Effect of sowing date on the optimum plant density of winter wheat

Pressure on financial margins in UK wheat production is driving a review of all inputs, and seed represents one of the largest financial inputs in wheat production. The potential savings through exploiting the crop's ability to compensate for reduced population are, therefore, attractive. Field experiments were carried out at ADAS Rosemaund (Herefordshire, UK) in 1996/97, 1997/98 and 1998/99 to investigate the effect of sowing date on dry matter growth and yield responses of winter wheat to reduced plant population. There were three target sowing dates (late-September, mid-October and mid-November), six seed rates (20, 40, 80, 160, 320 and 640 seeds m(-2)) and four varieties (Cadenza, Haven, Soissons and Spark). Grain yield was significantly affected by plant population with a mean reduction from 9.2 to 5.5 t ha(-1) as plant number was reduced from 336 to 13 m(-2). In addition, there was a significant interaction between plant density and sowing date. There was, however, no interaction between variety and plant population in terms of yield, except when lodging affected high plant populations of lodging susceptible varieties. The experiments demonstrated scope for reducing plant populations below the current target of 250-300 plants m(-2); however, the degree of reduction was dependent on sowing date. Over the three years, the average economic optimum plant density was 62 plants m(-2) for late-September, 93 plants m(-2) for mid-October, and 139 plants m(-2) for mid-November sewings. Compensation for reduced population was due to increased shoot number per plant, increased grain number per ear and to a lesser extent increased grain size. Higher economic optimum plant densities at later sowing dates were due to reduced tiller production and hence ear number per plant. The other compensatory mechanisms were unaffected by sowing date.