Quantitative analysis of pod formation, seed set and seed filling in winter oilseed rape ( Brassica napus L.) under field conditions

Pod formation, seed set and seed filling of winter oilseed rape ( Brassica napus L.) were studied in two field experiments with the variety Jet Neuf. Variations in total dry matter production (source size) and pod density (sink size) were established by varying seed rate, plant density, sowing date and the application of shading. od formation and seed set per pod were recorded on different branches, at various heights in the canopy and on pods initiated at different times during flowering. Pod density and seed density appear to be fully determined just after the end of flowering. The potential number of pods (i.e. the number of flower buds) did not appear to be a limiting factor, as at all positions in the canopy it was much higher than the number attained. Averaged over the whole canopy, pod set varied between 43 and 58%. Potential and actual pod density were linearly related to cumulative dry matter production of the crop until the onset of flowering and until the end of flowering, respectively, i.e. to total assimilate availability over that period. The number of seeds per pod did not vary with height in the canopy. Primary branch number and the time of pod initiation during flowering had some effect on the number of seeds per pod; however, this had little influence on the overall mean number of seeds per pod of the whole canopy and did not affect the differences among treatments. A linear relationship was found between seed density and dry matter production of the crop during flowering. Individual seed weight followed a sigmoid pattern when plotted against the temperature sum from the onset of flowering per flower bud until maturity. The pod hulls were the only competing sink during early seed fill and remobilization of reserve carbohydrates contributed at most 12% to final seed yield.