The effect of population density on floret initiation, development and abortion in winter wheat

There is little information in the literature concerning the physiological basis of the relationship between plant population density and kernel number in winter wheat (Triticum aestivum L.). Thus, two experiments were conducted to evaluate this relationship. Expt 1, involving three population densities, was carried out near Taian, China in 1982 and in Expt 2, two densities were evaluated near Lexington, Kentucky in 1986.Plants were sampled every 2 d in the spring, main stem spikes were dissected and florets were scored according to a 10-stage scale of development. The rate of primordia initiation increased as density increased until the point at which primordia numbers were equal in all treatments. After this point, an increase in density reduced the primordia initiation rate. In both experiments increasing density reduced the total number of floret primordia initiated and the number of kernels per spike. In Expt 1 the effect of density on kernel number per spike was accounted for approximately equally by the effect of density on number of primordia initiated and floret abortion. In Expt 2, however, floret abortion was influenced much less by density and accounted for only 7 % of the variation in kernel number per spike. The key result was that the effect of density was determined early in floral development. The data suggest that yield losses at high densities may be determined too early in development to be offset by N applications at the terminal spikelet stage.