The physiological processes determining grain yield potential in winter wheat
Winter wheat (Triticum aestivum L. ) is the most important crop grown extensively in the UK. There is a gap between yield production and world demand for wheat. So, there is a need to fill this gap. The overall objective of the present study is to investigate the physiological determinants of grain yield potential in winter wheat grown in UK conditions using doubled-haploid lines derived from a cross between winter wheat varieties Rialto and Spark. These varieties are known from previous work to contrast for source and sink type traits and for the presence/absence of the 1 BL/ IRS translocation and the Rht-Dl b semi-dwarf allele. Three field experiments (at Sutton Bonington, University of Nottingham 2003/4, 2004/5 and 2005/6) were conducted examining 25 DH lines of the Rialto x Spark population and the two parents. Two post-anthesis shading treatments (with and without shading) were applied to six genotypes (four DII lines and two parents) in 2003/4 and 2004/5. Two post anthesis de-graining treatments (with and without degraining) were applied to twelve genotypes (ten DH lines and two parents) in 2003/4 and 2004/5. Detailed measurements of grain growth were carried out on the two parents for five pre-determined grain positions within the ear. A range of physiological traits were measured, including developmental stages, light extinction coefficient, radiation interception, radiation-use efficiency (RUE), green area and biomass, stem water-soluble carbohydrate reserves, floret fertility, potential grain weight, grain filling rate and duration, final grain weight and combine grain yield. A source-sink balance model which quantified the source: sink balance during grain growth indicated that the four DH lines and the two parents were sink-limited during grain growth. The sink manipulation treatment (de-graining) generally confirmed the sink limitation in this population although small positive responses of grain growth to de-graining for a few lines indicated they may have been close to source limitation. Radiation-use efficiency measured as the regression slope of dry matter on accumulated PAR intercepted between onset of stem elongation and anthesis differed amongst genotypes in 2004,2005 and 2006. Pre-anthesis RUE was positively correlated with each of grains m2 and specific leaf weight (SLW). The 1BL/IRS translocation increased RUE significantly. The characterisation of the 25 DH lines in this study showed that the differences in grain yield were positively correlated with grains m'2 but not individual grain weight. There was positive correlation between stem WSC reserves and yield in one of the experimental seasons, 2006. The ten DH lines and the two parents differed in the length of the period between GS31 and GS61 by nine days. There was a positive correlation between the duration from GS31 to GS61 and radiation interception during this period which positively affecting grains m '2 Rate and duration of grain growth and final grain weight were assessed for five grain positions (G1 to G5) for Rialto and Spark under 50% shading and a control treatment. Rialto had heavier grains associated with a longer grain filling duration than Spark. The five grain positions had similar durations of grain filling but differed in final grain weight and rate of grain growth. Grains in the central spikelet (G1, G2 and G3) were sink-limited as they did not respond to de-graining. However, grains in the basal (G4) and apical (G5) spikelets were marginally source-limited since their final weight was increased by de-graining. These results suggest that breeders should consider selecting for extra grains in proximal grain positions in basal spikelets (e. g. G4) rather than in distal grain positions in central spikelets, because these grains in this position were heavier, had faster filling rate and had the ability to respond to extra assimilates later in the season. Harvest biomass was positively correlated with grain yield amongst the DH lines. So traits to improve biomass whilst maintaining harvest index may be important for future breeding progress. It is suggested that breeders might select for an extended duration between GS31 and GS61 and higher RUE (via high SLW) to improve grains M-2 and yield potential in future years. They also should select for higher stem carbohydrate reserves to increase source size alongside grain sink size.