34 



GEORGE HARRISON SHULL 



summary form. Some 73 ears were selected for planting, and 5,343 ears were 

 harvested. The complete grain-row distribution was as follows: 



Grain-rows 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 Total 



Frequencies ... . 401 812 1271 921 716476275 141 118 74 53 41 24 8 6 4 1 1 5343 

 Percentages. ... 7.5 15.2 23.8 17.2 13.4 89 5.2 2.6 2.2 14 10 0.8 0.5 0.2 0.1 0.1 00 0.0 100.0 



To save space and still indicate as completely as possible the significant 

 results of these studies in 1910, the data from the several kinds of families 

 of white dent corn grown at the Station for Experimental Evolution that 

 year are presented in the form of averages in Table 2.6. The several quanti- 

 tative indicators of physiological vigor, namely, the average number of 

 grain-rows, heights of stalks, and bushels of ear-corn per acre, can be readily 

 compared as follows: 



Types of Families 



Inbreds selfed .... 

 InbredsXsibs. . . . 



Crossbreds 



Fi between inbreds 

 F2 from Fi selfed. . 

 F2 from Fi X sibs . . 



Av. Heights 

 in Dms. 



19.3 

 19.8 



23.5 

 25.7 

 23.3 

 23.1 



Av. Yields 

 in Bu./A. 



25.0 

 28.7 

 63.5 

 71.4 

 42.6 

 47.9 



Si.x interesting comparisons can be made among these summaries: (1) 

 comparisons between inbreds selfed and inbreds crossed with pollen from 

 one or more of their sibs; (2) comparisons between inbreds and crossbreds 

 in which selfing has been completely prevented, but which still represent a 

 (fairly low) degree of inbreeding; (3) comparisons between inbreds and their 

 Fi hybrids; (4) comparisons between the crossbreds in which selfing has been 

 prevented through six generations and the Fi hybrids in which five successive 

 generations of selfing have been succeeded by a single cross; (5) comparisons 

 between the Fi and the F2 hybrids of the inbreds; and (6) comparisons be- 

 tween F2 hybrid families produced by selfing the Fi and those F2 families 

 produced by sibcrosses in the Fi. 



On making these comparisons we see that the evidence for residual hetero- 

 zygosity in the inbreds is indicated by excesses in the sibcrossed families of 

 the inbreds over the selfed inbreds of 8.7 per cent in grain-row number, 2.8 

 per cent in heights of stalks, and 14.7 per cent in yield of ear-corn. In the Fo 

 families (sections E and F, of Table 2.6) those produced from sibcrosses in 

 the Fi surpass those families produced from selfings in the Fi by 0.9 per cent 

 in grain- row number and 12.5 per cent in yield. 



The average heights of stalks reverse the expectation by showing an in- 

 significantly less height from the sibcrossed matings than from the selfings, 

 the difference being 0.9 per cent. The contrast between the results of six 

 successive selfings and the continued prevention of selfing for the same six 



