482 BOTANICAL GAZETTE [DECEMBER 



this hybridity. Previous conclusions that F x hybrids between self-fertilized 

 strains are on the average equal in yielding capacity, and in certain combina- 

 tions much superior, to strains cross-bred in the normal manner, have been 

 confirmed; also that reciprocal crosses are essentially equal. In addition it 

 is shown that the yield and quality of the crop are functions of the particular 

 hybrid combination, the results being the same whenever the cross is repeated. 

 The Fj was found no more variable than the pure self-fertilized parental strains, 

 but the F 2 was considerably more variable, the coefficients of variability in 

 number of rows on the ears in pure biotypes and in their F x hybrids being 

 respectively 9.081 per cent and 9.063 per cent, while that of the F 2 

 was 12.63 per cent. This greater variability in the F 2 , also noted by 

 EMERSON 9 in respect to size of seeds and height of stalks, though the 

 latter gives no coefficients, clearly demonstrates the segregation of different 

 grades of such purely quantitative characters. EAST has also presented 

 similar evidence of the segregation of a quantitative character (height of 

 stalk) in the F 2 , but he makes no reference to the reviewer's corresponding 

 results published a year earlier. He gives no coefficient of variability for 

 pure strains, but his coefficient for the F x was 8 . 68 per cent, while in the several 

 F 2 families it ranged from 12.02 per cent to 15. 75 per cent. 



The theory that the increased vigor of cross-bred maize plants is due to a 

 stimulation accompanying heterozygosis requires that crossing within the 

 same biotype or within the same F x shall give no advantage over self-fertiliza- 

 tion in the same biotype or in the same FL The reviewer 11 has reported on a 

 number of such sib-crosses in comparison with corresponding self-fertiliza- 

 tions, the advantage in favor of the crosses being so slight that they can be 

 fairly accounted for by the lack of complete genotypic purity in some of the 

 self-fertilized families. Crosses between sibs in ten self-fertilized families had 

 an average height of 20 dm. and gave an average yield per acre of 30. 17 bushels 

 as compared with a height of 19. 28 dm. and a yield of 29.04 bushels in the 

 offspring of self-fertilized parents. In the F 2 those families which sprang 

 from sib-crosses in the F x had an average height of 23 . 30 dm. as compared 

 with 23.55 dm. in families produced from self-fertilized parents, and the 

 corresponding yields per acre were 47.46 and 41.77 bushels respectively. 

 These results show that cross-fertilization is of no (or little) advantage except 

 when it brings together unlike hereditary elements. The relations of F z 

 and F 2 in regard to height of plants and yield per acre strikingly emphasize 

 the economic importance of using hybridized seed corn. Ten Fj families had 

 an average height of 25 dm. and produced an average yield of 68.07 bushels, 



' EMERSON, R. A., The inheritance of sizes and shapes in plants. Amer. Nat. 

 44:739-746. 1910. 



10 EAST, E. M., The genotype hypothesis and hybridization. Amer. Nat. 45: 

 160-174. figs. 6. 1911. 



11 SHULL, G. H., The genotypes of maize. Amer. Nat. 45:234-252. fig. i. 1911. 



