The literature suggests that grain number largely determines and as such limits yield in barley. Many of the reported studies were conducted in relatively low yielding environments and it is unclear if grain number is also a limiting factor in high yield potential climates. Nor is it known with certainty what physiological or morphological traits must be targeted in order to increase grain number. There may be a degree of trade-off between yield components whereby grain number is adjusted according to resource availability to the plant, either pre- or post-anthesis, in a way that ensures consistently well-filled grains at harvest. If mechanisms exist for adjusting grain numbers or grain storage capacity after anthesis to match assimilate availability, this may place limits on how far yield can be increased without increasing post-anthesis assimilate production. In order to determine the scope for increasing the yield potential of barley a more thorough understanding of the potential trade-offs between grain number, grain storage capacity and post-anthesis assimilate supply is required. The aim of research reported in this thesis was to establish what determines the yield of spring barley in Ireland and to investigate the timing and possible mechanisms involved in regulating grain number and grain storage capacity in relation to the supply of photoassimilates. Field experiments were carried out on spring barley (Hordeum vulgare L., cv. Quench) at several locations in Ireland from 2011 to 2013. A sub-set of experiments involving destructive sampling and in-field assessments on plots managed as per current best farm practice gathered crop growth, development, and yield component data across sites and seasons in order to establish what determines yield under typical crop production conditions. Separate experiments artificially manipulated the source:sink ratio of plots via shading and seed rate treatments to investigate in more detail the mechanisms determining grain number and grain weight and any potential trade-off between the two components. Grain number accounted for most of the variation in yield across 9 site/seasons of crops managed as per current best practice in Ireland (P < 0.001; R2 = 0.84) while grain weight remained relatively conserved. Ear number accounted for most of the variation in grain number (P = 0.002; R2 = 0.75) and ear number itself was largely determined by shoot survival from an early season peak through to harvest (P <0.001; R = 0.96). Shoot size and weight at the beginning of stem extension had the largest influence on shoot survival. Shading treatments were used to test whether there was a mechanism for adjusting grain numbers after anthesis to match the availability of assimilate for grain filling. Substantial post-anthesis reductions in assimilate supply during grain filling in 2011 and 2012 did not significantly reduce grain number (P > 0.05). A small reduction in grain number (8%) was found in response to shading for a two week period early post-anthesis in 2013, however this was likely a reduction in grain set in shoots or spikelets that reached anthesis after the treatment was imposed rather than a post-anthesis abortion or down-regulation of grain number. Percentage light interception by well managed (unshaded) canopies shortly after anthesis was generally greater than 93% across several sites and seasons, therefore increasing grain numbers to increase sink capacity would likely be associated with an unavoidable decrease in the amount of light intercepted per grain during the early grain development period. However, experiments showed that grain weight at harvest was neither reduced nor increased in response to variations in light interception during this period of endosperm development (P > 0.05), because soluble sugar concentrations in the grain were maintained at the expense of storage reserve deposition in the stems. Results suggest that grain number and grain storage capacity may both be determined pre-anthesis resulting in a trade-off during stem extension whereby grain numbers are adjusted in a way that helps conserve grain weight. A strong negative relationship between ear number and grain number per ear (P < 0.001; R2 = 0.81) across two sites of seed rate experiments in 2013 resulted in a plateau in overall grain number of approximately 18,000 grains m-2 suggesting that there may be a limit to how many grains can be established in a given environment; this was achieved with an ear number of approximately 1000 ears m-2. Yield potential for Irish conditions was estimated at 12.29 t ha-1 at 85% dry matter based on estimates of potential assimilate supply during grain filling; with a grain number of 26,481 m-2 required to utilise this. These estimates are both 44% higher than the mean yield and grain number achieved in crops managed as per current best farm practice. Once high potential ear numbers are secured (> 1000 m-2), breaking the negative relationship between ear number and grain number per ear may hold the key to further increasing grain number and hence yield potential. Increasing assimilate production and partitioning to ears during stem extension, either through increases in the duration of stem extension or solar radiation use efficiency, may enable larger grain numbers to be produced whilst maintaining or increasing individual grain storage capacity and deposition of stem storage reserves. Water and nutrient availability, as well as susceptibility to lodging may present further limitations to yield in the future.

Current trends in population growth suggest that global food production is unlikely to satisfy future demand under predicted climate change scenarios. Therefore, further improvements on wheat yield potential will be essential to meet future food demand. Thus, to further raise wheat grain yield it is critical to understand the physiological basis of grain number per m2 determination. Yield responsiveness to resource availability is usually related to grain number per m2 in most of grain crops, such as durum wheat. Under Mediterranean conditions, wheat yields vary widely, mainly in response to erratic rainfall, but it has been proposed that part of this variability may be reduced through nitrogen (N) management. Wheat yield responses to N fertilisation are usually related to those of grains per m2, which in turn is the consequence of processes related to floret development (floret initiation followed by floret death/survival) during stem elongation. However, there are almost no studies published relating floret primordia generation and survival in response to increases in N availabilities in wheat. Understanding the dynamics of floret primordia development as a physiological determinant of grain setting can be relevant to further increase grain yield in wheat. The focus of this thesis was to study the effects of N availability and the differences among modern cultivars (and, in some cases, combined with availability of other resources) on dynamics of floret development and survival determining the fate of these primordia and the generation of grain number in durum wheat. Six experiments were conducted (4 semi-controlled and 2 field conditions) during four consecutive growing seasons at Catalonia, NE Spain, with a combination of different N and water availabilities and contrasting modern durum wheat cultivars. Across all experiments, yield responsiveness to N was, in general, related to grain number increases as an indirect response to N through its effect on increasing growth. This is not only based on the fact that the relationship of grain number with spike dry matter was not improved if spike N were used instead, but also on the fact that detillering did increased grain number through the same relationships to spike dry matter. Also, improved spike fertility was due to both producing more fertile florets (in response to N and water or to the removal of competing shoots) and to reduce the percentage of failure of fertile florets in becoming grains. Responsiveness of the number of fertile florets was determined by a developmental response of floret primordia, which under high resource availability conditions continued developing normally in some distal florets of the spikelets, at any position of the spike, whilst in more stressful conditions their development stopped. Genotypic variation in fruiting efficiency was found for modern durum wheat cultivars and there was a tendency for some cultivars to have higher values of fruiting efficiency in most of the environments. Differences in fruiting efficiency were responsible for genotypic differences in grain number. A functional negative relationship was found between grain weight and fruiting efficiency, revealing a trade-off related to spike growth per floret developing normally.