Rice yield and yield components, number of panicle per m2
(Table 4), number of grain per panicle, and ripened grain ratio
(Table 5) were not different between VRF and FPM. Despite
high brown rice recovery and 1,000-grain weight of brown rice
compared to FPM, rice yield at VRF was not higher than FPM.
The N application rate influences the number of tillers formed
and hence the rice grain yield (Wells and Faw 1978). The
amount of N at panicle initiation affects the number of spikelets
per panicle and at full flowering affects 1,000-grain weight and
ripening percentage (De Datta 1981). The number of grain per
panicle, ripened grain ratio, brown rice recovery, and 1,000-
grain weight (Table 5) shows no evidence of nitrogen deficiency
from panicle initiation to grain filling as a result of the reduction
in N top-dressing in VRF at panicle initiation stage. However,
higher brown rice recovery and 1,000-grain weight of brown
rice explains better grain filling condition and could explain the