Greater embryo size in a large and carefully phenotyped mapping population was genetically associated with a greater number of longer seminal roots to increase grain yield in droughted field environments. Breeding modification of root architecture is challenging in field environments owing to genetic and phenotypic complexity, and poor repeatability with root sampling. Seeds from a large mapping population varying in embryo size were harvested from a common glasshouse and standardised to a common size before assessing in rolled germination paper at 12 and 20 °C for seedling growth. Differences in genotype means were large and heritabilities high (h2 = 0.55–0.93) indicating strong and repeatable genotypic differences for most root traits. Seminal roots 1 to 3 were produced on all seedlings, whereas growth of seminal roots 4, 5 and 6 was associated with differences in embryo size. Increases in seminal root number from 4 to 6 per plant were strongly, genetically correlated with increases in total seminal length (rg = 0.84, < 0.01). Multivariate analysis confirmed initiation and growth of seminal roots 1, 2 and 3, and of roots 4, 5 and 6 behaved as genetically independent (rPg = 0.15 ns) cohorts. Tails representing extremes in seedling root length and number were associated with significant differences in grain yield of up to 35% in droughted field environments but were not different in irrigated environments. Increases in grain yield were linked to greater lengths of seminal roots 4, 5 and 6 and were largely independent of plant height or development. This is the first report on the genetic relationship of seedling root architecture and embryo size, and potential in selection of seminal root size for accessing deep-soil moisture in droughted environments.

Increases in wheat performance under drought have been demonstrated in selection for factors contributing to greater water-use efficiency. Crop growth models have indicated the potential for significant increases in wheat yields across a wide range of wheat growing environments when both greater early vigour and higher integrated transpiration efficiency (TE) are selected together. However, greater early vigour comes with the possible trade-off of decreasing TE due to the production of larger, thinner leaves. A number of topeross-derived populations were specially developed for greater early vigour across a range of elite commercial backgrounds. These populations were assessed across multiple field environments and shown to vary significantly for early leaf area development (mean plant leaf area of 17-54 cm(2) at 3.5 leaf stage) and carbon isotope discrimination (20.8-22.4%), and exceeded the range for both traits across tested commercial varieties and parental genotypes. We demonstrate that it is possible to combine greater early vigour and greater integrated TE (measured as lower carbon isotope discrimination (CID)), and that increased integrated TE is associated, in part, with increases in photosynthetic activity. We also show that alternate GA-sensitive dwarfing alleles allow greater expression of early vigour when compared to traditional GA-insensitive dwarfing alleles. The potential exists for development of high water use efficiency wheat varieties combining alleles for greater early vigour and integrated transpiration efficiency. (C) 2015 Elsevier B.V. All rights reserved.

The breeding of wheat with greater early vigour has potential to increase water- and nutrient-use efficiency, as well as to improve weed competitiveness to raise crop yields profitably. Given that wheat is inherently conservative in its early growth, a sustained breeding effort was initiated to increase genetically seedling leaf area in developing novel high vigour germplasm. A recurrent selection programme was initiated by intercrossing a genetically diverse set of 28 vigorous wheat lines identified globally. These were intercrossed at random and S1:2 progeny with the largest leaf 1 and 2 widths were intermated to develop new populations for assessment of early growth. This procedure was repeated for up to 60 segregating families per cycle across six cycles over 15 years. Thirty random S1:2 progeny were retained from each cycle and seed-increased together to produce seed for early vigour assessment in multiple sowings. The most vigorous wheat seedlings were identified in later cycles, with some lines producing more than double the leaf area and biomass of elite commercial wheat varieties. Phenotypic selection for greater leaf width was associated with a realized significant (P<0.01) linear increase per seedling of 0.41mm per cycle (+7.1%) for mean leaf width, and correlated linear increases in total leaf area and biomass of 4.48cm2 per cycle (+10.3%) and 10.8mg per cycle (+5.3%), respectively. Genetic gains in widths of leaves 2 (+8.4%) and 3 (+11.5%) were significantly (P<0.01) greater than for leaf 1 (+5.3%). Selection for greater leaf width was associated with linear increases in coleoptile tiller leaf area, small curvilinear increases in leaf 1 length, and reductions in numbers of leaves and mainstem tillers. Genetic variances were large and heritabilities high for leaf width and total leaf area in each cycle, but reduced linearly in size with selection across cycles. Coupling diverse germplasm with a simple, inexpensive, and repeatable selection process has confirmed the value of recurrent selection in developing uniquely vigorous wheat germplasm for use as parents in commercial breeding.

Increasing climate variability, particularly variability in the timing and amount of soil water, means that breeding wheat (Triticum aestivum L.) varieties with stable high grain yields is increasingly more challenging. Changing environmental conditions in water-limited rainfed environments will alter genotype ranking to reduce confidence in the identification of consistently higher yielding performers. Greater early vigour (EV) and transpiration efficiency (TE) are two physiological traits that have demonstrated benefits as breeding targets for efficient water-use in Mediterranean in-season water and monsoonal stored water environments, respectively. This Perspective discusses the hypothesis that combining higher TE and greater EV will broaden the adaptation and increase grain yields for wheats grown across most rainfed environments. We examine the physiology underpinning adaptation with greater EV and higher TE, as well as the challenges and potential benefits of deploying these traits in combination. We then discuss how these two traits interact with different environments and, in particular, the different wheat-growing regions of Australia. We conclude that the combination of these two traits is genetically and physiologically feasible, as well theoretically beneficial to average yield in most rainfed environments. Hence, we suggest a strategy for reliably managing the complex genetics underpinning EV and TE when phenotyping and selecting both traits in commercial wheat breeding programs.

Background: The flag leaf of a wheat (Triticum aestivum L.) plant rolls up into a cylinder in response to drought conditions and then unrolls when leaf water relations improve. This is a desirable trait for extending leaf area duration and improving grain size particularly under drought. But how do we quantify this phenotype so that different varieties of wheat or different treatments can be compared objectively since this phenotype can easily be confounded with inter-genotypic differences in root-water uptake and/or transpiration at the leaf level if using traditional methods?Results: We present a new method to objectively test a range of lines/varieties/treatments for their propensity of leaves to roll. We have designed a repeatable protocol and defined an objective measure of leaf curvature called "rolled-upness" which minimises confounding factors in the assessment of leaf rolling in grass species. We induced leaf rolling by immersing leaf strips in an osmoticum of known osmotic pressure. Using micro-photographs of individual leaf cross-sections at equilibrium in the osmoticum, two approaches were used to quantify leaf rolling. The first was to use some properties of the convex hull of the leaf cross-section. The second was to use cubic smoothing splines to approximate the transverse leaf shape mathematically and then use a statistic derived from the splines for comparison. Both approaches resulted in objective measurements that could differentiate clearly between breeding lines and varieties contrasting genetically in their propensity for leaf rolling under water stress. The spline approach distinguished between upward and downward curvature and allowed detailed properties of the rolling to be examined, such as the position on the strip where maximum curvature occurs.Conclusions: A method applying smoothing splines to skeletonised images of transverse wheat leaf sections enabled objective measurements of inter-genotypic variation for hydronastic leaf rolling in wheat. Mean-curvature of the leaf cross-section was the measure selected to discriminate between genotypes, as it was straightforward to calculate and easily construed. The method has broad applicability and provides an avenue to genetically dissect the trait in cereals.

The green-revolution Rht-B1b and Rht-D1b dwarfing alleles are usually associated with increased wheat yields but are linked to reduced early growth and poor emergence if sowing conditions are unfavourable. Other dwarfing genes are available but not used in commercial breeding. The Rht13 bread wheat donor, Magnif M1, produces uniquely short peduncle and penultimate internodes to reduce plant height. A set of near-isogenic (NILs) and recombinant inbred (RILs) lines varying for height were developed from the cross of Magnif M1 and the Rht8c-containing Chuan-mai 18, and evaluated for a range of agronomic characteristics across favourable environments. Reductions in plant height were associated with increased grain number (r(2) = 0.35**) and harvest index (r(2) = 0.62**) in the NILs. Reduced-height RILs containing the Rht13-linked, Xgwm577(M) microsatellite marker were significantly shorter, produced greater biomass, yield and harvest index, and increased spike and grain number than lines without the marker. Approximately 74 and 7% of the total phenotypic variance in plant height was accounted for by allelic differences in Xgwm577 and Rht8 loci, respectively. The peduncle and penultimate peduncle internodes of Rht13-containing lines were proportionately shorter than Rht8c-containing sibs and lines containing the Rht-B1b dwarfing allele. The unique height-reducing phenotype, increased grain number and yield associated with Rht13 indicate considerable potential for use of this dwarfing allele for improving wheat performance. Crown Copyright (C) 2011 Published by Elsevier B.V. All rights reserved.

Improved ability to establish well when sowing at depth, into crop residues or hard soils should lead to increased yields in these situations. The semi-dwarfing Rht-B1b and Rht-D1b genes reduce plant height and increase grain number and yield in favourable environments. However, these genes are associated with reduced coleoptile length and leaf length extension to slow leaf area and biomass accumulation especially when seed are sown deep. Preliminary evidence indicates the potential of Rht4, Rht5, Rht8, Rht12 and Rht13 gibberellin-responsive (GAR) dwarfing genes to reduce plant height without affecting seedling vigour. Four large, inbred populations were generated varying for presence of one or more GAR-dwarfing genes. Lines were genotyped with molecular markers linked to each dwarfing gene and grown in multiple environments to evaluate seedling growth and agronomic performance. Genotypic variation was large for plant height, aerial biomass, grain yield and its components, grain number and size. Height reduction was greatest for Rht5 (−55%), Rht12 (−45%), Rht13 (−34%), Rht4 (−17%), and to a lesser extent Rht8 (−7%). In comparison, height reductions associated with Rht-B1b averaged 23%. Reduced height was genetically correlated with reduced lodging score (rg=0.84–0.93), increased dry-matter partitioning to grain (i.e. harvest index; rg=−0.46** to −0.86**) and increased grain number (rg=−0.22* to −0.73**). Most dwarfing genes were associated with increased grain number: Rht13 (+27%), Rht4 (+19%), Rht12 (+19%), and Rht-B1b (+9%). Rht8 had little effect on grain number (−1%) whereas later maturity associated with Rht5 contributed to reduced grain number (−66%). The influence of dwarfing genes on aerial biomass was negligible, with some Rht4, Rht12 and Rht13 semi-dwarf lines identified combining greater partitioning and aerial biomass to increase grain yield. Compared to tall siblings, coleoptile lengths and seedling leaf breadths were largely unaffected by GAR-dwarfing genes but leaf length was on average smaller in lines containing Rht5 or Rht12. These studies demonstrate the potential of GAR-dwarfing genes for increasing grain number and yield without compromising aerial biomass or coleoptile length in bread wheat.

INTRODUCTION Leaf rolling may be associated with improved grain yield in some drought situations. Evidence of varietal differences for leaf rolling in wheat (Triticum aestivum L.) has been reported [1], but studies investigating the amount and nature of genotypic variation in leaf rolling of wheat are rare [2]. The aim of this work was to determine the genetic system governing leaf-rolling in wheat using two separate approaches: 1 – by analysing generation means in two bi-parental crosses; and 2 – by using a diallel-mating design.

Selection for rapid leaf area growth has the potential to increase wheat biomass, and both water-use efficiency and weed competitiveness early in the season. Several morphological components contribute to increased seedling leaf area, including rapid seedling emergence and production of longer, wider leaves. Early emergence of a large coleoptile tiller has also been demonstrated to increase plant leaf area and biomass in wheat and other grass seedlings. Yet little is known of the extent and nature of genotypic variation for coleoptile tiller growth in wheat. A random set of 35 wheat, barley, and triticale genotypes was evaluated in glasshouse and outdoor studies for seedling characteristics, including coleoptile tiller growth and total plant leaf area. Coleoptile tillers were produced more reliably for seedlings grown outdoors and when supplied with additional soil nitrogen. Genotypic differences in coleoptile tiller frequency and leaf area were large, ranging from 0 to 78% and from 0.0 to 1.4 cm2, respectively at very early growth stages. Australian commercial wheats tended to produce fewer coleoptile tillers of smaller size than overseas germplasm where the coleoptile tiller accounted for up to 12% of total seedling leaf area. This compared favourably with mainstem tiller leaf area, which ranged from 0 to 3.5 cm2 and accounted for up to 16% of plant leaf area. Broad-sense heritabilities were high for coleoptile tiller presence and size in favourable conditions (c. 75%) but low (c. 40%) for seedlings evaluated across nitrogen content-varying soils. Generation means analysis was used to investigate genetic control for coleoptile tiller growth across multiple populations. Significant (P < 0.05) differences were observed among generations for coleoptile tiller frequency and growth (numbers of leaves, leaf area, and biomass). These differences reflected strong additive genetic control with little evidence for any gene action × year interaction. Increases in coleoptile tiller frequency and mass were correlated with larger embryo size and wider seedling leaves to increase seedling leaf area (rg = 0.89). Comparisons between reciprocal F1 and F2 generation means indicated strong maternal effects for coleoptile tiller growth in some but not all crosses. Screening in favourable environments will increase heritability and aid in selection for progenies producing large coleoptile tillers. Evidence for additive genetic control should permit early generation selection but not without some progeny-testing for coleoptile tiller growth together with other early vigour components associated with increased plant leaf area.

Wheat productivity is commonly limited by a lack of water essential for growth. Carbon isotope discrimination (Δ), through its negative relationship with transpiration efficiency, has been used in selection of higher wheat yields in breeding for rainfed environments. The potential also exists for selection of increased Δ for improved adaptation to irrigated and high rainfall environments. Selection efficiency of Δ would be enhanced with a better understanding of its genetic control. Three wheat mapping populations (Cranbrook/Halberd, Sunco/Tasman and CD87/Katepwa) containing between 161 and 190 F1-derived, doubled-haploid progeny were phenotyped for Δ and agronomic traits in 3–5 well-watered environments. The range for Δ was large among progeny (c. 1.2–2.3‰), contributing to moderate-to-high single environment (h 2 = 0.37–0.91) and line-mean (0.63–0.86) heritabilities. Transgressive segregation was large and genetic control complex with between 9 and 13 Δ quantitative trait loci (QTL) identified in each cross. The Δ QTL effects were commonly small, accounting for a modest 1–10% of the total additive genetic variance, while a number of chromosomal regions appeared in two or more populations (e.g. 1BL, 2BS, 3BS, 4AS, 4BS, 5AS, 7AS and 7BS). Some of the Δ genomic regions were associated with variation in heading date (e.g. 2DS, 4AS and 7AL) and/or plant height (e.g. 1BL, 4BS and 4DS) to confound genotypic associations between Δ and grain yield. As a group, high Δ progeny were significantly (P < 0.10–0.01) taller and flowered earlier but produced more biomass and grain yield in favorable environments. After removing the effect of height and heading date, strong genotypic correlations were observed for Δ and both yield and biomass across populations (r g = 0.29–0.57, P < 0.05) as might be expected for the favorable experimental conditions. Thus selection for Δ appears beneficial in increasing grain yield and biomass in favorable environments. However, care must be taken to avoid confounding genotypic differences in Δ with stature and development time when selecting for improved biomass and yield especially in environments experiencing terminal droughts. Polygenic control and small size of individual QTL for Δ may reduce the potential for QTL in marker-assisted selection for improved yield of wheat.

Crops reliant on rainfall for growth commonly experience water deficits that reduce grain yield and quality. Selection for genotypic adaptation to water-limited environments is challenging and progress is slow. Many factors contribute to slow breeding progress in dry, rainfed environments: (1) strict requirements for improved grain quality and disease resistance reduce genetic variance for performance under drought; (2) drought is a dynamic entity changing in timing and severity from one year/site to another; and (3) both the above contribute to large genotype × environment interaction to reduce repeatability of genotype performance and confidence in selection. The opportunity exists to compliment existing selection for yield with selection for novel trait variation to improve performance.

Timely sowing is critical for achieving high grain yields in winter cereals. However, inadequate seed-zone moisture for germination commonly delays sowing to reduce biomass and subsequent yield in semi-arid environments. Sowing deep to reach soil moisture is often avoided by growers of Rht-B1b and Rht-D1b semi-dwarf wheat as these wheat show poor emergence when sown deep. Their reduced cell elongation associated with insensitivity to endogenous gibberellins, results in shorter coleoptiles and smaller early leaf area. Alternative dwarfing genes responsive to endogenous gibberellins (e.g. Rht8) are available for use in wheat breeding. These reduce plant height without affecting coleoptile length and offer potential to select longer coleoptile wheat for deep sowing. Nine semidwarf (Rht8, Rht-B1b, and Rht-D1b) and seven tall (rht) wheat genotypes were sown at depths of 50, 80 and 110mm at three locations in 2 or 3 years. Coleoptile lengths measured in a growth cabinet at four temperatures (11, 15, 19 and 23°C) were strongly correlated with coleoptile length (rp=0.77–0.79**) and plant number (rp=0.49*–0.79**) in deep-sown plots in the field. Furthermore, differences in coleoptile length were genetically correlated with greater numbers of emerged seedlings (rg=0.97**), shallower crown depth (−0.58**), greater seedling leaf area (0.59**) and seedling biomass (0.44*). Wheat containing the Rht-B1b or Rht-D1b dwarfing genes produced significantly (P<0.01) shorter coleoptiles (97mm) than both Rht8 (118mm) and tall (117mm) wheat. In turn, compared with emergence from 50mm depth, the Rht-B1b and Rht-D1b wheat produced significantly fewer seedlings at 110mm sowing depth (−62%) than either Rht8 (−41%) or tall (−37%) wheat. Effects of deep sowing early in the season were maintained with reductions in spike number and biomass at both anthesis and maturity. Kernel number was also reduced with deep sowing leading to reductions in grain yield. Over all entries, genotypic increases in plant number were associated with increases in fertile spike (rg=0.61**) and kernel number (0.21*), total biomass (0.26*) and grain yield (0.28*). Reduction in spike number and grain yield with deep sowing was smallest for the Rht8 (−18 and −10%) and rht (−15 and −7%) wheat, and largest for the Rht-B1b/D1b (−39 and −16%) wheat. Plant height and coleoptile length were independent among Rht8 and tall wheat genotypes. This study demonstrates the importance of good seedling emergence in achieving high wheat yields, and the potential use of alternative dwarfing genes such as Rht8 in development of long coleoptile, reduced height wheat suitable for deep sowing.

Many candidate physiological traits have been suggested for wheat improvement under moisture stress and genetic diversity is present in the wheat gene pool for most of them. The objective of this study was to quantify the probable yield effect associated with variation in the expression of such traits within two sets of germplasm;Sisters(from an elite x elite cross) showing a range of drought adaptation, andDiverselines (representing synthetic-derived wheat as well as selected landraces) assembled for favourable expression of one or more of the following traits under moisture stress: final biomass, ability to extract water at depth from the soil, remobilization of stem soluble carbohydrates, and transpiration & water-use efficiency (WUE). Agronomic and physiological traits were measured in NW Mexico in 2005 under post-anthesis moisture stress that resulted in yield reductions of up to 65% compared with irrigated plots, depending on the genotype. There was a significant range of expression for all traits measured and calculations were performed to gain an idea of the relative potential contribution to crop performance if trait expression were maximized in the highest-yielding backgrounds. Theoretically, remobilization of stem carbohydrates would be associated with 7% yield gains forSisterand 20% forDiverse Lines. Maximizing WUE could achieve 16% and 4% gains in yield inSisterandDiverse Linesrespectively, despite the already strong association of the trait with yield, and similar results were found for transpiration efficiency. In both sets of germplasm, putting the best expression for water extraction (to 120cm) in the highest yielding background was associated with yield gains of 13%. Taken together these results suggest substantial yield gains under moisture stress are achievable if the genes representing these traits were to be combined using a complementary-trait based approach to breeding

Physiological research has shown that measurements on small plots of stomatal conductance, canopy temperature depression (CTD) or carbon isotope discrimination may be useful for screening breeding populations for yield potential, prior to the execution of expensive replicated yield trials. Such indirect selection criteria may be very effective as lower cost alternatives for estimating genetic gain for complex characteristics such as yield that are relatively expensive to measure accurately in the field. In the present paper, economic analysis is undertaken of the results of trials conducted at the International Maize and Wheat Improvement Center (CIMMYT) over three seasons to determine the value of the physiological traits being assessed. The results indicate that the economic value of incorporating these measurements into CIMMYT's breeding programme is potentially important. CTD and stomatal conductance are relatively cheap to measure and could be used to discard lines prior to extensive yield testing, whereas carbon isotope discrimination is relatively expensive and would not be economic for this purpose. The analysis indicates that the incorporation of physiological measurements is likely to provide important economic benefits to the programme. Indications are that other breeding programmes with similar breeding goals and comparable costing structures might also consider using such indirect selection traits.

More efficient wheat breeding methods are needed to meet demand for wheat from expected population growth in developing countries. This paper reports results from recent studies conducted at CIMMYT aimed at assessing the use of stomatal aperture-related traits (SATs) as indirect selection criteria for high yield-potential in bread wheat. Two classes of SATs were assessed: the instantaneous trait leaf porosity (POR), which is a close surrogate for stomatal conductance, and the integrative traits Δ13C of leaf and grain and Δ18O of grain. For 3 out of the 4 populations of breeding lines tested in 2001–02, the results indicated strong prospects for using SATs in screening for high yield potential. Grain Δ13C , leaf Δ13C and POR each showed promise, with moderate to high heritability, moderate to strong genetic correlations with yield, and yield gains from retrospective selection of about 40–50g m-2 at trial mean yields of 520g m-2 . Grain Δ18O showed less promise. Heritability of this trait was moderately high, but grain Δ18O was not strongly correlated with yield and retrospective gains from selection were small. For one population, none of the SATs showed any convincing association with yield. The reasons for this need further investigation as SATs are evaluated over additional seasons

Reliable selection of families with increased grain yield is difficult in breeding programs targeting water-limited environments. Carbon isotope discrimination (Δ) is negatively correlated with transpiration efficiency, and low Δ is being used for indirect selection of high wheat yield in rainfed environments. Yet little is known of genetic control and opportunities for improving selection efficiency of Δ in wheat. Half-diallel and generation means mating designs were undertaken to provide estimates of the size and nature of gene action for Δ in a range of wheat genotypes varying for this trait. Significant (P < 0.01) differences were observed for leaf tissue Δ among parents (19.3 to 20.7‰) and F1 progeny (19.4 to 20.9‰) in the half-diallel. General (GCA) and specific combining ability (SCA) effects were significant (P < 0.05), while Baker's GCA/SCA variance ratio of 0.89 was close to unity, indicating largely additive gene effects. GCA effects varied from −0.38 to + 0.34‰ for low and high Δ genotypes `Quarrion' and `Gutha', respectively. GCA effects and parental means were strongly correlated (r = 0.95, P < 0.01) while directional dominance and epistasis contributed to small, non-additive gene action for Δ. Smaller Δ in F1 progeny was associated with accumulation of recessive alleles from the low Δ parent. Narrow-sense heritability was high (0.86) on a single-plant basis. Generation means analysis was undertaken on crosses between low Δ genotype Quarrion and two higher Δ genotypes `Genaro M81' and `Hartog'. The F1, F2 and midparent means were not statistically (P > 0.05) different, whereas backcrossing significantly changed Δ toward the mean of the recurrent parent. Gene action was largely additive with evidence for additive × additive epistasis in one cross. Narrow-sense heritabilities were moderate in size (0.29 to 0.43) on a single-plant basis. Genetic gain for Δ in wheat should be readily achieved in selection among inbred or partially inbred families during the later stages of population development.