The effects of temperature (13–20 °C) and potential soil water deficit (0–350 mm on the leaf appearance rates of six contrasting clones of tea (Camellia sinensis) were studied over twenty-five months to provide guidance to tea growers in southern Tanzania on the choice of harvest intervals. Within the fully irrigated treatments (potential soil water deficit <65 mm) the mean time period between a shoot unfurling its second and third true leaf (a phyllochron) ranged from 5.8–7.9 d during the warm–wet season to 11–19 d during the cool-dry season. Between 61 and 79% of the weekly variation in the mean leaf appearance rate (1/phyllochron) for each clone could be explained by an asymptotic relation with the mean air temperature. Building on research in India which has suggested that appropriate harvest intervals for tea can be derived from measurements of leaf appearance rates, a simple procedure is described for determining harvest intervals from the daily mean air temperature. Corrections to account for the effects of drought and clonal differences in responses to temperature are also explained.

A three-year study was conducted on an Ultisol to determine the water requirement, yield and fruit-quality traits of three ratoon crops (R1, R2, R3) of ‘Grande Naine’ banana (Musa acuminata Colla, AAA group) subjected to four levels of irrigation. The irrigation treatments were based on Class A pan factors ranging from 0.0 (rainfed) to 1.0 in increments of 0.25. When needed, drip irrigation was supplied three times a week on alternate days. Results showed significant (p < 0.01) irrigation treatment and crop effects on bunch weight, yield, bunch mean hand weight, weight and fruit diameter of the third and last hands, and length of fruits of the third hand. Highest marketable yield (47.9 t ha−1) was obtained from the R2 crop with water application according to a pan factor of 1.0. It was concluded that irrigating the crop according to a pan factor of 1.0 was sufficient to justify the investment of a drip-irrigation system for a farm in the mountain region.

A farmer's intercropping system of white sorghum (Sorghum bicolor) and néré (Parkia biglobosa) was studied in Burkina Faso. In quadrats in three differing zones of tree influence, full sun (open field), semi-shade (under the edge of the tree crown) and full shade (under the crown), tree morphology and associated sorghum growth were recorded together with micro-meterological parameters. Mean photosynthetic photon flux density (PFD) was reduced to 51 and 26% of full sun levels, in the semi-shaded and fully shaded quadrats respectively. Soil temperature and moisture were also recorded. In the season studied, the principal effect of the trees on the crop was competition for light. For plants in full shade, total shoot and panicle mass were halved compared with the crop in full sun. However, plants in semi-shade were able to compensate for decreased PFD almost entirely. Measurable compensatory effects were decreased rate of leaf senescence, greater partitioning of resources into leaf production, and larger, thinner leaf laminae. It was concluded that the current farmers' practice of sowing under the crowns of trees produced a worthwhile, if somewhat reduced, crop yield, and that socially, economically and ecologically the sorghum–néré parkland system was worth retaining.

Leucaena (Leucaena leucocephala) is most preferred by development workers for agroforestry in the mid-altitude (600–1600 m) agro-ecological zone of North-West Cameroon because its seeds are common and inexpensive. This study was prompted by farmers' reluctance to accept leucaena for agroforestry because its early growth is poor on acid infertile soils prevalent in the area. Leucaena was planted using four-month-old potted plants and, from four months after planting, the shrub was cut five times in 20 months at 120 day intervals. Manuring improved leucaena's growth and biomass at sites with acid and non-acid soil alike. Total biomass responded linearly (R2=0.97) to manure rates at the acid and infertile site and curvilinearly (R2=0.95) at the non-acid site. Total leafy biomass from the first cut and subsequent prunings was highest at 12 months after planting. Manuring increased rooting depth and this contributed to the plant's observed ability to cope with moisture stress during the dry season. The economics of manure use for tree establishment on acid infertile soils for agroforestry systems needs further study.

In an on-station agroforestry experiment conducted in south-west Niger, the effects of seven windbreak species and of a soil mulch made from crop residue on wind erosion and pearl millet (Pennisetum glaucum) production were monitored. Within a distance of 20 m, strips of the perennial grass, Andropogon gayanus, reduced total annual soil flux by 6–55% and hedges of Bauhinia rufescens 2 m in height reduced soil flux by 47–77% compared with unsheltered control plots. No significant overall windbreak effect on millet stover and grain yields was found. In contrast, erosion reduction and yield increases due to mulch application were highly significant. Soil mulch is a promising alternative to complex windbreak–millet cropping systems in regions where direct economic benefits for farmers are not ensured by windbreaks.

Data from 110 on-farm vegetable trials conducted in Jamaica were examined to seek guidelines as to how on-farm trials should be executed in order to obtain quality data. The results show that while one-way blocking is useful in designing experiments it may sometimes be inefficient and that two-way blocking is generally preferable. If only a small number of trials can be conducted then it is important that up to four or five replications of each treatment are used at each site. Plot sizes should be larger than those used for on-station trials. Due to low correlations between results from different sites, experiments should be conducted at many sites. Thus a large number of trials should be planned to allow for treatment–environment interactions and trial failures.

Data from studies of growth and development, and response to plant density in common groundnut (Arachis hypogaea) cultivars were examined from published studies. Data were available from the humid tropics of Indonesia, the semi-arid tropics of north-west Australia and the humid coastal and inland elevated areas of north-east Australia. Temperature and irradiance played a major role in determining crop duration, individual plant size and partitioning of dry matter to pods across environments, and these plant characteristics provided the major determinants of pod yield and response to plant density. Crop duration was shortest in humid tropical and subtropical environments, with both high and low temperatures apparently delaying crop maturity. A relatively small individual plant size in humid tropical environments was due to a combination of low incident irradiance and short duration, with very high plant densities needed to maximize dry matter production. The progressive decline in harvest indices in more tropical environments was due to a decline in pod numbers per plant. Although increased plant density resulted in greater numbers of pod initials in the humid tropics, a high proportion of these pods did not contain developed seeds and pod yield at high densities remained relatively low at ≤2.5 t ha−1.

Previously published data from two tea experiments in Tanzania and Malawi were re-analysed to identify factors responsible for the marked seasonal peaks and troughs of production observed at the two locations. Simulation modelling indicated that the influence of seasonal variations in temperature or photoperiod on shoot growth rates alone were not able to explain the occurrence of a large peak in September/October followed by smaller and progressively wider peaks in December, February and April in well-irrigated, well-fertilized tea at the two sites. Various mechanisms to explain these seasonal variations in yield are discussed, but it is concluded that the best explanation is the regulation of the numbers of shoots available for harvesting through the effect of photoperiod on shoot activity. The significant production peak normally obtained in September/October at both locations can then be explained by the release of a large number of buds from their resting state around the time of the winter solstice in June. Subsequent peaks are due to successive cohorts of shoots released from apical dominance when the preceding peak is harvested.

Changes in soil pH water (pHw) were investigated on a sugarcane plantation in the Ramu Valley of Papua New Guinea. The plantation was established in 1979 from natural grassland and the dominant soils were Fluvents and Vertisols. Statistical analysis of the topsoil pHw data (n = 541) showed a significant (α = 0.01) decrease from 6.5 to 5.7 between 1979 and 1996. Based on samples from 80 fields at different sampling times, the average decrease in topsoil pHw was calculated to be 0.4 units after 10 years of continuous sugarcane cultivation (r2 = 0.481). Between 1986 and 1996, subsoil pHw also decreased significantly (α < 0.05) by 0.4 and 0.3 units in the 0.30–0.45 and 0.45–0.60 m horizons respectively. The acidification trend accelerated in the 1990s when trash-harvesting replaced pre-harvesting burning and sulphate of ammonia became the dominant nitrogen fertilizer. Between 1991 and 1995, average nitrogen application rates were 90 ka ha−1 a−1 which resulted in an annual addition of 11.6 kmol H+ ha−1. The soil buffer capacity was estimated to be 125 kmol H+ ha−1 pH−1 which implied that the pHw could further decrease by 0.5 unit after five annual applications of 90 kg N ha−1 pH−1 as sulphate of ammonia. Although these soils were young in pedological terms and they had been cultivated for less than 20 years, they had acidified significantly and this could affect sugarcane production adversely if such trend were to continue.

Experiments were conducted during the winter seasons of 1992–93 and 1993–94 at the Lumle Agricultural Research Centre and its off-station research site at Lopre in Nepal to study the agronomic performance and profitability of barley (Hordeum vulgare) and peas (Pisum sativum) mixed intercropping (that is, without any row arrangement) in the subsistence farming systems of the Nepalese hills. Sole crops of barley (cv. Bonus and Local) at 100kg seed ha−1 were compared with a pure stand of peas (cv. Tikot Local) sown at 60kg seed ha−1 and with peas intercropped with both varieties of barley at 100:40 or 100:20 kg seed ha−1. Results across two locations for two seasons showed that the barley+peas intercropping was advantageous in terms of overall grain yield, land equivalent ratio (LER), monetary advantage, economic return and dietary provision in the subsistence farming system. It also provided yield stability under adverse climatic conditions. Barley genotypes differed in their suitability for intercropping with peas. The yield of Local barley was reduced when peas were intercropped with it at both seed rates but this was not so with Bonus. The most appropriate combinations to realize the maximum advantage from intercropping were Bonus barley+peas at 100:20 kg seed ha−1 and Local barley+peas at 100:40 kg seeds ha−1. A combination of Bonus barley and peas at 100:20 kg seed ha−1 enabled farmers to harvest an additional 400 kg grains ha−1 from peas without significantly reducing the grain yield of barley. The study also verified through formal experimentation that the traditional practice of mixing barley and peas is advantageous.

When the same cultivars of groundnuts (Arachis hypogaea) were grown under a wide range of environmental conditions, temperature and irradiance played a major role in determining crop duration and partitioning of dry matter to pods, the latter assessed by harvest index. Utilizing published data for the Virginia groundnut cultivar Early Bunch under non-limiting conditions, we show that accumulation of thermal time using three cardinal temperatures (Tb=9 °C, To=29 °C and Tm=39 °C) has considerable potential for predicting crop maturity. In sixteen sowings ranging from the wet tropics in Indonesia to the elevated subtropics in Australia, harvest date for Early Bunch corresponded to the accumulation of 1808 (±23) degree-days after sowing. In all sowings except one in the semi-arid tropics, this value of thermal time was within eight calendar days of actual harvest maturity. Harvest index varied greatly with both location and sowing date, ranging from 0.31 (Indonesia) to 0.58 (subtropical Australia). Using total short-wave solar radiation incident during the growing season and calculated values of thermal time, the growing season for each sowing in each location was described in terms of a photo-thermal quotient (PTQ, MJ m−2 degree-day−1). Values for PTQ ranged from 0.99 (Indonesia) to 2.11 (subtropical Australia). Variation in harvest index could be explained largely by a curvilinear function of PTQ (R2=0.98), provided data were not confounded by the effects of photoperiod. In the semi-arid tropical environment, decreases in photoperiod associated with delayed sowing were the dominant factor controlling harvest index.

Cowpea (Vigna unguiculata) farmers from diverse geographical areas in northern Cameroon selected breeding lines from on-station trials for their own fields and explained their reasons (criteria) for making the selections. The average selection intensity employed by farmers varied from 6 to 17% and was similar to that employed by plant breeders. Their selection criteria fell into three broad categories relating to yield, preference and quality, and labour, which were very consistent over years, locations and gender. Selection criteria were strongly influenced by market preferences, reflecting the increasing role of cowpea as a cash crop. Implications of these findings for breeding strategies and genetic resource conservation are discussed.

The on-farm performance of more than 150 000 pre-basic potato (Solanum tuberosum) minitubers of 1–10 g was monitored over two years in contrasting environments in Nepal. Average minituber productivity (grams of basic seed produced per minituber planted) was 171 g in the southern plains and 116 g in the hills. Although the performance of minitubers varied between environments, minituber sizes and potato varieties, the overall results clearly indicated that minitubers can be successfully multiplied by small-scale farmers. This technology therefore allows for the highly decentralized production of basic seed which, in turn, will reduce farmers' dependence on an unreliable and costly supply of bulky conventional seed.