| Cotton harvest ing mechanization is the only way for sustainable developmen t o fcotton industry and to increase the farmers'income in Xinjiang. Breeding mechanicalharvesting varieties of cotton is a safeguard extend ed tool of mechanical harvest ingtechnology. Study on the genetic effects and the QTLs of mechanical harvesting cottonvarieties related to shape and properties would be helpful for selection of harmoniousrelationship between the mechanical picking, high yield and quality, and molecularbreeding through marker assisted selection (MAS) in cotton.In this study,6representative mechanical harvesting varieties and2non-mechanical harvest ing varieties of upland cotton,28F1and28F2hybrids wereproduced using diallel genetic mating design method. Field experiments were carriedout during2010, in two typical cotton growing regions of southern and northernXinjiang. Total of nine teen related traits were investigated. Statistical analysis wasconducted using genetic ADAA model and QTL analysis by Windows QTLCartographer5.0composite interval graphic method. The main results are as follows:(1) The results indicated that the dominance effect and additive×additiveepistatic effect mainly controlled the heredity of mechanically harvest ing relatedtraits of the off-spring. Significant additive×environment effect andadditive×additive epistat ic×environment effect was existed on first fruit branchheight, lint yield at p re-frost rate, day of flowering to boll opening, growth periodbetween southern and northern provinces of Xinjiang. Agronomic traits heritabilityis low. For agronomic harvest related traits, early generation selection in breedingprogramme is not suitable for mechanical harvesting cotton varieties.(2) Genetic correlation analysis show ed that genotype and phenotype correlationcoefficients of plant height, first fruit branch height and node, lint yield at pre-f ros trate are positively correlated. However, growth period and micronaire are negativel ycorrelated.Genotype correlation of first fruit branch node, growth period fiberlength, fiber uniformity degree are positively cor related. Decomposition ofcorrelation coefficients show ed that first fruit branch height and growth period bollweight, lint rate, lint yield significant additive×additive epistatic effect aranegatively correlated. Moreover, first fruit branch and growth period, micronaireadditive and additive epistatic×environment interaction effect are negativecorrelated.(3) The heterosis analysis revealed that heterosis over mean of parents and overbetter parents of main mechanical harvest traits in F1and F2are smaller. The heterosisover better parents of the most mechanical harvest traits is found negative. There areonly few combination hybrids that possessed perfect performance in mechanical harvest traits.(4) Results further showed that there might be pheno type contribution from5mechanical harvesting traits (plant height, first fruiting branch height, first fruintingbranch node, length of internode, percentage of crop harvested at pre-frost) to lin tyield (ranged from–20to–14%). The height of first fruit branch had largestdominance contribution to lint yield, while there was bigger restraining effect frompercentage of crop harvested at pre-frost to dominance effect of lint yield. Thelargest additive×additive epistatic effect contribution ratio was observed frompercentage of crop harvested at pre-frost to lint yield followed by internode lengthof first branch. The5mechanical harvesting trai ts had showed little dominancecontribution to lint yield, regardless percentage of crop harvested at pre-frost hadexisted largest dominance contribution to lint yield than other4mechanicalharvesting traits. In special environment, these existed diffe rences mainly inmechanical harvesting selection traits for the additive×additive epistatic effect of lintyield in different crosses.(5)Our studies concluded that plant height, node of first fruiting branch, leafsize, days of seedling to budding and da ys of budding to flowering w ere positivelycorrelated with the height of first fruiting branch. A total of5QTLs were detected b ycomposite interval mapping, including3QTLs for height of first fruiting branch onthe linkage group GL01. The variances exp lained by each QTL were ranged from0.04to8.18%. However, the remaining2QTLs of which1for first fruiting branch nodewas detected on the same area of the linkage group GL01, with0.01%variances,whereas, another1for days of seedling to budding was detected on the linkage groupGL02, with12.64%variances. The inheritance distance was observed142.05cM.
|
PDF Full Text Request