Straws and stovers are often called ‘by-products’ of grain production even though they are increasingly important, e.g. for animal feed, thatching, soil improvement, mushroom production and industrial use. As a result, plant breeders, agronomists, economists and animal nutritionists have to pay more attention than before to the total value of crops, i.e. whole plant value in which straws and grain both play a part. This paper reviews literature about the technical potential of breeding and/or management for more or better straw. It then discusses issues of the economic value (EV) and nutritional value (NV) of straw and stovers for livestock feed to guide research and development in cereal breeding and management. It is mainly based on experiences from the Indian subcontinent and semi-arid regions of the Near and Middle East. The paper shows that the quantity and quality of straw produced has changed considerably over recent decades as a result of breeding policies, new cultivation patterns and choice of cultivars. Both EV and NV depend on type of grain, animal production system and access to other feeds. A classification of these factors is provided and suggests that the EV of straw is particularly important in low-input systems with stovers from coarse grains.

An agro-ecological analysis of the maize component of an agroforestry system in the eastern mid-hills of Nepal was conducted in 1999. The aim of the study was to understand farmers' current maize cultivation practices as a prerequisite to development of maize varieties that are better adapted to local conditions than the improved varieties currently on offer from conventional maize breeding programmes in Nepal. Practices of 50 smallholder farmers were surveyed in three villages, and the response of the maize crop was recorded. The study revealed great variability in agronomic practices, and in particular, farmers' maize husbandry differed markedly from national recommendations for plant population, agronomic inputs and genotypes used. Farmers regularly employed practices that were not even officially recognised by the national research and extension services, such as thinning of maize for livestock fodder, growing the crop in association with trees for fodder, and relay cropping with finger millet, their priority being to optimize overall output of the farming system rather than maximizing maize productivity. Reasons for deviations from recommended practices are analysed and implications for maize crop improvement for heterogeneous conditions are discussed.

At its International Centres Week in October 2000 (ICW2000), the Consultative Group on International Agricultural Research (CGIAR) adopted a new Vision and Strategy. This paper is about Plank 4 of the CGIAR's Vision and Strategy which calls for the adoption, in collaboration with national and regional partners, of a regional approach to research planning, priority setting and implementation. Given the poverty and impact focus of international public goods research, both national agricultural research systems (NARS) and the CGIAR have advantages in pursuing a regional approach as a component of their respective activities. For the NARS in the region, this means seeking advantages at the regional level that they could not derive solely from a national-level approach. For the CGIAR, this means seeking complementary gains that it could not achieve exclusively through a global or ecoregional approach. These mutual advantages open the door for partnerships in regional research between NARS and their regional organizations, and the CGIAR. The paper highlights the advantages as well as risks and limitations of a regional approach to research. Since ICW2000, all regional and sub-regional organizations and CGIAR Centres have taken action to facilitate consultation processes that could eventually lead to the establishment of a regional approach to research for the CGIAR and NARS. The paper notes some emerging lessons, and takes a forward look.

As part of the reform process of the Consultative Group on International Agricultural Research (CGIAR), its Technical Advisory Committee is being transformed into a Science Council. An interim Science Council has been operating since January 2002. The primary role of the Council will be to enhance and guard the relevance and quality of science in the CGIAR System. Relevance is viewed as researching the ‘right things’ to address the System's goals, and deals with the relevance of identified research priorities and strategies and the nature of the planning and consultation process. Ex-ante analyses of need and potential impact to formulate the priorities and strategies are important elements in defining relevance. Quality is viewed as researching the ‘right things well’ and deals with the efficient use of state-of-the-art knowledge, research methods and protocols. The paper outlines the interim Council's suggestions for an operational framework of the Science Council. It serves as a starting point for the development of specific criteria, procedures and guidelines for carrying out the envisioned responsibilities. The Science Council must act in a strategic advisory role, basing its advice on: (i) planning and strategy development in the context of CGIAR goals; (ii) internal self-assessments by CGIAR Centres and Programmes and independent external monitoring and evaluation; and (iii) impact assessments. The paper discusses these three complementary functions. Each depends on the other in a dynamic feedback process that is essential if the CGIAR is to remain at the frontiers of science and development, maintain its science relevance and quality, and contribute in the most effective and efficient manner to the mission and goal of the Group.

The effects of on-farm seed priming (i.e. seed soaking) on the emergence, growth and yield of cotton and maize were studied in the field in the south-eastern lowveld of Zimbabwe. Experiments were conducted on both crops in the 1999/2000 and 2000/2001 seasons and, in the 2001 winter season, on maize only. The interaction of priming with tillage (ox-drawn ploughing or hand-hoeing) and simulated sowing rainfall regimes (irrigations of 15 mm, 30 mm or 45 mm at planting) was studied. Priming usually increased the rate of emergence in maize, but always decreased final percent emergence in cotton. In the 2000/2001 season, there was an interaction between priming and simulated sowing rainfall regimes such that the 15 mm treatment gave a smaller adverse effect of priming in cotton than the 30 and 45 mm treatments. In maize, however, the 15 mm treatment gave an adverse rather than a positive effect of priming on emergence. There was little effect of tillage on emergence or growth. Priming did not affect the relative growth rate of cotton or maize, although plants grown from primed maize seed were consistently larger at any given date throughout the 2001 winter season. Plants from primed seed also flowered and matured earlier in the winter 2001 season. There were no significant effects of priming on yield, except in the 1999/2000 season, where priming decreased yield in cotton. It was concluded that the effect of priming can depend on crop species.

Institutional changes in Benin have brought to light farmers' demand for varieties better suited to local growing conditions than existing ones. In response, we initiated a participatory cotton breeding experiment in 1996 to evaluate the relevance of such a methodology for the improvement of a commercial crop grown under rain-fed, semi-intensive cropping systems. This paper compares the performance of the first four mass-selection cycles, implemented by three farmer-breeders (F-B) and one formal breeder, with the original population and two commercial controls over three sites and two years. First results show that genetic changes occurred in all the F-B populations. The highest yielding F-B population (Savalou) was also more exuberant and later maturing than the others. Within the relatively narrow range of environments considered in the trial, there is no evidence that decentralized breeding results in better local adaptation. In Benin, participatory cotton breeding may be considered as complementary to formal on-station breeding and useful for enlarging the genetic variability offered to the farmers. Although the farmers want the approach to be scaled-up, its sustainability relies on a formal partnership between research and farmers institutions.

A survey of barley and barley-based farming systems was carried out in Palestine during the 2000 growing season to assess traditional cultivation methods and to analyse developments in the production systems. Seventy-six farmers in the three barley growing agro-ecological zones in the West Bank were included in the study. The research area encompasses a range of agro-ecological conditions, from high to low potential environments, strongly related to precipitation. Estimated grain yields varied from below 1000 kg ha−1 in the driest part of the Eastern Slopes zone to above 4000 kg ha−1 in the more fertile areas in the Central Highlands and Semi-Coastal zones. Barley was found to be closely linked to the dominant crop-livestock farming systems. Thus 84% of the farmers interviewed grew barley as a feed for their own livestock during the dry season. In the more fertile agro-ecological zones, the traditional farming system is shifting towards more intensive production methods with increased mechanization, mainly for land preparation, sowing, harvesting and use of fertilizer. Here animal production is becoming more intensive and a larger proportion of the farm products, including barley, is now produced for the market. Despite the relatively high yields, in comparison to other barley growing areas in the region, local barley landraces, particularly 6-rowed types, persist well and were preferred by 86% of the farmers interviewed. However attempts to modernize agriculture may result in the replacement of traditional barley landraces, and even the barley-livestock cropping system in the more fertile agro-ecological zones. An effort should therefore be made to conserve this valuable genetic resource.

In this paper, we present the case for making better use of the theory, concepts, language and experience from the field of industrial and commercial ‘new product development’ within agricultural research. The argument is that by conceptualizing, organizing and managing applied and adaptive research activities as new product development, agricultural research will be better able to respond to enlarging agendas, increased expectations of impact and declining budgets. The role of the design function in innovation development is highlighted, and the major implications of a new product development stance for the organization and management of agricultural research are identified.

Farmers grow vegetables widely during the dry season in wetlands known locally as dambos in southern Africa. Declining soil fertility is one of the major factors limiting smallholder vegetable production in the dambos of eastern Zambia. An experiment was initiated with 43 farmers with the objective of assessing the agronomic and economic feasibility of foliar biomass of gliricidia (Gliricidia sepium) and leucaena (Leucaena leucocephala) for production of cabbage, onion and a subsequent maize crop during the dry season. The treatments were, on a dry-matter basis, 8 and 12 t ha−1 gliricidia, 12 t ha−1 leucaena and 10 t ha−1 manure+half the recommended fertilizer rate, inorganic fertilizer at recommended rate, and a control without any inputs. Direct field measurements and informal enquiries were used for evaluating the effects of different treatments. The highest cabbage and onion yields were obtained from manure+half-rate fertilizer application. The gliricidia biomass transfer technology produced cabbage, onion and maize yields comparable with the full fertilizer application. In both cabbage and onion, manure+fertilizer gave generally higher net incomes. Biomass transfer also recorded higher net incomes than the control, and required lower cash inputs than the fully fertilized crop. Net incomes of the biomass treatments were substantially reduced by the labour costs for pruning and incorporation of the biomass. The results indicate that the gliricidia biomass transfer technology could be used as an alternative to inorganic fertilizers for vegetable production in dambos.

In this study, different levels of defoliation were imposed on a determinate species (maize) and a relatively indeterminate species (cotton). The aim was to quantify the effects of defoliation on plant growth and production, under either optimum or water-stressed conditions. Under well-watered conditions, 33% defoliation twice (conducted 28 and 35 days after emergence) resulted in a 16% reduction in grain yield of maize while 67% defoliation once (conducted 28 days after emergence) had no significant effect on yield. Under water stress, the grain yields of maize plants with 33% (twice) and 67% defoliation were 13.5% and 25% greater than that of non-defoliated control plants, respectively. For cotton, the reproductive yields (seed and lint) with 33% and 67% defoliation (conducted 43 days after emergence) were reduced, under well-watered conditions, by 28% and 37% of that of the non-defoliated control, respectively. Defoliated cotton plants lost less fruiting forms (squares and young bolls) than non-defoliated plants during water stress. Therefore, under water stress the harvestable product of cotton plants with 67% defoliation was double that of non-defoliated control plants. In non-defoliated cotton plants, a second flush of flowering after release from water-stress permitted further compensatory fruit set and boll harvest. Defoliated plants did not show such levels of compensation. Defoliation significantly reduced water use by maize and cotton. The relative yield advantage of defoliated plants under water-stress conditions can be attributed to defoliation-induced improvement in water status as reflected in measures of photosynthetic rate and stomatal conductance. Under anticipated drought stress, defoliation could be an important management practice to reduce drought-induced yield decrease, but this needs to be tested under field conditions.

Different fruit thinning methods were tested on ‘Sensation’ mango trees over two seasons in a trial conducted in Northern Province, South Africa. The objective of the study was to select the best thinning intensity and method based on different parameters. Thinning was carried out in October prior to the occurrence of excessive natural fruit drop. When fruits were thinned to one and two fruits per panicle, fruit number, weight and yield per tree at harvest were all increased. When 50% of the panicles were thinned there was an increase in the fruit retention potential of the trees as well as in the size and quality of the fruit produced. Large fruits were, however, prone to certain physiological problems. Chemical thinning with Corasil.E produced very small fruits with an increased percentage of ‘mules’ (fruit without seed). Trees subjected to severe thinning intensities showed earlier recovery of starch reserves and better vegetative growth than the control.

Cassava (Manihot esculenta) returns organic matter and nutrients to the soil through leaf litter and these amounts need to be quantified to help understand and design sustainable cropping systems. Our objectives were to estimate dry matter and nutrient contents in cassava leaf litter and to derive relationships between litter fall and easily measurable cassava yield components. Litter traps (1 m2) were placed in farmers' fields for monthly monitoring during a three year period. Maximum monthly leaf litter production ranged from 0.5 to 1.0 t ha−1, and occurred at the end of the first rainy season and at the onset of the next rainy season. In the first year, the mean dry matter of leaves collected during 12 months of growth was 3.4 t ha−1 for the unamended treatment, and 4.1 t ha−1 when N-P-K fertilizer was applied. The totals were 2.4 and 3.0 t ha−1 in 2000–2001 and 1.6 and 2.5 t ha−1 in 2001–2002, respectively. Annual differences were apparently related to rainfall. The relationship with fresh root yield was best described using one slope and yearly intercepts giving an r2 of 0.63. This relationship can be exploited for estimating litter dry matter in agronomic experiments when rough estimates are sufficient, keeping in mind that the relationship may not be the same for cultivars of differing architecture. Otherwise, the use of litter traps gives the best estimate of annual litter production.

Growth and water use of sole crops and intercrops of morphologically contrasting maize and pea cultivars were measured in two years. The maize cultivars were Nancis with erectophile and Sophy with planophile leaves and the pea cultivars Maro a leafy pea and Princess a semi-leafless pea. In the first part of the season water use was lower for sole maize but intercrops and sole pea used similar amounts of water. By 90 days after sowing, when peas had matured, all crops had used similar amounts of water. Maize had slightly greater water use efficiency than peas. Cultivars Nancis and Princess tended to have greater water use efficiency than Sophy and Maro respectively. Intercrops produced more dry matter than sole crops and therefore had consistently greater water use efficiencies.

Nitrogen (N) is a major nutrient that limits crop production in southern Africa. We hypothesized that coppicing tree legumes, which are integrated in cropping systems, would intercept leaching nutrients and could also increase topsoil N in nutrient-depleted soils. This hypothesis was verified in three ongoing experiments at Msekera (experiments 1 and 2) and Kagoro (experiment 3) in Zambia. Planted tree fallows of Gliricidia sepium, Leucaena leucocephala, Acacia angustisma, and Sesbania sesban were compared with natural fallows and with continuous maize cropping with or without fertilizer (no-tree) controls. Top and subsoil samples were taken in the tree treatments and in the no-tree controls to establish short and long-term tree effects on soil N dynamics. 15N was introduced at various soil depths down to 2 m to determine the vertical root-reach of coppicing trees. Samples taken on two different dates showed that planted trees are capable of capturing subsoil N. The amounts retrieved by trees in experiment 2 did not vary with depth or dates except for A. angustisma which retrieved more N from the top 0.20 m than from the subsoil. L. leucocephala and G. sepium had similar characteristics in terms of coppice biomass production and N content, and both species rooted to at least 2 m. G. sepium in a mixture with S. sesban, retrieved more applied N than when planted alone, implying that mixed fallows may be effective in resource capture. There was more inorganic-N in the topsoil of coppiced fallows was significantly higher than in unfertilized maize plots. Subsoil N accumulation was evident under fertilized maize plots. There was less subsoil nitrate-N beneath planted trees than beneath mono-cropped maize plots indicating that trees probably retrieved subsoil N. Maize yields subsequent to coppicing tree fallows were at least 170% higher than unfertilized controls indicating improved soil fertility status in the tree systems.

Twelve seedling cocoa families were evaluated as rootstocks in Sabah, Malaysia using three commercial cocoa clones as scions. The average yield was about 3 t dry cocoa beans ha−1. Yields on pure Scavina rootstock were about 10% above average and those on pure West African Amelonado about 10% below average. However, the effects of rootstock on yield were correlated with those on vigour and there was no effect on the ratio of yield to continuing vegetative growth. Rootstock did not influence bean weight or number of beans per pod or the uniformity of the trees. There was no indication of an interaction between rootstock and scion for any of the traits that were studied. The rootstock effect is considered large enough to warrant its control in critical work, especially field experiments with budded cocoa, but development of high performance scions is a higher priority in cocoa than intensive work with rootstocks.

On a large coconut plantation planted on a deep peat soil in Sumatra, Indonesia, hybrid coconut yields are stabilizing at values well below their potential, even in plots where known production factors have been mastered since the outset: land preparation, water control, and mineral fertilization. To explain this situation, five hypotheses have been proposed. They were tested one by one in field trials. The authors have demonstrated that the main reason for stabilized yields was the Lepidoptera root pest Sufetula spp. and conclude by considering lines of research for controlling this pest in the context of rehabilitation and replanting.

The parasitic angiosperms, Striga hermonthica and S. gesnerioides, obligate root parasites endemic in sub-Saharan Africa, constitute severe constraints to cereal and legume production in West and Central Africa. Over the years, a range of effective component technologies has been identified for Striga control in Africa. The potential of these technologies has been demonstrated under researcher-managed conditions. To promote farmer testing of the technologies, community workshops were conducted in 42 rural communities in Kaduna State, northern Nigeria. These revealed that agriculture was the main source of livelihood for most households. The three most important crops, maize, sorghum and pearl millet are attacked by S. hermonthica, regarded as the major constraint to crop production, often causing 70–100% crop loss. Farmers recognised two types of Striga damage (underground and aboveground), with greater damage being caused by underground Striga. Farmers attributed increasing incidence and severity of Striga damage to lack of capital, poor soil fertility, infestation of previously uninfested land by Striga seeds, and continuous cropping of host crops. The most widely used among the 15 existing Striga control techniques identified by the farmers were hoe weeding and hand pulling, application of inorganic fertilizer and manure, crop rotations, fallowing, and early planting. In assessing possible control measures farmers considered increased crop yield, reduced Striga reproduction and Striga emergence, greater crop vigour, and increased soil fertility as positive attributes. Negative attributes comprised increased labour requirement, higher costs, increased risk of crop damage or yield reduction, and lower quantity and quality of produce. Overall, a legume-cereal rotation was the most highly rated control option for S. hermonthica management evaluated by the farmers. The implications of these results are examined with respect to farmers' adoption and adaptation of Striga control options beyond the experimental plots.

The objective of the research was to test whether safflower (Carthamus tinctorius) could be grown successfully and profitably as a rain-fed, winter, oil-seed crop at a high-elevation site in a semi-arid area. A three-year field experiment was conducted in the Bekaa Valley of Lebanon at a site 995 m asl with a mean annual precipitation of 518 mm. Two safflower lines, Syrian Hama Local and PI301055, selected from earlier screenings, were compared with barley and lentil, the two dominant rain-fed crops, together with chickpea. Safflower flowered and matured more than a month later than the other three crops, and gave the highest dry matter at flowering. Safflower yielded more than chickpea and lentil, and there were no significant differences in yields between the safflower lines and barley, which is the highest yielding winter crop in semi-arid northern Bekaa. The straw yield of safflower was higher than that of barley, which was in turn higher than those of chickpea and lentil. According to calculations based on seed yield, safflower gave the highest net revenue of $191 ha−1, followed by lentils ($160 ha−1), chickpea ($97 ha−1), and then barley ($27 ha−1). Farmers in the semi-arid Bekaa Valley of Lebanon and in other areas with similar conditions are recommended to plant safflower to meet a regional vegetable oil deficit and to diversify cropping systems.

A field experiment in an alluvial sandy loam saline soil was conducted during the winter (rabi) season from 1997–98 to 1999–2000 at the Central Institute for Research on Buffaloes, Hisar, to study the effect of saline drainage water (EC=3.6 –7.4) on five (rabi) forage crops: oat (Avena sativa), rye grass (Lolium rigidum), senji (Indian clover) (Melilotus indica) berseem (Egyptian clover) (Trifolium alexandrinum) and shaftal (Persian clover) (Trifolium resupinatum). All the crops were established using canal water as pre-sowing irrigation and the various irrigation strategies were imposed subsequently. Irrigation with canal water resulted in a 115% increase in forage yield compared with the saline drainage water. The results suggested that alternate irrigation with saline drainage water increased the yields of all the forage crops compared with using saline drainage water only. Further, alternate irrigation, starting with canal water, was superior to alternate irrigation starting with saline drainage water because less salt was added in total. Oat produced the largest green-forage yield (32.3 t ha-1) in the first year while rye grass gave its maximum in the second (34.6 t ha-1) and third years (37.0 t ha-1). Persian clover performed better than did Egyptian clover in all the three years. Interaction between species and irrigation treatments was significant. In comparison with canal irrigation water, there was a 36 %, 42 %, 54 %, 68 %, and 85 % yield reduction in rye grass, oat, Persian clover, Egyptian clover and senji, respectively when only saline drainage water was used for irrigation reflecting their relative tolerances of salinity. Yields declined linearly for all crops with increases in the quantity of salt applied.

Gliricidia sepium features prominently as a soil replenishment tree in planted coppicing fallows in eastern Zambia. Its usual method of propagation, through nurseryseedlings, is costly and may possibly hinder wider on-farm adoption. We compared fallows propagated by potted and bare root seedlings, direct seeding and stem cuttings, in terms of tree coppice biomass production, soil inorganic N availability and post-fallow maize yields under semi-arid conditions. We hypothesized that cutting fallows initially in May (off-season) would increase subsequent seasonal coppice biomass production as opposed to cutting them in November (at cropping). The tree survival and biomass order after two years was: potted = bare root > direct > cuttings. The post-fallow maize productivity sequence was: fertilized maize = potted = bare root > direct > cuttings = no-tree unfertilized controls, across seasons. However, farmers may prefer directly seeded fallows owing to their cost effectiveness. Soil inorganic N and maize yield were significantly higher in May-cut than in November-cut fallows. Preseason topsoil inorganic N and biomass N input correlated highly with maize yields. This implies that bothparameters may be used to predict post-fallow crop yields.