LAUSR

Abstract Understanding the differences in yield traits of rice among pothole seedling of mechanical transplanting (PSMT), carpet seedling of mechanical transplanting (CSMT) and mechanical direct seeding (MDS) is of great importance not only for rice scientists but also for rice farmers to develop a high-yield production system under mechanical conditions in a rice-wheat rotation system. However, such traits are yet to be studied among rice varieties of japonica-indica hybrid rice (JIHR), japonica conventional rice (JCR) and indica hybrid rice (IHR). Field experiments were conducted in 2014 and 2015, where six cultivars of the three rice types JIHR, JCR and IHR were grown individually with PSMT, CSMT and MDS methods, under respective managements for each method to achieve the maximum attainable yield. Results showed that (i) the PSMT significantly increased grain yield of JIHR by 22.0 and 7.1%, of JCR by 15.6 and 3.7% and of IHR by 22.5 and 7.4%, compared to MDS and CSMT on average across the two years, respectively. The highest yield was produced by the combination of JIHR and PSMT; (ii) high yield under PSMT was mainly attributed to large sink capacity and high-efficient dry matter accumulation. With sufficient panicles per hectare, the increase of spikelet number per panicle, especially the increase in spikelet number of the secondary rachis-branches was determined to be the optimal approach for developing a large sink capacity for rice under PSMT. The optimal tillers development, large leaf area index at heading stage, and high leaf area duration, crop growth rate and net assimilation rate during grain-filling phase could be the cause of sufficient dry matter accumulation for rice under PSMT; (iii) moreover, the PSMT favored plant growth as well as enriched the stems plus sheaths during grain-filling phase, as compared with CSMT and MDS. These results suggest that PSMT may be an alternative approach to increasing grain yield in a rice-wheat rotation system in the lower reaches of the Yangtze River in China.