392 



Mr. C. C. Hurst. 



[Feb. 1, 



ct»at presents a considerable range of tint. The differentiating pair of 

 Mendelian characters is thus apparently the presence or absence of black 

 pigment in the " points." 



Summary. 



In modern thoroughbred horses chestnut colour is a Mendelian recessive 

 to bay and brown, which are dominant characters. 



The consideration of other colours being excluded, bays and browns are of 

 two kinds : — (a) Those that when mated with chestnuts will give no chestnut 

 offspring, (b) Those that when mated with chestnuts will give, on an 

 average, half their offspring chestnuts and the remainder bays or browns. 

 Similarly, the recessive chestnuts, variously extracted from the dominant bays 

 and browns, breed true, as a rule, when mated together, without reversion to 

 their bay or brown ancestors. To this rule, 9 exceptions were found in 

 1104 cases, and it is not impossible that these may be due to errors in the 

 records. 



Note added January 31, 1906. 



In the paper read on January 18, Professor Weldon disputes these 

 conclusions, while admitting that, contrary to Professor Pearson's statement, 

 chestnuts breed true within a very small percentage of error. Since my 

 paper was read, a fresh tabulation has been made of a sample including 

 more than 2500 offspring recorded from chestnut mares, taken without 

 selection. 



Omitting greys, two sires of doubtful identity and sires with less than 

 10 foals each, it appears that — 



30 RR sires give 455 K + 3D, 

 25 DD „ 450 D + 411, 

 54 DP „ 496 D + 507 R. 



This evidence, with that previously collected and with Professor Weldon's 

 fresh evidence, points to about 1 per cent, of exceptions in the records of 

 both the chestnuts and the pure dominants. Professor Weldon's argument 

 is based entirely on the alleged existence of exceptions. The Stud Book is 

 remarkably accurate, but there is a sufficient margin of demonstrable error 

 in the returns to make it possible that the few exceptions which cannot be 

 eliminated are due rather to mistake than to any physiological peculiarity in 

 the animals concerned. As a matter of fact, a sensible proportion of the 

 original records are afterwards corrected, either in the Stud Book itself or in 

 the Racing Calendar. In illustration of this, it may be mentioned that 



