16 CROSSBREEDING CORISr. 



seed produced the better, and in four the yield was the same. The 

 1908 seed produced 1,797 pounds of ears and the 1909 seed produced 

 1,938 pounds, a decreased yield of 7 per cent, due, perhaps, to the 

 poor development or loss of A^igor of the 1908 seed. The 1909 seed 

 was groA\ii from the 1908 seed under conditions that prevented any 

 crossing with other varieties. Its greater productiveness therefore 

 can not be attributed to mixture with other varieties. 



While the 1908 seed of Selection 119 produced 7 per cent less than 

 its progeny seed grown in 1909, it is not certain that the age of the 



1908 seed was the cause of the decreased productiveness as the seed 

 germinated perfectly the spring following its maturity, and also 

 showed a germination of 98 per cent in the spring of 1910. The 



1909 seed showed a germination of 100 per cent in the spring of 1910. 

 A comparison of the productiveness of the 1908 and 1909 seed of 



the female parents shows sufficient instances in which the 1908 seed 

 produced better than the 1909 seed to make the average production 

 of the 1908 seed equal to that of the 1909 seed. 



CROSSES COMPARED IN PRODUCTIVENESS WITH THE MALE PARENT. 



In computing the per cent of increased 3'ield over the male parent 

 in Table II the seed of the male parent grown the same 3'ear the cross- 

 ing was accomplished is considered. As none of the female parents 

 consistently produced better than the male parent, the male parent 

 of all the crosses. Selection 119, is taken as the basis with which to 

 compare the crosses. In all cases the average yield of two rows, the 

 nearest one on either side of the cross, is compared with the cross. 



Considering that contradictory results from any of the four tests 

 is sufficient cause for ignoring all four of the tests, we have remain- 

 ing five crosses which produced uniformly less than their male parent, 

 and two comparisons in which the cross uniformly produced better 

 than either parent. 



The five first-generation crosses (x Selection 119) that uniformly 

 produced less than the better of the two parents are Illinois Lea- 

 rning, Sturges Hybrid Flint, Silvermine, Golden Eagle, and Selec- 

 tion 77. 



The two firstr-generation crosses (X Selection 119) that uniformly 

 produced better than either parent are Cross 120 and Hickory King. 



COMBINED RESULTS OP THE FOUR MARYLAND TESTS. 



In Table III the four separate Maryland tests are combined. In 

 the Derwood test some of the female-parent varieties were represented 

 by both 1908 and 1909 seed. In such instances the average of the 

 two has been used. In these combined results this comparison of 

 the crosses with the ]>etter yielding of the two parents is a straight 

 comparison of the crosses with their male parent, for it is more pro- 



218 



