Superior Grains Determined By Grain Weight are not Fully Correlated with the Flowering Order in Rice

Superior Grains Determined By Grain Weight are not Fully Correlated with the Flowering Order in Rice PENG Ting 1, 2*, LV Qiang 3*, ZHAO Ya-fan 1, 2*, SUN Hong-zheng 1, 2, HAN Ying-chun 1, 2, DU Yan-xiu 1, 2, ZHANG Jing 1, 2, LI Jun-zhou Li 1, 2, WANG Lin-lin 1, 2 and ZHAO Quan-zhi 1, 2 1 Collaborative Innovation Center of Henan Grain Crops, Henan Agricultural University, Zhengzhou, P.R.China China 2 Research Center for Rice Engineering in Henan Province, Henan Agricultural University, Zhengzhou, P.R.China 3 College of Agronomy, Henan University of Science and Technology, Luoyang, P.R.China Abstract Rice panicles are composed of many branches with two types of extreme grains, the superior and the inferior. Traditionally, it has been well accepted that earlier flowers result in superior grains and late flowers generate inferior grains. However, these correlations have never been strictly examined in practice. In order to determine the accurate relationship between superior and inferior grains and the flowering order, we localized all the seeds in a panicle in four distinct rice species and systematically documented the rice flowering order, flower locations and the final grain weight for their relationships. Our results demonstrated that the grain weight is more heavily determined by the position of the seeds than by the flowering order. Despite earlier flowering has a positive correlation with the grain weight in general, grains from flowers blooming on the second day after anthesis generally gained the highest weight. This suggests earlier flowers may not result in superior grains. Therefore, we concluded that superior and inferior grains, commonly determined by grain weight, are not fully correlated with the flowering order in rice. Following the order of the grain weight, the superior grains are generally localized at the middle parts of the primary branches, whereas inferior grains were mainly on the last two secondary branches of the lower half part of the panicle. In addition, the weight of inferior grains were affected by spikelet thinning and spraying with exogenous plant growth regulators, indicating that physiological incompetence might be the major reason for the occurrence of the inferior grains. Key words: rice, flowering order, grain weight, grain position, inferior grains, superior grains 1 INTRODUCTION Rice (Oryza sativa) is the most important food crop around the world, which provides over 21% of the energy of the world s population and even 76% of the energy intake of the population of South East Asia (Fitzgerald et al. 2009). Due to dwarfing breeding and hybrid breeding, rice yield has more than doubled in most parts of the world and even tripled in some countries and regions (Xing and Zhang 2010). However, these cultivars or combinations (e.g., New Plant Type rice, hybrid rice and super rice) have encountered the trouble of poor Correspondence ZHAO Quan-zhi, Tel: +86-371-63558293, Fax: +86-371-63558122, E-mail: qzzhaoh@126.com * These author contributed equally to this work. 1

grain-filling which leads to a low seed setting rate (Yang 2010; Yang and Zhang 2010), and blocked their high yield potential and affected grain quality seriously (Yang 2010). In general, the grains usually located on apical primary branches in a panicle have earlier flowering time, fill faster and produce heavier grains defined as superior grains. The later-flowering inferior grains which are usually located on proximal secondary branches, which fill slowly and poorly, consequently produce lighter grains (Yang 2010; Yang and Zhang 2010). Recently, large quantitative efforts have been made to study the mechanism underlying the poor filling of rice inferior grains, but the results are diverse because of the different sampling location of superior and inferior grains within a rice panicle (Wang et al. 2006, 2007, 2012; Wei et al. 2011; Zhang et al. 2012a, b). Furthermore, it is still elusive up to now whether the maldevelopment or the low physiological activity of the inferior grains that leads to their poor plumpness. Therefore, clarifying the relationships between the position of superior and inferior grains, the time of flowering within a rice panicle, grain weight distribution of different positions in a panicle, and blossom characteristics is essential to reveal the inner mechanism involved in poor filling of rice inferior grains. production scientifically and practically. RESULTS This is also beneficial for realizing the goal of high yield and quality, efficient Various anthesis of different grains in a panicle The anthesis of different grains in the panicle of the four cultivars was investigated, and it was completed within a period of 7 d in Xingfeng 2 and Y liangyou 2, 6 days in Fangxin 1 and II you 838 (Table 1, Fig. 1). The blooms peak of the conventional japonica rice cultivars ( Xingfeng 2 and Fangxin 1) were 2nd, 3rd and 4th d after flowering, while hybrid indica rice II you 838 bloomed at 2nd, 3rd and 5th d after flowering, and Y liangyou 2 reached the peak of blossom at 2nd, 3rd, 4th, and 5th d after flowering. However, all cultivars in present experiment had less grains flowering at the first and 6th d, and there were scarce grains blossomed on the 7th d, and there was no big difference on grain flowering order patterns among different cultivars (Fig. 1). It was found that grains on primary branches and secondary branches abaxial panicle axis flowered first, then the grains on the upper section of secondary branches in the middle part of a panicle, grains on the primary branches, and the top grains on secondary branches in the lower part of a panicle flowered. Grains on secondary branches in the basal part of a panicle blossomed at last. For example, more than 80% seeds flowered on the 1st and 2nd d during anthesis positioned on upper part primary branches and the remaining 20% seeds concentrated on the upper part in a panicle. Among the seeds blossomed on the 3rd and 4th d during anthesis, there were about 65% grains positioned on the basal primary branches and 35% grains located on the lower part of the panicle, whereas seeds blossomed on the 5th to 7th d mostly positioned on the middle and basal secondary branches. Flowering order of grains within the panicle was in a basipetal manner, that is, from the upper to the basal branches. It was the same for the flowering order of secondary branches on primary branches. As for the grains on the same branch, blossom order was identical for the primary and secondary branches. The distal grains on the branch flowered firstly, and then it was turn to basal grains and the penultimate grain from the tip of the branch flowered finally. 2

Distribution characteristics of grain weight on the panicle According to the flowering time of a panicle, grains flowered on the 2nd d had the greatest grain weight, and those flowered on the 1st d had a lighter grain weight, and the later flowered grains had a decreasing grain weight except Y liangyou 2. Y liangyou 2 grain weight order was 1 DAF>3 DAF >4 DAF >2 DAF >5 DAF >6 DAF >7 DAF (Table 1). Grains on the primary branches had a heavier grain weight than the ones on the secondary branches with a general tendency of grain weight decreasing from the outside to the inside, from the upper part to the lower part within a panicle (Fig. 2). Seeds of 1.1 and 1.2 presented the smallest grain weight, while seeds of 1.3, 1.4 and 1.5 of conventional Japonica rice cultivars and 1.4, 1.5 and 1.6 of hybrid indica rice cultivars presented the heaviest grain weight in the primary branches. With regard to the grains on the secondary branches, the top kernel was the heaviest, and 1.2 was the lightest one. Grain weight negatively related with its flowering order in the panicle As shown in Fig. 3, the grain weight of different genotypes had a significantly negative correlation with flowering order (days after flowering). The coefficient correlation index of Xingfeng 2, Fangxin 1, II you 838, and Y liangyou 2 were -0.684, -0.530, -0.689, and -0.420, respectively. Nevertheless, seeds blossomed earlier such as those located at 1.5 and 1.6 on primary branches did not presented the heaviest grain weight in the primary branch, which was inconsistent with the idea that seeds that blossomed earlier had heavier grain weight, and the four cultivars showed the similar tendency. The distribution characteristics of grain plumpness Grain plumpness varied between genotypes or locations on the branches. In general, more than half seeds had grain plumpness higher than 90% as for conventional japonica rice cultivars (Xingfeng 2 and Fangxin 1). However, seeds of hybrid indica rice cultivars (II you 838 and Y liangyou 2) presented different grain plumpness. The grain plumpness of 36.29% seeds of II you 838 was higher than 90%, and 13.31% seeds was less than 70%. As for Y liangyou 2, only 4.12% seeds had a grain plumpness higher than 90%; 58.08% seeds had grain plumpness between 80 and 90%, and 12.37% seeds was less than 70%. These results suggested that the grain plumpness of conventional japonica rice cultivars bearing less seed number showed no obvious difference. However, the grain plumpness of large panicle hybrid rice cultivars showed apparent difference (Table 2). In addition, the distribution characteristics of grain plumpness were consistent with the grain weight of seeds according to the position on the panicle. Distribution characteristics of superior and inferior grains in the panicle In order to clarify the distribution characteristics of superior and inferior grains, following the order of the grain weight, we selected the superior grains to the heaviest seeds occupying 20% of the total weight of the entire panicle seeds, and the inferior grains to the lightest seeds consisting of 20% of the total panicle grain weight. The remaining seeds of 60% total weight were between the superior and inferior grains. 3 Fig. 4 showed that

distribution of superior and inferior grains correlated with the location of grains and the four rice cultivars showed similar tendency. In general, superior grains distributed at the middle and basal part of primary branches, whereas inferior grains were always localized on the last two secondary branches of the lower half part of the panicle. In addition, the flowering time of superior grains for conventional cultivars mainly concentrated on the 2nd d and then on the 3rd d during anthesis, while for hybrid cultivar Y liangyou 2, superior grains flowered at 1st-4th d and concentrated on the 4th day after anthesis. As for inferior grains, the blossom time distributed from the 3rd day after anthesis and mostly focused on 3rd-6th d after anthesis (Table 3, Fig. 4). The plasticity of grain weight on inferior grains Thinning spikelets and exogenous hormone treatments could significantly increase the grain weight of inferior grains (Fig. 5). The grain weight of inferior grains treated with spikelet thinning was 22.49 mg, which was 91.24% higher than the control (P<0.01). When treated with the two kinds of compound plant growth regulators, the grain weight of inferior grains improved 17.48 and 16.38%, significantly higher than the control (P<0.01), respectively. DISCUSSION Previous studies have described that grains in the same panicle had the different flowering time and grain weight (Mohapatra et al. 1993; Tsutomu and Toshiaki 2003; Yang et al. 2006). The present study, for the first time, investigated the flowering order and weight of all the grains in a panicle (Figs. 1 and 2). The grains of different rice cultivars blossomed from the outside to the inside in a basipetal way. According to the position of grains on branches (including primary and secondary branches), the distal grains flowered first and subsequently flowered from the basal to the upper part of the panicle, followed by the penultimate one flowered last (Fig. 1). The single grain weight of grains located on the primary branch was greater than that of grains on the secondary branch in the same branch, and grains on primary branch near the axis and grains on distal secondary branch presented the highest grain weight and those on sub-distal part presented the smallest grain weight (Fig. 2). Rice grain quality and grain weight had a close relationship with the location of grains (Tsutomu and Toshiaki 2003; Yang 2010). Usually, rice superior and inferior grains were defined according to their flowering time and grains position on the panicle (Wang et al. 2012; Zhang et al. 2012a, b). Superior grains referred to the grains that blossomed on the 1st and 2nd d during anthesis or located at the upper primary branches of a rice panicle (Wang et al. 2006, 2007) and the distal grains on the primary branches (Wei et al. 2011), whereas inferior grains referred to the ones on the basal secondary branches (Wang et al. 2007; Wei et al. 2011). Our research also showed that the flowering time and grain weight of the seeds on the rice panicle existed in a location-dependent manner, and the single grain weight had an obvious negative correlation with time of flowering (days after flowering). However, some seeds flowered earlier didn t have the higher grain weight. For example, seeds that have the maximum grain weight were the ones flowered on the second day since onset of anthesis, and the seeds flowered on the 1st d had smaller grain weight than the previous ones. Grain weight of the rest seeds decreased with the delaying of flowering time. According to our research, superior grains of conventional japonica rice 4

cultivars often located at 1.4 and 1.5 on the primary branches and those of hybrid indica rice cultivars almost positioned at 1.5 and 1.6 on primary branches in a rice panicle. Inferior grains distributed from the 3rd d since onset of anthesis and located at two paraxial secondary branches on the lower part of the panicle except the distal grains (Fig. 4). Grain filling of rice is a complex process, and the grain weight is negatively correlated with time of flowering significantly (Fig. 3). Grains blossomed and fertilized earlier fill faster and always have heavier grain weight and have better grain plumpness. Grains fertilized later fill slower and keep a quite lower filling rate, even zero rate for a long period after fertilizing, so these seeds have more probability to be sterile (Mohapatra et al. 1993; Yang et al. 2000). But the grain weight of these seeds blossomed on the 2nd day was higher than that of seeds blossomed on the 1st d (Table 1). In addition, some seeds on other position were also likely to have higher grain weight and even develop into superior grains although they flower late (Fig. 3). These results indicated that grain development was one reason caused the poor grain plumpness, but it was not the only one. researchers suggested that the assimilate supply was not the reason leading to the poor grain plumpness, because sucrose content in inferior grains was higher than that of superior ones during the early rice grain filling stage (Yang et al. 2006; Yang 2010; Yang and Zhang 2010). The content of ABA, IAA, Cytokinins, Polyamines, and the activity of some key enzymes of starch synthesis (such as SuS, AGP, SS, SBE) and their gene expression in inferior grains was lower than that of superior ones, whereas the content of ethylene showed the opposite trend during the early and middle stage of grain filling. Some All these factors above had a positive correlation with rice grain filling rate and division rate of endosperm cells except ethylene (Panda et al. 2009; Wang et al. 2012; Yang et al. 2000, 2006; Zhang et al. 2009a, b, a; Zhu et al. 2011). It is implied that the poor physiological activity in the early filling stage and the low rate of conversion rate of sucrose to starch in inferior grains contributed most to their poor grain filling (Yang 2010; Yang and Zhang 2010). The grain weight of inferior grains increased nearly half under spikelet thinning treatment, which was lower 3.37 mg than superior grains (Fig. 5-A). This may be attributed to the improvement of sugar content and starch-metabolizing enzyme activity in inferior grains and then increased the physiological activity of sink organs (Ishimaru et al. 2005; Tang et al. 2009). In addition, methods of cultivation such as spraying exogenous hormones (Yang et al. 2008; Zhang et al. 2009b; Zhang et al. 2012a) and wetting and mild soil drying (Yang and Zhang 2010; Zhang et al. 2012a) during grain filling could increase the sink physiological activity and enhance the plumpness of inferior grains. Therefore, the increasing grain weight of inferior grains by spraying exogenous compound plant growth regulators might (Fig. 5-B) result from enhancing sink physiological activity of inferior grains. All the above evidence indicated that the low physiological activity of inferior grains in early filling stage is the main cause of its poor grain plumpness. potential of rice could be fully exploited by means of genetic improvement or cultivation to enhance the grain filling of inferior grains in the future. Yield CONCLUSION Flowering time and grain weight of the seeds on the rice panicle existed in a location-dependent manner, and the single grain weight had an obvious negative correlation with their flowering order (days after flowering). However, superior and inferior grains commonly determined by grain weight, are not fully correlated with the 5

flowering order in rice. Based on the order of the grain weight, the superior grains are generally localized at the middle parts of the primary branches, whereas inferior grains were mainly on the last two secondary branches of the lower half part of the panicle. spikelet thinning and exogenous hormone treatment. Furthermore, the grain weight of inferior grains significantly increased under MATERIALS AND METHODS Plant materials and cultivation The experiment was conducted at the research farm of Henan Agriculture University, Henan Province, China (34 53 N, 113 35 E, 94 m altitude). Two conventional Japonica rice varieties Xingfeng 2, Fangxin 1 and two hybrid indica rice varieties II you 838, Y liangyou 2 were selected as research materials. Rice seeds were germinated in plastic breeding trays on 30 April and seedlings were transplanted into 80 plastic buckets with 20kg sifted dry topsoil per bucket on 14 June. The specification of bucket was 34 cm (inside diameter) 35 cm (height). Each cultivar had 35 buckets, and there were three hills in each bucket with one seedling per hill. All the buckets were buried into soil in order to assimilate farm condition with general field management. Flowering order and grain weight distribution Ten buckets were transferred to greenhouse under the daily period of 14 h light at (30±2) C and 10 h dark at (25±2) C at booting stage to investigate flowering order. In order to collect representative panicles used for sampling, the panicles flowered on the same day at initial anthesis were chosen. Then, the chosen panicles having average primary branches were selected as candidate for sampling. In total, thirty representative panicles were chosen and tagged for each cultivar, respectively. The flowering order was recorded as follows: fifteen representative panicles were selected from the onset of anthesis (the first day of flowering), then marked the seeds flowered on the same day from 2 to 4 p.m. every day using marker pen with different colors until the end of anthesis. Another selected fifteen panicles were harvested in mature stage, and the single grain weight of grains on different position was measured by electronic balance (1/10000, Sartorius, Goettingen, Germany). To draw the diagram of each cultivar, thirty representative panicles of each cultivar were used after harvest. The position of grains used in distributing diagram was the one existing in more than half of the representative panicles, and the unfilled grains were ignored. The determination of grain plumpness was referred to Zhu et al. (1995) and added some judicious improvement. It was expressed by the relative plumpness of grains on different positions and reflected the filling condition of each spikelet on the panicle. Grain plumpness (%)=Fertilized grain weight/maximum grain weight 100 Spikelet thinning In this study, rice cv. Xingfeng 2 planted in the buckets was employed to examine the effect of spikelet thinning treatment on the grain weight of inferior grains. At full heading stage, we tagged single panicle flowered on the 6

same day. Then the upper two-thirds of the panicle were cut off leaving the basal rachis seeds according to Ishimaru et al. (2005) as spikelet thinning treatment (Fig. 6, T1), while panicles containing all the seeds as control. In each treatment, 50 panicles that headed on the same day were chosen, thinned and tagged. Water management and insect pest control were managed according to field environment. Superior grains and Inferior grains were separated from the tagged panicles in mature period and weighed to get the single grain weight. Exogenous compound plant growth regulators spray applications Seedlings of Xingfeng 2 were transplanted into paddy field in the rice growing season (from early May to mid October) at a hill spacing of 0.3 m 0.13 m with one seeding per hill. The area of each plot was 20 m 2 and each treatment comprised three replications. Two kinds of compound plant growth regulators 1 and 2 (constitute by different concentration of exogenous hormone and microelement) chosen by our lab previously were used, and all the regulators contained ethanol and Tween 20 at final concentrations of 0.1% (v/v) and 0.01 (v/v), respectively. The same volume of water containing the same concentrations of ethanol and Tween 20 was applied to the control plants. In each pot, 100 panicles that headed on the same day were chosen and tagged. Plant growth regulators were foliar-sprayed at the day after tagged. Water management and insect pest control were managed according to general rules. Superior grains and inferior grains were separated from the tagged panicles in mature period and weighed, respectively. References Fitzgerald M A, Mccouch S R, Hall R D. 2009. Not just a grain of rice: the quest for quality. Trends in Plant Science, 143, 133-139. Ishimaru T, Hirose T, Matsuda T, Goto A, Takahashi K, Sasaki H, Terao T, Ishii R, Ohsugi R, Yamagishi T. 2005. Expression patterns of genes encoding carbohydrate-metabolizing enzymes and their relationship to grain filling in rice (Oryza sativa L.): comparison of caryopses located at different positions in a panicle. Plant Cell Physiolog, 464, 620-628. Mohapatra P, Patel R, Sahu S. 1993. Time of flowering affects grain quality and spikelet partitioning within the rice panicle. Functional Plant Biology, 202, 231-241. Panda B B, Kariali E, Panigrahi R, Mohapatra P K. 2009. High ethylene production slackens seed filling in compact panicled rice cultivar: Plant Growth Regulation, 582, 141-151. Tang T, Xie H, Wang Y, Lu B, Liang J. 2009. The effect of sucrose and abscisic acid interaction on sucrose synthase and its relationship to grain filling of rice (Oryza sativa L.). Journal of Experimental Botany, 609, 2641-2652. Tsutomu and Toshiaki. 2003. Morphological development of rice caryopses located at the different positions in a panicle from early to middle stage of grain filling. Functional Plant Biology, 30, 1139-1149. Wang F, Chen S, Cheng F, Liu Y, Zhang G. 2007. The differences in grain weight and quality within a rice (Oryza sativa L.) panicle as affected by panicle type and source-sink relation. Journal of Agronomy and Crop Science, 1931, 63-73. 7

Wang F, Cheng F, Zhang G. 2006. The relationship between grain filling and hormone content as affected by genotype and source-sink relation. Plant Growth Regulation, 491, 1-8. Wang Z, Xu Y, Wang J, Yang J, Zhang J. 2012. Polyamine and ethylene interactions in grain filling of superior and inferior spikelets of rice. Plant Growth Regulation, 1-14. Wei F, Tao H, Lin S, Bouman B, Zhang L, Wang P, Dittert K. 2011. Rate and duration of grain filling of aerobic rice HD297 and their influence on grain yield under different growing conditions. Journal Science Asia, 37, 98-104. Xing Y and Zhang Q. 2010. Genetic and molecular bases of rice yield. Annual Review of Plant Biology, 61, 421-442. Yang J. 2010. Mechanism and regulation in the filling of inferior spikelets of rice. Acta Agronomica Sinica, 3612, 2011-2019. Yang J, Peng S, Visperas R M, Sanico A L, Zhu Q, Gu S. 2000. Grain filling pattern and cytokinin content in the grains and roots of rice plants. Plant Growth Regulation, 303, 261-270. Yang J, Yunying C, Zhang H, Liu L, Zhang J. 2008. Involvement of polyamines in the post-anthesis development of inferior and superior spikelets in rice. Planta, 2281, 137-149. Yang J, Zhang J. 2010. Grain-filling problem in 'super' rice. Journal of Experimental Botany, 611, 1-5. Yang J, Zhang J, Wang Z, Liu K, Wang P. 2006. Post-anthesis development of inferior and superior spikelets in rice in relation to abscisic acid and ethylene. Journal of Experimental Botany, 571, 149-160. Zhang H, Li H, Yuan L, Wang Z, Yang J, Zhang J. 2012a. Post-anthesis alternate wetting and moderate soil drying enhances activities of key enzymes in sucrose-to-starch conversion in inferior spikelets of rice. Journal of Experimental Botany, 631, 215-227. Zhang H, Tan G, Wang Z, Yang J, Zhang J. 2009a. Ethylene and ACC levels in developing grains are related to the poor appearance and milling quality of rice. Plant Growth Regulation, 581, 85-96. Zhang H, Tan G, Yang L, Yang J, Zhang J, Zhao B. 2009b. Hormones in the grains and roots in relation to post-anthesis development of inferior and superior spikelets in japonica/indica hybrid rice. Plant Physiol Biochem, 473, 195-204. Zhang Z, Chen J, Lin S, Li Z, Cheng R, Fang C, Chen H, Lin W. 2012b. Proteomic and phosphoproteomic determination of ABA's effects on grain-filling of Oryza sativa L. inferior spikelets. Plant Science, 185-186, 259-273. Zhu G, Ye N, Yang J, Peng X, Zhang J. 2011. Regulation of expression of starch synthesis genes by ethylene and ABA in relation to the development of rice inferior and superior spikelets. Journal of Experimental Botany, 6211, 3907-3916. Zhu Q, Wang Z, Zhang Z, Hui D. 1995. Study on indicators of grain filling of rice: Journal of Jiangsu Agricultural College, 162, 1-4. (in Chinese) 8

Fig. 1 The flowering order of seeds on the panicle. F1 (black), F2 (pink), F3 (red), F4 (yellow), F5 (blue), F6 (orange), F7 (grey) represent the seeds flowered on the1st, 2nd, 3rd, 4th, 5th, 6th, and 7th d during anthesis, respectively. 9

Fig. 2 Distribution characteristics of grain weight on the different position. Different colors (black, pink, red, orange, and grey) represent different ranges of grain weight on the panicle. 10

32 Xinfeng 2 Fangxin 1 32 Grain weight (mg) 28 24 20 24.93 28.60 25.45 28.78 28 24 20 Grain weight (mg) 16 r=-0.684**, n=147 r=-0.530**, n=152 16 Grain weight (mg) II you 838 Y liangyou 2 36 32.01 30 24 25.57 18 12 6 r=-0.689**, n=248 0 0 1 2 3 4 5 6 7 8 Days after initial flowering 30 23.92 25 20 20.47 15 r=-0.420**, n=291 10 0 1 2 3 4 5 6 7 8 Days after initial flowering Grain weight (mg) Fig. 3 Correlation analysis for the flowering time and grain weight of different genotypes ( was the dividing line of the minimum grain weight of superior grains, was the dividing line of the maximum grain weight of inferior grains). Fig. 4 Distribution characteristics of superior and inferior grains. inferior grains, respectively. Black and grey represent superior grains and 11

Grain Weight (mg) 30 25 20 15 10 5 a b 30 25 20 15 10 5 Grain Weight (mg) 0 CKs Cki T1 CK T2 T3 0 Fig. 5 The effect on grain weight of inferior grains under spikelet thinning treatment (A) and spraying exogenous hormone (B). CKs and CKi represent grain weight of superior and inferior grain weight in control planted in the buckets condition, respectively. CK refers to the grain weight of inferior grains grow in the field condition. T1, T2, and T3 represent grain weight of inferior grains treated by spikelet thinning, spraying compound chemical regulator 1, and spraying compound chemical regulator 2, respectively. Fig. 6 Diagram of seeds on a rice panicle. The rice panicle consists of a main axis and primary, secondary branches, and seeds. The primary branches are attached at the main axis, the secondary branches grow on the primary branch, and the seeds are attached at the primary, secondary branches. In this study, the positions of seeds are defined as the order of itself on the main axis, primary branch, and secondary branch from top to bottom. For example, 1.1.3 represent the third seed located on the first primary branch, 5.3.2 refer to the second seed located on the second secondary branch of the 5th primary branch as showed in the Fig. 12

Table 1 Grain number and average grain weight (mg) in different period of blossom of the four genotypes Cultivar F1 F2 F3 F4 F5 F6 F7 Total GN GW GN GW GN GW GN GW GN GW GN GW GN GW GN Average GW Xinfeng 2 10 28.07 34 28.28 30 27.50 36 26.69 14 24.96 16 24.26 7 24.06 147 26.76 Fangxin 1 9 27.88 56 28.21 40 27.34 23 25.82 11 25.05 13 23.42 0-152 26.96 Ⅱyou 838 26 30.53 76 31.13 95 28.91 5 26.15 44 22.55 2 13.53 0-248 28.45 Yliangyou 2 21 23.08 45 22.39 77 22.54 77 22.46 37 20.72 29 20.12 5 20.31 291 22.02 F1, F2, F3, F4, F5, F6, and F7 represent days after initial flowering. That is, grains flowered on the 1st, 2nd, 3rd, 4th, 5th, 6th, and 7th d. GN and GW represent grain number and grain weight, respectively. The same as below. Table 2 Grain number and percentage (%) of rice grains having different grain plumpness (GP) of various genotypes. Grain plumpness Xinfeng 2 Fangxin 1 II you 838 Yliangyou 2 GN Percentage GN Percentage GN Percentage GN Percentage GP 90% 77 52.38 84 55.26 90 36.29 12 4.12 80% GP<90% 52 35.37 50 32.89 90 36.29 169 58.08 70% GP<80% 18 12.24 11 7.24 35 14.11 74 25.43 60% GP<70% 0 0.00 5 3.29 14 5.65 23 7.90 GP<60% 0 0.00 2 1.32 19 7.66 13 4.47 Total 147 100.00 152 100.00 248 100.00 291 100.00 Table 3 Seed number of superior and inferior grains flowering on different day during anthesis Cultivar Superior grains Inferior grains F1 F2 F3 F4 F5 F6 F7 F1 F2 F3 F4 F5 F6 F7 Xinfeng 2 2 13 9 5 1 0 0 0 0 1 4 8 11 6 Fangxin 1 0 20 7 2 1 0-0 1 5 9 6 9 - Ⅱyou 838 4 24 22 0 0 0-1 1 17 2 27 2 - Yliangyou 2 8 7 10 24 8 1 0 1 8 8 12 11 15 3 13