Grain Production and Dry Matter Partition in Rice (Oryza sativa L.) in Response to Water Deficits during the Whole Grain-Filling Period

Abstract

The grain growth of rice is highly tolerant to water stress throughout the grain-filling period. Despite a large reduction in dry matter production, the grain growth during both the period of active cell division and expansion, and the subsequent period of rapid starch deposition was little affected by water deficits. Reduction in dry matter production due to water stress was almost completely compensated by the increased transfer of reserved assimilates to the grain. But, the plants which experienced drought during the early stage of grain growth and were relieved from the stress thereafter yielded less than the plants which were well watered throughout the grain-filling period. Similar response was observed in wheat, and a deteriorated root function in stressed plants is suggested as a possible factor for this reduced grain yield. It is therefore probable that the yield reduction in rice associated with prolonged drought during the period of grain-filling is a result of the indirect effects of water stress through the function of roots with regard to the production of hormonal substances. However, in view of the highly tolerant nature of the short term grain growth to water deficits, it is more likely that the lack of assimilate supply to the grain is responsible for reduced grain yield in plants under prolonged stress, since there must be an upper limit to the amount of pre-anthesis reserves that can be available for grain growth. This situation has been clearly demonstrated with maize. The objective of this study was to examine the above premise, i.e., the total assimlate supply is the primary factor that controls grain yield in rice droughted during most of the grain-filling period. Rice plants (Oryza sativa L. cv. Nipponbare) were grown in wooden trays (1.05×1.20×0.15 m) containing sandy silt soil. The trays were placed in a vinyl-covered house located in a field and so arranged to form a block of miniature crops that simulates a field density of 95, 200 plants/10a. Part of the block was adequately watered throughout the whole growing period, while water application to the remaining block was restricted after ear emergence so that leaf water potential around midday was maintained at -10 to -15 bars. At intervals during the treatment period, water potential was measured on the flag leaves between 1300 and 1500 hours. During this period, areas of green leaves and grain growth were also monitored. Dry weights of grains, shoots and roots were determined both at the time of ear emergence and at maturity. Daytime water potential in the leaves of stressed plants declined rapidly after the start of treatment, reaching -12 bars 10 days after ear emergence (Fig. 2). Leaf water potentials in the control plants remained at -3 to -6 bars except at the later stage of the growth. Leaf senescence was accelerated in stressed plants, LAI reaching less than 50% of the initial value by the 20th clay after ear emergence, whereas the controls still retained the green leaves of about 60% of the initial value until the full-ripe stage (Fig. 3). Water deficits reduced the dry matter production by approximately 40% (Tables 2 and 3) through the reduction in both the leaf area duration and the average rate of dry matter production (Table 4). Despite this, the grain yield of desiccated plants was only reduce by 16% (Tables 1 and 2), due to the increased assimilate transfer from the shoots to the grain (Tables 2 and 3). Thus the apparent contribution of reserves to grain yield was 34% in the desiccated plants, about three times as high as in the controls (Table 3.). These findings are in agreement with our earlier studies; the grain growth for the first 10 to 20 days after ear emergence was only slightly affected by the water stress (Fig. 4). However, the rate of grain growth declined as the water stress was prolonged. [the rest omitted]