SYSTEMATIC IMPROVEMENT OF ANIMALS 193 



reasons well understood by the student. I attribute this failure 

 to our universal desire to experiment in something striking. 



If crossing has any value, it is not to improve the herd of a 

 farmer, but to afford material for improving the breed as a whole, 

 and even this is a long, tedious, and expensive undertaking 

 because of the operation of Mendel's law. Farmers who have 

 tried it will say that crossing produces some good animals, but 

 they are worthless as breeders. This is because of the principle 

 just mentioned and the erratic behavior of characters dominant 

 and recessive, as explained in Chapter XI. 



The practical farmer should have clearly in mind what he 

 desires to do. If he very much wants to improve the breed as 

 a whole, then well and good. All breeds need it, but he may 

 as well understand that he has undertaken a Herculean task 

 that will take much time and no little money. 



Most men are rightly after herd improvement merely ; that 

 is, to bring into their own herds the most they can afford of 

 the best that has already been accomplished in improvement. 

 Now the least that such a man can afford to do is to buy a sire 

 of the desired breed and begin at once to improve his own herd. 

 Then later he can improve the breed, if he is able. 



Rational improvement. The rational procedure for the man 

 who would improve his live stock is to secure a well-bred young 

 male of the breed he prefers and " grade up," beginning with 

 the females he has on hand, or such other common stock as can 

 be bought on reasonable terms. Let him then raise several 

 generations of grades, and later, if inclination offer and money 

 permit, he can put in a few pure-bred females with his high 

 grades and begin the production of a pure-bred herd ; or he 

 may go on with high grades indefinitely, well knowing that/<?r 

 market purposes the high grade is as good as the full blood. 



Starting in this way he will have several substantial advan- 

 tages, which may be enumerated as follows : ( 1 ) he will start 

 cheaply ; (2) he will produce relatively large numbers, making 

 rigid selection possible ; (3) he will discover the special breed 



