An experiment is described in which the productivity of maize-bean and maize-potato mixtures in alternate rows was compared with pure stands of the component species in four cropping seasons. In seasons with low rainfall, yield from the mixtures fell short of that from pure stands, but in one exceptionally wet season a yield advantage was found for maize-bean mixtures. The poor performance of mixtures in low rainfall seasons appeared to result from large reductions in maize yield attributable to competition from the potato or bean. In the one high-rainfall season, no such reduction in maize yield occurred. An implication of these findings for farming systems with mixed cropping is discussed. Webster and Wilson (1966) concluded that in most experiments on mixed cropping in the tropics, more than one acre of pure stand was required to produce the yield of one acre of the mixed crop and concluded that for the tropical small-scale farmer, there was no advantage to be gained by replacing the traditional practice of mixed cropping. Reports from Africa appearing since 1966 have generally substantiated Webster and Wilson's conclusions. Willey and Osiru (1972) found that yields of mixtures of maize and beans were up to 38% greater than could be achieved from pure stands in an area of Uganda with a short but fairly reliable rainy season and the advantages were even greater with mixtures of beans and sorghum (Osiru and Willey, 1972). These results are important because the yield of all pure stand comparisons was high, thus confounding the opinion that mixed cropping is only of value at low levels of crop management or fertility. In a climate with a long rainy season in N. Nigeria, Andrews (1972) showed total yield advantages of up to 80% from mixtures of sorghum with millet or cowpea. This paper presents the yield results of trials carried out over four seasons with maize-bean and maize-potato mixtures in alternate rows in the Highlands of Kenya. The area has two rainy seasons per year; the long rains usually begin at the end of March, continue until the end of May and are followed by a cool, cloudy, dry period and the short rains begin in late October, continue until the end of the year and are followed by a warm, bright dry season. Both seasons are highly unreliable, both in the quantity of rain and its duration. MATERIALS AND METHODSThe experiment was located at the Kabete Field Station (i° 15'S, 36 0 44'E) of the Faculty of Agriculture, University of Nairobi at an altitude of 1815 m. 12

SUMMARYAn experiment at the Kabete Field Station of the University of Nairobi compared maize–bean mixtures with pure stands of the two crops at three plant densities. Although the mixtures gave an apparent yield advantage over pure stands, this could be explained solely by the increased population pressure in the mixtures. The implication of this finding for other published work with cereal–legume mixtures is discussed, and a distinction is drawn between those sites in East Africa where a conclusive yield advantage from mixed cropping has been found and those where any apparent advantage might be explained by the higher population pressure.

No abstract

In contrast to earlier results, a clear advantage from mixed cropping of maize and beans (Phaseolus vulgaris) was found in the long-rain seasons (March to September) of 1976 and 1977. Pure stand yields were low due to drought in 1976 and excessive rainfall in 1977. These and other results for long-rain seasons in Kenya suggest that mixtures are more efficient where pure stand yield levels are low, but there is little difference between cropping systems where levels are high. Light interception by maize leaves was quite small until late in the life of the bean crop.Earlier results (Fisher, 1977a) showed that maize-bean mixtures at Kabete produced less per unit of land area than their pure stands in one short-rain season, but in two other seasons there was little difference between pure stands and mixtures. An apparent advantage from the mixtures in another season seemed to result from a higher population pressure in the mixture rather than from the intrinsic advantage of mixing (Fisher, 1977b). The trials now reported were primarily designed to provide further information on these seasonal differences.Problems arise in summarizing results from mixed cropping experiments where pure stand yields of the component crops differ as greatly as maize and beans. Following Evans' (i960) reasoning, I previously used the term 'equivalent area' for the area of pure stands required to give the same production as a unit area of the mixture (Fisher, 1977a). Several contributors to a recent symposium (Monyo et al. t 1976) expressed the same concept by the term 'land equivalent ratio', which is inaccurate because the value is actually a sum of two (or more) ratios. A term, numerically equal to both of these variants and widely used by pasture agronomists, is 'relative yield total' (RYT) (Trenbath, 1974), which I now propose to adopt since it expresses most precisely how the index is computed.Because the distribution of RYT on a plot basis tends to be skewed, my previous procedure for statistical analysis of 'equivalent areas' slightly exaggerates but does not invalidate the differences reported (Fisher, 1977a). It seems therefore preferable to calculate the RYT from the experimental mean yields rather than on a plot basis, and informally to assess the significance of a result by reference to the magnitude of standard errors for yielu in the particular experiment.Since maize production is marginal in the short-rain season at Kabete, these studies were confined to the long rains. Replacement designs (Willey and Osiru, t Present address:

The effect of plant density, date of apical bud removal and leaf removal on the growth and yield of single-harvest Brussels sprouts (Brassica oleracea var. gemmifera D.C.) II. Variation in bud sizeSUMMARY A method is suggested for estimating the mean bud size and standard deviation of a sample of sprout buds, separated into size grades by the use of riddles. From the standard deviation a potential size fraction is estimated, being the fraction of the buds occurring in a size grade of 10 mm width if the mean bud size were located at the mid-point of the grade. T.his estimate of bud size uniformity was independent of date of sampling for a period of at least 4 weeks.The potential size fraction was increased by stopping (apical bud removal) and in some cultivars an optimum stopping date was indicated. Increased plant density also increased the potential size fraction, particularly in unstopped crops. Between-plant variation in bud size was significantly greater for the hybrid Thor than the outpollinating Seven Bills but within-plant variation was the more important component of overall bud size variation and was greatly reduced by stopping and by increased plant density in unstopped crops. Mean bud size at any harvest date was increased by early stopping and lower plant density.It is concluded that the date of stopping should be chosen to minimize bud size variation and the plant density varied to spread maturity.