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.

Agricultural research has contributed enormously to poverty reduction and increased food security worldwide. Wheat crop improvement is a good example of this contribution. Public investments in wheat research from the Green Revolution onwards led to significant productivity increases: following the widespread adoption of semi-dwarf varieties, annual yield growth rates peaked at 2·75% p.a. in the 1980s. Since then, public and private investments in crop (including wheat) research have been modest despite the potential of such research to contribute substantially to the first Millennium Development Goal (MDG) of halving hunger and poverty by 2015. Drawing on a wide spectrum of recent literature, the present paper broadens the usual frame of reference for diagnosing the adoption of improved technology and measuring impact. The adoption of improved varieties and management practices is influenced on the supply side by the nature and performance of the input delivery pathway from research to the farm (input value chains), and on the demand side by the characteristics of the farm household system and the marketing or value-adding chains from the farm to the consumer (output value chains). These three elements (input value chains, farm household system characteristics, and output value chains) can be viewed as a U-impact pathway. This pathway determines the rate and extent of adoption of improved varieties and practices, the magnitude of direct and indirect impacts, and the potential for feedback loops leading to improved functioning of the input and output value chains. The U-impact pathway provides a framework to identify an expanded set of beneficiaries from crop improvement which extend beyond the common focus on producers and final consumers; conventional surplus analysis can then be used to estimate the wider benefits to crop improvement. Additional metrics may be needed to estimate impact related to non-economic benefits, such as poverty, health and social capital. The implication of this fuller accounting of impacts is that the benefits accruing to agricultural research may be greater, and more widely distributed across the economy, than previously recognized by research managers and policy-makers. This strengthens the case for maintained or increased public and private sector investment in crop improvement.

The Green Revolution era focused on enhancing the production and productivity of rice and wheat. New challenges demand that the issues of efficient resource use and resource conservation receive high priority to ensure that past gains can be sustained and further enhanced to meet the emerging needs. Extending some of the resource-conserving interventions developed for wheat to rice culture is a major challenge for researchers and farmers alike. The present paper shares recent research experiences on resource conservation technologies involving tillage and crop establishment options and associated agronomic practices which enable farmers in reducing production costs, increase profitability and help them move forward in the direction of adopting conservation agriculture.

Wheat nitrogen-use efficiency in the Yaqui Valley has been estimated at about 0·31. The nitrogen that is not recovered by the crop has important environmental costs that have regional and global consequences. In addition, these nitrogen losses represent an important reduction in farm income. The objective of the present work was to validate a technology that includes the use of N-rich strips together with the GreenSeeker™ sensor and a crop algorithm in farmers' fields with the ultimate goal of improving nitrogen-use efficiency through site-specific nitrogen management in irrigated spring wheat. During the wheat crop cycle 2002/03 and 2003/04, 13 validation experiments of c. 1 ha each were established in farmers' fields in the Yaqui Valley. After the validation phase, during the wheat crop cycle 2005/06, eight technology transfer trials were established in farmers' fields; these had on an average an area of 10 ha each. Both the validation and technology transfer trials compared the farmers' conventional nitrogen management use v. the use of the N-rich strip together with the Green Seeker™ sensor and a crop algorithm to derive N recommendations for each individual field. The results of the validation trials showed that on an average over all locations, farmers were able to save 69 kg N/ha, without any yield reduction. At the price of US$0.9 per unit of N in the valley when these experiments were established, this represented savings to the farmers of US$62/ha. The technology transfer trials demonstrated that, in large commercial areas with an average size of 10 ha, farmers could improve their farm income by US$50/ha, when using sensor based N management. The combination of the N-rich strip, together with the use of the sensor and a crop algorithm to interpret the results from the sensor, allowed farmers to obtain significant savings in N use and thus in farm profits. Farm income was increased by US$56/ha, when averaged over all trials in all years.

Wheat is grown on 210 million ha throughout the world producing approximately 600 million tonnes of grain (10 year average; FAO 2005) and providing on average one fifth of the total calorific input of the world's population (FAO 2003). For some regions such as North Africa, Turkey and Central Asia, wheat provides half of total dietary energy intake. Of the cultivated wheat area, half is located in less developed countries where there have been steady increases in productivity since the green revolution, associated with genetic improvements in yield potential, resistance to diseases and adaptation to abiotic stresses (Reynolds & Borlaug 2006a, b) as well as better agronomic practices (Derpsch 2005). Nonetheless, challenges to wheat production are still considerable, especially in the developing world, not only because of increased demand but also because of the increased scarcity of water resources (Rosegrant 1997; WMO 1997), ever more unpredictable climates (Fischer et al. 2002), increased urbanization and loss of good quality land away from agriculture (Hobbs 2007), and decreased public sector investment in agriculture and rural affairs (Falcon & Naylor 2005). To meet demand in a sustainable way, more resources are required to breed a new generation of genetically improved cultivars as well as implement resource-conserving agronomic management practices.

The introduction and use of herbaceous legumes may contribute to agricultural intensification, especially in the context of sustainable crop and livestock production systems. In the context of evaluating different legume species for these systems in moist savanna zone of West Africa, the present study involved the evaluation of 46 accessions of Lablab purpureus (L.) Sweet between 2000 and 2002 at Samaru, Zaria in the northern Guinea savanna of Nigeria to identify accessions with the potential to contribute to grain or forage production and those with the potential for multiple use. Grain yields (0·6–2·4 t/ha) with a mean seed crude protein and phosphorus content of 25·3 and 0·46 g/kg, respectively, were obtained. Dry matter yields (leaf: 0·3–3·1, stem: 0·2–4·6, root: 0·03–0·3 t/ha) were produced between 40 and 140 days after planting. Within this same period up to 1152 kg shoot crude protein/ha, equivalent to 184 kg N/ha, were recorded for accessions. Ten accessions that may contribute to grain production, eight accessions that may contribute to forage production and six accessions with drought tolerance were identified. Some white-seeded accessions (Grif 1246, ILRI 4612 and PI 183451) with good grain and forage yield and high-protein content have the potential to provide more and higher quality food for people and feed for livestock. The observed potential to contribute to grain production for protein-rich food, feed for livestock and green manure for soil N improvement suggests lablab may be an acceptable legume option for use in cereal-legume-livestock systems in the moist savanna zone of West Africa.

The objectives of the present study were to evaluate heterosis for grain yield and yield components in durum wheat, and to assess the prediction potential of amplified fragment length polymorphism (AFLP) based and agronomic trait based genetic distances (GD and MD, respectively) to F1 performance, mid parent heterosis (MPH), and specific combining ability effects (SCA) under well-watered and moisture stress conditions. Six parental genotypes with different responses to moisture stress and their 15 F1 crosses were evaluated for their responses to moisture stress conditions in a glasshouse. Some cross combinations showed significant MPH for grain yield and yield components. The expression of heterosis for grain yield was greater under moisture stress conditions than under well-watered conditions. Cluster analysis of the parental lines based on agronomic performance under stress conditions was similar to cluster analysis result based on AFLP marker profiles. F1 performance was strongly correlated to both SCA effects and MPH under both stress and well water conditions. The correlation between SCA and MPH was very high under both treatment conditions for all traits. Correlation between GD and MD was significant only under stress conditions. Positive correlation was found only for the association between GD v. F1 performance and GD v. SCA effects for harvest index (HI) under well-watered conditions. None of the correlations between MD and SCA effects were significant. The absence of association between GD and heterosis for yield and most agronomic traits implied that heterozygosity per se diversity is not a good predictor of heterosis or F1 performance under both well-watered and stressed conditions.

Decision-making processes in agriculture often require reliable crop response models. The Fujian province of China is a mountainous region where weather aberrations such as typhoons, floods and droughts threaten rice production. Agricultural management specialists need simple and accurate estimation techniques to predict rice yields in the planning process. The objectives of the present study were to: (1) investigate whether artificial neural network (ANN) models could effectively predict Fujian rice yield for typical climatic conditions of the mountainous region, (2) evaluate ANN model performance relative to variations of developmental parameters and (3) compare the effectiveness of multiple linear regression models with ANN models. Models were developed using historical yield data at multiple locations throughout Fujian. Field-specific rainfall data and the weather variables (daily sunshine hours, daily solar radiation, daily temperature sum and daily wind speed) were used for each location. Adjusting ANN parameters such as learning rate and number of hidden nodes affected the accuracy of rice yield predictions. Optimal learning rates were between 0·71 and 0·90. Smaller data sets required fewer hidden nodes and lower learning rates in model optimization. ANN models consistently produced more accurate yield predictions than regression models. ANN rice grain yield models for Fujian resulted in R2 and RMSE of 0·87 and 891 vs 0·52 and 1977 for linear regression, respectively. Although more time consuming to develop than multiple linear regression models, ANN models proved to be superior for accurately predicting rice yields under typical Fujian climatic conditions.

An understanding of genotype by environment (G×E) interaction would be useful for establishing breeding objectives, identifying the best test conditions, and finding areas of optimal cultivar adaptation. Data from field assays including eight environments and eight elite lines were analysed to identify environmental and genotypic variables related with G×E interaction for yield in sesame multi-environment trials in Venezuela. Both predictable and unpredictable environmental variables were recorded. Yield components were recorded as genotypic variables. Yield and yield components were used to perform additive main effect and multiplicative interaction (AMMI) analysis. Significant differences (P<0·01) for G×E interaction were observed for all variables examined, except for the number of branches per plant. For yield, 0·28 of the total sum of squares corresponded to G×E interaction. Using environmental and genotypic data, correlation analysis was carried out between genotypic and environmental scores of the first interaction principal component axis (IPCA 1) for all variables examined. Significant correlations (P<0·05) were observed between IPCA 1 for yield and content of sand and silt in soil. No significant correlation was found between IPCA 1 score for yield and genotypic variables. These results indicate that edaphic properties at the trial locations play an important role in yield G×E interaction in Venezuelan sesame. These results should help select test sites for sesame in Venezuela to minimize G×E interaction and make selection of superior genotypes easier. Two strategies can be recommended: multi-environment trials at sites with average, not extreme, sand and silt content, or stratification of sites according to sand and silt content.

The estimation of dry matter intake (DMI) using the alkane pair technique has been validated in ruminants, but not in equids. The current paper reports the finding of three comparative validation studies carried out using a total of 12 cattle, 29 donkeys and 10 horses during which directly measured intake was compared to estimated intake using the alkane pair technique. Two methods were developed to dose the even chain alkanes that were used as external markers. Study I, carried out in Zimbabwe, compared the accuracy of estimated intake with measured intake in cattle and donkeys using hexatriacontane (C36) as the external marker. Studies II and III were carried out in the UK with horses and donkeys and compared the accuracy of estimated intake with measured intake using dotriacontane (C32) as the external marker. Study III also tested the effect on the accuracy of intake estimates of two marker dosing levels (mean daily dose of 224 mg per animal and 448 mg per animal) and two dosing frequencies (2× and 3× daily). Twice daily dosing of even-chain alkane at the lower dose level provided an estimate of DMI similar to that obtained by thrice daily dosing at this low level. The higher dose level given twice daily tended to produce large variation in faecal concentrations of dosed even-chain alkanes, this variation was reduced when dosing frequency was increased to thrice daily. The accuracy of estimated intake improved progressively as the number of faecal sampling days was increased from one to six with no significant difference between estimated intake based on day 5 or 6 of faecal sampling.

The results of all three studies indicate that the alkane pair technique provides a robust method of estimating intake in equids with no significant difference between measured and estimated values in all but one case. Using C31 as an internal marker provided a more accurate estimated intake than using C33 as the internal marker in all cases. Faecal recoveries of alkanes in equids do not appear to show the same influence of carbon chain length that has been observed in ruminant studies.

The physical properties of hardness and dry matter and the histological characteristics of dermal and epidermal tissues were investigated in lateral and medial claws obtained post mortem from mature crossbred female cattle. Claws containing pigmented (n=22) or non-pigmented (n=22) coronary wall horn were studied. Impression hardness, measured on the dorsal border of claws, proximo-distal at intervals of 5 mm from the lower perioplic line was shown to increase to the 20 mm lower site. Similarly, in comparison with values for pigmented horn, pooled measurements for non-pigmented horn showed greater hardness from the lower perioplic line to the 40 mm lower site but were not different for subsequent measurements towards the distal edge. Differences in values recorded for sole and heel and for coronary wall horn measured under the distal edge were not different for claws varying in coronary horn pigmentation. The absence of differences in dry matter at measurement sites suggested that differences in impression hardness values were not caused by variation in water content of horn. Histological examination of pigmented claw wall horn and underlying soft tissue showed pigment-containing cells in the coronary epidermis but not coronary dermis or in dermis or epidermis of laminae. Pigmentation appeared concentrated in melanosomes in cells along the basement membrane and was present uniformly throughout the epidermis and in cells lining horn tubules. There was no pigmentation detected in the soft tissue epidermis of claws selected visually for non-pigmented wall horn. The presence of melanosomes in epidermal cells from the pigmented coronary region was demonstrated by electron microscopy. The results suggest that pigmentation in the coronary region is associated with reduced initial development of hardness. The possible antagonism between antioxidant properties of melanins and requirement for an oxidative environment for disulphide bond formation in the cytoskeleton of horn cells is discussed.