Germplasm for genetic improvement of lint yield in Upland cotton: genetic analysis of lint yield with yield components

Determination of genetic effects for lint yield and yield components in cotton (Gossypium hirsutum L.) germplasm is critical for its utilization in breeding programs. This study was designed to apply the conditional approach and an additive and dominant model to analyze genetic effects for lint yield and yield components. Forty-eight F-2 populations derived from crosses between four existent Upland cotton cultivars as female parents and 12 germplasm lines as male parents were evaluated at two locations in 2008 and 2009. Conditional and unconditional variance components were estimated by the mixed linear model based conditional approach. Lint yield and yield components were mainly controlled by genotypic effects, i.e., additive variance and dominance variance (a parts per thousand yen66 % of total phenotypic variation). Lint percentage and lint index had the highest proportions of additive variance component to the total phenotypic variances. SP156 and SP205 had positive additive effects for lint yield and yield components, and were also parents of the most hybrids with positive predicted dominant effects. Therefore, these two lines are good combiners for development of both pure lines and hybrids. Positive additive contribution effects to lint yield from lint percentage, boll number, boll weight, and seed index were detected in different parents. Adding seed index to boll number and lint percentage increased additive contribution effects to lint yield from these two components relative to the contribution effects from either boll number or lint percentage alone. Results in this study suggest that boll number, lint percentage, and seed index should be balanced in pure line development.