Some factors causing variation in the yield of individual plants of wheat

Abstract

Seeds of wheat (cv. Warimba) were sown by broadcasting to achieve a crop density of 160-180 plants m-2. The emergence date of each seedling was recorded. For each plant, measurements were made of the distance to neighbouring plants, and the weights and yields of these neighbours. Broadcasting the seed had ensured that there was variation in the distance to neighbours and also variation in the area postulated to be occupied by each plant. These areas varied in shape. Regression analysis was used to evaluate the effect of rapidity of emergence, the area available to a plant and the effect of competition from neighbours on the yield of each plant. Those plants that emerged 5 days after seeding on average had three times the yield of plants that emerged on day 12; however, there was very great variation around the regression, and emergence accounted for only 3.8 % of the variation in yield. Variation in the area available to a plant accounted for 14.3 % of the variation in yield and a competitive index made up of the size and distance away of neighbours accounted for 8.3 %. Each of these factors had a statistically significant effect on final yield. When they were combined in a multiple regression they accounted for only c. 20% of the variation, and 80% remains unexplained. Competition was shown to vary from plant to plant; it was not solely a communal effect reducing plants similarly. The shape of the area postulated as being available to a plant could not be shown to have an effect on its yield. The results are discussed in reference to measures of competition and the difficulties experienced by plant breeders when attempting to select among single plants for yield.