The genus Vigna includes seven species of pulse, two African and five Asian. While the Asian species are within the subgenus Ceratotropis, the two African species are in different sections of the subgenus Vigna, namely Catjang (V. unguiculata) and Vigna (V. subterranea). Limited gene flow is possible within subgenus Ceratotropis, but none has as yet been demonstrated between subgenera nor between sections within subgenus Vigna. Parallel evolutionary trends are apparent in all species, with considerable increase in seed and pod size. Erect bush forms have evolved in all cultigens except V. aconitifolia and V. subterranea. The former has not apparently changed much in vegetative form under domestication, while the geocarpy of the latter precludes erect growth. Future prospects for the Asiatic grams would be better if yield levels could be improved. The cowpea is likely to remain an important crop at present levels of performance although there is room for improvement. There seems little immediate prospect of expanded ground bean production under existing conditions.

A consideration of the apparent lack of general success of minor pulse crops can lead to a better appreciation of the factors which have made the major pulses successful. The relevant factors are both biological and economic. High productivity based on an efficient physiology, amenability to local agronomic practices and a satisfactory range of ecological tolerance are prime prerequisites for success. Aspects of quality, palatability, lack of toxic factors (or ease of detoxification), ease of preparation and cooking and the more imponderable questions of consumer preference are equally important. If any of these prerequisites are not adequately met success in cultivation is improbable. At the present time the lack of more extensive success of the winged bean appears to be due to some deficiency in ecological tolerance, and of the horse-gram and jack and sword beans to a lack of palatability; the Hausa groundnut (Kersting's groundnut) suffers from a probably inefficient physiology while limited consumer acceptance has possibly restricted production of pigeonpea and hyacinth bean. The rather limited evolutionary advance achieved by this group of pulses is probably directly related to their somewhat restricted production and the relatively low level of selection pressures to which they have been subjected.

The genus Phaseolus as currently recognized contains four pulses: common bean (P. vulgaris), runner bean (P. coccineus), lima bean (P. lunatus) and tepary bean (P. acutifolius). Although these all have their origins in tropical and sub-tropical latitudes, they have evolved in different ecological zones. The runner bean has evolved at higher altitudes, the common bean at intermediate levels and the lima bean at lower altitudes. The tepary has a specialized desert-annual life-form. The occurrence of day neutral genotypes in all species has permitted their spread into cool and worm temperate zones (the common and runner beans) and the warm temperate zone (lima bean). It is possible therefore to grow one or other Phaseolus species in most areas of the world where cultivation can be practised. The common bean has evolved the widest range of growth forms, seed and pod sizes, pod forms and textures, and seed and pod colors. World-wide an enormous primary gene pool has been produced. The common bean pool is the secondary gene pool for the runner bean and vice versa, since partially fertile interspecific species produce sterile or inviable F1 hybrids so that potential for interspecific gene transfer is limited. Thus no known secondary gene pools exist for Phaseolus species other than P. vulgaris and P. coccineus.

Although the groundnut and soyabean are not close taxonomic relatives they have a number of features in common. Both are polyploid (2n = 4x = 40 chromosomes) and have arisen from the domestication of wild polyploids. Polyploidy has occurred independently at least twice in the genus Arachis, while in Glycine the genus includes both tetraploids and octoploids, an unusually high ploidy level in a legume. Both crops are oilseeds although oil content varies widely between the two species, ±50% in groundnut and ±20% in the soyabean. The soyabean produces ±35% protein content of good quality whereas that of the groundnut is ±30% and less well balanced nutritionally. The soyabean originated in the Far East whereas the groundnut originated in the eastern Andes foothills (N.W. Argentina to S. Bolivia). The evolutionary changes under domestication in the soyabean are typical; the cultigen has become a compact, erect, self-supporting plant with reduced pod dehiscence and loss of seed dormancy. The evolutionary response pattern of the groundnut is similar but modified on account of its geocarpy. This necessitates the maintenance of adequate contact between basal branches of the plant and the soil for effective fruiting. Notwithstanding this, strongly ascending growth forms have become established. The greatest contrast between the two species is in photoperiod sensitivity; the groundnut is day-neutral while the soyabean is a short-day plant with genotypes varying widely in the day length tolerated without inhibition of flowering.

A serious problem in long-term field experiments is that soil is moved from plot to plot as a result of cultivations. The problem is often overlooked and very few attempts have been made to quantify this movement under varying conditions.

A simple, mathematical model for approximating the movement of soil in one horizontal dimension by repeated tillage with one or more tillage implements is presented. The model describes the development with time of a concentration gradient of substance, by means of the solution of a diffusion equation. The model is in agreement with ‘the central limit theorem’, when it is used for the situation where the same cultivations are repeated many times in alternating directions.

The model fitted well to data taken from the literature.

Four plant densities and two genotypes of potato intercropped with sugarcane were examined at Lucknow, India and compared with potato–sugarcane sequential cropping. The best yields and monetary returns were obtained with one sugarcane row to two rows of potato (Kufri Chandramukhi) at 80 × 103 potato plants ha−1; this treatment combination gave virtually a full yield of both potato and cane. Sequential cropping of potato and sugarcane with 80 × 103 potato plants ha−1 in rows 50 cm apart gave better yields than intercropping potato with sugarcane in a 1:1 row arrangement. Kufri Chandramukhi, a short duration variety of potato, was more suitable for intercropping with sugarcane than the long duration Kufri Sindhuri. Growing potato either in sequence or as an intercrop with sugarcane decreased soil bulk density and increased its available nitrogen content.

The development and influence of crop drought stress was examined in winter-planted small-grain cereals under Mediterranean environmental conditions. In two average or wetter than average years crop drought stress in the grain-filling period was greatly exacerbated by the addition of fertilizer (N and P) and this usually resulted in significantly lower kernel weights. Yet this reduction was not so substantial as to seriously affect crop yields. Total grain yields were largest in treatments receiving fertilizer, due to greater spike numbers and numbers of kernels per spike. But heavy nitrogen applications may promote the risk of a large proportion of small or shrivelled grains in a dry year.

Grain yield of maize (Zea mays L.) was decreased by 30, 38 and 34 kg ha−1 for each day by which sowing was delayed after the first planting in March in 1978, 1980 and 1981, respectively. Yield reduction was associated with reduced plant size (plant and ear heights, tassel weight and branch numbers), ear number, ear weight, grain weight, ear length, ear diameter and kernel weight. High yields can thus be obtained by planting with the first rains of the year, especially if a good rate of emergence and seedling vigour under relatively hot, dry soil conditions can be ensured either culturally or genetically.

Simple linear correlations, stepwise multiple regressions and path-coefficient analyses were used to determine the relation between grain yield of maize (Zea mays L.) and weather factors in a three year study involving several planting dates within each year. Maximum and minimum relative humidity, which demonstrated negative relationships with yield, were the most reliable factors, both directly and indirectly, for predicting yield. Temperature (including accumulated heat units), sunshine hours and total and effective rainfall generally showed negligible direct effects on yield. Potential evaporation, which showed positive correlation, had a negative direct influence on grain yield. We conclude that, whenever possible, path analysis should be used as well as correlation and regression analyses in explaining the complex multiple interactions of yield and weather factors in crop production.

A previously presented model for one-dimensional soil movement by repeated tillage is extended for use under the more complex conditions of long-term field experiments, for (i) following the two-dimensional movement (along and across the plots) of accumulating substances (nutrients, disease organisms, etc.) caused by tillage; (ii) determining the smallest acceptable plot size from given conditions, when planning new long-term field experiments and (iii) determining the size of the discard relative to the size of the treated plot.

Diagrams are given from which rough estimations can be made of both the movement of accumulating substances in long-term field experiments and the smallest acceptable plot size, when a computer with the necessary facilities is not available.

The effects of iron pyrites, organic materials and micronutrients on the yield of rice and wheat and on the amelioration of saline sodic soil were studied at Kanpur. The grain yield of the first crop of rice increased from 1.37 to 2.97 t ha−1 following application of pyrites at 50% of the estimated gypsum requirement. The grain yield of succeeding wheat and rice crops also increased significantly with pyrites treatment. These increases were associated with marked decreases in pH, electrical conductivity, exchangeable and soluble sodium content and bulk density of the soil and with an increase in its hydraulic conductivity. The beneficial effect of pyrites was increased with the application of rice straw, Sesbania as green manure and micronutrients, the pyrites plus Sesbania green manure treatment being the most effective.

Examples are taken from three biometrical texts, two unpublished experiments and Yates (1937) to illustrate quantitatively how the lay-out of the plots can affect the estimate of error variance in field experiments and how confounding of treatment and positional effects can arise. In two examples in which the plots did not run the whole length or width of the block, the estimates of error were reduced by one half and one quarter respectively by allowing for sub-blocks within blocks or for rows and columns. In split-plot experiments where the main plots were split breadthwise, the sub-plot error was about the same size as the main-plot error or larger whereas it is expected to be at most one third that of the main plots when these are split lengthwise. Serious bias can arise in the estimate of treatment effects due to confounding with positional effects; this is illustrated.

Checker-board field layouts are used to screen sorghum genotypes for resistance to Striga. A systematic check entry is used to monitor crop growth conditions and methods to analyse the results are assessed.

Apparatus was built to screen sorghum and pearl millet for seedling emergence through a hot soil surface. Seeds were sown in soil in long clay pots arranged in a steel water tank so that the top 7 cm of the pots was above the water level. The soil in the pots was heated with infra-red lamps fitted to a frame above the tank. By adjusting the height of the frame the temperature of the soil could be changed. The system allows emerging plumules to be subjected to high soil temperatures (up to 50°C) but without water stress. Both crops exhibited genotypic differences in emergence.

Intercropped corn and soyabean were grown in a randomized block design (RBD) and a systematic design (SD) in 1981 and 1982 to evaluate total yield, yield complementarity, crop competition and land-use efficiency as affected by corn and soyabean densities. In the RBD, three corn density treatments were applied to corn–corn–soyabean–soyabean, corn–corn–soyabean, and corn–soyabean intercrop row patterns and to monocultures. In the SD, 9 or 10 corn densities were factorially combined with up to 13 soyabean densities in a corn–soyabean row pattern (91 cm spacing). In the RBD, total intercrop yields were similar to corn monoculture yields (60.8 to 62.7 t ha−1 in 1981) in the corn–corn–soyabean and corn–soyabean patterns at high corn densities but were less at low density. The corn–corn–soyabean–soyabean pattern yielded 87% of the corn monoculture at high density in 1982. These results were confirmed in the SD, where total intercrop yields were highly responsive to corn density and poorly responsive to soyabean densities. Both experiments indicated that high corn row densities are needed to maintain yields in corn–soyabean intercrops, possibly higher than the levels studied. Land equivalent ratios were above 1.0 for all but the low density corn–corn–soyabean–soyabean pattern in the RBD. The ability of an intercrop with as many as half the corn rows replaced by soyabean to produce yields similar to corn monoculture indicates a biological complementarity for dry matter production in the mixtures.

It is proposed that two distinct objectives should be recognized in the evaluation of intercropping advantages: (i) a biological objective to determine the increased biological efficiency of intercropping and (ii) a practical objective to determine the advantages that are likely to be obtained by a farmer. The sole crop systems with which intercropping must be compared to satisfy these objectives are defined. Evaluation in relative, absolute, monetary and nutritional units is discussed and some aspects of presenting intercropping data in graphical form are illustrated.

The capacity of sorghum lines to germinate at high soil temperatures (above 40°C) is clearly related to the temperature sensitivity of embryo protein synthesis. The development and optimization of a simple laboratory technique based on this correlation for the rapid large-scale screening of sorghum lines is described. Embryo-containing half seeds are incubated in the presence of 14C-labelled amino acids at different temperatures and the resulting radio-labelled proteins extracted and bound to nitrocellulose discs for counting. The extent to which incorporation of 14C-label into proteins is inhibited by elevated temperatures can be used to predict the likely temperature sensitivity of a given line during germination.

Twenty improved upland rice cultivars were evaluated in yield trials at three locations for three successive years in south-western Nigeria. Highly significant genotype x environment interactions were observed, indicating that some cultivars yielded better in some locations than others. The use of regression analysis adequately identified the cultivars best suited to each environment. S2d values (measures of stability) for the cultivars ranged from 0.0007 to 0.2196, the cultivars with smaller values being more stable at the respective locations. The more adaptable cultivars have high mean yields and a deviation from regression as small as possible. The cultivar best adapted to good environments was code 3 (an improved selection from the cross Lac 23 x IR 528) with a yield of 3.36 t ha−1, while code 9 (a sister selection from the same cross) was best adapted to poor environments. Code 8 (a selection from the cross IR 154/O.S. 6 and 63–83) showed general adaptability. Once grain quality has been evaluated some of these cultivars might therefore replace code 18, the currently recommended upland rice cultivar, in the rain-forest zone.

The effects of intercropping, mulching and staking on growth and yield of three tomato varieties were evaluated in four field trials in 1982 and 1983. Fruit yields of the improved varieties were significantly reduced by intercropping with okra but the yield of a local variety was unaffected. Though the yield of okra when grown with tomato varieties was less than that of a sole crop, the combined yield of the two crops in mixtures was more than the means of the species in monoculture. Relative Yield Totals (RYT) increased to maxima of about 1.42 and 1.39 when the local variety was grown with okra in 1982 and 1983 trials, respectively. Mulching and staking significantly increased vegetative growth, yield and yield components of the tomato plants. Mulched plants grew taller and had more branches and a greater number and weight of fruits than staked plants. Improved varieties responded better to mulching and staking than the local variety. Growing improved varieties of tomato in sole stands with proper mulching and staking is therefore recommended.

The relationship of leaf weight ratio (LWR), leaf area ratio (LAR), stomatal frequency and dry matter (DM) production with yield and with photosynthetic efficiency, as reflected by 14CO2 fixation rate, was studied in fox-tail millet (Setaria italica) for three consecutive years. Several genotypes were identified which combined small leaf area and/or low stomatal numbers per plant with productivity under rain-fed conditions equal to that of genotypes with large stomatal numbers. 14CO2 studies in these genotypes showed that the rate of carbon fixation by the leaves was greater, at any canopy position, than in the corresponding large LA/large DM types. Genotypes with consistently small stomatal numbers per plant, associated with good dry matter accumulation and productivity, may also show small transpirational water loss under rain-fed or drought conditions. It is concluded that such genotypes are desirable for better productivity under these conditions.