138 Bulletin 251. 



If one is to use the Nursery method, the plants must be especially 

 planted. While the nursery method certainly allows the breeder to distin- 

 guish the individual plants more clearly, in crops like wheat, oats, and so 

 on, which are sown broadcast or drilled, it entails very much extra work 

 and is probably to be recommended only for the use of experimenters 

 who are giving their entire time to the work. 



In selecting the best plants in any crop the breeder must aim to examine 

 a very large number of plants and carefully compare their important 

 characters. To know what the important characters are, it is neces- 

 sary to be familiar with the crop and have a thorough knowledge 

 of those qualities wdiich go to make up a plant of the greatest intrinsic 

 value. In some cases breeders have given primary attention to some 

 quality which is largely secondary in nature. Corn-breeders, as an 

 illustration, have given great attention to getting ears well filled out over 

 the tip. This character is of no value except to produce a good ear for 

 exhibit and would be of no value there if the ordinary score card did not 

 require it. Such a character is of no value, unless it is correlated with 

 heavy yield, and the writer knows of no evidence to show that this is the 

 case. What the corn-breeder desires is the variety that will give the 

 largest yield per acre of good grain. If this variety happens to bear ears 

 well filled over the tip, well and good. The filling of the tip is not a 

 detrimental character. If, however, this heavy yielding capacity is found 

 in a variety in which the tip of the ear is not so well filled it does not 

 materially matter, as this in general is not a detrimental character. 



In making the first selections it is usually the best policy to make a 

 preliminary selection of a much larger number of plants than are actually 

 desired. The breeder can then examine these selections with greater care 

 and discard the poorest from among them retaining only the superior 

 individuals. 



Careful breeders have found it very desirable to have a clearly defined 

 ideal type which they are striving to produce. In selections within the 

 race the breeder should have all of the characters of the race which he 

 is breeding clearly in mind in order to adhere strictly to the type of the 

 variety ■ in the selections. In making selections of new variations, 

 mutations, etc., in attempting to secure new races, naturally no one type 

 can be adhered to. In testing these different individuals, however, the 

 characters of a certain type should be borne in mind and deviations from 

 this type in the progeny should be weeded out. 



The individual the unit of breeding. 



The unity of the individuals is also an important factor in plant-breed- 

 ing. If, for instance, one is attempting to produce a seedless fruit, it is 

 important that he discover a plant which shows a tendency to produce 



