Search

Lentil is a cool season grain legume produced primarily in the Mediterranean and temperate regions of the world where water and heat stress in critical developmental windows constrain yield. The effect of stress on yield depends on the timing, intensity and duration of stress; here we focus on timing relative to phenological development. To determine the critical period for grain yield and grain protein, we compared unshaded controls and crops exposed to sequential 10–14 d shading periods using two locally adapted varieties at Roseworthy, south-eastern Australia, and Valdivia, southern Chile. Yield of unstressed controls varied from 1 t ha−1 at Roseworthy to 7 t ha-1 at Valdivia, and grain protein from 22.0 % at Valdivia to 27.6 % at Roseworthy. Irrespective of growing conditions, the most sensitive period was at 50–126 °Cd after flowering, around pod emergence. Grain number and biomass accounted for most of the variation in yield. Grain protein concentration varied according to a bi-linear model, with minor reduction (

Grain weight (GW) is an important component of grain yield and quality of sunflower and the improvement of this trait has been proposed as an aim for breeding of this crop. There is strong evidence suggesting the pre-flowering period as key to determining GW in different crops; however, the key traits and mechanisms controlling GW during this phase are still largely unknown in sunflower. This study conducted a detailed survey of how the pre-flowering period determines grain growth dynamics by assessing the early developmental process during the growth of reproductive organs and the sensitivity of GW, grain number (GN) and grain oil concentration to different source–sink manipulations immediately before flowering. Two experiments were carried out under field conditions in two growing seasons. In Exp. 1 two sunflower oilseed hybrids contrasting in GW (Alybro and SG) were sown at the Experimental Station of the Universidad Austral de Chile in Valdivia (39°47′S, 73°14′W), Chile. In Exp. 2, the same hybrids plus a confectionery genotype (RHA) were evaluated in the Experimental Station where two source–sink treatments were performed: a control (without manipulation) and a shading treatment starting 16 days previous to anthesis, i.e. during the R2-R5 period. Crop development, average and individual GW, GN and oil concentration of grains were assessed at harvest. Also, flower and grain traits and their dynamics were measured from early reproductive stage (ovary at R3) to harvest in different grain positions from the capitulum (e.g., fresh and dry weight and dimensions of grains, pericarp and embryo and grain oil concentration). GW was significantly affected by genotype and shading treatments imposed at pre-anthesis, but no statistical effect of the lower source–sink ratio was found on GN. Contrary to GW, grain oil of oilseed genotypes was highly conservative under the source–sink reduction in our study. Positive associations were found between final GW and: ovary weight at R3, at anthesis (R5), ovary growth rate, grain dimensions and water content. Final GW also showed a linear association with ovary growth rate. This study is the first to evaluate the ovary weight from the early development stage and the ovary growth rate during the pre-anthesis period in sunflower and the effect of source reduction at pre-anthesis on grain oil concentration, supporting the importance of the maternal tissues on the setting of potential and actual GW in sunflower.

Grain yields in wheat can be limited by the assimilate supply (source) or by the carbohydrate demand of the grains (sink). Recently, there have been questions regarding the capability of crop models to simulate the physiology of source-sink interactions in crops; however, crop models have never been tested with source-sink manipulated data. We tested the Nwheat model with detailed measured field experimental data with treatments of manipulated source (i.e., assimilate supply), sink (i.e., kernel number), and their combinations. In general, the model could reproduce observed effects of shading before and after anthesis as well as the additional impact of halving the spikes. A 90% shading during grain filling reduced individual grain weights drastically, with the remaining yield mostly determined by carbohydrate remobilisation, which the model reproduced. The model also reproduced the decline of biomass accumulation due to shading, but was not sensitive enough to simulate the observed reduction of kernels per m2 from a 90% reduction in solar radiation between booting and the beginning of grain filling, resulting in an overestimated grain yield. The model reproduced the positive impact of a 7% genetically increased radiation use efficiency (RUE) on growth and yield. A sensitivity analysis indicated that the yield response to increased RUE can vary among environments. The yield impact can be positive in many environments, but negative in terminal drought environments. In these environments, stimulated early growth from higher RUE can cause accelerated water deficit during grain filling and reduced yields. The model adequately simulated source-sink interactions of most of the treatments, but there were obvious shortcomings in simulating kernel set and final grain size. Improving these will be critical for estimating crop-environmental interactions that affect assimilate supply, including breeding, industrialisation-induced or geo-engineered solar dimming, genetically and atmospheric CO2-related increased RUE, and source manipulations, such as pest and disease impacts.

Annual crops accommodate environmental variation through grain number, whereas grain weight is more stable. Grain number is determined in a species-specific window which has been established for many crops, but not for oat. Field trials were established at two sites in southern Australia and in one site in southern Chile where successive, single 14-d shading periods were applied from crop establishment to maturity to identify the developmental window when the crop is most responsive to stress. Three oat varieties were compared in Australia (Mitika, Williams and Wintaroo) and two in Chile (Mitika and Yallara). Unshaded controls yielded from 327 g m −2 in Australia to 747 g m −2 in Chile. The overall pattern of yield response to time of stress was similar to that of wheat; it spanned the period from stem elongation (GS31) to about 10 days after anthesis. In line with theory, most of the yield response was mediated through response in grain number; further, the two environments in Australia where reduction in grain number in response to stress shortly before anthesis was larger, individual grain weight increased with shading. Grains per panicle was more responsive to stress than panicles per m 2 , in contrast to other cereals. The critical period is often assumed to be species-specific. However, our limited comparison of varieties suggests that there might also be varietal differences in oat. Interaction between time of shade and variety was significant for harvest index in all locations. Hence, we propose genotype-dependent response to time of stress is worth exploring.

Breeding heavier seeds while maintaining seed number has been proposed as an effective strategy to improve oil crop production. Recent studies assessing rapeseed source-sink (S-Sratio) reduction after flowering suggest a window in the early seed-filling period that can compensate, either partially or totally, for a decrease in the number of seeds. However, little evidence has been reported in relation to the key window for seed plasticity in rapeseed. Furthermore, such studies have only assessed spring rapeseed genotypes and there is a lack of information on winter rapeseed. This study aims to elucidate the occurrence of a narrow window when seed plasticity is greatest, which is hypothesized to occur during the early stages of seed filling after the beginning of flowering in both winter and spring rapeseed genotypes. To asses this hypothesis, two winter and two spring rapeseed hybrids were evaluated under three S-Sratio treatments in two experiments performed in field conditions in Valdivia, Chile. The winter hybrids Trust and Mercedes were assessed under different nitrogen levels, and the spring hybrids Lumen and Solar were assessed using two sowing dates. S-Sratio treatments in both rapeseed genotypes consisted of a control without S-Sratio manipulation, a reduced S-Sratio from the beginning of flowering [BBCH 61] to 15 days after flowering (DAF) and a reduced S-Sratio from 15 to 30 DAF. The S-Sratio was reduced by shading the crop with black nets to intercept 75 % of the incoming solar radiation. Shading treatments decreased seed yield relative to the control between 13 and 42 % for the winter genotypes and from 23 to 44 % for the spring genotypes. The negative effect of shading on seed number under both S-S reduction timings explains the seed yield reduction. Interestingly, different windows of time for seed weight plasticity were found in response to the lower S-Sratio for winter and spring genotypes (i.e., 15–30 and 0–15 DAF, respectively). In the winter-type genotypes, seed weight increased between 28 and 33 % under the 15–30 DAF shading treatment, while seed weight in spring-type genotypes increased from 15 to 39 % during the 0–15 DAF treatment. Seed weight change was almost negligible outside of the sensitivity window as this trait increased only 2% at 0–15 DAF in winter-type genotypes and 7% at 15–30 DAF in spring-type genotypes, across treatments. In parallel, the S-S reduction decreased seed number by 23.6 and 40 %, at the 0–15 and 15–30 DAF treatment, respectively. The study showed statistical differences for seed oil and protein concentration across treatments, but both quality traits remained highly stable to the S-S reduction and showed similar behavior in winter and spring rapeseed types. Seed oil concentration changed little while seed protein concentration was more sensitive to seed weight response. The findings of the present study suggest that seeds per silique and seed weight are determined by the plant assimilate S-Sratio; therefore, seed number and seed weight determinations overlap as has been demonstrated in other crops like wheat, barley, sorghum and sunflower.

The cellular expansion of maternal tissues has been suggested as strongly associated to grain volume determination in wheat and is therefore a key process to understand potential grain weight determination. The dynamics of grain growth in seed size-contrasting diploid, tetraploid and hexaploid genotypes of wheat were evaluated to explore the relationships between the final grain weight and: a) grain volume, b) water content, c) epidermal pericarp cell size and d) the rate of cell area (area per cell) increase. The assessment of cell expansion was reinforced by the study of the expression of the XTH5 gene in pericarp tissues, as this gene codifies a cell wall-remodeling agent. Differences in grain weight across ploidies and experiments (greenhouse and field evaluations) were attributable to the grain-filling rate (ANOVA p

Understanding the response of the number of seeds and seed weight to the availability of assimilates is crucial for designing breeding strategies aimed to increase seed and oil yield in oilseed rape. This study aims to answer the questions: i) do seed number and seed weight in oilseed rape differ in their plasticity in response to the availability of assimilates at flowering? and ii) how sensitive are oil and protein concentrations to the availability of assimilates during flowering? A spring oilseed rape hybrid was sown in two field experiments and the treatments were combinations of i) two plant densities and ii) shading or no shading between the beginning and end of flowering. Seed yield was not affected by plant density or by shading. Lower plant density was compensated by an increase in seed number per plant, without effects on single-seed weight. However, the negative effect of shading during flowering on seed number per area was fully compensated by an increase in single-seed weight by 47–61%. The plasticity of single-seed weight observed in the present study of oilseed rape has never been reported for annual seed crops. Shading at flowering increased both the seed filling rate and the duration of the seed filling period at all positions in the canopy. We also observed that the reduction of the source-sink ratio at flowering increasing seed weight does not necessarily modify oil or protein concentrations. Thus, single-seed weight could be targeted to increase seed yield in oilseed rape without compromising oil content.

Faba bean is a cool-season, indeterminate grain legume grown in Mediterranean and temperate environments where yield loss associated with abiotic and biotic stress is common. Yield responses depend on the timing of stress relative to the species-specific vulnerable stages in the crop lifecycle. To determine the critical period for yield determination in faba bean, we used sequential 14-d shading periods in locally-adapted varieties grown in five environments with yield of unshaded controls from 2.3 to 6.8 t ha−1. Yield of shaded treatments diverged from the controls around 450 °Cd before flowering, reached the largest difference around 100 °Cd after flowering, and converged with controls towards 700–800 °Cd after flowering; the most critical stage aligned with pod emergence. Seed number accounted for most of the variation in yield response to shading. Shading increased seed size when stress severely reduced seed number. Pod number accounted for most of the variation in seed number for shading before flowering, and both pod number and seeds per pod contributed to the reduction in seed number in crops shaded after flowering. The defined critical period is a useful reference for breeding and agronomic solutions to improve yield under stress.

Previous yield gains in wheat have been achieved through increased expression of grain number per unit area, while fruiting efficiency has also been shown to explain improved grain number. However, combining a high grain number and weight in the same genotype is a difficult task in practical breeding. This study evaluated the progeny of a cross between two elite parents that contrast in grain number and grain weight, as a way to better understand how the two desired traits might be combined to achieve a significant boost in yield. The two parents are both high biomass elite spring wheat lines that achieve high yield through contrasting strategies: Bacanora achieves high grain number with a high density of relatively small spikes. Weebil has a lower density of larger spikes in which larger grain size is expressed. The specific objectives of this research were: (i) to evaluate grain yield performance of DH lines and parents over two seasons, (ii) to identify promising physiological traits explaining high yield performance and (iii) to determine whether these traits are also expressed at the low plant densities. Two experiments were carried out under field conditions in southern Chile where the highest yield potential of wheat has ever been recorded. In Exp. 1 the performance of the 105 spring wheat DH lines, their parents and a Chilean spring cultivar (Pandora-INIA) with similar phenology were assessed together. Results showed that it is possible to increase yield potential by combining high grain number and grain weight, thereby reducing the trade-off between both components. 42 and 50% of the DH lines showed transgressive segregation in S1 and S2, respectively. 3 and 4% of the DH lines yielded between 15,000 and 16,000 kg ha−1. Grain yield was highly correlated with above-ground biomass (R2 = 0.78, p

There is a noticeable gap in knowledge about the key phases for grain yield determination in temperate legume crop species such as lupin and pea. For this reason, the aim of the present study was to comparatively assess the critical period for grain yield and grain number determination of narrow-leafed lupin ( Lupinus angustifolius L.) and pea ( Pisum sativum L.). Two field experiments were conducted during the 2007–08 (Exp. 1) and 2009–10 (Exp. 2) growing seasons at the experimental station of the Universidad Austral de Chile. Treatments were the combination of (i) two crops (narrow-leafed lupin and pea) and (ii) seven shading treatments (control without shading and six shading treatments imposed for 15 days throughout the crop cycle). Lupin showed a longer ( P P P

Among the yield components, grain weight is considered a conservative trait whose determination is still beyond our complete understanding. Crop physiology uses a whole approach to study this complex trait, which can provide helpful information to plant breeders and molecular biologists. This study emphasizes the understanding of pre- and post-anthesis determinants of final grain weight. A field experiment was carried out in two growing seasons evaluating two wheat cultivars contrasting in grain weight potential. Carpel weight at pollination, grain dimensions, grain water, dry matter and volume dynamics were assessed. Among grain dimensions, grain length was the trait, which explained final grain weight (r2 = 0.78; P

The impacts of acidic soils and Al toxicity on wheat nutrient economy have been scarcely researched under field conditions even though these soils are widely spread in wheat production areas around the world. The main objective of this study was to quantitatively evaluate the element (N, P, K, Ca and Al) economy of an Al-sensitive and an Al-tolerant wheat cultivar growing under different soil Al concentrations at field conditions. To reach this objective, two field experiments were conducted in an Andisol in Valdivia (39°47′18″S, 73°14′05″W), Chile. Treatments were a factorial arrangement of: (i) two spring wheat cultivars (Al-sensitive, Domo.INIA and Al-tolerant, Dalcahue.INIA) and (ii) five exchangeable Al levels (0–2.7 cmol(+) kg−1) with three replicates. At harvest, plant biomass was sampled and divided into 5 organ categories: ears, grains, blade leaves, stems plus sheath leaves and roots. The element content (N, P, K, Ca and Al) in each organ was measured to quantify element uptake and concentration, nutrient uptake efficiency (UPE) and nutrient utilization efficiency (UTE). Element uptake (N, P, K, Ca, and Al) was negatively affected by the increased soil Al concentration in above-ground and root biomass in both cultivars (R2 = 0.61–0.98, p

Soil acidity and Al toxicity are highly extended in agricultural lands of Chile, especially where wheat is widely sown. To evaluate quantitatively the response of wheat biomass and its physiological determinants (intercepted radiation and radiation use efficiency) to Al toxicity, two field experiments were conducted in an Andisol in Valdivia (39°47′S, 73°14′W), Chile, during the 2005–2006 and 2006–2007 growing seasons. Treatments consisted of a factorial arrangement of: (i) two spring wheat cultivars with different sensitivity to Al toxicity (the sensitive cultivar: Domo.INIA and the tolerant cultivar: Dalcahue.INIA) and (ii) five exchangeable Al levels (from 0 to 2.7 cmol(+) kg−1) with three replicates. Crop phenology and intercepted radiation (IR) were registered during the entire crop cycle, while 10 samples of above-ground biomass were taken at different stages between double ridge and maturity. Both biomass and leaf area index (LAI) were recorded in these 10 stages. Radiation use efficiency (RUE) was calculated as the slope of the relationship between accumulated above-ground biomass and accumulated photosynthetically active radiation intercepted by the canopy (IPARa). Crop phenology was little affected by soil Al treatments, showing only up to 17 days delay in the Al-sensitive cultivar under extreme Al treatments. Above-ground biomass at harvest was closely associated (R2 = 0.92) with the crop growth rate but no relationship (R2 = 0.14) was found between the crop cycle length. IPARa explained almost completely (R2 = 0.93) the above-ground biomass reached by the crop at harvest under the wide range of soil Al concentrations explored in both experiments. On the other hand, a weaker relationship was found between above-ground biomass and RUE. The effect of soil Al concentration on IPARa was mainly explained by LAI as a single relationship (R2 = 0.93) between IR (%) and LAI at maximum radiation interception showing a common light attenuation coefficient (k = 0.33).

Several studies have been conducted to evaluate the response of crops, especially temperate cereals, to different source–sink ratios during grain filling. However, there is much less information about temperate legumes and even less work comparing the two. The objective of this study was to evaluate the response of both grain yield and grain nitrogen concentration of wheat ( Triticum aestivum L.), narrow-leafed lupin ( Lupinus angustifolius L.) and pea ( Pisum sativum L.) to similar source reduction during grain filling. Two field experiments were conducted in a high yielding environment of Southern Chile. In experiment 1 wheat and narrow-leafed lupin were grown for two consecutive years. Experiment 2 evaluated wheat and pea on two sowing dates. In both experiments a reduction in the source–sink ratio was imposed by using black nets that intercepted 90% of the incident solar radiation from the commencement of the linear dry matter accumulation to physiological maturity. Grain yield was differentially ( p p vs. temperate legumes.

Considering the importance of acidic conditions and Al toxicity in arable soils of Chile, 2 field experiments were conducted in the 2005-06 and 2006-07 growing seasons in Valdivia (39°47′18′′S, 73°14′05′′W), Chile in an Andisol. The objective of this experiment was to quantitatively evaluate the effect of different soil exchangeable Al levels on grain yield, and above-ground and root biomass of Al-tolerant and Al-sensitive wheat cultivars under field conditions. Treatments were a factorial arrangement of: i) two spring wheat cultivars (Al-sensitive, Domo.INIA and Al-tolerant, Dalcahue.INIA) and ii) five exchangeable Al levels (0–2.7 cmol(+) kg−1). The experimental design consisted of a randomized complete block design with three replicates. At harvest, grain yield, grain number, thousand grain weight and above-ground biomass were recorded. At the same time, root samples were taken with the pinboard monolith method. Afterwards root biomass, root length density and specific root length were measured. Both above-ground and below-ground traits showed a wide range of values (e.g., between 10 to 2618 g m−2 and between 8 to 117 g m−2, for above-ground and root biomass, respectively) under the soil Al concentrations used in this study. Al tolerant and Al sensitive cultivars showed different sensitivities to Al toxicity. Interestingly, linear associations were found between grain yield or above-ground biomass and soil Al concentration in Al-sensitive (R2 = 0.95 p

Nitrogen (N) uptake at heading and at maturity was evaluated under four N treatments (20, 50, 110 and 160 kgN ha−1 at sowing) in four malting barley cultivars released to the market from 1944 to 1998 in Argentina. Grain N concentration was slightly lower in the newest cultivar than in the oldest one. Grain N yield showed cultivar × N interaction: modern cultivars tended to increase their grain N yield in response to N fertilization more than old cultivars. Grain N yield was correlated to grain yield as well as to total above-ground biomass. Nitrogen fertilization affected total N in above-ground biomass at maturity, but no differences among cultivars were found. Nitrogen harvest index of the most modern cultivar (0.76) was higher than that of the oldest cultivar (0.67). Modern cultivars tended to have a higher N content in ears at heading than old cultivars, and the magnitude of the differences increased with N availability. Physiological nitrogen efficiency for grain yield was significantly higher in the newest than in the oldest cultivar (57 and 43 g grain g N − 1 , respectively).

In temperate cereals are commonly accepted that determination of grain number (GN) and grain weight (GW) scarcely overlap during the crop cycle. However, the assumption that GW is determined exclusively after anthesis needs to be critically reviewed in the light of reports published over the few years where temperature treatments imposed before anthesis decreased GW of bread wheat. Although these evidences suggest that both GW and GN could be affected by environmental conditions before anthesis little is known about the effect of pre-anthesis temperature on these two main yield components in wheat, barley and triticale at field conditions. In addition, the effect of temperature on GW and GN at different stages prior to anthesis has been scarcely evaluated. The objectives of the current study were: (i) to evaluate the overall response, and specific differences, of GN and GW to pre-anthesis temperature, and (ii) to study the effect of different timings of high temperature at pre-anthesis on GN and GW in wheat, barley and triticale. Three fully irrigated field experiments were carried out in three successive seasons. At each season, a wheat, barley and triticale high yielding cultivar was evaluated at three temperature regimes: control, and two timings of heating before anthesis. During the first and second seasons, the timings of heating were booting-anthesis and heading-anthesis. In the thirst season, the timings were beginning of stem elongation-booting and booting-anthesis. Plots were arranged in a split-plot design with three replicates, where the main plot was assigned to thermal regime and the sub-plots to crop species. To apply heat, transparent chambers equipped with thermostatically controlled electric heaters were used. The thermal regime was controlled by sensors connected to a temperature regulator and recorded using data loggers. Temperature within the chambers was stable across developmental stages, crops, and seasons; it averaged 5.5 °C higher than air temperature. Thermal treatments consistently reduced grain yield (p < 0.05), the magnitude of the effect ranged between 5 and 52%. The highest effect was found when temperature increased during stem elongation (yield decrease: 46%), lowest when treatments were imposed during heading-anthesis (15%) and intermediate for treatments imposed during booting-anthesis (27%). Most effects of thermal treatments on yield were due to parallel effects on GN. However, thermal treatments significantly (p < 0.05) decreased GW during the three seasons. The most effecting treatment on GW was when the crops were heated during the B-A period, i.e. GW decreased up to 23%.

Two field experiments were carried out to evaluate the effect of the genetic improvement of barley on leaf and tiller appearance. Four malting barley cultivars were used to represent the main genotypes released from 1944 to 1998 in Argentina. One experiment was carried out with 110 kg N ha −1 of nitrogen availability at sowing, while in the other experiment cultivars were grown with 20, 50, 110 or 160 kg available N ha −1 . There was no difference in final leaf number among cultivars, which ranged from 9.5 to 11.2 leaves per main stem. Differences in phyllochron between old and modern cultivars were only observed for the early leaves and under treatments of low nitrogen availability (20 and 50 kg N ha −1 ). Cultivars began tillering at a similar time. Nitrogen treatments modified the range of maximum tiller number per plant from three (N 20 ) to seven (N 160 ) across cultivars. In addition, higher nitrogen availability delayed the time to maximum tiller number. Differences in the number of tillers per plant were explained by the rate of tiller appearance. High levels of tillering were accompanied by high levels of intercepted radiation and associated with modern cultivars growing under high levels of nitrogen. Accumulated intercepted radiation during the jointing–heading period had a trend to increase with years from release, at a mean rate of 1.17 MJ m −2 per year.

Field experiments were conducted during 1991 and 1992 to identify the effect of wheat breeding on crop biomass production and its physiological determinants, i.e. radiation interception and radiation-use efficiency. To address this objective, biomass accumulation and its physiological attributes of seven cultivars (including a commercial hybrid) released between 1920 and 1990 were compared in fertilised and irrigated plots. Weeds, pests, diseases and lodging were controlled or prevented. Crop radiation interception was calculated from measurements of incident and transmitted radiation at different times after sowing. Above-ground dry matter was determined at particular stages of development of the cultivars. Radiation-use efficiencies and crop growth rates were calculated for each cultivar for pre-anthesis (in both years) and post-anthesis (in 1992) periods. Biomass at anthesis tended to be less in most recently released cultivars than in older materials. Accumulated intercepted radiation at similar developmental stages also differed between cultivars. These differences, as well as the trend of biomass at anthesis, were caused by differences in the length of developmental phases between cultivars rather than by changes in the architecture of the canopies. The most recently released cultivars had shorter seedling emergence-anthesis periods than older cultivars. Moreover, all cultivars had similar canopy light extinction coefficients ( k ), pre-anthesis radiation-use efficiencies, and crop growth rates. After anthesis, the two oldest cultivars accumulated the least biomass and their radiation-use efficiencies and crop growth rates were smaller than those of modern cultivars. Remarkably, modern cultivars maintained during post-anthesis almost the same levels of crop growth rates and radiation use efficiencies reached during the pre-anthesis period.

Past breeding programs for increased wheat grain yield have increased the number of grains per m 2 , associated with a higher spike dry weight at anthesis. Other workers have shown that shading during the preanthesis period strongly reduced the number of grains per m 2 of the crop via reductions in the spike dry weight at anthesis. However, these effects have not been determined for cultivars of different potential number of grains per m 2 . A factorial field experiment was executed with three bread wheat cultivars released in 1920 (Klein Favorito), 1940 (Eureka FCS) and 1980 (Buck Pucara) in Argetina and two shading treatments (shading to 50% of the incoming solar radiation from the beginning of stem elongation to heading or no shading). Both cultivars and shading significantly affected the number of grains per m 2 , but their interaction was not significant. Cultivars differed only in number of grains per spikelet, while shading reduced both the number of spikes per m 2 and the number of grains per spike. The lack of a significant interaction indicates that the absolute reduction in grain number was similar for all cultivars, but the relative effect of shading was much greater for the old (low number of grains) cultivar than the modern one (62 and 42%, respectively). Preanthesis shading reduced number of grains per m 2 on main shoots less (45%) than on tillers (65%). Therefore, the relative contributions of these shoot categories to total number of grains per m 2 were strongly modified by shading. An evaluation of published data with conflicting findings on this shading effect suggests that the degree of modification observed has been related to the proportions of tiller and main shoot spikes.