The effects of various intercropping patterns of maize and melon on soil temperature and moisture content and crop growth and yield were compared in field trials over a two year period. Intercropping raised soil temperature in the upper 10 cm at 0600 h and reduced it at 1000, 1400 and 1800 h, and also increased soil moisture content by about 30%, compared with maize sole cropping. Melon biomass and seed yields were reduced by intercropping but maize growth characters were unaffected. However, the reduction in melon yields was less when grown with maize in widely spaced rows. Maize intercropped with melon in alternate pairs of rows (2:2) gave the best land equivalent ratio (LER).

The effects of phosphorus fertilizer and the insecticide carbofuran on the growth and N2-fixation of Azolla pinnata and on the growth, grain yield and nitrogen uptake of intercropped rice were examined in a wet and a dry season. Treatment with phosphorus or carbofuran increased the biomass of Azolla and the amount of nitrogen fixed (nitrogen yield) in both seasons, but the response was much better in the dry season. Azolla inoculation at 1.0 t ha−1 resulted in a greater bio mass and nitrogen yield than inoculation at 0.5 t ha−1. In the dry season, a combination of phosphorus and carbofuran enhanced the growth and N2-fixation of Azolla more than either treatment alone. Carbofuran treatment slowed the rate of decomposition of Azolla, particularly in the dry season. The plant height, leaf area index and dry matter production of rice at flowering time were increased in the plots treated with phosphorus or carbofuran in the wet season and these treatments increased rice grain yield and nitrogen uptake in both the wet and dry seasons.

The Agricultural Technology Improvement Project (ATIP) has been conducting on-farm research in Botswana with the goal of identifying improved arable production technologies. In an effort to increase the role of farmers in technology design and assessment, ATIP set up farmer groups that meet on a regular basis to discuss farming problems and on-farm trials. This article reviews and evaluates ATIP's experiences with farmer groups. A typology of technology development groups is proposed and group management issues are discussed.

Experiments on okra to determine the economic loss caused by insect damage to the leaves or by harvesting young leaves as a vegetable are described. Leaf harvests at 1-weekly intervals yielded 30-50% more fresh leaf than harvests at 2 or 3-weekly intervals. Removing a quarter of each leaf or single leaves from the upper or lower parts of the plant at 3-weekly intervals did not affect marketable fruit yield, but yield was reduced by about 30% when one young leaf was harvested at 1 or 2-weekly intervals or when two or three leaves were harvested at S-weekly intervals. Removing more than one quarter of each leaf or two or three fully expanded leaves from the lower three-quarters of the plant at 3-weekly intervals also caused up to 40% yield loss.Harvesting single leaves at S-weekly intervals is therefore recommended if okra is to be grown for both leaves and pods. The plant can tolerate at least 25% defoliation, as can happen with insect damage, before any yield loss occurs.

F. O. Olasantan: Efecto de la deshojadura sobre el crecimiento y rendimiento del gombo (Abelmoschus esculentus,) y su importancia para modelos de cosecha de hojas y dano causado porplagas.

Genotypes of mung bean commence flowering at very different times depending on sowing date and location, but relatively little is known about the modulation of flowering by environmental factors. Previous and frequently cited conclusions have been that: (a) genotype, photo-period and temperature all interact to determine relative earliness to flower, and that (b) the genetic control of these photothermal responses is seemingly complex. However, our reanalyses of original data in terms of rates of progress towards flowering, 1/f, rather than the traditional approach based on days from sowing to flowering, f, show that the photothermal modulation of flowering in mung bean can be described by a series of simple, linear models, and that interaction terms involving photoperiod and temperature are often insignificant. The merits and implications of this alternative analysis and interpretation of original data are discussed.

Root and tuber crops make a major contribution to the food supply in many developing countries. Most of the relatively limited research effort on tropical root crops has focussed on improved production technologies, especially with regard to genetic improvement. Interest in post-harvest technologies and crop use has increased recently with the growing recognition of their importance in the marketing of these perishable crops. This paper reviews recent advances in storage research for three of the principal root crops: cassava (Manihot esculenta C), yam (Dioscorea spp.) and the edible aroids (Colocasia spp.).

R. D. Cooke, J. E. Rickard y A. K. Thompson: El almacenamiento de tubérculos tropicales - mandioca, ôame y aráceas comestibles.

Four genotypes of groundnut grown with limited irrigation in a medium depth Alfisol in Central India transpired similar total amounts of water (220–226 mm) over the season, but produced different amounts of shoot dry matter (390–490 g m−2). The extraction front of Kadiri 3 moved most rapidly down the soil profile which may have enabled it to maintain the fastest rates of transpiration when soil water depletion was greatest. Tap root extension rates of Kadiri 3 in the first 32 days after sowing were also the fastest. NC Ac 17090 was more efficient than the other genotypes in extracting water immediately after irrigation from the upper 40 cm of the soil, but this had little value in determining the pattern of water availability in this experiment. Differences in the water extraction characteristics of these genotypes explain little of the variation in dry matter:water ratio, and do not account for the major variation in harvest index associated with drought.

Most commonly, on-farm research has sought to validate and demonstrate technologies developed under controlled experimental conditions. It is argued here that on-farm research has a role in all stages of technology development, and that methodologies and objectives transplanted from the research station are inappropriate, particularly given the restrictions they impose on farmers' capacity to experiment with new technology. Drawing on experience with alley farming in W. Africa, it is shown how the farm and farm family can be incorporated fully throughout the process of technology development. The example is used to clarify the role of researchers in participatory technology development.

Line sowing of jute has been advocated by some workers for the past 40 years as being more productive than the traditional broadcast method. But most cultivators still prefer broadcast sowing. A comparative study employing identical management practices showed that there is no significant difference in fibre yield per unit area between the two sowing methods. Under similar population pressures self-thinning is greater in line sown than in broadcast crops. Thus, provided similar seed rates and cultural conditions are maintained, farmers may use either sowing method.

P. Palit, A. C Bhattacharyya y B. K. Samanta: Re-evaluación de los efectos de la siembra al voleo y en lineas sobre el rendimiento defibra del yute.

Four genotypes of groundnut grown with limited irrigation during the post-rainy season in Central India produced similar amounts of dry matter per unit of intercepted solar radiation (e) before pod-filling, although different e values were observed during pod-filling. The relation between cumulative transpiration and intercepted radiation was similar for all genotypes. When drought became severe, fractional radiation interception (f) was reduced by folding of leaves, with little decrease in leaf area (L). The ratio f/√L was used as an index of the degree of leaf folding and was correlated with leaf water potential. The degree of folding varied with genotype and may have contributed to the observed differences in e and the dry matter:water ratio (q). The genotype EC76446(292) had the smallest q and largest f/√L ratio (the poorest radiation avoidance), while Kadiri 3 had the largest q and smallest value of f/√L.

Good data handling is a prerequisite for good statistical analysis and interpretation. This paper begins by distinguishing types of numerical data. Care for original data and the risks involved in any process of copying data are emphasized. After discussion of the entry of data into a computer file, problems of scrutiny and checking are discussed at length. An attempt is made to suggest what action should follow when such scrutiny discloses anomalies in some values; if explanations can be found, or if clear evidence indicates corrections that are needed, no problem arises, but otherwise various subjective responses need consideration. Advice is offered on the numbers of digits to be retained in computations and reports. A final section presents ideas on the filing of data and analyses from an experiment.

A long-term field study was conducted from 1979 to 1985 in the Karnal district of Haryana, India to evaluate the effect of three rates of gypsum (0, 5.2 and 10.4 kg ha−1) on the performance of five grasses (Diplachne fusca, Panicum laevifolium, P. antidotale, Chloris gayana and Cynodon maritimus) on an extremely sodic soil with an initial pH of 10.6 and an exchangeable sodium percentage (ESP) of 94. Diplachne, Chloris and Cynodon usually gave their maximum yield in the second year of planting, P. antidotale in the third year and P. laevifolium in the fourth year. The results showed that Chloris can be recommended for such situations as it maintains a higher productivity over a longer period than the other grasses. Alternatively P. antidotale and P. laevifolium can be planted in association with Diplachne to provide a steady forage supply for a long time. Increasing doses of gypsum resulted in decreases in soil pH, exchangeable sodium and ESP and increases in calcium, magnesium, organic carbon and the infiltration rate of soil.

Maize yields where the nitrogen source was sulphur-coated urea combined with ordinary urea were significantly greater than those where an equivalent amount of ordinary urea was applied at identical times. Urea blended with neem cake was less effective.

K. P. Prabhakaran Nair y P. B. Sharma: Eficacia comparada de la urea normal y revestida y urea tratada con inhibidores de la nitrificación como fuente de nitrógeno para el maiz

The pressure chamber method for measuring leaf water potential in peanuts was compared with the thermocouple psychrometer technique. The pressure chamber over-estimated leaf water potential by an average of 0.4 MPa over the range −0.5 to −5.0 MPa. It is suggested that the pressure chamber technique may be appropriate in comparative studies of peanut water stress. However, where absolute measurements are required, as in the calculation of leaf turgor potential, either a correction factor should be applied or, preferably, the thermocouple psychrometer technique used. If no account is taken of the discrepancy between techniques, erroneously large negative turgor potentials will result.

The tuber yield of potatoes in a field experiment conducted at Jalandhar between 1981 and 1984 was greater following green manuring with either Crotalaria juncea or Sesbania aculeata than after fodder crops of Zea mays or Pennisetum typhoides. The residual effect of phosphorus applied to potato on subsequent wheat crops varied with the previous green manuring and fodder cropping treatments but was always sufficient for the wheat when applied at an economic rate; the lack of an application of phosphorus fertilizer to the wheat did not result in a decline in the availability of soil phosphorus. The use of C. juncea as a green manure increased the availability of phosphorus but decreased that of potassium. Pennisetum typhoides also depleted soil potassium reserves. Sesbania aculeata green manure improved the organic carbon status of soil whereas P. typhoides depleted it.

Rates of leaf development in four groundnut genotypes were found to depend primarily on temperature, although there was a linear relation between the thermal time to produce one leaf and mean leaf water potential below −0.6 MPa. Flower, peg and pod production was analysed in relation to leaf number, since this integrated effects of temperature and water status.

When water was not limited, the ratio of pod number: peg number was about 0.8 for all four genotypes. During drought only one genotype (TMV 2) maintained this value, while for the other genotypes the ratio was approximately 0.15. Genotypes also showed different patterns of development during recovery from drought. TMV 2 maintained relatively high values of partitioning factor throughout the pod-filling period, resulting in the largest harvest index and pod yield. Kadiri 3 achieved the second largest harvest index and pod yield by maintaining production of pegs during drought and only forming pods when stress was relieved.

The rate of regrowth in realtion to total dry matter yield is important for smallholders fodder crop production. A simple regression technique is used to compare the rates of regrowth of eight perennial fodder crops cut at the same time. A brief interpretation and discussion of results from data illustrative of the model' use in given.

The measurement of leaf water and osmotic potential using thermocouple psychrometry was investigated in field grown stands of sunflower and lucerne. For the chamber size and tissue dimensions used, vapour pressure equilibrium was achieved after 4 to 6 h for leaf water and osmotic potential samples. In lucerne, as the time delay between excising the leaf and sealing it in the chamber was increased from 30 s to 8 min, water apparently evaporated from the leaf sample and leaf water and osmotic potential markedly decreased. For routine psychrometer determinations of leaf water and osmotic potential in lucerne (and presumably other species) leaf samples should therefore be sealed in the chambers as quickly as possible to minimize rapid water loss and associated errors in water potential determination. Vapour pressure equilibration for sunflower capitulum and root water and osmotic potential samples was achieved only after 6 and 24 h, respectively.

G. C. Wright y D. M. Whitfield: Medición de los potentiates totales y osmóticos en tejidos de alfalfa y girasol usando psicrómetros termopar.

The contents of the volatile flavour compounds and the flavour index of black teas vary with localities. The magnitude and order of the variations are clonally dependent, the relative performance of clones varying with location. For a comprehensive breeding programme to improve black tea quality, it is suggested that tea clones should be evaluated in the areas of potential release before they are made available to tea growers.

In spite of large year-to-year variations in clove yields, production in the islands of Zanzibar and Pemba is often in synchrony and the annual harvests of the two islands are correlated with similar variables. Clove harvests are positively correlated with rainfall in October to February two years before the harvest and negatively correlated with the harvest in the previous year and rainfall around July, and October to December of the year before the harvest. It is likely that rainfall in these periods affects vegetative growth of the clove tree and clove bud differentiation.