484 ANNUAL REPORT OF THE Off. Doc. 



can be depended upon. Under ordinary farm conditions the ability 

 of individuals to reproduce themselves is not tested, and furnishes 

 a very important reason why little progress has been made in the 

 improvement of field crops. Take timothy, for example. A casual 

 inspection of a field of timothy will show that there is a great varia- 

 tion in the length of head, the length of stem, the amount of leaves 

 and the number of stalks per stool. Under the usual method no 

 selection is exercised and no test of the power of transmission of 

 characters is possible. A few experimenters have selected plants 

 (stools) having different characteristics and by planting 100 seeds 

 from each plant in rows, one seed at a place, have obtained remarka- 

 able results. After the ability of the plant to transmit its charac- 

 ters has been demonstrated, the seed can be rapidly mutilated for 

 field purposes. 



It is well understood by live stock breeders that the best individual 

 does not always produce the best progency. It is a common ex- 

 pression that this animal is a good breeder or that animal a poor 

 breeder. 



At the Ohio State University in 1902, fourteen ears of maize of a 

 given variety were selected and two rows of fifty hills each were 

 planted from each ear. The smaller ear containing next to the 

 smallest weight of corn produced the heaviest yield of corn. This 

 ear weighed 14 per cent, less than the average w^eight of the fourteen 

 ears and yielded 32 per cent, more than the average yield of the same 

 fourteen ears. This testing of the power of plants to transmit their 

 characteristics is painstaking work and will form a large part of the 

 work of the successful plant breeder. 



The Impwtance of Large Numher^: If a thousand persons stand in 

 a row it will be found that most of them are nearly the height of 

 the average, while a few are considerably shorter and a few con- 

 siderably taller than the average. The length or weight of a number 

 of ears of maize will vary in the same manner. This chart shows the 

 yield of 509 hills of Early Ohio potatoes grown this year at the Cor- 

 nell Agricultural Experiment Station. Most of these it will be 

 noticed varied uniformly and to a small extent from the average of 

 all, while a few hills yield much more and a few less than the aver- 

 age. The average yield of the ten poorest hills was 44.4 grams, and 

 of the ten best hills was 1035.4 grams, or on the basis of 80 per cent. 

 of a stand 14 and 320 bushels respectively. No such difference in 

 hereditary power can be expected, since part of the difference, how 

 much we do not know, is due to environment. There will be a ten- 

 dency on the part of the poorest and best when planted to return to 

 the common or average type. What it is desired especially to em- 

 phasize is the fact that it seems to be a universal law of organic 

 being for most of them to breed true to type while a few of them vary 

 considerable from that type. In order, therefore, to make progress 

 in breeding, it is necessary to find the organisms that have the ten- 

 dency to vary as desired. Among a million organisms there may be 

 only one that possesses the required characteristics. The chances 

 of finding the desired individual increase as we increase the num- 

 ber from which selection is made, the chances of securing satisfac- 

 tory results are increased many-fold if 5,000 seeds are planted in- 

 stead of 500. 



Plant Breeder's Advaritages ;• It has been shown that the breeder 



