PRINCIPLES AND PEACTICB OF BEEEDING. 103 



assimila,ted ancestors. This occasional breeding back and 

 consequent divergence from the existing type, is liable to 

 continue for a great number of generations ; and it can only 

 be repressed by a long and uniform course of breeding, and 

 by a rigorous " weeding out " — that is, exclusion from 

 breeding — of every animal exhibiting a tendency toward 

 such divergence. 



We cannot always, among either pure bloods or mongrels, 

 breed from perfect or approximately perfect individuals, or 

 those which are alike m their structure and properties. 

 Necessity sometimes, and economy frequently, requires us to 

 make use of materials which we would not voluntarily select 

 for'the purpose. In such cases, it should always be the aim 

 of the breeder to counteract the imperfection of one parent by 

 the marked excellence of the other parent in the same point. 

 If, for example, a portion of the ewes of a flock are too short- 

 wooled, they should, ' other things b^ing equal, be coupled 

 with a particularly long-wooled ram. 



The hereditary predispositions of breeding animals are 

 also to be regarded, as well as their actual existing charac- 

 teristics. In the case just given, if the long-wooled ram was 

 descended from uniformly short-wooled ancestors, his length 

 of wool would be what is termed an " accidental " trait or 

 property; and there would be little probability of his 

 transmitting it with uniformity and force to his ofispring out 

 of short-wooled ewes. There would be no certainty of his 

 doing so, even among long-wooled ewes. 



What are considered accidental characteristics are them- 

 selves generally the result of breeding back to a forgotten 

 ancestor, but sometimes they are purely spontaneous. In 

 such cases, they are exceptions, not to be accounted for by 

 any of the known laws of reproduction. As a general thing 

 they are not transmitted to posterity. In other cases they 

 are feebly transmitted to the first generation and then 

 disappear. But occasionally they are very vigorously repro- 

 duced, and if cultivated by inter - breeding, the related 

 animals possessing them soon become fixed in their de- 

 scendants apparently as firmly as the old and long -established 

 peculiarities of breed.* The following is an instance of this,. 



* It is claimed that artificial peculiarities even — those produced hy external 

 causes after birth— are sometimes inherited, as for example, a limb distorted by 

 accident. To this extent, I suspect the genuine cases of inheritance, are very rare. 

 But habitual artificial properties, and to some extent, structures, marks etc., not unfre- 

 quently become hereditary. If, for example, men or brutes are kept healthy and 

 viEorous for several generations, by proper food and exercise, they will have more 

 Tigorons ofi'spring than the descendants of the same ancestors improperly fed and 



