Farmers' Week in Agricultural College. 



211 



some very remarkable variations in yield, as shown by the following 

 table : 



The best pure strain has given an average yield of 40:75 bushels for 

 four years, while the poorest strains have averaged 28.88, a difference 

 of 12 bushels to the acre. It is understood that the check plat has yield- 

 ed 35 bushels, or is just about half way between the highest and lowest 

 strains. In other words, an ordinary wheat field seems to be made 

 up of a mixture of a great many strains of wheat, some of which are good 

 yielders and some of which are poor yielders. By selecting out single 

 plants and keeping the product of each plant separate, we are able to 

 isolate these poor strains and after tests are made we can rogue out 

 poor yielders and save only the good yielders. 



It will be understood that by these methods of plant breeding, little 

 attention is paid to outward appearance. For example, by the ear row 

 selection of corn breeding, all types of ears are planted and there seems 

 to be no relation between the outward appearance of an ear and its 

 ability to yield. It is the same way with wheat plants — there is no 

 way of telling by the appearance of a wheat plant whether it will make 

 a good yielding or a poor yielding strain. The only sure way, apparent- 



