SELECTION 585 



in type. Enthusiastic amateurs have often attempted to maintain 

 a " small herd of exceptional excellence." Such attempts have 

 always failed, and must fail, for the reason that such a herd 

 affords too little material for selection, and therefore the impos- 

 sibility of maintaining its type is a bar not only to progress, but 

 even to the bare maintenance of the initial excellence. Suppose, 

 for example, that a small herd of exceptional animals be brought 

 together, say, three cows and a bull, the choice from many 

 of the best herds. What are the mathematical probabilities of 

 their being able to reproduce their own number of equal excel- 

 lence before the original herd disappears ? It is very slight 

 indeed, unless one of the number proves to be a phenomenal 

 individual breeder, which, in truth, occasionally happens. 



Value of the exceptional breeder. The more the matter is 

 studied the more it will be found that the excellence of any herd 

 or of any breed is sustained and advanced, not by the general 

 mass, but by a few exceptional, not to say phenomenal, breeders. 



The trotting blood of to-day owes its high development to a 

 very few foundation animals, coming to us through Hamble- 

 tonian 10, and sustained and developed by an insignificantly 

 small percentage of the general mass of stallions.' 1 



The excellence of the Shorthorn is maintained in the same 

 way, and it is not too much to say that in all probability there 

 are never living at any one time in any breed more than a score 

 or so of animals that produce anything like a real and positive 

 advance in the breed, 2 " The line of descent runs not through 

 their veins." 



The exceptional breeder not necessarily the exceptional indi- 

 vidual. Neither Hambletonian 10 nor Geo. Wilkes was the 

 best trotter in the breed ; indeed, the highest performers have 

 contributed little. They were the offputs, but accidentally, or 

 of necessity, they were not of the main line of descent. In the 

 corn-breeding experiments already cited, the ear to which the 

 present high-protein stock all traces was not one of the originally 

 highest protein ears. The great sires and the great dams that 



1 See chapter on " Heredity," table of the Big Ten, p. 555. 



2 This number is too high. The Shorthorn breed probably never saw twenty 

 such bulls as Champion of England. 



