138 PLANT-BREEDING 



menting the yield is the most essential purpose of all breed- 

 ing. The comparison, of course, suffers from the cross- 

 pollinated condition of the mother ears, but, as we have seen, 

 as a rule, not to any noxious degree, and the plants grown 

 from such hyl^rid kernels will probably be thrown out by 

 the first selection. The plot should be protected as effect- 

 ively as possible from contamination by pollen from unbred 

 varieties. As a rule, it will hardly be possible to place it on 

 good soil at a sufficient distance from the remaining fields, 

 and a protection by hedges or timber will equally be too 

 cumbersome in ordinary cases. The best plan is to place 

 it in the midst of a large field of a selected strain and to sur- 

 round it by three or more rows, sown with the seeds of the 

 selected ears which remain after the preparation of the seed 

 corn for the main rows. The first contrivance will, of course, 

 not be available in the first season after starting the breeding 

 plot, but from the third year it will always be practicable. 

 In the choice of the best place, attention is to be given to 

 the direction of the prcvaihng winds, that they may carry as 

 few pollen grains from the adjacent fields as possible. It 

 has often been ascertained that pollen has drifted over a 

 quarter of a mile, and by the planting of stray plants of sweet 

 corn and the estimate of the number of starchy grains pro- 

 duced on them, some knowledge concerning this transpor- 

 tation of pollen could easily be secured. Dr. Hopkins has 

 pointed out the dangers of repeated self-pollination in the 

 breeding plot and recommended the detassehng of alter- 

 nate rows and the harvesting of the ears of these rows only, 

 in order to meet this difficulty, but with this question we are 

 not here concerned. In the first breeding year close-pollin- 

 ation of the progeny of the same mother ear should be appre- 

 ciated as a means of shortening the period of subsequent 

 selection; it will probably prove harmless even if a repeate 1 

 close or self fertilization should prove objectionable. 



