A better understanding of the agronomic and economic benefits of integrating organic and inorganic fertilizers is essential for their wide-scale dissemination and adoption in smallholder farming systems in developing countries. Field studies were conducted for 10 cropping seasons (1996–2000) on a Kandiudalfic Eutrudox soil to compare the effects of Minjingu phosphate rock (MPR) and triple superphosphate (TSP) on the yield of maize (Zea mays), in combination with three nitrogen (N) sources: urea, Tithonia diversifolia and Sesbania sesban leafy biomass. Urea and tithonia were applied to provide 60 kg N ha−1. Phosphorus (P) from either MPR or TSP was added either once at 250 kg P ha−1 at the beginning of the experiment or annually at 50 kg P ha−1 for five years, the two rates representing two P recapitalization strategies. From the second year, potassium (K) was added to half of each plot to correct for the deficiency that emerged. Over the 10 cropping seasons, the agronomic and economic benefits of the two P recapitalization strategies were similar and were not influenced by the P sources used. With N and K application, both P sources resulted in at least a two-fold maize grain yield increase over the control with no P application that averaged 1.1 tons ha−1. Comparing the N sources, although urea and tithonia had similar net benefits when P was applied, the total cost associated with tithonia was considerably higher. Without P application, sesbania fallow was the most financially attractive option. Sesbania fallows as an N source were also least sensitive to increases in the price of fertilizers and the cost of labour although the financial benefits of this system remained low in the absence of P and K application.

Chicken manure is a potential source of plant available phosphorous (P) and this field study explored its use for dry land maize production in a soil with very low inherent P levels. The treatments consisted of applying 0, 20, 40 and 80 kg P ha−1 each from chicken manure (2.3% P), inorganic fertilizer (single super phosphate, 10.5% P) and a combination of manure and fertilizer. The treatments were replicated four times in a randomized complete block design over two crop seasons. The results revealed that maize growth and yield responded to the P from both manure and fertilizer compared to the control. There were significantly (p < 0.05) taller plants and yield (grain and biomass) in maize supplied with P than control. At each level of P application however, the response was higher for fertilizer than manure. The combination of manure and inorganic fertilizer produced results that were in the middle. There were strong indications that application of chicken manure was associated with increased lodging possibly due to the additional N in the manure, attack by stalk borer (Buseola fusca) and reduced seedling emergence. The application of chicken manure on the other hand led to increased pH and greater residual P content of the soil than fertilizer. It is concluded that chicken manure presents a viable option for supplying P to sustain maize crops in the smallholder farming sector of South Africa and its management warrants further investigation in view of the potential adverse impacts on soil and water resources that can accrue when large amounts are used for a long period.

Field experiments were conducted in 2005 and 2006 to investigate the impacts of alternative rice cultivation systems on grain yield, water productivity, N uptake and N use efficiency (ANUE, agronomic N use efficiency; PFP, partial factor productivity of applied N). The trials compared the practices used with the system of rice intensification (SRI) and traditional flooding (TF). The effects of different N application rates (0, 80, 160 and 240 kg ha−1) and of N rates interacting with the cultivation system were also evaluated. Resulting grain yields with SRI ranged from 5.6 to 7.3 t ha−1, and from 4.1 to 6.4 t ha−1 under TF management. On average, grain yields under SRI were 21% higher in 2005 and 22% higher in 2006 than with TF. Compared with TF, SRI plots had higher harvest index across four fertilizer N rates in both years. However, there was no significance difference in above-ground biomass between two cultivation systems in either year. ANUE was increased significantly under SRI at 80 kg N ha−1 compared with TF, while at higher N application rates, ANUE with SRI was significantly lower than TF. Compared with TF, PFP under SRI was higher across all four N rates in both years, although the difference at 240 kg N ha−1 was not significant. As N rate increased, the ANUE and PFP under both SRI and TF significantly decreased. Reduction in irrigation water use with SRI was 40% in 2005 and 47% in 2006, and water use efficiency, both total and from irrigation, were significantly increased compared to TF. With both SRI and TF, the highest N application was associated with decreases in grain yield, N use efficiency and water use efficiency. This is an important finding given current debates whether N application rates in China are above the optimum, especially considering consequences for soil and water resources. Cultivation system, N rates and their interactions all produced significant differences in this study. Results confirmed that optimizing fertilizer N application rates under SRI is important to increase yield, N use efficiency and water use efficiency.

New technologies often lack a developed supply market, which makes it problematic to estimate future profitability. This paper develops a simple methodology to circumvent this problem based on the concepts of break-even and expected marginal rate of return. The methodology estimates ex ante a range of target threshold prices: the range of maximum prices that farmers are likely to be willing to pay. Only if the future market is able to supply the new technology within or below the threshold price range is significant adoption likely. The target threshold prices can be used to reorient technology development and delimit recommendation domains. The paper applies the methodology to the case of rock phosphate in upland rice in western Cote d'Ivoire. The paper shows that for rock phosphate to be viable for upland rice farmers its on-farm price should generally not surpass Euro cents 5.7–8.1 per kg of rock phosphate.

In agricultural innovation, the success of widely used technologies is often assumed to have been inevitable. Conversely, the blame for the failure of new technologies that researchers, policy makers and extensionists consider superior to existing solutions is often placed on farmers. However, these assumptions can be challenged by taking a social-constructivist view of on-farm innovation to examine how and why farmers made sense of new technologies and how this sensemaking shaped their use of these technologies over time. The present study took such an approach in its analysis of Australian woolgrowers’ adoption, abandonment, implementation and use of new wool-testing technologies that highlighted the social and dynamic nature of innovation on-farm. On-farm innovation in this case was an evolving, dynamic process that changed over time as woolgrowers made sense of new technologies. The primary message to agricultural innovation researchers, technology developers, policy makers and extensionists is that successful on-farm innovation requires the active, ongoing engagement of industry participants. In order to engage industry participants in the innovation process, sensemakers’ personal identity frames and social context, and how these interpretation frameworks relate to the new technology need to be understood.

Conflict is the most common cause of food insecurity. Foreign aid to countries emerging from conflict often allows a funded but brief window for the confirmation-testing and diffusion of agricultural innovation in affected areas. This paper asks the question: what lessons has agricultural research learned through its involvement in this process in countries emerging from conflict? Drawing on experience from Afghanistan and other countries, this paper documents some cases in which it has been possible to inject an element of simple hypothesis testing, often in farmer-managed trials, into post-conflict plans leading to useful lessons. Agricultural researchers need to be cognizant of this approach so that the practice becomes more widely used and lessons recorded for future use.

Traditional cultivars of arabica coffee (Coffea arabica) are susceptible to coffee leaf rust (CLR, Hemileia vastatrix) and coffee berry disease (CBD, Colletotrichum kahawae). CLR is of worldwide importance, while CBD is still restricted to Africa. Host resistances present in C. canephora to both destructive diseases have been successfully introgressed into C. arabica, after several decades of breeding and selection notably in Brazil, Colombia, India and for CBD resistance in East Africa. CLR resistant cultivars have since been grown on hundreds of thousands of hectares in Latin America and elsewhere, while CBD (and CLR) resistant cultivars are now increasingly being planted in East and South Eastern Africa. They contribute to ecologically sustainable coffee production and to considerable socio-economic benefits for the coffee growers. Nevertheless, some representatives of the international coffee trade continue to be sceptical about the cup quality of these modern cultivars. This paper presents an overview of the considerable amount of scientific evidence accumulated over the years showing that, with all environmental factors at optimum level, disease-resistant cultivars can produce quality coffees as good as those from the best traditional varieties.

The results of research on the water relations and irrigation need of banana are collated and summarised in an attempt to link fundamental studies on crop physiology to irrigation practices. Background information on the ecology of the banana and crop development processes, with emphasis on root growth and water uptake, is presented, followed by reviews of the influence of water stress on gas exchange (stomatal conductance, photosynthesis and transpiration), crop water use, and yield. Emphasis is placed on research that has international relevance and, where appropriate, three geographical areas (the tropics, subtropics and Mediterranean climates) are considered. Although roots can extend to depths of 1.0–1.5 m, the ‘effective’ depth of rooting is usually taken to be 0–0.40 m, sometimes extending to 0.60 m. Stomatal conductance is a sensitive measure of soil water availability and plant water status, whilst transpiration rates can be limited by dry air (saturation deficits >2.0 kPa). In the subtropics, there are seasonal differences in the crop coefficient (Kc) with values ranging from 0.6 in the winter months to about 1.0 Epan in the summer. It is difficult to draw generic conclusions with wide applicability from the irrigation experiments as they were reported. All the components of marketable yield can be enhanced by irrigation whilst applying insufficient water delays crop development. Annual yield responses to irrigation are variable, but water use efficiencies of 40 kg ha−1 mm−1 (fresh fruit/water applied) have been achieved in the tropics and subtropics (and elsewhere up to 80 kg ha−1 mm−1 with ‘partial’ replacement of the soil water deficit). To ensure large yields of (marketable) fruit, soil water deficits must be kept low (Ψm > −20 kPa at 0.2 m depth). In the subtropics, this means irrigation intervals should not exceed 2–3 d during the summer. The cooling effect of irrigation with micro-sprinklers on the soil and pseudostem temperatures, compared with drippers, can delay crop development and reduce annual yields by 30%. There is some (limited) evidence that the presence of the B genome contributes to drought tolerance. Yield response factors to irrigation for different growth stages have yet to be confirmed. Opportunities to improve the water productivity of the many, diverse banana cultivars need to be explored further.