28 GEORGE HARRISON SHULL 



the end of 1907 only ready to make a beginning on the problems of the rela- 

 tionship between pure lines and their hybrids, which I now saw was the cru- 

 cial field that needed exploration. 



As a first step in that direction, but without as yet a full comprehension 

 of its importance, I made in July, 1907, pollinations between plants of C4, 

 which I later designated "Strain A," and a plant of C6, which later became 

 my "Strain B." I also made two sib crosses within these two strains. The 

 cross of Strain A X Strain B, which gave rise in 1908 to Fi family, D9, in- 

 volved an 8-rowed ear of the former strain (from an original selection for 

 12 grain-rows) and a 12-rowed ear of Strain B which had originated in a selec- 

 tion for 14 grain-rows. The near-reciprocal cross (Fi family, D13) resulted 

 from the application of pollen from a 12-rowed plant of Strain A to silks of 

 the same plant of Strain B, which supplied the pollen for the near- reciprocal 

 cross. 



At the time when these two near-reciprocal crosses were made between 

 Strains A and B, the truth had not yet dawned upon me that I should do the 

 same with all of my other selfed families. Aside from these two sets of crosses, 

 the handling of the cultures was the same as in previous years. The results 

 of the grain- row counts are given in Table 2.4. Unfortunately, there was con- 

 siderable damage from crows, and failures to germinate for unknown reasons. 

 The missing hills were replanted on June 8, 1908, and all of the new plantings 

 made on this date seem to have reached maturity. To overcome the suggested 

 deteriorating effect of soil depletion, the cultures were grown this year on the 

 area east of the laboratory building (occasionally referred to in subsequent 

 notes as "Fast Garden"). 



In summarizing the results for the year 1908, it may be noted first that 

 the tendency to concentrate the frequencies of the grain-rows in the extremes 

 of the range, at the expense of those in the middle, has continued strongly. 

 As before, the most noteworthy concentration is at the lower extreme. All 

 classes below 16 are considerably stronger in 1908 than in 1907 and the 

 maximum frequency is now on 12 instead of 16. This is in part due to the fact 

 that several of the lower-class families were grown in duplicate. Between 

 classes 14 and 26 the relative strength of the classes was lessened in 1908. 

 Above class 24 the frequencies were increased, there being 84 ears above 

 class 24 in 1908 and only the equivalent of about 50 in the same region in 

 1907, when raised to the same total number. The highest number of grain- 

 rows noted was 42. 



The important new features brought in by the near-reciprocal crosses be- 

 tween Strain A and Strain B and a sib cross in Strain A are presented in my 

 report to the American Breeders' Association at Columbia, Mo., in January, 

 1909, on "A pure line method in corn breeding." I find a discrepancy in that 

 the 78 ears produced by the sib cross weighed only 16.25 pounds instead of 

 16.5, as stated in my 1909 paper. Whether by an oversight or intentionally, 



