Producing more food per unit of water has never been as important as it is at present, and the demand for water by economic sectors other than agriculture will necessarily put a great deal of pressure on a dwindling resource, leading to a call for increases in the productivity of water in agriculture. This topic has been given high priority in the research agenda for the last 30 years, but with the exception of a few specific cases, such as water-use-efficient wheat in Australia, breeding crops for water-use efficiency has yet to be accomplished. Here, we review the efforts to harness transpiration efficiency (TE); that is, the genetic component of water-use efficiency. As TE is difficult to measure, especially in the field, evaluations of TE have relied mostly on surrogate traits, although this has most likely resulted in over-dependence on the surrogates. A new lysimetric method for assessing TE gravimetrically throughout the entire cropping cycle has revealed high genetic variation in different cereals and legumes. Across species, water regimes, and a wide range of genotypes, this method has clearly established an absence of relationships between TE and total water use, which dismisses previous claims that high TE may lead to a lower production potential. More excitingly, a tight link has been found between these large differences in TE in several crops and attributes of plants that make them restrict water losses under high vapour-pressure deficits. This trait provides new insight into the genetics of TE, especially from the perspective of plant hydraulics, probably with close involvement of aquaporins, and opens new possibilities for achieving genetic gains via breeding focused on this trait. Last but not least, small amounts of water used in specific periods of the crop cycle, such as during grain filling, may be critical. We assessed the efficiency of water use at these critical stages.

Pearl millet, a key staple crop of the semi-arid tropics, is mostly grown in water-limited conditions, and improving its performance depends on how genotypes manage limited water resources. This study investigates whether the control of water loss under non-limiting water conditions is involved in the terminal drought tolerance of pearl millet. Two pairs of tolerant×sensitive pearl millet genotypes, PRLT 2/89-33–H77/833-2 and 863B-P2–ICMB 841-P3, and near-isogenic lines (NILs), introgressed with a terminal drought tolerance quantitative trait locus (QTL) from the donor parent PRLT 2/89-33 into H77/833-2 (NILs-QTL), were tested. Upon exposure to water deficit, transpiration began to decline at lower fractions of transpirable soil water (FTSW) in tolerant than in sensitive genotypes, and NILs-QTL followed the pattern of the tolerant parents. The transpiration rate (Tr, in g water loss cm−2 d−1) under well-watered conditions was lower in tolerant than in sensitive parental genotypes, and the Tr of NILs-QTL followed the pattern of the tolerant parents. In addition, Tr measured in detached leaves (g water loss cm−2 h−1) from field-grown plants of the parental lines showed lower Tr values in tolerant parents. Defoliation led to an increase in Tr that was higher in sensitive than in tolerant genotypes. The differences in Tr between genotypes was not related to the stomatal density. These results demonstrate that constitutive traits controlling leaf water loss under well-watered conditions correlate with the terminal drought tolerance of pearl millet. Such traits may lead to more water being available for grain filling under terminal drought.

Studies in A. thaliana reported a genomic region responsible for drought avoidance co-130 mapping with a region contributing to a constitutively lowered Tr and simultaneously 131 leading to an enhanced TE (Masle et al. 2005;McKay et al. 2008). Following these 132 examples, our current hypothesis is that only critical portions of the large DT-QTL, 133 linked to specific mechanisms, matter for the terminal drought tolerance of pearl millet 134 and these needs to be accurately mapped to enhance the precision of marker assisted 135 introgression. In addition, Tr is only one component of plant water use, which likely 136

IntroductionPearlmillet is themain subsistence crop for smallholder farmers systemswhere it is grown at low plant density. Intensifying pearl millet cultivation could boost productivity although it may have trade-offs. Increasing planting density would indeed increase the leaf area and the related water budget, whereas a denser canopy could create a more favorable canopymicroclimate to the benefit of the water use efficiency (WUE) of the crops. The first aim of this work was to test the yield response of popular pearlmillet varieties to an increased density and to assess possible genotypic variation in this response. The second aim was to measure the water use and the WUE of the crop in different densities.MethodTo this end we designed several field and lysimetric experiments To increase the robustness of the results, these trials were carried out in India and Senegal, using two independent sets of genotypes adapted to both sites.ResultsIn the field, the higher sowing density significantly increased yield in all genotypes when trials were carried out in high evaporative demand conditions. There was no genotype x density interaction in these trials, suggesting no genotypic variation in the response to density increase. The high-density treatment also decreased the vapor pressure deficit (VPD) in the canopies, both in the field and in the lysimeter experiments. In the lysimeter trials, although the higher density treatment increased water use, the resulting increase in biomass was proportionally higher, hence increasingWUE of the crops in all genotypes under high density. The increase in yield under high density was closely related to the increase in WUE, although this link was more tight in the high- than in the low evaporative demand seasons. This confirmed a strong environmental effect on the response to density of all genotypes tested.DiscussionAlthough they did not open a scope for breeding density tolerant cultivars, these results highlight the possibility to improve pearl millet yield by increasing the density, targeting specifically areas facing high evaporative demand.

Indoor experiments with individual plants often show that transpiration rate is restricted under high vapor pressure deficit (VPD), resulting in a plateau of transpiration that increases water use efficiency (WUE) of some genotypes. We tested this hypothesis outdoors during dry or rainy seasons of India and Senegal, based on the response of the transpiration of canopy-grown sorghum plants to the reference evapotranspiration that takes both light and VPD into account. This response showed no plateau at high evaporative demand in 47 genotypes, but a large genetic variability was observed for the slope of the relationship over the whole range of evaporative demand. Unexpectedly, this slope was genetically correlated with WUE in two experiments with high evaporative demand: genotypes that most transpired had the highest WUE. Conversely, a negative correlation was observed under low evaporative demand. Genotypes with high WUE and response to evaporative demand were also those allowing maximum light penetration into the canopy. We suggest that this caused the observed high WUE of these genotypes because leaves within the canopy had sufficient light for photosynthesis whereas we observed a lower VPD in the canopy than in open air when leaf area index reached 2.5-3, thereby decreasing transpiration.