| A population of 188 BC2F2 and BC2F2-derived lines developed from a cross between the Colorado winter wheat cultivar 'Ankor' (recurrent parent) and the CIMMYT synthetic-derived spring wheat line 'Sokoll' was grown in replicated field trials in Fort Collins (2007 and 2008) and Greeley (2009) Colorado.;Moisture stress affected grain yield in side-by-side wet and dry treatments. The 2008 dry trial in Fort Collins experienced a 14.7% reduction in grain yield, the 2009 Greeley field trials experienced a 23.2% yield reduction, and the 2011 dry trial experienced an 18.1% yield reduction. Grain yield, biomass, and kernel characteristics (average kernel number, total spike number, and total kernel number) generally had higher mean values under wet compared to dry treatments in the different growing seasons. Test weight and average kernel weight had higher values in the wet treatments in two of three locations where there were side-by-side treatments, and spike: stem ratios had higher mean values under the dry than the wet treatment in 2011. A range of moderate to high heritability estimates for grain yield, test weight, and days to heading indicates much of the expression of these traits is genetically controlled. Heritability estimates were low to moderate for grain fill duration, plant height, and average kernel weight, indicating that environmental variation for these traits was more important than genetic variation. Heritability values were generally higher in wet than dry environments. Pearson correlation coefficients revealed positive and consistent correlations between grain yield and important yield components.;Forty-one linkage groups spanning 1414.4 cM and covering parts of all 21 chromosomes were constructed from 141 microsatellite markers and three gene-specific markers (Ppd-D1, Glu-A1, and Glu-D1) in an F2 mapping population developed from the same cross. Seventy-six of these markers were genotyped in the BC2F 2:3 lines. The population varied for the proportion of lines carrying Sokoll alleles at these markers, from 10.0 to 47.1%, with an average of 24.7%. Single-factor QTL analysis was used to detect significant (P≤0.01) markers, the size of their effects, and the source of favorable alleles. A total of 88 QTL were identified in the BC2F2-derived population. Individually, the QTL explained from 3.6 to 14.6% of the phenotypic variation. Sokoll contributed the favorable allele for 34 (38.6%) QTL (mostly on the A and B genomes), while Ankor contributed the favorable allele for 54 (61.4%) QTL. The Sokoll allele added a 235 kg/ha phenotypic effect at a QTL for grain yield located on chromosome 6B. Regions on chromosomes 1A, 1B, 2A, 2B, 2D, 3A, 4B, 5A, 5B, 5D, 7A, and 7B contained QTL for multiple traits. This co-localization of QTL for multiple traits suggests that the effects may be due to pleiotropy.;Distribution of QTL was unbalanced between irrigation treatments; 68 QTL (71.6%) were detected under full irrigation and 27 QTL (28.4%) were identified under limited irrigation. In general, QTL for most of the traits were detected in both soil moisture levels. While some QTL were detected in two to three environments, most QTL were detected only in one environment. Other markers (e.g. Xgwm484, Xgwm499, and Xbarc108 ) were consistently detected for multiple traits and under both fully and partially irrigated treatments, indicating broad adaptability and potential use of these markers marker-assisted breeding.;Mean performance of high versus low yielding lines was compared for all traits in 2010 and 2011, when the lines were grown in larger plots with more accurate yield estimates. There were significant (P≤0.05) differences between high and low lines for traits such as grain yield, test weight, and total spike number. However, there was not a significant separation in high and low means for other traits such as plant height, spike length, biomass, and spike:stem ratio. Traits with differences between high and low lines help explain the basis of yield differences in this population. The hypothesis that biomass was a good predictor of yield was confirmed in the Dailey 2010 and Fort Collins 2011 dry and wet environments, and the hypothesis that the spike: stem ratio was a good predictor of yield was rejected in all environments and at multiple growth stages in 2010 and 2011. (Abstract shortened by UMI.)... |
