Comparison of heterosis among hybrids as a measure of hybrid relatedness with that to be expected on the basis of pedigree

--J. S. C. Smith and O. S. Smith

The use of heterosis data to measure germplasm associations among hybrids was proposed by Troyer et al. (Crop Sci. 28:481-485, 1988). Comparisons of the associations between hybrids using heterosis, isozymic, and zein HPLC data have been shown to give broad agreement (Smith et al., Proc. Ann. Corn Sorghum Ind. Res. Conf. 42:187-203, 1987). However, no comparison has been reported for the associations that would be shown by heterosis data with those that would be expected on the basis of known pedigree. This comparison is important to make in order to test the validity and accuracy of the assumptions upon which the use of heterosis to estimate relatedness among hybrids is founded.

In this study, we measured distances among 10 commercial hybrids currently released by Pioneer Hi-Bred International, Inc. Relationships on the basis of pedigree between pairs of hybrids ranged from 99% (Wright's coefficient) or 50% (Malecot's coefficient) to 6% (Wright) or 4% (Malecot). Yield data were collected at 3 reps per location (60 plants per rep) with 7 locations during 1988; all generations of seed were made simultaneously at Johnston in 1987. Distances between hybrids on the basis of heterosis were calculated according to the formula proposed by Troyer et al. (Crop Sci. 28:481-485, 1988). A correlation of r2 = 0.81 was found between inter-hybrid distances calculated from the heterosis and pedigree data. These preliminary data (a second round of heterosis measures will be made in 1989) suggest that heterosis could be utilized to estimate germplasm and pedigree similarity among hybrids. However, RFLP data (reported by us in this Newsletter) provide estimates of relatedness that are more expeditiously arrived at and which can show a high degree of correlation both with pedigree and with agronomic performance (F1 yield); the correlation between F1 yield and RFLP distance data being greater than that between F1 yield and pedigree distance data. Thus, RFLP or 2-dimensional gel electrophoresis data (see Higginbotham, Smith, and Smith in this Newsletter) may be more useful and reliable means to characterize germplasm as aids in breeding and agriculture.

Please Note: Notes submitted to the Maize Genetics Cooperation Newsletter may be cited only with consent of the authors

Return to the MNL 63 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage