The source–sink ratio during grain filling is a critical factor that affects crop yield in wheat, and the main objective of this study was to determine the source–sink relations at both the canopy scale and the individual culm level under two nitrogen (N) levels at the post-jointing stage.  Nine widely-used cultivars were chosen for analyzing source–sink relations in southwestern China; and three typical cultivars of different plant types were subjected to artificial manipulation of the grain-filling source–sink ratio to supplement crop growth measurements.  A field experiment was conducted over two consecutive seasons under two N rates (N+, 150 kg ha^–1; N–, 60 kg ha^–1), and three manipulations were imposed after anthesis: control (Ct), removal of flag and penultimate leaves (Lr) and removal of spikelets on one side of each spike (Sr).  The results showed that the single grain weights in the three cultivars were significantly decreased by Lr and increased by Sr, which demonstrated that wheat grain yield potential seems more source-limited than sink-limited during grain filling, but the source–sink balance was obviously changed by climatic variations and N deficient environments.  Grain yield was highly associated with sink capacity (SICA), grain number, biomass, SPAD values, and leaf area index during grain filling, indicating a higher degree of source limitation with an increase in sink capacity.  Therefore, source limitation should be taken into account by breeders when SICA is increased, especially under non-limiting conditions.  Chuanmai 104, a half-compact type with a mid-sized spike and a long narrow upper leaf, showed relatively better performance in source–sink relations.  Since this cultivar showed the characteristics of a lower reduction in grain weight after Lr, a larger increase after Sr, and a lower reduction in post-anthesis dry matter accumulation, then the greater current photosynthesis during grain filling contributed to the grain after source and sink manipulation. 

   Good crop stand establishment and root system development are essential for optimum grain yield of dryland wheat (Triticum aestivum L.). At present, little is known about the effect of tillage and straw mulch on the root system of wheat under dryland areas in southwestern China. The aim of this study was to evaluate the effect of three tillage treatments (no-till, NT; rotary till, RT; conventional till, CT) and two crop residue management practices (straw mulch, ML; non-straw mulch, NML) on stand establishment, root growth and grain yield of wheat. NT resulted in lower soil cover thickness for the wheat seed, higher number of uncovered seeds, lower percentage of seedling-less ridges and lower tiller density compared to RT and CT; ML resulted in higher tiller density compared to NML. Straw mulching resulted in more soil water content and root length density (RLD) at most of the growth stages and soil depths. The maximum RLD, root surface area density and root dry matter density were obtained under NT. In the topmost 10 cm soil layer, higher RLD values were found under NT than those under RT and CT. There were no significant differences in the yield or yield components of wheat among the tillage treatments in 2011–2012, but NT resulted in a significant higher yield compared to RT and CT in 2012–2013. Grain yield was significantly higher in ML compared to in NML. A strong relationship was observed between the water-use efficiency and the grain yield. Both NT and ML proved beneficial for wheat in term of maintaining higher tiller density, better soil water status and root growth, leading to a higher grain yield and enhanced water-use efficiency, especially in a low rainfall year.

Synthetic hexaploid wheat (SHW) represents a valuable source of new resistances to a range of biotic and abiotic stresses. A recombinant inbred line (RIL) population with 127 recombinant inbred lines derived from a SHW-derived variety Chuanmai 42 crossing with a Chinese spring wheat variety Chuannong 16 was used to map QTLs for agronomic traits including grain yield, grains per square meter, thousand-kernel weight, spikes per square meter, grain number per spike, grains weight per spike, and biomass yield. The population was genotyped using 184 simple-sequence repeat (SSR) markers and 34 sequence-related amplified polymorphism (SRAP) markers. Of 76 QTLs (LOD>2.5) identified, 42 were found to have a positive effect from Chuanmai 42. The QTL QGy.saas-4D.2 associated with grain yield on chromosome 4D was detected in four of the six environments and the combined analysis, and the mean yield, across six environments, of individuals carrying the Chuanmai 42 allele at this locus was 8.9% higher than that of those lines carrying the Chuannong 16 allele. Seven clusters of the yield-coincident QTLs were detected on 1A, 4A, 3B, 5B, 4D, and 7D.