Tengfei Lyu ,Jie Shen ,Jun Ma *,Peng Ma ,Zhiyuan Yang ,Zou Dai ,Chuangang Zheng ,Min Li

a Rice Research Institute,Sichuan Agricultural University,Chengdu 611130,Sichuan,China

b Crop Ecophysiology and Cultivation Key Laboratory of Sichuan Province,Chengdu 611130,Sichuan,China

c Xichang University,Xichang 615000,Sichuan,China

d Guizhou Academy of Agricultural Sciences,Guiyang 550006,Guizhou,China

ABSTRACT Machine transplanting and the application of slow-release nitrogen(N)fertilizer(SRNF)have played vital roles in the modernization of rice production.We aimed to determine the effects of potted-seedling transplanting—a new machine-transplanting method—and SRNF on hybrid rice yields.A 2-year splitplot experiment (2016–2017) was conducted in Meishan,Sichuan province,China,using two machinetransplanting methods (potted-seedling and blanket-seedling) and three N treatments.Total green leaf area,high-effective leaf area and its rate at heading,net photosynthetic rate of flag leaves 7 days after heading,glutamate synthase (GOGAT) and glutamine synthase (GS) activity after heading,dry matter production,and N accumulation at heading and maturity increased under the potted-seedling method or 70% SRNF as a base+30% urea application at the panicle initiation stage (SBUP).Stem diameter and number of small and of all vascular bundles at the neck–panicle node in potted-seedling plants increased as a result of increasing numbers of effective panicles,secondary branches,and spikelets.In pottedseedling plants,treatment with SBUP increased the number of large and total vascular bundles at the panicle–neck internode and the number of differentiated and surviving secondary branches and spikelets and decreased the number of ineffective tillers and degenerated secondary branches and spikelets.We conclude that the potted-seedling machine transplanting method and SRNF combined with urea topdressing can strengthen the source–sink relationship in rice,resulting in higher yields.

Keywords:Potted-seedling machine transplanting Slow-release nitrogen fertilizer Urea topdressing Branches and spikelets Yield

Abbreviations:SRNF,slow-release N fertilizer;GOGAT,glutamate synthase;GS,glutamine synthase;SBUP,70%SRNF as a base+30%urea application at the panicle initiation stage;NUE,nitrogen-use efficiency;SBUB,105 kg SRNF ha-1+45 kg urea ha-1as a base;LSD,least significant difference;LAI,leaf area index;Pn,net photosynthetic rate;DMPH,dry matter production at heading stage;DMPAH,dry matter production after heading;DMPM,dry matter production at maturity stage;NAE,N accumulation at elongation stage;NAH,N accumulation at heading stage;NAM,N accumulation at maturity stage;SPB,surviving primary branch;SB,secondary branch;DiSB,differentiated SB;SSB,surviving SB;DSB,degenerated SB;PDSB,percentage of degenerated SB;PS,primary spikelet;SS,secondary spikelet;TS,total spikelet;DiPS,differentiated PS;SPS,surviving PS;DPS,degenerated PS;PDPS,percentage of DPS;DiSS,differentiated SS;SSS,surviving SS;DSS,degenerated SS;PDSS,the percentage of DSS;DiTS,differentiated TS;STS,surviving TS;DTS,degenerated TS;PDTS,percentage of DTS.

1.Introduction

China is the most populous country in the world and the most agriculturally productive [1],with rural farm workers accounting for more than 70% of the country’s population.Rice (Oryza sativaL.) is one of the most important crops and is a major food source for more than 60% population of China [2].However,with rural labor migration and the aging of rural populations,fewer people are available to work in agriculture.The development of mechanized planting is thus crucial to China’s future agricultural development.

Currently,there are two main methods of machine transplanting for rice.One is blanket-seedling,which uses hard plastic or calcium plastic paper seedling trays matched with blanket seedling-raising machinery and a transplanter.This method has been developed into a high-yield cultivation technique with the integration of a variety of indigenous cultivation patterns[3].However,it has several problems,including short seedlings at the age suitable for transplanting,high seedling density,and severe damage to shoots and roots during the transplantation process.This damage subsequently leads to a longer revival stage and yield losses[4].In 2010,a new type of machinery that uses seedling trays matched to the seeder and transplanter was developed [4].This new method includes 448 evenly distributed cells in a tray,with 3–5 seeds placed into each cell by the seeder.In comparison with the blanket-seedling method,potted-seedling transplantation reduces seed consumption by 33%–50%.It also extends by more than 10 days the seedling age suitable for transplanting,protects the root system from damage incurred during transplanting,and assists in the production of a high-yield rice group[5,6].It is considered one of the best methods for mechanized rice cultivation in China.

Slow-release nitrogen (N) fertilizer (SRNF) has been used as an effective way to reduce fertilizer consumption,improve N-use efficiency(NUE)and rice yield,and minimize environmental pollution[7–10].However,a single basal application of SRNF may not be sufficient to meet the N requirements of rice throughout the entire growth period[11].To solve this problem,a novel SRNF application method,using SRNF as a basal application combined with topdressing of urea,has been proposed to further increase NUE and rice yield [12].However,there is limited information available on the combination of SRNF and urea as topdressing with pottedseedling machine transplanting,and its effects on rice yield.Moreover,although this transplanting method combined with SRNF application could improve rice yields,the mechanisms underlying this increase remain largely unknown.

2.Materials and methods

2.1.Experimental conditions

In a split-plot study,the main plots were two machinetransplanting methods:potted-seedling (M1),as the method of interest,and blanket-seedling (M2),as a reference.The sub-plots were four N treatments:zero-N,as a control (N0);150 kg SRNF ha-1as a base (N1);105 kg SRNF ha-1+45 kg urea ha-1as a base(N2,SBUB);and 105 kg SRNF ha-1as a base+45 kg urea ha-1at the panicle initiation stage (N3,SBUP).The experiments were conducted in paddy fields at Meishan (30°12′N,103°81′E),Sichuan province,China,during the 2016–2017 rice growing seasons(March to August).The soil properties were 124.16 mg kg-1available N,22.89 mg kg-1Olsen-P,106.97 mg kg-1exchangeable K,36.93 g kg-1organic matter and pH 7.0 in 2016,and 100.36 mg kg-1available N,31.93 mg kg-1Olsen-P,96.81 mg kg-1exchangeable K,27.87 g kg-1organic matter and pH 6.7 in 2017.

The seedling trays,rice seeders,and transplanters were provided by local cooperatives;the potted-seedling raising machinery and transplanter were 2BD-600 (LSPE-60AM) and 2ZB-6A (RXA-60 T) and were made by AMEC,Changzhou,China;the blanketseedling raising machinery was LX-SYB-2 (Zhejiang Lv Xin,Taizhou,China) and the transplanter was Yanmar VP6D (Yanmar,Japan).The seeding rates per tray of potted and the blanket seedlings were respectively 35–40 and 70–75 g.The SRNF was resincoated urea provided by Kingenta (Linyi,Shangdong,China),and N content was 46.0%.The F-you 498 (F32A × Shuhui 498) cultivar was used,as it is the most popularindicahybrid rice in China.Basal P (75 kg P2O5ha-1) and K (150 kg K2O ha-1) were applied in all treatments 1 day before transplantation.The planting density was 20.8 × 104hills ha-1,with a 33-cm row spacing and 14.5-cm intra-row spacing for the potted seedling method and a 30-cm row spacing and 16-cm intra-row spacing for the blanket seedling method.The unit plot size was 27 m2and plots were separated by a 1-m-wide alley.A plastic film was inserted into the soil to a depth of 0.5 m.Water management was based on rice-field dry–wet alternate irrigation:when the number of tillers reached 20 per plant,the plot was drained until the jointing stage,and then drained again one week before harvest.Weeds,insects,and diseases were controlled using standard herbicides and pesticides to avoid yield loss.

2.2.Growth measurements

2.2.1.Dry matter and N content determination

Three hills with the average number of tillers for each treatment were selected for sampling at the jointing,heading,and maturing stages in 2016 and 2017.The samples were divided into stem and sheath,top three leaves (high-effective leaves),other leaves,and panicle portions,exposed to 105 °C for 1 h,and then dried at 80 °C to constant weight.The dry matter of each portion was determined,and finally the total N concentration was determined using an automatic Kjeldahl apparatus (Kjeltec-8400,Foss,Sweden).

2.2.2.Key enzyme activities for N metabolism determination

In 2017,10 flag leaves from each treatment were selected for sampling after heading at 0,10,20,and 30 days and then froze for 1 min in liquid nitrogen and stored at-80°C.Following Umemoto [13],glutamate synthase (GOGAT) and glutamine synthase(GS) activities were determined.

2.2.3.Leaf area and net photosynthetic rate

After the samples were divided into four parts,the area of the top three leaves(high-effective leaves)and other leaves were measured with an LI-3100 leaf area analyzer (LI-COR Biosciences,Lincoln,NE,USA) in 2017.The net photosynthetic rate of the flag leaf was determined using an LI-6400XT Portable Photosynthesis System (LI-COR) 7 days after heading in 2017.The measurement conditions and methods followed Wang et al.[14].

2.2.4.Neck panicle node

Three hills with the average number of effective panicles for each treatment were selected for sampling at the mature stage.Fresh segments of six main stems that were consistently eared were cut approximately 1 cm below the panicle base node.After fixation and desilication,the stems were embedded in paraffin and stained with safranine and fast green for observation.The number of vascular bundles and inside and outside diameters of the stem were determined under a microscope and photographed in 2017.

2.2.5.Panicle branching and spikelets

In 2016,in each treatment,three hills with the average number of effective panicles were assessed at the first heading stage,and following Matsushima [15],the number of surviving primary branches,surviving and degenerated primary spikelets,and surviving and degenerated secondary branches and spikelets were determined for every stem.The percentages of degenerated branches and spikelets were calculated as follows:

Differentiated number=surviving number+degenerated number;

Percentage of degenerated=degenerated number/differentiated number × 100%.

2.2.6.Yield and yield components

Fig.1.Photosynthetic area and net photosynthetic rate (Pn) of rice under different machine transplanting methods and N applications.High-effective LAI:Top three leaves area index.Machine-transplanting methods:potted-seedling(M1)and blanket-seedling(M2).N treatments:zero-N(N0),CRNF 150 kg ha-1 as a base(N1),CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 as a base(N2),and CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 at the panicle initiation stage(N3).Different letters indicate significant difference at the 0.05 probability level.

Rice was harvested at the maturity stage,and the yield of each plot was recorded after measurement of the moisture content and removal of impurities.The grain yield was adjusted to a moisture content of 13.5%.Before harvesting,the number of effective tillers per hill was determined using 60 plants from each plot.Ten marked plants were separated into signal tillers according to the date of marking and used to measure 1000-grain weight,seed setting rate,and grain number per panicle.The actual yield was harvested from other plants in each plot and did not include edge-row plants.

2.3.Data analysis

Statistical analyses were performed using DPS 7.05 (Zhejiang University,Zhejiang,China).Means for cultivation methods were compared using the least significant difference (LSD) test at the 0.05 and 0.01 probability levels.Graphs were constructed with Microsoft Excel 2007(Microsoft Corporation,Redmond,WA,USA).

3.Results

3.1.Effective leaf area index (LAI) and net photosynthetic rate (Pn)

Fig.2.Grain weight per panicle for two machine-transplanting methods in 2016 and 2017.Machine-transplanting methods:potted-seedling(M1)and blanket-seedling(M2).N treatments:zero-N(N0),CRNF 150 kg ha-1 as a base(N1),CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 as a base(N2),and CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 at the panicle initiation stage (N3).Different letters indicate significant difference at the 0.05 probability level.

The total effective LAI,high-effective LAI,high-effective leaf area rate at heading,andPnof the flag leaves 7 days after heading were generally higher in the potted plants (M1) than in the blanketseedling(M2)plants in 2017(Fig.1).Among the N application treatments,the total effective LAI and high-effective LAI were significantly higher with N3than with N1for both transplanting methods.In M1plants,total effective LAI and high-effective LAI were also significantly higher with the N1than with the N2treatment,and the high-effective leaf area rate with N2was significantly lower than that with N1or N3.For both transplanting methods,Pnwith N2was significantly lower than that with N1or N3.

3.2.Dry matter production

The mean dry matter production amounts at heading and maturity were respectively 13.3% and 4.3% higher in M1than in M2plants in both years (P<0.05;Table 1).With respect to N treatments,N3plants showed the highest dry matter production at heading and maturity under both transplanting methods in both years,significantly greater than that under N1and N2.The grain weight per panicle at maturity of M1plants was higher than that of M2plants,and showed the trend N3>N1>N2for both transplanting methods;grain weight per panicle at maturity was significantly higher under N3than under N1or N2in both years (Fig.2).

Table 1 Effects of machine transplanting method and N application on dry matter transportation of rice (t ha-1).

Table 2 Effects of machine transplanting methods and nitrogen applications on N accumulation of rice (kg ha-1).

3.3.N accumulation

The mean N accumulations at heading and maturity were respectively 3.7%–22.0% and 2.1%–8.8% higher in M1than in M2plants in both years (P<0.05;Table 2).With respect to N treatments,N accumulation at elongation showed the trend N2>N1>N3,and the differences among them were significant;N accumulation of N2plants at heading and maturity was significantly lower than those under N1and N3.Compared with N1,N accumulation with N3increased by 2.3% and 1.8% under the potted-seedling method and increased by 14.8% and 5.4% under the blanket-seedling method,in 2016 and 2017,respectively.

Table 3 The structure of the neck–panicle node.

3.4.Key enzyme activity in N metabolism after heading

The activity of GOGAT and GS in flag leaves gradually decreased after heading (Fig.3).Compared to that with blanket-seedling,GOGAT and GS activity after heading under the potted-seedling method were increased to different degrees.Under both machine-transplantation methods,GOGAT and GS activity of flag leaves in every stage after heading showed the trend N3>N1>N2,and the differences among them were significant.

Fig.3.Glutamate synthase (GOGAT) and glutamine synthase (GS) activity in flag leaves after heading in 2017.Machine-transplanting methods:potted-seedling (M1) and blanket-seedling (M2).N treatments:zero-N(N0),CRNF 150 kg ha-1 as a base (N1),CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 as a base(N2),and CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 at the panicle initiation stage (N3).AH:After heading.Different letters indicate significant difference at the 0.05 probability level.

3.5.Structure of the neck–panicle node

Table 3 and Fig.4 show that the number of all and of small vascular bundles and the total area of the neck–panicle node were significantly greater in M1than in M2.Differences in the size and total number of vascular bundles were observed between N treatments in M1,whereas in M2,differences were seen in the cavity and total area.

3.6.Differentiated and degenerated branches and spikelets

The number of surviving total spikelets,secondary branches and spikelets,and differentiated secondary spikelets in M1were significantly higher than those in M2,but the number of differentiated and surviving primary spikelets were significantly lower in 2016(Tables 4 and 5).N treatments significantly affected differentiated,degenerated,and surviving branches and spikelets(Tables 4 and 5).In general,the number of differentiated and surviving total spikelets,and secondary branches and spikelets showed the trend N3>N1>N2,and the difference between N2and N3was significant for both transplanting methods.In M1plants,degenerated total and secondary spikelets,as well as their percentages,were higher in N1than in N2and N3,whereas the degenerated primary spikelets and percentages were higher in N3than in N1and N2;the percentage of degenerated secondary branches was significantly lower in N3than in N1.In M2plants,the differentiated and surviving primary spikelets were significantly higher in N3than in N1,and the degenerated secondary spikelet percentage was significantly lower.

Table 4 Effects of transplanting method and N application on differentiated,degenerated branches in machine-transplanted rice.

3.7.Grain yield

Fig.4.Structure of the neck-panicle node.Machine-transplanting methods:potted-seedling(M1)and blanket-seedling(M2).N treatments:zero-N(N0),CRNF 150 kg ha-1 as a base (N1),CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 as a base (N2),and CRNF 105 kg ha-1 as a base+urea 45 kg ha-1 at the panicle initiation stage (N3).

Rice yield was significantly (P<0.05) affected by transplanting method and N treatment in both years (Table 6).Disregarding N treatments,the mean yield of potted seedlings was significantly higher than that of blanket seedlings in 2016 and 2017.This improvement was attributed to the increased number of spikelets per panicle and the number of panicles.Sink filling and 1000-grain weight under the potted-seedling method were also significantly(P<0.05) higher than those under the blanket-seedling method in 2016.Yield and number of spikelets per panicle under both transplanting methods in both years showed the same trend(N3>N1>N2);the difference between N2and N3was significant(P<0.05).

Table 6 Rice yield and its components under different machine-transplanting methods and N treatments.

4.Discussion

The source-sink relationship plays an important role in rice yield [16].The ‘‘source”refers to the organs or tissues of rice that produce and output assimilates [17].The rice ‘‘sink”refers to the organs or tissues that assimilate and store assimilates,the spikelets of the panicles being the most important [18].‘‘Flow”occurs through the vascular tissue system in rice and the vascular bundles in the neck-panicle internode can be regarded as the channels from‘‘source”to‘‘sink.”High-yielding rice tends to have a strong source and sink and efficient flow.In this study,the yield of the potted rice plants was 8.8% higher on average in both years than that of the blanket-seedling plants.More importantly,the yield of potted rice plants reached 12.4 t ha-1on average when 70% SRNF of total N was applied as a base and 30%urea was applied at the panicle initiation stage (N3,SBUP),and was significantly higher than that with 100% SRNF as the base (N1).These results suggest that the potted machine-transplanting method and SBUP can increase rice yield.

Rice yield formation includes dry matter production and distribution,and biological yield depends on the photosynthetic intensity,duration,and area [19].Zhou et al.[20] showed that nutrients transferred to the panicle are obtained mainly from two sources:dry matter production before heading and photosynthates in functional leaves after heading.In the present study,dry matter production before heading,active photosynthetic area at heading,net photosynthetic rate of flag leaves 7 days after heading,total green leaf area,and high-effective leaf area proportion were higher in the potted-seedling transplanted rice than in the blanket-seedling rice.Of the three N application treatments,these indicators were highest under the N3(SBUP)treatment,explaining the higher dry matter production and yield at maturity under this N treatment.

Plant N uptake and accumulation are the intrinsic driving forces for rice yield formation [21].In this study,the mean N accumulations at heading and maturity in potted-seedling plants were 12.6% and 5.2% higher,respectively,than that in blanket-seedling plants in both years;moreover,compared to that with 100% SRNF as the base,N accumulation with SBUP at heading and maturity increased by respectively 2.3% and 1.8% in potted-seedling plants.Because the potted-seedling hybrid rice was subjected to almost no injury when transplanted,its roots and tillers developed early and rapidly,and SBUP effectively reduced the occurrence of ineffective tillers.This reduction in turn reduced the loss of nitrogen,expanded the tillers,and improved N uptake capability from elongation to heading [22],and these were also the important characteristics of N absorption and accumulation in the differences between the potted-seedling and the blanket-seedling groups.GOGAT and GS are key enzymes in the process of N assimilation[23]that directly affect N absorption and utilization in the process of rice grain formation[24].The present study showed that GOGAT and GS activities of potted-seedling plants after heading at 0–30 days were higher,by different amounts,than those under the blanket-seedling method,indicating that N assimilation and absorption capacity of the potted-seedling hybrid rice was higher than that of the blanket-seedling rice.Moreover,compared to that with 100% SRNF as the base,GOGAT and GS activity with SBUP were significantly improved after heading at 0–30 days,and this result was consistent with that of Du et al.[25].Above showed that urea topdressing of SBUP,which provided nutrients,coincided with the demand of hybrid rice at the young panicle differentiation stage,and thus,SBUP solved the problem of insufficient N nutrition at the young panicle differentiation stage when N is applied as SRNF alone.It also promoted N absorption and assimilation ability,thereby improving the yield and N utilization efficiency of machine-transplanted hybrid rice.

‘‘Flow”is the development of vascular tissue systems in the rice[17].In this system,the vascular bundles in the neck-panicle internode can be regarded as the total channel from ‘‘source”to‘‘sink”.Liu et al.[26]reported that the number of spikelets per panicle was positively correlated with the outside diameter of the neck–panicle node.In our study,this outside diameter was on average 6.6% higher in potted-seedling than in blanket-seedling plants.The numbers of large,small,and total vascular bundles in the neck-panicle node were 8.4%,13.3%,and 10.9% higher,respectively,under the potted-seedling than the blanket-seedling treatment.The numbers of large and total vascular bundles were significantly higher with N3than with N1and N2,and the phloem area with N2was smaller than that with N1and N3in the potted rice plants.That the phloem parenchyma cells of the neckpanicle internode are the main tissues that transport assimilates may explain the low grain weight under N2.

The rice sink is usually determined by the total number of spikelets and their weight.The total number of spikelets is determined by the effective panicles and the number of spikelets per panicle [27].Our study showed that compared with the blanketseedling treatment,the potted-seedling treatment resulted in more early-emergence tillers but fewer later-emergence tillers,and increased the later effective panicles and their proportions.The total effective tillers and percentage,spikelets per panicle,and yield consequently increased.In addition,N2plants had more emergent tillers than N1and N3plants,especially tillers that emerged 18 days after transplantation,but most were ineffective.However,excessive numbers of ineffective tillers can lead to small panicle size and reduction in grain yield[28],suggesting one expla-nation for the higher spikelets per panicle and yields under N1and N3than under N2.

The surviving spikelets at maturity are the consequence of the differentiated and degenerated spikelets at the young panicle differentiation stage [29].This study showed that the number of differentiated and surviving secondary spikelets,as well as total surviving spikelets in the potted-seedling plants,was higher than that in the blanket-seedling plants.Yao et al.[30] showed that the final number of spikelets is influenced mainly by the number of surviving secondary spikelets.Our results were in accord with this finding,and the proportion of surviving secondary spikelets reached respectively 67.3% and 64.9% under the potted-and blanket-seedling treatments.In a previous study [20],we showed that the number of surviving secondary spikelets depended on the surviving secondary branches;in the present study,the number of surviving secondary branches in potted-seedling plants was significantly higher than in blanket-seedling plants,with the same trend for surviving secondary and total spikelets.

Many studies [7,10,30,31] have shown that SRNF as a basal application or urea as a topdressing increases rice yields by increasing spikelets per panicle.Ma et al.[32]found that topdressing with urea increased grain weight.Compared with 100%SRNF as a base (N1),30% urea topdressing (N3) significantly increased the differentiated secondary and total spikelets,and their degenerated and percentage of degenerated secondary and total spikelets decreased,so that the surviving secondary and total spikelets also increase.In addition,the spikelets per panicle and grain weight of N3were 4.1% and 1.4% (potted-seedling),and 3.4% and 0.7% (blanket-seedling) higher on average than those of N1.Consequently,the grain weight per panicle of N3was 5.9%(potted-seedling) and 2.4% (blanket-seedling) higher than that of N1.

A coordinated source–sink relationship can increase the processing speed of assimilation products,the number of spikelets per panicle,sink filling,grain weight of rice,and rice yield [33].This study showed that the potted-seedling machinetransplanting method could improve the rice source-sink relationship and flow efficiency,especially when 70% SRNF was used as a base with 30% urea applied at the panicle initiation stage.

CRediT authorship contribution statement

Jun Ma,ZhiyuanYang,and Chuangang Zheng conceived the original screening and research plans;Tengfei Lyu,Jie Shen,Peng Ma,and Zou Dai performed the experiments and analyzed the data;Tengfei Lyu wrote the article with contributions of all authors.Jun Ma and Min Li provided editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Key Research and Development Program of China (2017YFD0301701 and 2017YFD0301706),and National Natural Science Foundation of China (31660369).

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2020.08.013.