54 A PURE-LINE METHOD IN CORN BREEDING 



cent more grain by weight than the self-fertilized family. In the self- 

 fertilized family, 73 ears were produced, weighing 12 Ibs., and in the 

 cross between sibs the 78 ears weighed i6y 2 Ibs. There was also a 

 striking difference between these two families as regards variability in 

 the number of the rows on the ear, as may be seen in this table : 



Unfortunately the parents of these two families were not identical 

 in the number of rows, the mother of the self -fertilized family having 

 8 rows and that of the cross-fertilized family 10. The greater height, 

 greater weight of grain produced, the higher number of rows on the 

 ears, and the greater variability in the number of rows, in the cross- 

 fertilized family, all point to the same conclusion, namely, that my self- 

 fertilized Strain A was not yet reduced completely to a homozygous 

 condition, and that the parents, or at least one of them, of my cross- 

 bred family was heterozygous. 



The two families which were the product of reciprocal crosses be- 

 tween Strain A and Strain B, have proved of great interest, for al- 

 though the individuals of both Strain A and Strain B were small and 

 weak, and the self-fertilized families of these produced respectively only 

 12 Ibs. and 13 Ibs. of ear-corn, the hybrid family in which Strain A 

 supplied the mother and Strain B the father, produced 92 ears weighing 

 48 Ibs., and the reciprocal cross produced 100 ears weighing 55 Ibs. 

 Typical ears of Strain A and Strain B, and of their reciprocal hybrids, 

 may be compared in fig. 2. If we reduce these results to bushels per 

 acre on the basis of 10,000 ears per acre and 70 Ibs. per bushel, it is 

 found that Cross AXB has produced 74.4 bushels per acre and Cross 

 BXA has produced 78.6 bushels per acre, the average for the two 

 families being nearly 77 bushels per acre. The two families which I 

 have kept continuously cross-bred during the period in which these 

 experiments have been in progress, and which have been likewise con- 

 tinually selected to 12 and 14 rows of grains, may be properly taken as 

 controls. These two families together produced 203 ears weighing 

 107^ Ibs., or at the rate of 75 bushels per acre, and when the compari- 

 son is extended so as to include my other continuously crossed fami- 

 lies 8 families in all it is found that these produced collectively at 

 the rate of a little less than 75 bushels per acre. 



