208 



HORTICULTURAL SOCIETY OF NEW YO3& 



white dent, which occasionally produced a red ear, but was never before 

 known to produce ears bearing spots. The red kernels of this ear were 

 uniformly red with a small light spot at the cap of each kernel. The kernels 

 composing the white spot, although appearing very white in contrast with 

 the red ones, when examined closely proved to have very fine red lines or 

 streaks, radiating from the caps down the sides of the kernels. These two 

 types of kernels were planted separately. The red ones yielded a crop com- 

 posed of eighty-four red ears and eighty pure white ears; while the white 

 kernels with fine red stripes produced thirty-nine ears with kernels like those 

 planted, and thirty-six pure white ears. In the illustration is shown the 

 red ear with the spot at the base. The two ears labeled 70-1 represent the 

 pure red ears and the pure white ears that were produced in about equal 

 quantities by planting the red kernels of the spotted ear. The two ears 

 labeled 70-1^ represent the pure white ears and the ears bearing striped 

 kernels like those from the white portion of the spotted ear. 



It should be borne in mind that these special cases cited are illustrations 

 of the selection of individual variations, and not variations purposely pro- 

 duced by hybridization. They are given, not because they show progress in 

 desirable lines of corn improvement, but because they illustrate how readily 

 the corn plant yields to modifications of stalk, ear, or kernels, through seed 

 selection, in accordance with individual variation. The conclusions that 

 necessarily follow are these : If stalks with certain qualities are wanted, 

 seed must be taken continuously from such stalks; if ears with certain 

 qualities are wanted, seed must be taken continuously from such ears; if 

 kernels with certain qualities are wanted, such kernels must be planted ; and 

 furthermore, if stalks with certain qualities, bearing ears with certain quali- 

 ties, containing kernels with certain qualities, is what is wanted, attention 

 must be given to stalks, ears, and kernels when selecting seed. To fulfil these 

 requirements seed ears cannot be selected from a corn crib, but must be se- 

 lected from standing stalks. Every seed corn grower should have a seed 

 patch and an increase patch. In the seed patch the very best ears should be 

 planted, each row being planted with seed from a single ear. The crop from 

 each row should be weighed so as to obtain the performance record of the 

 seed ears. From the best rows a few of the choicest ears should be taken 

 and used in planting the seed patch the next year, while the other good seed 

 ears can be shelled together and planted in the increase patch, which should 

 be of sufficient size to yield enough good seed for planting the general cron. 



The many needed improvements, the possibility and method of their 

 accomplishment, have been briefly discussed without speaking of hybridiza- 

 tion. The possibilities of corn improvement are so great and are so certain 

 of accomplishment by isolation and rigid selection that little need be said 

 concerning hybridization in this connection. For one with an abundance of 

 time for keeping records, the corn plant offers one of the best fields for the 

 study of the problems of heredity. While the plant breeder has learned 

 much and will learn much more by corn hybridization, the corn grower has 

 improved his corn much and will improve it much more by seed selection. 



Breeders make use of hybridization to cause increased variation of char- 

 acters, but for the present this is not necessary in the case of corn. We 

 now have many types each exhibiting variations to a high degree, so that for 

 practical results the augmentation and fixation of desirable characters is far 

 more important than is an increase in the number of characters- 



