Orderly Mechanical Seedling-Throwing: An Efﬁcient and High Yielding Establishment Method for Rice Production

: Developing an efﬁcient and high-yielding mechanical rice establishment system is one of the most important approaches for intensive and large-scale rice production. Recently, an orderly mechanical rice seedling throwing system (OMST) was successfully developed; however, the performance of this system is unknown. In the present study, a two-year ﬁeld experiment was carried out in a split-plot design with three establishment methods arranged in the main plots, and two elite rice cultivars arranged in the sub-plots. The grain yield and growth-related traits were then determined. The results showed that the grain yield of OMST was signiﬁcantly higher than manual seedling throwing, and was equivalent to that of manual transplanting, which was mainly due to the variances in panicle number and total spikelet number. Further analysis suggested that the orderly mechanical seedling throwing takes advantage of higher biomass accumulation after heading, increased leaf area index and decreased leaf senescence rate against manual seedling throwing, and more tillers and biomass accumulation at vegetative growth stage as compared to manual transplanting. The present study showed that the OMST is an efﬁcient and high-yielding rice establishment method that may be a promising option to replace traditional manual seedling throwing in rice production.


Introduction
Rice is the dominant staple food in China.In recent years, the labor scarcity in rural areas and the increase in rice production costs resulting from rapid urbanization and industrialization has greatly challenged rice production in China [1].Meanwhile, continuous population growth requires a further improvement in both rice yield and total rice production.It has been estimated that China will need to produce approximately 20% more rice by 2030 to meet its domestic needs [2].Therefore, the systematic upgrading of rice production from the manual, high input and low-profit method to a simplified, mechanized and efficient system is desperately needed.Over the last decades, a variety of approaches have been adopted in Chinese rice production system, including the development of resourceefficient rice cropping systems [3,4], the promotion of tillage and harvest machines [5], and innovation in precise fertilizer management [6], which have greatly mitigated the disparity between rice food demand and the scarcity in labor and resources.However, the labor demand and the production costs during the rice establishment process are still relatively high and have become a major constraint for rice production in China [7,8].To solve this problem, simplified and mechanized rice establishment methods need to be addressed.
In traditional rice production, the rice seeds are sown on a nursery to raise rice seedlings, and then the rice seedlings are manually transplanted to the puddled field, which requires high labor input and is no longer suitable for large-scale rice production in China [9].In recent years, several simplified or mechanized rice establishment methods, such as rice direct-seeding (DSR) and machine transplanting, have been extended rapidly to replace manual transplanting (MT), and the characteristics of these establishment methods have also been intensively studied [9][10][11][12].Meanwhile, seedling throwing (MST) is a simplified rice establishment method in which rice seedlings with soil on their roots are thrown by hand into fields [13].In comparison with MT, MST greatly reduces the labor input during transplanting, which increases the economic benefits.Significant variances in growth performance between MT and MST were observed.The setback caused by uprooting and transplanting was either smaller or non-existent in MST than that in MT, which may shorten the growth duration and is regarded as an ideal trait for multiple rice cropping where the rice growing season is strictly limited.Moreover, higher tiller generation, vigorous root growth and more panicle numbers were observed in MST system [14,15].However, unlike the MT system where rice seedlings are transplanted in lines and rows, the rice seedlings in the MST system are randomly and unevenly distributed in the field.Such traits may reduce the ventilation ability and increase the relative air humidity of the MST population especially under high nitrogen input and high planting density condition, which increases the risk of pest/disease infection and lodging and thus lead to a decrease in grain yield.In addition, although the rice transplanting machine has been developed and adopted widely in China, little attention has been paid to the mechanical seedling throwing system.
Recently, an orderly mechanical rice seedling throwing system (OMST) was developed using a seedling-throwing machine (manufactured by Zoomlion Co., Ltd., Changsha, China).The OMST throws the rice seedlings to the field by machine but the seedlings distribution is similar with MT arranged in lines and rows.The OMST has been rapidly promoted in several provinces of China including Hunan, Anhui, Jiangxi, and its advantages in economic profits have been proved in farmers' practices.However, the grain yield performance and the growth traits of the OMST are unknown.It can be speculated that OMST might combine the advantages of MST regarding seedling vigor and fast tiller generation, as well as the advantages of MT regarding orderly population distribution.To test this hypothesis, a two-year field experiment was carried out with three rice establishment systems incorporated, namely orderly mechanical seedling throwing (OMST), manual transplanting (MT) and manual seedling throwing (MST) (Figure 1).The objectives of the present study were to compare the variances between OMST, MT and MST in grain yield, yield components and growth characters, and to determine the extension potential of OMST system in China.
requires high labor input and is no longer suitable for large-scale rice production in China [9].In recent years, several simplified or mechanized rice establishment methods, such as rice direct-seeding (DSR) and machine transplanting, have been extended rapidly to replace manual transplanting (MT), and the characteristics of these establishment methods have also been intensively studied [9][10][11][12].Meanwhile, seedling throwing (MST) is a simplified rice establishment method in which rice seedlings with soil on their roots are thrown by hand into fields [13].In comparison with MT, MST greatly reduces the labor input during transplanting, which increases the economic benefits.Significant variances in growth performance between MT and MST were observed.The setback caused by uprooting and transplanting was either smaller or non-existent in MST than that in MT, which may shorten the growth duration and is regarded as an ideal trait for multiple rice cropping where the rice growing season is strictly limited.Moreover, higher tiller generation, vigorous root growth and more panicle numbers were observed in MST system [14,15].However, unlike the MT system where rice seedlings are transplanted in lines and rows, the rice seedlings in the MST system are randomly and unevenly distributed in the field.Such traits may reduce the ventilation ability and increase the relative air humidity of the MST population especially under high nitrogen input and high planting density condition, which increases the risk of pest/disease infection and lodging and thus lead to a decrease in grain yield.In addition, although the rice transplanting machine has been developed and adopted widely in China, little attention has been paid to the mechanical seedling throwing system.
Recently, an orderly mechanical rice seedling throwing system (OMST) was developed using a seedling-throwing machine (manufactured by Zoomlion Co., Ltd., Changsha, China).The OMST throws the rice seedlings to the field by machine but the seedlings distribution is similar with MT arranged in lines and rows.The OMST has been rapidly promoted in several provinces of China including Hunan, Anhui, Jiangxi, and its advantages in economic profits have been proved in farmers' practices.However, the grain yield performance and the growth traits of the OMST are unknown.It can be speculated that OMST might combine the advantages of MST regarding seedling vigor and fast tiller generation, as well as the advantages of MT regarding orderly population distribution.To test this hypothesis, a two-year field experiment was carried out with three rice establishment systems incorporated, namely orderly mechanical seedling throwing (OMST), manual transplanting (MT) and manual seedling throwing (MST) (Figure 1).The objectives of the present study were to compare the variances between OMST, MT and MST in grain yield, yield components and growth characters, and to determine the extension potential of OMST system in China.

Site Description
The present study was conducted at Qianshanhong Town (29 • 19 N, 112 • 15 E), Yiyang county, Hunan Province, China during 2021 and 2022 growing season (Figure 2).Before the experiment, the soil from the 0-20 cm layer was randomly collected and the soil nutrients content were then determined by the automatic Kjeldahl apparatus method (nitrogen), microwave digestion-flame photometry method (phosphorus and potassium), potassium dichromate volumetric method (organic matter).The soil pH was determined by an au-tomatic soil pH meter (FieldScout PH400, manufactured by SPECTRUM Techno.Inc., Haltom City, TX, USA) The soil was clay loam with soil pH = 6.77.The available nitrogen, phosphorus, potassium, and organic matter were 117.37 mg kg −1 , 15.57 mg kg −1 , 111.07 mg kg −1 and 27.9 g kg −1 , respectively.The temperature (daily maximum, daily average, and daily minimum) and rainfall during the rice-growing season at the experimental sites in 2021 and 2022 are shown in Figure 3.

Site Description
The present study was conducted at Qianshanhong Town (29°19′N, 112°15′E), Yiyang county, Hunan Province, China during 2021 and 2022 growing season (Figure 2).Before the experiment, the soil from the 0-20 cm layer was randomly collected and the soil nutrients content were then determined by the automatic Kjeldahl apparatus method (nitrogen), microwave digestion-flame photometry method (phosphorus and potassium), potassium dichromate volumetric method (organic matter).The soil pH was determined by an automatic soil pH meter (FieldScout PH400, manufactured by SPECTRUM Techno.Inc., Haltom City, TX, USA) The soil was clay loam with soil pH = 6.77.The available nitrogen, phosphorus, potassium, and organic matter were 117.37 mg kg −1 , 15.57 mg kg −1 , 111.07 mg kg −1 and 27.9 g kg −1 , respectively.The temperature (daily maximum, daily average, and daily minimum) and rainfall during the rice-growing season at the experimental sites in 2021 and 2022 are shown in Figure 3.

Experimental Design
The experiment was arranged in randomized complete block design under split plot arrangement with three replications.Three different rice establishment methods, viz., orderly mechanical seedling throwing (OMST), manual transplanting (MT), and manual seed-ling throwing (MST) were arranged in the main plot, and two elite inbred varieties

Experimental Design
The experiment was arranged in randomized complete block design under split plot arrangement with three replications.Three different rice establishment methods, viz., orderly mechanical seedling throwing (OMST), manual transplanting (MT), and manual seed-ling throwing (MST) were arranged in the main plot, and two elite inbred varieties that are widely adopted by local farmers, Xiang-zaoxian24 (XZX24) and Zhongzao39 (ZZ39) were arranged in the sub-plot.
In MT plots, the rice seeds were sown in nursery bed and 25-day-old seedlings were transplanted at a hill spacing of 25 cm × 12 cm with 4 seedlings per hill.In MST and OMST plots, the rice seeds were sown into the holes of the specialized plastic seedling trays with 4 seeds per hole.After covering mud on the surface, the plastic trays were transferred to nursery bed.At 25 days after sowing, the seedlings for MST were manually thrown into the plot at the planting density of 33 hills per m 2 , while the seedlings for OMST were mechanically thrown to the plots at hill spacing of 25 cm × 12 cm.Before transplanting, the seedling numbers per hole in the plastic trays were carefully checked and adjusted to ensure the same seedling numbers in each plot.The rice seeds for all the plots were sown on 24 March and transplanted on 18 April in both 2021 and 2022.A fertilizer dose of 150:32:124 of N:P:K kg ha −1 was equally applied to each plot.All the P, one third of the N, and half of the K were applied as a basal starter dose, while the residual N was equally split at the middle tillering stage and the panicle initiation stage, and the other 50% of the potassium was top-dressed during panicle initiation.Both basal and top-dressing fertilizers were manually broadcasted to each plot.All the plots were flood irrigated immediately after transplanting at the water depth of 1-3 cm, and at 7 days before harvesting, the field was drained.At early tillering, 25% cyhalofop-butyl was spray-applied to control weed infection.20% abamectin, 10% imidacloprid, 20% tricyclazole and 5% validamycin were sprayed applied at middle tillering and heading to prevent pest and disease infection.

Observations 2.3.1. Grain Yield and Yield Components
At maturity, 15 plants near the center of each plot were sampled to determine yield components, aboveground total biomass.Panicle number was counted in each sample to determine the panicle number per m 2 , and then plants were separated into straw and panicles.Through hand-threshing, all spikelets parted from the rachis were submerged into the tap water to separate the filled grains from the others.Further screening was completed by a winnowing cleanliness instrument (FJ-1; China Rice Research Institute, Hangzhou, China) to separate the half-filled spikelets from the unfilled spikelets.Three sub-samples of 30.0 g of filled spikelets, 2.0 g of unfilled spikelets, and all of the half-filled spikelets were taken to count the number of spikelets.After oven-drying at 70 • C to constant weight, dry weight of straw, rachis and filled, half-filled and unfilled spikelets were determined.Grain yield was determined from a 5 m 2 area in each plot and adjusted to the standard moisture content of 0.14 g H 2 O g −1 fresh weight.

Growth Analysis
Aboveground biomass was determined at middle tillering, panicle initiation, heading and maturity stage by collecting samples of 12 plants.Samples oven-drying at 70 • C to constant weight.Leaf area index (LAI, which was calculated as the total leaf area of the plants divide the land area that growth the plants) was assessed manually in 2022 at middle tillering, panicle initiation and heading stage.The leaf area was measured as the length of green leaf (from leaf base to leaf tip) multiplied by the maximum width of the blade and an empirical shape factor (0.75).A chlorophyll meter (SPAD-502, Manufactured by Konica Minolta Ltd.Tokyo, Japan) was used to obtain SPAD values of intact flag leaf at heading and maturity.Three SPAD readings were taken around the midpoint of each leaf blade.15 SPAD readings from the main stem of the five plants that near the plot center were averaged to represent the mean SPAD value of each plot.

Tiller Numbers
Excluding the three border plants, 10 hills were labeled in each plot to count tillers at fixed intervals from 12 days after transplanting to 54 days after transplanting.

Statistical Analysis
Data were analyzed by analysis of variance using Statistix 9.0 (manufactured by Analytical Software, Tallahassee, FL, USA).The differences between treatments were separated using the least significance difference (LSD) test at the 0.05 probability level.Graphical representation of the data was performed using Sigmaplot 12.5 (manufactured by Systat Software, Inc., Palo Alto, CA, USA).

Grain Yield and Yield Components
The grain yield of OMST were similar or higher than MT and was significantly increased compared with MST (except for XZX24 in 2022 where the yield difference did not reach significant level) (Figure 4).Averaging across years, the grain yield observed in OMST system was 7.17 and 7.18 t ha −1 , which was increased by 16.1% and 15.3% for XZX24 and ZZ39, respectively, as compared with MST.When compared with MT, the grain yield of OMST showed no significant difference for XZX24 in 2021, and for both varieties in 2022.However, for XZX24 in 2021, OMST significantly increased the grain yield by 7.1% as compared with MT.Above results indicated the better yield performance of OMST than MST.The grain yield components of the three establishment methods in 2021 and 2022 were shown in Tables 1 and 2, respectively.The panicle number of OMST was significantly higher than that observed in MT and MST (except for the MST of ZZ39 in 2022, which did not reach significant level).For spikelet number per panicle, significant variances between years were observed.The spikelet number per panicle in OMST was significantly lower than MT in 2021 but did not show a significant difference with MT in 2022.Except for XZX24 in 2021, the lowest spikelet number per panicle was observed in MST among the three establishment methods.In addition, although the grain filling percentage did not vary across the three establishment methods in 2021, the grain-filling percentage showed a trend of MT > OMST > MST (p < 0.05) in 2022.Besides, the grain weight did not show difference among establishment methods, which was consistent across years.The yield component analysis indicated that the OMST showed an improved panicle formation ability than MT and MST.Note: In each column, different lowercase letters denote statistical differences between treatments of a variety at the 5% level according to LSD test.** denotes the differences were statistically significant across treatments, ns denotes the differences across treatments did not reach significant level.OMST: Orderly mechanical seedling throwing, MT: Manual transplanting, MST: Manual seedling throwing, XZX24: Xiangzaoxian24, ZZ39: Zhongzao39.

Aboveground Biomass
The aboveground biomass of the three establishment methods at middle tillering stage (T), panicle initiation stage (PI), heading stage (HD) and physiological maturity stage (PM) were presented in Figure 5.When compared with MT, the OMST showed a better bio-mass accumulation ability before grain filling, which was significantly increased by 67.4% and 48.5% at T, by 49.8% and 25.3% at PI, and by 4.3% and 12.3% at HD, for XZX24 and ZZ39, respectively, averaged across years.At PM, although the aboveground biomass in OMST was significantly higher than MT in 2021, no significant difference was observed between the two treatments in 2022.Mean-while, the biomass accumulation of MST from T to HD was similar or lower than that observed in OMST, and was similar or higher than that of MT.In maturity, the lowest aboveground biomass was observed in MST (except for ZZ39 in 2021, the biomass in MST did not vary with MT).
The OMST significantly improved the biomass accumulation after heading as compared with MST (Figure 6).When averaged across years, the biomass accumulation after heading of the OMST was significantly increased by 50.6% and 24.2% for XZX24 and ZZ39, respectively, as compared with that of MST.Meanwhile, the biomass accumulation of OMST was significantly by increased 28.6% and 29.5% for XZX24 and ZZ39 in 2021, respectively as compared with MT.In 2022, however, no significant variance was observed between OMST and MT on biomass accumulation after heading.

The Dynamics of Tiller Number
Significant faster tiller generation speed on OMST than that in MT was observed in both varieties and in both years (Figure 7).At 18 days after transplanting (DAT), the tiller number of OMST was 194.7 per m 2 and 216.7 per m 2 for XZX24 and ZZ39 in 2021, and was 336.6 per m 2 and 246.6 per m 2 for XZX24 and ZZ39 in 2022, which was increased by 21.3-71.0%as compared with that of MT.The advantages of OMST in tiller generation was maintained from 18 DAT to 54 DAT, in which the tiller number per unit area of OMST was increased by 9.6% and 11.2% for XZX24 and ZZ39, respectively, as compared with MT, averaged across years.In MST system, similar or even higher tiller generation speed than that of OMST was observed at the early growth stage.However, when the tiller number reached the maximum level, the tiller number decreased faster in MST than that in OMST.At 54 DAT, 8.8% and 9.5% higher tiller numbers in OMST than in MST was observed for XZX24 and ZZ39, respectively, when averaged across years.

Leaf Area Index
The leaf area index (LAI) of the three establishment methods at different growth stages are shown in Figure 8.For both varieties, the LAI of OMST was significantly higher than that of MT and did not show significant difference as compared with that of MST at middle tillering stage.At the PI and HD stage, the LAI of the OMST was highest amongst the establishment methods, which was in-creased by 5.4% and 17.6% at PI, respectively, as compared with MT and MST, and was increased by 15.1% and 29.0% at HD, respectively, as compared with MT and MST.

The Dynamics of Tiller Number
Significant faster tiller generation speed on OMST than that in MT was observed in both varieties and in both years (Figure 7).At 18 days after transplanting (DAT), the tiller number of OMST was 194.7 per m 2 and 216.7 per m 2 for XZX24 and ZZ39 in 2021, and was 336.6 per m 2 and 246.6 per m 2 for XZX24 and ZZ39 in 2022, which was increased by 21.3-71.0%as compared with that of MT.The advantages of OMST in tiller generation was maintained from 18 DAT to 54 DAT, in which the tiller number per unit area of OMST was increased by 9.6% and 11.2% for XZX24 and ZZ39, respectively, as compared with MT, averaged across years.In MST system, similar or even higher tiller generation speed than that of OMST was observed at the early growth stage.However, when the tiller number reached the maximum level, the tiller number decreased faster in MST than that in OMST.At 54 DAT, 8.8% and 9.5% higher tiller numbers in OMST than in MST was observed for XZX24 and ZZ39, respectively, when averaged across years.

Leaf Area Index
The leaf area index (LAI) of the three establishment methods at different growth stages are shown in Figure 8.For both varieties, the LAI of OMST was significantly higher than that of MT and did not show significant difference as compared with that of MST at The leaf area index (LAI) of the three establishment methods at different grow stages are shown in Figure 8.For both varieties, the LAI of OMST was significantly high than that of MT and did not show significant difference as compared with that of MST middle tillering stage.At the PI and HD stage, the LAI of the OMST was highest among the establishment methods, which was in-creased by 5.4% and 17.6% at PI, respective as compared with MT and MST, and was increased by 15.1% and 29.0% at HD, respe tively, as compared with MT and MST.

Leaf SPAD Reading
The SPAD reading of the flag leaf in OMST was lower than that in MT at HD stage (Figure 9).While no significant difference was observed between OMST and MST treatment at HD.At maturity, the SPAD value of OMST did not vary with that of M, and was 16.8% and 5.3% higher than that of MST for XZX24 and ZZ39 respectively (although the variance in ZZ39 did not reach a significant level).The above results suggested that the OMST may have better photosynthetic ability than MST during grain filling stage.

Leaf SPAD Reading
The SPAD reading of the flag leaf in OMST was lower than that in MT at HD s (Figure 9).While no significant difference was observed between OMST and MST t ment at HD.At maturity, the SPAD value of OMST did not vary with that of M, and 16.8% and 5.3% higher than that of MST for XZX24 and ZZ39 respectively (although variance in ZZ39 did not reach a significant level).The above results suggested tha OMST may have better photosynthetic ability than MST during grain filling stage.

Discussion
Developing an efficient and high-yielding mechanical rice establishment syste one of the most important approaches for intensive and large-scale rice production general, rice can be established by either transplanting, direct seeding or seedling thr ing [16].In recent years, several machine transplanting and machine direct-seeding tems have been successfully developed, and the yield and growth performance, the o mum agronomic practices, such as nitrogen rate, and planting density, have also b widely elucidated [9][10][11][12].In rice seedling throwing systems, however, the mechan equipment and the relative technology is still lacking.The results of the present st showed that the newly developed orderly mechanical seedling throwing system re sented significant yield advantages compared to traditional manual seedling throw suggesting the potential of OMST system as a replacement to traditional MST for rice duction in the future.In present study, significant variances in yield components w observed between OMST and MST.Although the tiller generation speed at early gro

Discussion
Developing an efficient and high-yielding mechanical rice establishment system is one of the most important approaches for intensive and large-scale rice production.In general, rice can be established by either transplanting, direct seeding or seedling throwing [16].In recent years, several machine transplanting and machine direct-seeding systems have been successfully developed, and the yield and growth performance, the optimum agronomic practices, such as nitrogen rate, and planting density, have also been widely elucidated [9][10][11][12].In rice seedling throwing systems, however, the mechanical equipment and the relative technology is still lacking.The results of the present study showed that the newly developed orderly mechanical seedling throwing system represented significant yield advantages compared to traditional manual seedling throwing, suggesting the potential of OMST system as a replacement to traditional MST for rice production in the future.In present study, significant variances in yield components were observed between OMST and MST.Although the tiller generation speed at early growth stages was similar between OMST and MST, the final tiller number, panicle number and total spikelet number in OMST was significantly higher than in MST, which resulted in the yield advantages in OMST.It has been reported that the uneven distribution of the seedling in MST population inhibit the tiller generation rate and accelerate the death of existed tillers [17].The higher panicle-bearing tiller rate and sink size in orderly distributed rice population have also been observed as compared with that in in-orderly ones [18].
The shift from the in-orderly population in MST to the orderly ones in OMST is beneficial for the improvement in field micro-environment.It has been suggested that the orderly distribution of rice seedlings in OMST greatly improved the wind velocity as compared with MST [19].The light transition rate was improved in orderly distributed rice population [20].Such changes might be responsible for the changes of growth characters in OMST.Our study found that the total biomass and the biomass accumulation after heading was significantly higher in OMST than in MST, which contribute to similar grain filling percentage and grain weight between the two system even though the sink size in OMST was significantly improved.The larger sink size generally requires better dry matter assimilation or translocation ability during grain filling [21], which is influenced by the canopy structure, the light interception rate, as well as the leaf function duration [22][23][24].While in present study, the improved dry matter accumulation after heading in OMST than in MST might be due to: (1) the higher LAI that might increase light interception and improve radiation use efficiency; (2) the higher SPAD value at HD and the lower SPAD decrease from HD to PM that might enhance the leaf photosynthetic ability and reduce the leaf senescence rate during grain filling.Previous research found that the root dry matter and root activity was higher in seedling throwing system than in seedling transplanting [25], and the vigor of the root system are highly correlated with the leaf senescence [26].However, the differences between OMST and MST with regards to root traits, leaf photosynthetic ability, as well as the light interception and radiation use efficiency remained to be unknown, which need to be addressed in future studies.
The grain yield of the OMST showed no significant difference or even higher than that of traditional manual transplanting, which partially reflected the yield stability and the extension potential of this system.The yield component analysis suggested that the OMST possessed better panicle generation ability than MT, which was mainly attributed to more tillers and biomass accumulation especially at vegetative growth stages.This result was consistent with previous research in which the rice seedling throwing system was characterized as earlier tillering, higher plant density and more panicles in comparison with the seedling transplanting system [27,28].However, the significant higher spikelet number in 2021 and significantly improved grain filling percentage in 2022 in MT system compensated the decrease in panicle number as compared with OMST, which let to equivariant grain yield between MT and OMST.In addition, the grain yield of the MST showed no difference or lower than MT in present study, which was in consistent with several research studies [29][30][31].However, the research of Sanbagavalli denoted that the grain yield in MST was higher than that in MT [32].The variances between different research studies might be attributed to the differences in climate condition, genotype and agronomic practices [27,33] and the yield performance of OMST against MT and MST needs to be examined further.Significant variances in spikelet per panicle between years and treatments were observed.In 2021, The spikelet per panicle in MT was higher or similar than OMST and MST, while the grain-filling percentage showed no difference between treatments, which resulted in increased filled spikelet number per panicle in MT.While in 2022, although the spikelet number per panicle of MT showed no difference to MST and OMST, the significantly improved grain-filling percentage also resulted in improved filled spikelet number per panicle in MT.Nevertheless, the mechanisms underlying the variances in yield components among different establishment methods responding to years and locations remain unknown and need to be addressed in the future.Moreover, the comparison between OMST and other rice establishment methods, such as machine transplanting, direct seeding remain unknown and need to be addressed in future studies.
The growth and the yield formation characters of OMST showed significant variances to that of MT and MST, indicating that the optimum crop management methods for OMST might be varied with the other crop establishment methods.However, the responses of the OMST system to different planting density, fertilizer and water management strategies remains unknown and high-yielding and efficient cultivation technology for OMST need to be developed in the future.

Conclusions
The grain yield of the newly developed orderly mechanical seedling throwing system was significantly higher than manual seedling throwing and was equivalent to that of manual transplanting.The yield components analysis showed that the differences in grain yield among the three establishment methods were mainly attributed to the variances in panicle number and total spikelet number.Further analysis suggested that the orderly mechanical seedling throwing takes the advantages of higher biomass accumulation after heading, increased leaf area index and decreased leaf senescence rate against manual seedling throwing, and more tillers and biomass accumulation at vegetative growth stage against manual transplanting.Such advantages might be attributed to the combined effects of seedling throwing and orderly seedling distribution.The present study demonstrated that the orderly mechanical seedling throwing is an efficient and high-yielding rice establishment method that might be a promising option to replace traditional manual seedling throwing in rice production.

Figure 1 .
Figure 1.Pictorial illustration of the field performance of orderly mechanical seedling transplanting (a,b); manual transplanting (c); and manual seedling throwing (d).

Figure 1 .
Figure 1.Pictorial illustration of the field performance of orderly mechanical seedling transplanting (a,b); manual transplanting (c); and manual seedling throwing (d).

Figure 2 .
Figure 2. The location of the experimental site.Figure 2. The location of the experimental site.

Figure 2 . 14 Figure 3 .
Figure 2. The location of the experimental site.Figure 2. The location of the experimental site.Agronomy 2022, 12, x FOR PEER REVIEW 4 of 14

Figure 3 .
Figure 3. Temperature (daily maximum, daily average and daily minimum) and rainfall during the rice-growing season at the experimental sites in 2021 (a,c); and 2022 (b,d).

Figure 9 .
Figure 9.The SPAD reading of the flag leaf under different rice establishment methods: (a) X zaoxian24; and (b) Zhongzao39.OMST: Orderly mechanical seedling throwing, MT: Manual t planting, MST: Manual seedling throwing, XZX24: Xiangzaoxian24, ZZ39: Zhongzao39, T: M tillering stage, PM: Physiological maturity stage, HD: Heading stage.Different lowercase letter note statistical differences among treatments of a cultivar at the 5% level according to LSD test.E bars above mean indicate standard error (n = 3).

Table 1 .
The grain yield components under different rice establishment methods in 2021.In each column, different lowercase letters denote statistical differences between treatments of a variety at the 5% level according to LSD test.** denotes the differences were statistically significant across treatments, ns denotes the differences across treatments did not reach significant level.OMST: Orderly mechanical seedling throwing, MT: Manual transplanting, MST: Manual seedling throwing, XZX24: Xiangzaoxian24, ZZ39: Zhongzao39.

Table 2 .
The grain yield components under different rice establishment methods in 2022.