Responses of crop yield to water stress are classified on the basis of physiological considerations. Crop growth is sensitive to water stress, particularly because of the physiological effects of stress on the production and maintenance of photosynthetic tissue. Crop yield, depending on the nature of the harvested organ and the origin of its constituents, may be more or less sensitive to water stress than crop growth. In applying this information to irrigation management, there is a need for simple quantitative measures of plant water status. The problem of avoiding water stress during critical ontogenetic stages and, conversely, the possibility of using moderate water stress at certain times to improve yield and the efficiency of water usage, are discussed.

A survey of citrus groves was conducted by analysing leaf and soil samples from the main citrus growing areas of the island. The leaf data, evaluated by comparing them with the California leaf analysis standards for assessing the nutrient status of citrus leaves, indicated that the California standards are of limited value in attempting to assess the nutritional status of citrus in Trinidad, particularly with reference to the elements zinc and magnesium. The possibility of using soil analyses as an index to the mineral nutrition of citrus is worthy of consideration. Although symptoms of zinc and magnesium deficiencies had often been observed, these elements had not been given the attention they deserved in the regular fertilizer programmes.

A NPK factorial experiment was conducted during two successive cropping seasons to study direct and residual effects of various treatments on the flower initiation and yield of soybeans. Plants were slower to reach the maximum flowering stage in the year of fertilizer application than in the following year. During the first year potassium in association with phosphorus delayed the flower forming process, whereas potassium alone hastened flowering in the second year unless associated with a high dosage of nitrogen. All the treatments gave far greater yields in the first season, suggesting that fertilizers should be applied during every cropping season. Nitrogen and phosphorus at 20 pounds of N2 and P2O5, per acre gave the biggest yields, but potassium failed to show any response. The residual responses to potassium and phosphorus during the second year were significant but not to nitrogen.

Lagière's B9 from Allen 51–296 (Gossypium hirsutum) was found to be non-homologous with Knight's B genes B1–B7, B9 and Bherb. To differentiate between this gene and Knight's B9 from G. herbaceum the symbols have been changed to B9L and B9K respectively. It was impossible to test Knight's B10 from Kufra Oasis G. punctatum and Lagière's B10 from Allen 51–296, but to avoid confusion it is proposed that these genes should be known as B10K and B10L.

Experiments involving N, P, and K fertilization, and height and frequency of cutting, were carried out on a mixed pasture of giant stargrass (Cynodon plectostachyus) and centro (Centrosema pubescens) at the University of Ibadan during the growing seasons 1951–61. For the first cut after fertilization yield responses of 17 and 24 per cent resulted from applications of 50 and 100 lb N per acre respectively in various experiments but the effect was transient and recovery of fertilizer nitrogen probably did not exceed 30 per cent. Nitrogen fertilization of sown pastures is unlikely to be an economic practice in Nigeria in the forseeable future. Yield responses of 30 per cent resulted from applications of 2½ and 5 cwt superphosphate per acre with a marked residual effect two years after application. Significantly higher concentrations of extractable inorganic soil P at the higher P fertilization rate after 2 years were largely a reflection of the levels of the iron phosphate fraction, but similar trends were evident in the aluminium and calcium phosphate fractions. No differences were found in total, organic, or residual P. The proportion of centro in one 1959 cut for the 0 and 5 cwt rates of superphosphate were 46 and 38 per cent whereas N had no effect, the reverse of the response normally encountered in a temperate mixed pasture. No response to K was obtained. Increased frequency and height of cutting both resulted in decreased yields of dry matter but cutting at 2–4 inches at 8-weekly intervals appears to be the most satisfactory method of managing this pasture for hay production, with yields of approximately 2¼ tons dry matter per acre.

After a review of previous work, experiments on cocoa seed storage at the West African Cocoa Research Institute are described. Although no definite recommendations are made as to the optimum methods of storing cocoa seed, the effects of temperature, fungicides, pre-storage in the pod after harvest, air-drying, and some minor factors are described and discussed.

The results are reported of three field experiments conducted from 1955 to 1963 to investigate the effects of sunn green manuring in the sugar cane rotation. In a three-year rotation with sugar cane and wheat, green manuring twice was not superior to green manuring once, either in terms of sugar cane yield or the nitrogen and organic carbon status of the soil. Two experiments showed that the yield increase of cane was proportional to the weight of the green manure crop, irrespective of how the latter was disposed of, suggesting that the only factor with which yield increase was correlated was the weight of roots of sunn which were incorporated into the soil. The green manure effect is resolved into two components namely ‘green matter effect’ and ‘legume effect’, both of which are additive and nearly equal.

A detailed soil map of the Island of Barbados has recently been completed, and a number of factors make this area an interesting ‘laboratory‘ in which to investigate the uses to which such a map may be put. The map has already proved to be of considerable value in extension work by providing a more rational basis for advice about cultivations, growth and ripening patterns, crop diversification and the organization of simple dispersed-block trials. Because of the importance of available soil moisture and drainage, techniques have been developed to investigate these aspects in relation to the different soil types. The most rewarding feature, however, has been the enthusiastic response of plantation managers to the soil map as a basis for the exchange of information, the communication of the results of experimental work, the reorganization of field boundaries, and the introduction of new crops.

Groundnuts (Arachis hypogaea L.) have been found to be one of the most promising additional cash crops for diversifying Sudan agriculture. Three hundred and thirteen agronomically distinct forms have been separated from 171 varietal samples, and 121 short-maturing upright bunch types and 49 long-maturing spreading bunch types have been tested in a series of trials over several years in four climatic regions. Among upright bunch varieties, PJ121070 and Large Natal 142 appeared to be the best selections as oil and confectionery types respectively, and among spreading bunch varieties, MH.372, Beladi Bunch and NJ37 for oil, confectionery kernels and unshelled nuts respectively. A hybridization programme was started in 1959 to improve the existing varieties of groundnuts and to produce non-germinating upright bunch varieties for growing in the heavy clay soils. Inheritance studies indicated the dominance of an alternate pattern of branching, prostrate plant habit and long maturity. Dominance of testa colours was in the order of purple–red–russet brown–light brown–white. Recessive albino seedlings were observed only in crosses involving different branching pattern. There appeared to be no association between the various characters studied except maturity period, which was related to branching pattern and growth habit.

In a trial carried out in 1958, aldrin applied as a fog with a swingfog machine at concentrations of 10 and 20 per cent in gas oil was more effective than dieldrin or chlordane at similar concentrations for control of Atta cephalotes. Two applications of 10 and 20 per cent benzene hexachloride and toxaphene failed to give complete control. In a subsequent trial, aldrin in concentrations ranging from 2·5 to 8·0 per cent was also found to be effective against A. cephalotes but against A. laevigata, aldrin and dieldrin were more effective than chlordane at concentrations of 5·0 and 2·5 per cent. In campaigns carried out from 1959 to 1963, 5000 nests were destroyed by fogging with 2·5 per cent aldrin in gas oil.