1906.] On the Inheritance of Coat Colour in Horses. 



393 



Ben Battle, a. sire recorded in the Stud Book as chestnut, appears in the 

 Eacing Calendar several times as bay or brown, and Mr. G. H. Verrall, who 

 has kindly given help in this matter, writes that it is practically certain that 

 Ben Battle never ran as a chestnut. 



Several alleged chestnuts, ascribed to Hackler, Wolf's Crag, and other pure 

 dominant sires can be similarly corrected. Among the residual exceptions 

 are some which were born dead or died unnamed. Very few indeed can be 

 proved to have appeared in public uncorrected. Other errors probably arise 

 through an incorrect return of the sire's name. For instance, a certain sire 

 appears to give a total of 43 dominants and three chestnuts from chestnut 

 mares ; a scrutiny revealed that two of these chestnuts (which happen to be 

 among the four mentioned above) were from one chestnut mare. Moreover, 

 the same mare is credited with a hay foal by a chestnut sire. These three 

 exceptions occur in four consecutive returns from one stud. Considering the 

 extreme rarity of any exception, the coincidence seems to point to inaccuracy 

 in the returns of the breeder in question. In Professor Weldon's tabulation 

 such a sire would be ranked as a DE, and the 43 bays and browns he gave 

 would go to create the excess of dominants which Professor Weldon found to 

 result from the mating EE x DE — an excess obviously due to the inclusion 

 of such cases in that category. It is, of course, not impossible that genuine 

 exceptions do occur. They must, however, be exceedingly rare in any case, 

 and I am disposed to doubt whether the returns made to the Stud Book 

 have the extreme precision which would be required to establish such 

 occurrences. 



Finally, it would appear that the distinct properties of chestnuts must be 

 ascribed to segregation rather than to ancestry, seeing that their behaviour in 

 heredity is entirely different from that of bays and browns, 1 though their 

 ancestral composition may for several generations have been the same. 



This analysis was undertaken without any knowledge of previous work on 

 similar lines, but several papers stating more or less concordant conclusions 

 have since been discovered. The memoir by Crampe,* mentioned by 

 Professor Weldon in the discussion on his paper of January 18, gives, as 

 I now find, extensive tables drawn from German sources, showing that 

 within a small margin of error chestnut (fuchs) breeds true. Of the excep- 

 tions, several, as he shows, are probably mistakes, and the rest he regards as 

 dubious. Wilckens,f a s the result of a similar analysis on a large scale, 

 found 24 recorded exceptions to the purity of " fuchs " per 1000 matings. 

 The absolute purity of chestnuts, however bred, is asserted by Mr. Wilfrid 



* Landw., J. B., 1888, vol. 17, especially p. 828. 

 t Ibid., p. 575. 



