Much has been written and said about genotype-environment (GE) interactions and the particular problems which they pose for plant breeders. It is not the purpose of this article to dwell upon every aspect of this story, but rather to discuss how these problems came to be recognized, to comment upon the various techniques which have been employed in seeking a solution to them and to suggest what developments might lie ahead.

SUMMARYThe method of investigating interactions in two-way tables by the regression analysis introduced by has been applied to data from competition diallel experiments with plant species reported by and Norrington-Davies (1968). Arithmetic and logarithmic scales were used in both experiments and the relative advantages of these are briefly discussed.Significantly high proportions of the interactions between species (row) and associates (column) effects were explained as differences between the linear regressions of individual performance on the associate values. Consequently the performance of the species in competition could largely be specified by three parameters. These were the species mean (v), the regression coefficient (b) and the mean effect of associates (a), which respectively measured the general vigour of the species, its sensitivity to competition and its aggressiveness. The parameters were used to derive formulae which provide descriptive and predictive measurements of the competitive advantage of species in particular combinations, and of the mixture performances relative to the performance of other mixtures or monocultures. The types of competition phenomena which could derive from a situation involving only general competitive abilities were shown to vary greatly and depended on the correlations between the three parameters in the experimental material.The possible types of interactions between associated genotypes (competi. tion, co-operation, antagonism, etc.) can be defined in terms of the general competitive ability parameters, or recognised as specific competitive abilities. It is thus suggested that the regression technique forms a useful approach to the discovery and classification of these effects among competing species.The second experiment (Norrington-Davies, 1968) involved competition between grass species under four different treatments. Common regression lines constructed over all treatments indicated that response to competitive stress was to some extent similar to the response to other kinds of environmental stress. This raised the concept that some aspects of general competitive abilities could be determined from general response to limitation in environmental factors. The plant breeding implications of this are briefly discussed, particularly the possibility of predicting performance under competition from performance as spaced plants.

The role of competition, coexistence and co-evolution in the formation of plant communities is discussed, particularly in relation to the breeding of improved grass/legume mixtures. Competition occurs whenever the demand for a particular resource outstrips supply, with the pressures generated within a species expected to exceed those between species. These pressures must be withstood before populations can coexist within a community. This is accomplished by a process of niche diversification, arising from temporal or spatial differences between the populations, that enables them to draw on resources not readily available to their competitors. Coexistence is crucial to the success of any breeding programme designed to raise the productivity of grass/ legume pastures, because it enables components to adapt not only to the environment which they share, but also to each other. A strategy that improves the "general ecological combining ability" of one or both components by a process of recurrent or reciprocal recurrent unilateral adaptation may prove successful, particularly if existing niche differences are increased thereby. Although both processes may give rise to populations which have apparently coevolved, only those resulting from reciprocal recurrent selection will meet the criteria of specificity and reciprocity.

A diallel arrangement, which incorporated the essential features of the de Wit density replacement series, was employed to study the effects of competition amongst five genotypes of perennial ryegrass (Lolium perenne). Of the five genotypes concerned two were derived from S. 24, two were collected from natural populations in South Wales, while the remaining genotype originated from S. 23. These five genotypes were grown as monocultures and in all ten binary combinations. Within each combination there were three mixture proportions, namely 75:25, 50:50 and 25:75. All mixtures and monocultures were represented by two boxes, one of which was cut at 3-week intervals (frequent cutting) the other being cut at 6-week intervals (infrequent cutting). At each cut all plants within the appropriate mixtures and monocultures were harvested individually and their dry weight recorded.The results obtained over the first 18 weeks of the experiment (i.e. the first three complete growing periods) establish that competition is occurring in nine of the ten binary combinations. Within these nine combinations competition may be classified into one of three groups: first, it may be compensatory, in which the gains and losses incurred by the two components counterbalance; secondly, it may be positive complete complementation, where the advantage gained by the stronger component is such that the mixture performance matches that of the better monoculture, and thirdly, it may be positive over-complementation, where the yield of the better monoculture is surpassed by the mixture. Further tests disclose that a long-leaved S. 24 genotype is the strongest competitor, while a short-leaved, prostrate, indigenous genotype proves to be by far the weakest competitor.Estimates of the equilibrium proportions for each genotype combination suggest that most combinations are expected to become monocultures of the strongest component, with only the combination between the long-leaved indigenous and longleaved S. 23 genotypes remaining a mixture at equilibrium. None of these equilibria coincides with the proportions required to achieve maximum productivity from a particular combination. The results are considered in relation to the known characteristics of these five perennial ryegrass genotypes, while the. wider agronomic implications are also discussed.conditions of competitive stress the individual organisms or components are reacting towards Competition arises where two or more organisms each other's presence in order to secure for themseek a common resource whose supply falls below selves a sufficient share of the environmental their combined demands (Donald, 1963). Hence resources. From an agronomic standpoint such competition may be regarded as a medium through reactions are of considerable importance because which the environment regulates the balance of the impact they can have on the composition between the components of a mixture. Under and yield of herbage mixtures, regardless of the * Present address: Laboratory of Industrial Crops, degree of ...

No abstract