1726 The Cornell Reading-Courses 



The question of numbers is very important. One should not start 

 with too few, but should grow as many plants as possible. In the case 

 of corn, wheat, or oats, for example, it would be unwise to begin with 

 less than 50 heads or ears. If the crop is potatoes, surely not less than 200 

 tubers should be in the first choice. If a large number of individuals is 

 chosen for the beginning, the chance of finding more desirable types is 

 correspondingly greater. 



A complicated system of records, with the pedigree of every plant grown, 

 is unnecessary. The scientific man should keep such records for the pur- 

 pose of studying principles which may be put to practical use. After the 

 farmer has obtained a type, or strain, of plants superior in comparison 

 with others, he needs only to increase the seed of that strain, protecting 

 it if necessary from mixtures, until he has enough for his whole farm 

 and perhaps a surplus to sell. There is an ever-increasing demand for 

 well-bred seed of all crops. 



It is remarkable, too, how rapidly seed can be increased from a small 

 beginning. A single head of wheat may yield 150 to 200 bushels within 

 three years time. The farmer should take advantage of this generous quality 

 of nature. Seed selected from important crops will yield an abundant 

 increase in a surprisingly short time, and the yield of crops on the farm 

 will be materially increased by the use of highly bred seed. 



biometry 



The writer has said that it is not necessary for the practical man to 

 keep a complicated system of records. It is desirable, however, that he 

 have some means of sizing up the situation in order to determine varia- 

 tion and type and to see what progress he is making. For this purpose 

 the breeder, both of plants and animals, has had to call on the mathema- 

 tician for help. As a result of this, a subject called biometry has been 

 developed. The term means measure of life. To go into this subject 

 deeply would require special training. But there are some practical 

 uses to which it can be applied with but little study, no more, indeed, 

 than is needed to run a new type of farm implement. 



To determine amount of variation 



It might be supposed, for example, that a farmer is selecting corn rather 

 carefully, paying especial attention to length of ears. He should proceed 

 by measuring a random sample of 300 to 500 cars of each crop, arrang- 

 ing the results in a distribution table as shown in Fig. 76. If one person 

 measures the ears and another makes the marks in the proper square, 

 the task is comparatively short. 



