deep, clear black to 
seal-brown, but very little is known about the relations of the different 
shades to one another. 
Bay is black with a dominant restriction factor B which confines 
the expression of black to the mane and tail and the extremities in general, 
the rest of the body being covered with bay hairs. Since the factor B 
acts only on C' it may be present in a latent condition in chestnut horses. 
This conception, originally suggested by Wentworth, differs from other 
hypotheses in that it accounts for the fact that black x chestnut matings 
give a high proportion of bay offspring. Castle suggests that chestnut 
horses carrying the bay factor be called sorrel and those lacking it chest- 
nut; but the data presented by McCann indicate that the terms should 
be reversed, if it be desired to bring the terminology into conformance 
with common practice. 
The position of the so-called brown horses in the Mendelian scheme 
is a matter over which there has been much speculation. Seal-brown 
appears to be merely a shade of black; but mahogany-brown, 7.e., bay with 
black patches alternating with bay on the sides of the body, seems to be 
bay heterozygous for B. The data of Table LXI may be explained by 
such a formulation, but actual experimental investigations should be 
carried out, if it be desirable to determine the relations accurately. 
Gray is a color in which black and white hairs are intermingled in the 
coat. Gray foals at birth are very dark, but with age they become pro- 
gressively lighter until in old horses the color is almost white. As a color 
gray is not much favored in any except the Percheron breed. The gray 
factor G is a dominant factor, and its relations to those which have just 
been mentioned are such that when it is present the coat color is gray 
irrespective of which of the other factors may also be present. In the 
Clydesdale gray is tabooed, consequently all gray stallions are castrated, 
and gray mares are bred to stallions of a different color. According to 
Digitized by Microsoft® 
