Although viewed as important for on-farm adaptive research, farmer-participation has often been superficial and has not led to meaningful incorporation of farmer perspectives. This paper describes the use of a method involving the understanding of farmers' perceptions in determining priorities for research which integrates the different concerns and contributions of farmers and scientists with the transfer of technology from farmer to farmer.

Simple methods are proposed for measuring the relative importance of grain and fodder produced jointly from maize fields in four zones in northern Pakistan. These methods suggest that the joint production of grain and fodder from maize is important in all four zones and that fodder accounts for between one-third and one-half of the total value of the crop. The relative value of fodder production is particularly high in two of the zones (the irrigated mid-altitude Swat Valley and the rainfed low altitude Islamabad District), both characterized by a high ratio of animal numbers:maize area. The results suggest that maize research and extension recommendations that do not take into consideration the fact that farmers produce maize for fodder as well as grain will often not be accepted by farmers.

Nodulated plants of Stylosanthes hamata, S. guianensis, S. humilis, S. scabra and S. fruticosa were grown in controlled environments with varied root and shoot temperatures. Measurement of dry matter and nitrogen content suggested that shoot temperature may be more important than root temperature in controlling growth and nitrogen fixation. There were strong interactions with variety. A fall in relative growth rate with increase in shoot temperature was least for S. guianensis and greatest for S. hamata and S. scabra. The optimum root temperature for growth and nitrogen fixation was approximately 30°C. Ninety percent maximum yield was achieved between root temperatures of 15–36°C for growth and 23–34°C for nitrogen fixation but varied with variety. Nitrogen fixation was more sensitive than dry weight to root temperature. The pattern of response of percentage nitrogen and nitrogen fixation efficiency reflected those for dry weight and nitrogen yields. Shoot to root ratios decreased toward the optimum root temperature then increased at the highest temperature. The reaction of varieties to root and shoot temperatures may be an important factor in determining their suitability for new regions.

Nodulated plants of six varieties of Stylosanthes were grown at a constant temperature of 30°C for 21 days in a controlled environment and then for a further 21 days at a range of day/night root temperatures. Dry matter and nitrogen content of S. hamata cv. Verano and CPI40264A were measured after growth at day/night root temperatures of 30/30, 40/30, 50/30, 60/30 and 60/35°C. Similarly, growth and nitrogen content were measured for Verano, CPI40264A, S. gvianensis cv. Oxley and cv. Cook, S. humilis cv. Lawson and S. scabra cv. Seca after growth at root temperatures of 25/20, 25/25, 30/25, 30/30, 35/30 and 40/30°C. There were strong interactions of root temperature with variety and harvest time (31 and 42 days) but all varieties grew well even at the highest temperatures (50–60°C) provided they were supplied with nitrogen. Symbiotically dependent plants of S. hamata cv. Verano were more tolerant of high root temperatures and maintained better growth and nitrogen fixation at 40/30°C than those of CPI40264A, especially after the 31 day harvest. All six varieties responded similarly, but to differing extents, to the lower range of root temperatures. Nitrogen fixation was more severely affected than dry weight by high root temperature.

A method of obtaining the standard error and some other moments for the ratios of biological measurements is given, with the conditions to be satisfied by the variables used to form such ratios. Some applications of the distributional properties of the ratios in the analysis of results from experiments are discussed.

Azolla and blue-green algae (BGA) inoculated in a rice field produced a higher biomass and fixed more nitrogen in transplanted than in direct-seeded rice. Application of Azolla, BGA, composts of Eichhomia and Azolla and green manuring with Sesbania increased tiller numbers, height, grain and straw yields, and nitrogen uptake by rice in both the transplanted and the direct-seeded crop. The growing and incorporating of three Azolla crops with the rice crop produced more grain than the use of other nitrogen sources but the rice crops showed a poor response to inoculation with BGA. The yield of rice following green manuring with Sesbania or the use of two crops of Azolla after transplanting was similar to that of a crop receiving 60 kg N ha−1 as urea, but smaller yields resulted from the use of composts of Azolla and Eichhornia.

Effects of simulated and actual grazing at the tillering stage on tiller production and other agronomic characters were studied over a number of seasons at two sites in northern Syria with two local barley landraces. In general, grazing encouraged tiller production but reduced head number per unit area because of a higher tiller mortality or infertility. Seed number per head and seed weight were not affected by grazing except in one season. Grazing delayed heading and maturity and reduced plant height. Simulated and actual grazing produced similar effects on the different characters investigated.

The effect of a single grazing at the tillering stage on barley yield and economic returns was studied with two widely grown local landraces over five seasons at Tel Hadya and two seasons at Breda in northern Syria. In general, grazing reduced both grain and straw yield, particularly the latter. The percentage of grain yield reduction was similar at the two sites. Grazing increased net revenue on average, though there were marked seasonal differences. This increase was higher at Breda, the drier site, a finding which contradicts the commonly-held belief that grazing is unprofitable in dry areas.

Growth and development of the tropical grain legumes are generally highly sensitive to photo-thermal regime, so that seasonal and regional effects on phenology and yield potential can be large. Yet failure adequately to recognize and fully exploit the consequences of genotype × latitude/sowing date × density interactions has frequently constrained both agronomic and genetic advance with these species. Thus there is opportunity for short term productivity improvements through agronomic strategies which accept the implications of phenological plasticity, and seek to optimize management practices such as sowing date and sowing density, to exploit more effectively the yield potential and broaden the adaptation of existing cultivars. The greatest physiological potential for genetic improvement in the productivity of the tropical grain legumes lies not with increasing total biomass, but with increasing the proportion of bio-mass partitioned into seed, i.e. with improved harvest index (HI). There are difficulties in selecting for higher HI, including its association with phenology, although the latter provides the most powerful tool for manipulating HI in the short term. As with the cereals, more productive genotypes are likely to be characterized by reduced sensitivity to photothermal conditions, shorter growth duration, a more synchronous reproductive ontogeny, and greater HI, than is typical of traditional cultivars. Accordingly, optimal sowing densities will be higher, and crops will need more inputs, better management and more effective protection, than afforded to them in subsistence systems of production. Moreover, the increases in productivity will be achieved, at least partially, at the expense of the potential for yield homeostasis in adverse environments, and of non-seed components such as forage, fuel or tubers - attributes that are often highly valued in subsistence agriculture where many of the tropical food legumes are presently grown. Improvement programmes will therefore need to adopt strategies in particular situations which reflect local perceptions of the importance of productivity improvement relative to that of minimizing inputs and risk.

A field study was conducted in 1985 at two locations in Maryland, USA, to test the influence of environment (location), input level (N fertilization), management (component density combinations), and measure of performance on recommendations for a maize/soyabean intercrop. Location affected the degree of intercrop yield depression of each crop; the effect of N on relative competitive abilities; the maximum Land Equivalent Ratio (LER) and maximum modified Staple Land Equivalent Ratio (SLER); the difference between LER and SLER; the optimum density combinations and change in optima in response to N; and the N level at which the greatest land use efficiency was achieved. Given the complexity of the environmental effects, recommendations based on the trials have a very narrow range of applicability.

A trial was carried out at Pantnagar in India during 1984–85 and 1985–86 to study the productivity of Egyptian clover and oats under intercropping. Cross sowing of both the crops either at full or half seed rates and line sowing of oats in broadcast Egyptian clover was better than other intercropping systems or sole cropping in respect of herbage production, digestible dry matter yield and land equivalent ratio but crude protein yield from the mixed crops was similar to that from sole cropped Egyptian clover.

A simple Stress Time Index (STI) for predicting yield loss in the tea crop due to drought is proposed, based on the daily summation of the difference between the potential soil water deficit and a specified limiting value. Validation of the technique with results from a line-source irrigation experiment with a single clone in the Southern Highlands of Tanzania suggests that there is a linear relationship between STI and relative yield loss during the warm dry season. As a result of changes in the composition of the shoot population at each harvest the apparent critical deficit at which shoot growth is restricted increases from below 20 to 300 mm as the dry season progresses. The rate of yield loss with increasing STI also varies through the dry season for the same reason.