| Rice panicle, which is the final destination of about half photosynthetic products, reveals the quality of rice population and indicates its grain yield directly. In the progress of rice cultivars improvement and grain yields increasement, total spikelets per unit area growing caused by spikelets per panicle expansion plays a key role. So spikelets per panicle expansion and panicle weight growth are regarded as important targets in rice breeding and cultivation for further increasement of grain yield. Large panicle rice and its high-yielding culture technique promoted the grain yield per unit area in Sichuan Basin, but its grain yield potential and nitrogen use efficiency were restricted by the specific ecological conditions. In 2013 and 2014,2 large panicle and 2 medium panicle hybrid rices which were different in panicle size were chosen as trial cultivars to create a range of rice population with varied grain yields. Several agronomic measures, like nitrogen (N) application mode, planting geometries, ecological sites, panicle N application modes and sowing date were used in this progress. Features of grain yield formation and N accumulation and utilization of these rice population were studied to supply theoretical and practical directions for the synchronous enhancement of grain yield and N use efficiency of rice production in Sichuan Basin. The major results were as follows:1. The grain yields of large panicle hybrid rice were significantly higher than those of medium panicle hybrid rice, the average grain yield of high-yielding treatments of the former was 17.02% higher than that of the latter. However, the excellent grain yield performance of large panicle hybrid rice must be based on an outstanding rice population structure, while for the medium panicle hybrid rice, the grain yield could maintain stable though its population structure varied on a greater scale. Obtaining larger panicles with a certain number of effective panicles is crucial for large panicle hybrid rice to acquire high grain yields, an excessive peak tillering possibly resulted in more effective panicles, but the panicle size would decrease sharply, and the high grain yield potential could not fully develop. Compared with large panicle hybrid rice, gaining a large number of effective panicles was a key procedure for medium panicle hybrid rice to become high yielders, as its effective panicle rate was less than large panicle hybrid rice, the peak tillering number of medium panicle hybrid rice was significant more than that of large panicle hybrid rice.2. The spikelets per panicle difference between large and medium panicle hybrid rice was mainly decided by spikelets differentiation on secondary branches. large panicle hybrid rice, if with an excessive peak tillering, would be poor in showing its large panicle superiority, and there would be little difference between large and medium panicle hybrid rice in differentiated and survived spikelets on primary and secondary branches. However, if with a proper peak tillering (about 240×106ha-1), large panicle hybrid rice would perform better in showing its large panicle superiority, and its differentiated and survived spikelets on secondary branches would be significantly more than those of medium panicle hybrid rice. At santai site, whose ecological conditions were favorable for rice, both large and medium panicle hybrid rice developed tertiary branches and spikelets on them consequently, but the difference between them for the spikelets on tertiary branches were not as much as those on secondary branches.3. There was not an explicit relationship between large and medium panicle hybrid rice for the ratio of non-structural carbohydrates (NSC) content in stem and sheath at heading stage to total spikelets, but the ratio of NSC output during grain filling stage to total spikelet of large panicle hybrid rice was significantly more than that of medium panicle hybrid rice, implying an fluent output of photosynthate from stem and sheath to grains. On the contrary, the ratio of NSC reservation at maturity stage to total spikelets of large panicle hybrid rice was significantly smaller than that of medium panicle hybrid rice, which was probably one of the physiological bases for the fluent transport of photosynthate for large panicle hybrid rice.4. For large panicle hybrid rice, a decreased rice population quality resulted from an excessive peak tillering at elongation stage would weaken photo synthetic capacity after anthesis, but it could still show robust grain filling capacity with the help of photosynthate transport accumulated before anthesis, therefore large panicle hybrid rice could show the grain yield advantage to medium panicle hybrid rice though their sink size (total spikelets) were similar. With a proper peak tillering (about 240×104ha-1) and the large panicle advantage were fully developed, the spikelet-root activity advantage of large panicle hybrid rice would enhance photosynthetic capacity after anthesis effectively, which helped to produce similar or larger 1000-grain weight as compared to medium panicle hybrid rice, though the total spikelets of large panicle hybrid rice were much greater.5. Different panicle weight type of hybrid rice absorbed N dissimilarly.15N tracer showed that large panicle hybrid rice was in great demand of panicle N, while medium panicle hybrid rice absorbed tillering N more efficiently. Hence, the optimal panicle N ratio of large panicle hybrid rice was larger than that of medium panicle hybrid rice, the optimal panicle N ratio of the former was 32.46%, whereas it was 26.26% for the latter. Under triangle planting geometry (TPG), the effective panicle rates and effective panicles of large and medium panicle hybrid rice were higher than those under other planting geometries (wide and narrow row planting geometry and seedling-throwing planting geometry), besides, spikelets on primary branches, spikelets per panicle and total spikelets all showed similar trends. In addition to the great advantage in gaining a large sink size, rice under TPG could also fill grains efficiently, so rice under TPG produced significantly more grains than those under the other planting geometries. Rice transplanted with double seedlings per hole could obtain more effective panicles, but its disadvantage of less spikelets per panicle counteracted the advantage of more effective panicles, and their sink sizes and grain yields had no significant differences at last. For large panicle hybrid rice, the same amount of spikelet-promoting N and spikelet-developing N was in favor of developing a great sink size, while for medium panicle size, a relative smaller panicle N was proper, if panicle N ratio was fixed (30%), more spikelet-promoting N should be applied as spikelet-developing N or grain N. An delayed sowing date would shorten the vegetative growth duration of large panicle hybrid rice, resulting in a small rice population at elongation stage. A small rice population was poor in photosynthetic capacity in booting stage and could not meet the demands of developing large panicles, and the total spikelets were relative less therefore. As for medium panicle hybrid rice, a delayed sowing date showed little influence, the sink size almost maintained the same, but rice population transplanted with younger seedlings embodied a great advantage in photosynthetic capacity after anthesis, which was beneficial for producing more grains.6. In this research,11300kg ha-1 and 9800kg ha-1 was defined as high-yielding floors of large and medium panicle hybrid rice respectively, which were 45% and 25% more than the average grain yield of rice in Sichuan. For large panicle hybrid rice, the grain yield range of high-yielding treatments was 11348-12152kg ha-1. Peak tillering number, effective panicle rate, and heading uniformity were 222.69-253.88×104ha-1, 73.19%-90.61% and 81.27%-94.73%, respectively. Effective panicles, panicle weight, total spikelets and dry matter accumulation at maturity stage were 177.00-202.20×104ha-1, 5.86-6.66g,421.06-480.78×10-1, and 18320-19660kg ha-1, respectively. For medium panicle hybrid rice, the grain yield range of high-yielding treatments was 9802-10362kg ha-1. Peak tillering number, effective panicle rate, and heading uniformity were 250.73-289.18×104ha-1,68.29%-81.94% and 69.02%-88.70%, respectively. Effective panicles, panicle weight, total spikelets and dry matter accumulation at maturity stage were 177.95-211.92×104ha-14.75-5.51g,347.58-418.77×106ha-1, and 16880-17829kg ha-1, respectively. Among high-yielding treatments, there were no sharp contradictions between effective panicles and spikelets per panicle, and peak tillering, effective panicle rate, effective panicles, total spikelets, heading uniformity and panicle weight were all in large variation ranges, so in high-yielding treatments, effective panicles and panicle weight could complement each other to some extent. However, the panicle weights of high-yielding treatments in this research have remained at high levels, so we should deeply ponder over that if the further enhancement of grain yields could rely on the panicle weight growth which was closely related to spikelets per panicle development.7. Peak tillering-heading uniformity curve and peak tillering-panicle weight curve all taken on parabola. For the relationship between peak tillering number and heading uniformity, the formulas were y=-0.00317*X2+1.20052*X-18.63679 (R2=0.91922**) for large panicle hybrid rice and y=-0.00314*X2+1.20883*X-20.39413 (R2=0.89107**) for medium panicle hybrid rice. For the relationship between peak tillering number and panicle weight, the formulas were y=-0.00317*X2+1.20052*X-18.63679 (R2=0.91922**) for large panicle hybrid rice and y=-0.00314*X2+1.20883*X-20.39413 (R2=0.89107**) for medium panicle hybrid rice. In this research,87% and 81% were the mean values of heading uniformity for large and medium panicle hybrid rice respectively in high-yielding treatments, and 80% and 75% were the mean values of effective panicle rate. With these indexes and the aforesaid formulas, the grain yields of high-yielding treatments could be calculated accurately, so heading uniformity could be regarded as an index for optimizing rice population.8. An efficient dry matter production efficiency of accumulated nitrogen (PE) at elongation stage was favorable for large panicle hybrid rice to establish a sound rice population containing enough biomass, preventing the inefficient photosynthetic capability in booting stage and the failure of developing large panicles. N content of rice plants (the reciprocal value of PE) at heading stage had significant positive correlation with total spikelets, indicating that N content of rice plants at heading stage should be as low as possible. PE at maturity stage was positively correlated with the grain yield, the formulas were y=88.2021*X+1958.00532 (R2=0.74522**) for large panicle hybrid rice and y=54.95833*X+4190.33439 (R2=053509**) for medium panicle hybrid rice, a robust PE at maturity stage was obviously crucial for obtaining high grain yields. |
PDF Full Text Request