The limited fossil fuel reserve available for manufacturing fertilizer nitrogen and the adverse effects of continued use of high fertilizer nitrogen doses on the environment call for a more efficient use of indigenous soil nitrogen. This paper presents several ways of enhancing soil nitrogen use in wetland rice. These involve utilizing nitrogen present in the deeper soil layers, increasing soil nitrogen mineralization rate, decreasing the loss of mineralized nitrogen from the rooting zone, and adjusting rice variety, soil flooding, and transplanting time. To sustain nitrogen fertility and productivity of ricelands, however, the original soil nitrogen levels must be maintained through natural resources like recycled crop residues and enhanced biological N2 fixation. Various ways of stimulating biological nitrogen fixation by both indigenous and exogenous agents are discussed. Since enhancement of soil nitrogen use in rice and maintenance of the original nitrogen level in soil by stimulating biological nitrogen fixation have not been examined together in the field, elaborate field experiments should be conducted to assess the impacts of such combined practices on long-term soil nitrogen fertility and productivity.

This paper describes the construction, usefulness and randomization of several designs for field experiments in which there is more than one set of blocks, namely: (a) row-and-column designs, in which there are two crossing sets of blocks, treatments being applied to the plots formed by their intersections; (b) row-and-column designs in which factors are applied to complete rows or complete columns, that is, criss-cross (or strip-plot) designs; and (c) split-plot designs, in which the plots in a study of one factor are used as blocks in the study of another. All are examples of a wider class of designs, with many ramifications, said to have ‘simple block structure’. It is suggested here that some of the assumptions underlying row-and-column designs are questionable. Some alternative approaches are indicated.

Seeds of wheat (Triticum aestivum L.) were coated with a range of vegetable oils before sowing, in an effort to provide a hydrophobic coating that would only dissolve or wash away in the presence of sufficient water to maintain continued seedling growth. Germination of oil-treated seeds was reduced by 20–50% and time to 50% emergence was increased by two to five days. Treating pesticide-coated seeds with vegetable oil further decreased germination and extended time to emergence. Oil treatment did not decrease water uptake or affect seed respiration. Degree of saturation and linoleic acid:oleic acid ratio were not correlated with the effects of the various oils. The decreased germination and delayed emergence associated with the oil treatment were probably due to physiological rather than physical factors.

Linear covariance analysis of crop yield data using plot stand as the covariate is not satisfactory when the missing plants are eliminated before maturity. This is because the resulting relationship between yield and plot stand is non-linear. In this paper, a transformation of plot yield data to adjust for sub-optimal stands in circumstances involving interplant competition is examined. A biomathematical derivation and interpretation of the proposed model is provided. An experiment involving two varieties of maize (Zea mays) grown at three locations provided the data used to validate the model. Adjusted genotype mean yield values obtained using the proposed exponential transformation were in agreement with observed values.

The tea industry in Malawi is the oldest in Africa but in the past had always produced poor quality tea. A plant improvement programme was therefore started in 1956 at the Tea Research Stations in Mulanje. The improvement in made tea quality was assisted by an investigation into the biochemistry of tea fermentation. Early success was obtained with polyclonal seed varieties and the development of clones from field bushes. These were used in cross breeding to produce new progeny clones which are showing a steady improvement in quality and yield. The need for improved yield was recognised as the programme progressed and appropriate selection procedures adopted. Techniques for improved vegetative propagation for field planting, grafting ofseed bearer trees and composite plants were developed. Liaison with the tea industry has been good and all new plantings in southern Africa since about 1975 have been with improved material. The programme is continuing and future prospects are good.

Phosphorus, followed by potassium, were the plant nutrients that most limited yields during the third decade of a long-term fertilizer experiment on groundnut, maize and cassava conducted on an Ultisol in Ghana. Application of sulphate of ammonia significantly decreased soil pH, thus reducing crop yields. Superphosphate or mulch applied alone consistently increased the yields of all three crops. Application of potash increased the yield of cassava throughout the cropping period, but the potash × mulch interaction depressed yields in two of the three cropping seasons. Use of crop rotation, grass mulch and fertilizer, including small amounts of lime, enabled crop yields and soil fertility to be maintained at a fairly high level under continuous cropping.

Nitrate-nitrogen was analysed over two cropping seasons in 57 farmers' fields in the northern Guinea savanna of Nigeria. Differences between fields were at least five-fold and often ten-fold irrespective of fertilization rates. Average concentrations were highest at the beginning of the rainy season. Loamy soils had a later peak for nitrate release and maintained high concentrations for a longer period after the start of the rains than the more sandy soils. Nitrate-N was not correlated with soil organic carbon or total nitrogen content. Grain yield of maize was closely associated with nitrate-N in the soil but not with organic carbon or total N. Four patterns of nitrate-N release over the season could he differentiated using cluster analysis. Soil texture, soil pH, soil organic carbon, stover management and cropping history contributed most to a differentiation of the four cluster groups in a discriminant model. There was wide variability in the inherent soil-nitrate level and in its importance in explaining differences in yield among farmers' fields. The development of technologies resulting in improved nitrogen availability or better nitrogen utilization should be based on research of inherent soil processes. For the transfer of such technologies, recommendation domains should be defined based on different native patterns of soil-nitrogen availability.

Hesperaloe funifera (Agavaceae) is an experimental fibre crop which produces strong, fine fibres with potential uses in the specialty-paper industry. Hesperaloe's agronomic potential was evaluated in two long-term studies. The plant is a perennial with a rosette habit. Growth occurred by production of new leaves from the primary rosette until Year 3 and from secondary rosettes from Year 3 on. Rates of biomass growth increased over the five years studied. High-density stands (27 000 plants ha-l) produced 190 t fresh weight ha-1 after five years. Plots harvested once, after Year 5, produced more biomass than plots harvested twice, after Years 3 and 5. Increased biomass production can be achieved by improved stand establishment and nutrient management.

A field experiment was conducted from 1989 to 1992 to investigate the effect of sewage sludge applications on wheat yield under different irrigation regimes. The sewage sludge was applied at rates equivalent to 0, 20 or 60 t ha-1 whenever the available soil moisture reached 30 or 70%. The results showed a significant response in grain yield and its nutrient content with no significant differences between rates of sewage sludge applied. Much higher responses were obtained under the wet than under the dry regime.

Permanent pastures were successfully established with mixtures of bulks of crosses of Hordeum vulgare ssp. vulgare with ssp.spontaneum (given the name Mia Milia) and with mixtures of natural crosses involving H. vulgare ssp. agriocrithon (given the name Akhera).These crosses were made to exploit the brittle rachis gene of wild barley in order to develop self-reseeding pasture barley. No re-seeding of the pasture was necessary but a light cultivation to cover the seed in the first year of sowing improved plant establishment. No cultivation was necessary in the following years. Genotypes with a tough rachis were rapidly eliminated from the Milia pastures (F2 bulks) because of their poor seed dispersal. Nitrogen and phosphorus fertilizer increased dry matter yield. Levels of seed dormancy in the wild barley were adequate to provide a safeguard against extreme dry weather conditions and other natural hazards. Forage quality was very high, with a crude protein content of 18% and digestible organic matter of 80% in harvested dry matter. Permanent barley pastures that exploit the brittle-rachis gene of indigenous genetic material have considerable potential for increasing the animal-carrying capacity of permanent pastures and marginal cropped lands.

The manure that is returned to croplands each year is an important resource in the subsistence farming systems of eastern Kenya. Measurements on several farms have shown that the manure being used is of very poor quality. Analyses of soil samples from beneath the bomas (small enclosures) where animals are kept indicate that substantial losses of nutrients occur through leaching; ammonia volatilization and denitrification may also be involved. Current practice is to apply the manure at rates that appear to make poor use of this scarce resource. The findings are discussed in terms of what changes to the system may be feasible to reduce losses of nutrients, and to improve the quality of the manure and the effectiveness of its use.