104 EVOLUTION AND ANIMAL LIFE 



speed in their various classes and ages, and the experiment 

 was in the highest degree successful. In one sense such at- 

 tempts are not experiments. The skillful breeder knows that 

 out of the many combinations possible in crossing, some few 

 will fall in line with his plans. He has only to preserve these, 

 and to clinch them by in-and-in or segregated breeding to 

 bring about a result he may have deemed possible or desir- 

 able. It is possible, by intentional selection, to turn a non- 

 essential or race character into a selective or adaptive one. 

 The Hampshire sheep have black ears, but by persistent se- 

 lection the ears could probably be made white. Probably also 

 the horns of the Dorsets could be bred on Hampshires by 

 making use of possible occasional reversions to the horned 

 stock. This result could be attained very rapidly by a cross- 

 ing with Dorset stock, but this triumph of the breeder's art 

 has rarely any homologue in the wild state or in the condition 

 of unconscious selection. 



When selection ceases, the adaptive characters are likely to 

 decline or disappear. Under cessation of selection, called by 

 Weismann panmixia, no premium is placed on traits of excel- 

 lence, from the human standpoint, such as long wool, plump- 

 ness or symmetry of form; and only the purely vegetative ad- 

 vantages of the individual count. But while the traits of 

 excellence disappear, the race traits or nonadaptive characters 

 persist unchanged. A herd of neglected Hampshire sheep is 

 still a herd of Hampshires. The black face, ears, and legs 

 remain black, with no tendency to fade. 



When the worst individuals are selected for breeding, we 

 have the reversal of selection. A flock of Hampshire culls, 

 feeble, loose-jointed, scant-wooled, unsymmetrical, could be 

 used in breeding, and the adaptive characters usually sought 

 for could be bred out of them. But they would still be Hamp- 

 shires, for the hereditary characters which had persisted with- 

 out the aid of selection would persist after selection ceases or 

 even if it is reversed. When these same characters are made the 

 object of selection, they are subject to the same laws as ordinary 

 adaptive characters. 



What is true of a breed of sheep a product of geographical 

 isolation with segregative breeding is true in a general way of 

 any wild species of animals or plants. Its adaptive characters 

 are due to natural selection. These change more rapidly than 



