Grain Yield Response to Planting Density in Forage Rice with a Large Number of Spikelets

Because forage rice (Oryza sativa L.) must be cost-competitive with other forage crops, high grain yield is essential. The objective of the present study was to determine the effects of planting density (11, 25, or 100 hills m(-2)) and N application method (the 79-21N method, with 79% of the N applied before the panicle initiation stage and 21% after, or the 36-64N method, 36% before and 64% after) on grain yield and its components in the cultivar Mizuhochikara, which produces an unusually large number of spikelets per square meter. Grain yield was not different between N application methods but increased from 799 g m(-2) to 870 g m(-2) with increasing planting density. The number of spikelets per square meter was not strongly influenced by the planting densities. The percentage of filled spikelets and the 1000-grain weight was higher at 100 and 25 hills m(-2) than at 11 hills m(-2). The amount of nonstructural carbohydrate (NSC) in the leaf sheaths plus stems per square meter at the full heading stage was larger at 100 and 25 hills m(-2) than at 11 hills m(-2). The percentages of both primary and secondary spikelets increased with increasing planting density, whereas that of tertiary spikelets decreased. Grain weight per spikelet was generally higher at primary and secondary spikelets than at tertiary spikelets. At each position in the panicle, grain weight per spikelet was heavier at 100 and 25 hills m(-2) than at 11 hills m(-2). Thus, dense planting may increase the grain yield, the percentage of filled spikelets, and the 1000-grain weight by increasing NSC at full heading stage and decreasing the number of tertiary spikelets even at a large number of spikelets per square meter in Mizuhochikara.