The relationships between maize(Zea mays L.)lodging resistance and yield formation depend on dry matter allocation to ear and stem

Ping Zhng,Shungcheng Gu,Yunyun Wng,Chenchen Xu,Yting Zho,Xioli Liu,Pu Wng,*,Shouing Hung,*

a College of Agronomy and Biotechnology,China Agricultural University,Beijing 100193,China

b College of Agronomy,South China Agricultural University,Guangzhou 510642,Guangdong,China

Keywords:Corn Lodging Yield formation Physical traits Dry matter allocation

ABSTRACT Lodging is a critical constraint to yield increase.There appear to be tradeoffs between yield formation and lodging resistance in maize.Hypothetically,it is feasible to reduce lodging risk as well as increase grain yield by optimizing dry-matter allocation to different organs under different environments.A three-year field experiment was conducted using four maize cultivars with differing lodging resistances and five growing environments in 2018–2020.Lodging-susceptible(LS)cultivars on average yielded more than lodging-resistant(LR)cultivars when lodging was not present.The yield components kernel number per ear(KN)and thousand-kernel weight(TKW)were both negatively correlated with lodging resistance traits(stalk bending strength,rind penetration strength,and dry matter weight per internode length).Before silking,the LR cultivar Lishou 1(LS1)transported more assimilates to the basal stem,resulting in a thicker basal stem,which reduced dry matter allocation to the ear and in turn KN.The lower KN of LS1 was also due partly to the lower plant height(PH),which increased lodging resistance but limited plant dry matter production.In contrast,the LS cultivars Xianyu 335(XY335)and Xundan 20(XD20)produced and allocated more photoassimilates to ears,but limited dry matter allocation to stems.After silking,LS cultivars showed higher TKW than LR cultivars as a function of high photoassimilate productivity and high assimilate allocation to the ear.The higher lodging resistance of LS1 was due mainly to the greater assimilate allocation to stem after silking and lower PH and ear height(EH).High-yielding and high-LR traits of Fumin(FM985)were related to optimized EH and stem anatomical structure,higher leaf productivity,low assimilate demand for kernel formation,and assimilate partitioning to ear.A high presilking temperature accelerated stem extension but reduced stem dry matter accumulation and basal stem strength.Post-silking temperature influences lodging resistance and yield more than other environmental factors.These results will be useful in understanding the tradeoffs between KN,KW,and LR in maize and environmental influences on these tradeoffs.

1.Introduction

Maize(Zea mays L.)is grown on more than 194 million hectares worldwide with a global mean yield of 5.64 t ha-1[1].Given that maize has the largest yield potential in comparison with other staple crops including wheat(Triticum aestivum L.)and rice(Oryza sativa L.)[2],raising its yield is a global priority for food and feed security[3,4].Lodging of maize has been estimated to cause annual global yield losses of 5%–20%[5,6].Lodging also reduces grain quality and impedes mechanical harvest[7,8].Lodging will become a larger challenge as plant density continuously increases to increase yield in maize production[9].

Maize stalk lodging[13]normally occurs at the basal stem(such as by breakage at the third or fourth stem internode above the ground)from early growing season to maturity[10–12].Several indicators including stalk bending strength(SBS,in N m),rind penetration strength(RPS,in MPa),and dry matter weight per unit length(DWUL,in g cm-1)are closely associated with maize stalk lodging resistance[12,14–16],and widely used as assessment of maize stalk lodging resistance[14,17,18].Maize stalk strength is affected by many morphological traits.Stalk mechanical strength decreases with stalk internode diameter and as the length of the basal internode increases[19].Maize stalk lodging resistance also depends on plant architecture traits including plant height(PH),ear height(EH),and height at center of gravity[12].Reductions in PH,EH,and height at center of gravity as a result of plant growth regulator application can increase stalk lodging resistance[20,21].Changes in plant architecture and stem are associated with dry matter accumulation and allocation.Stem elongation determines PH and EH depends on photosynthates from maize leaves[22].Stems will become thinner and weaker under shaded or highplant-density conditions as leaf productivity decreases[21,23,24].Lodging resistance depends on assimilate partitioning to roots and stems that can affect synthesis of structural substances such as lignin and cellulose.As estimated by Foulkes et al.[25]and Pi?era-Chavez et al.[26],a lodging-proof wheat ideotype required 7.9 t ha-1stem biomass and 1.0 t ha-1root biomass in the top 10 cm of soil,which resulted in strong basal stems and root systems.However,previous studies have focused mainly on the relationship between lodging resistance and physical traits,rarely considering relationships between lodging-associated physical traits and dry matter accumulation and partitioning to different organs.

There are tradeoffs between KN formation,kernel weight,and lodging resistance[26,27],but they have not been well characterized.Maize kernel formation takes place in a critical period during anthesis(～15 d pre-to 15 d post-anthesis)[28–30],mostly coincident with stem elongation.KN formation depends on the plant growth rate(PGR,the ratio of biomass accumulated during the period to the duration of the period[31])and the proportion of dry matter that is allocated to the ear during the critical period[29,32,33].Vega et al.[32]reported that the PGR threshold was 1 g plant-1d-1,below which no maize kernels were set.Thus,the development of KN may compete for assimilates with stem elongation.The competition intensity becomes larger with time before flowering,because root growth,stem elongation,ear growth,and silk extension all accelerate in this period[34,35].Assimilate allocation between stem and ear can affect both kernel weight and lodging resistance.More than 50% of post-anthesis assimilates will be allocated from leaves to the ear during grain filling period[36–38].Pre-anthesis nonstructural carbohydrates that are stored in the stem can be transported to the ear during grain filling,reducing stalk quality[12].Therefore,relationships between yield formation and lodging resistance need to be carefully understood.

The growth environmental conditions affect dry matter accumulation and partitioning,yield formation,and lodging resistance[14,17,37].Temperature and solar radiation(SR)are the primary environmental factors that affect crop development and biomass accumulation.Pre-flowering high temperature shortens the vegetative growth period,reducing solar radiation accumulation[38,39],plant dry matter weight,kernel set,and grain yield.During grain filling,variations in temperature and radiation strongly affect growth rate and duration of grain filling and thereby kernel weight and maize yield[40].Hou et al.[41]found that 1 °C increase in mean temperature and 1% decrease in cumulative photosynthetically active radiation resulted in respectively 0.83 and 0.15 t ha-1maize grain yield reduction,in agreement with findings in wheat[42].With respect to lodging resistance,quantitative analysis[14]showed that for every 1 MJ m-2reduction in total intercepted photosynthetically active radiation,the SBS,RPS,and DWUL of the basal stem decreased by respectively 0.0667 N,0.075 N,and 0.06 mg g-1.Likewise,high temperature has negative effects on root lodging resistance in canola(Brassica napus L.),as indicated by a large reduction in root anchorage[43].However,there have been few studies focusing on how maize lodging resistance is affected by growth environment(as measured by effective cumulative temperature and the cumulative SR).And although it is known that growth environment affects yield formation,few studies have considered the effects of environment on both yield and lodging resistance from the perspective of dry matter accumulation and partitioning.

The objectives of this study were(i)to identify the relationship between yield components(KN and TKW)and lodging resistance,(ii)to explain the relationship between yield and lodging resistance in terms of dry matter accumulation and partitioning,and(iii)to evaluate the effects of growth environment on the formation of yield and lodging resistance in maize.

2.Materials and methods

2.1.Experimental site and design

Experiments were arranged in a randomized complete block design with two factors(environment and maize cultivar)and were conducted in 2018,2019,and 2020 at Wuqiao Experimental Station(37°41′N,116°37′E)of China Agricultural University,Hebei province.To assure a wide range of environmental conditions,five different sowing dates in three years were used(Table 1):two dates in 2018 and 2019(April 28,2018 and May 10,2019)and three dates in 2020(May 3,May 23 and June 13).These five sowing dates represented five different environments and were named En1 to En5.The daily minimum and maximum temperature(Tminand Tmax),rainfall,and wind gust speed(at 2 m height)during the growing season are shown in Fig.1.Based on our previous results[44],two lodging-susceptible(LS)cultivars Xianyu 335(XY335)and Xundan 20(XD20)and two lodging-resistant(LR)cultivars Lishou 1(LS1)and Fumin 985(FM985)were used.All cultivars were planted at a density of 75,000 plants ha-1and the plot sizes were 45.6 m2(9.5×4.8 m)in 2018–2019 and 103.2 m2(12.9×8 m)in 2020.

Table 1 Timings of sowing date,growth stages,durations of pre-silking and silking-harvest phase,and growing degree days(GDD)and solar radiation(SR)during growth periods of four cultivars in five environments in 2018–2020.

The soil in the experimental field was a sandy loam,and the 0–20 cm soil profile contained 9.7 g kg-1organic matter,0.6 g kg-1total nitrogen,10.8 mg kg-1available phosphorus,and 141.9 g kg-1available potassium.At sowing,60 kg N ha-1,120 kg P2O5ha-1,and 100 kg K2O ha-1were applied,and an extra 180 kg N ha-1was added at the V13 stage(when the collar of the 13th leaf was visible).Sufficient water was applied to prevent water stress.Flood irrigation of approximately 60 mm was applied to each plot one week before sowing to achieve uniform seedling emergence,and a further 50 mm flood irrigation was applied at the V13 stage after fertilization in all three years.Weeds,insects,and diseases were well controlled during the growing season.

2.2.Experimental sampling and measurements

2.2.1.Environmental factors

Daily weather data(Tmax,Tmin,and sunshine hours)at the experimental site during maize growing seasons in 2018–2020 were recorded(Chinese Meteorological Administration,2018–2020).

The effective cumulative temperature(GDD)during different growth periods was calculated as follows:

where GDDk-iis the GDD during the growth period from days k to i;Tbaseis the basic effective temperature(10 °C for maize)[45,46].

The cumulative SR during the growth period from days k to i was calculated as follows:where Q is the astronomical radiation,Q1is the proportion of sunshine(actual sunshine hours divided by possible sunshine hours),and α and β are correction coefficients[40].

The five environments had differing temperature,rainfall,and SR during the maize growing season(Fig.S1;Table 1).En2 had the highest mean daily temperature and Tmax,and En5 had the lowest mean daily temperature and Tmin.En1 had the highest cumulative rainfall(531.6 mm),compared to that(334.0–374.6 mm)in En2–5.

On average across cultivars,the mean growth period of En1 was larger than that of other environments(Table 1).The pre-silking GDD varied from 956.1°C(En4)to 1117.2°C(En2),and the cumulative SR varied from 1044.8 MJ m-2(En5)to 1543.6 MJ m-2(En1).During the post-silking period,En5 displayed the lowest GDD,En4 displayed the lowest SR,and En1 had the largest GDD and SR compared to the other environments.During the entire growth season,the GDD was similar between En1 and En2 which was higher than that of the other environments.The total SR was highest in En1 and lowest in En4 and En5.

2.2.2.Grain yield,yield components,harvest index and lodging proportion at harvest

The measurement procedures for yield,yield components,harvest index,and lodging percent were as previously described[44].At harvest,all ears in the two middle rows were collected from each plot.KN per ear and the TKW of 20 representative ears were measured.TKW was determined based on 500-kernel weight of three samples dried at 80 °C to constant weight.KN per unit area was calculated from KN and ear number m-2.Harvest index(HI)was estimated as grain weight divided by aboveground biomass weight at harvest.Plants were considered to be stem-lodged when the stem was broken at or below the ear node and to be rootlodged when leaning more than 45° from the vertical[47].The number of lodged plants in the two middle rows of each plot at harvest was recorded.The lodging proportion was calculated as the lodged-plant number divided by the total plant number of the two central rows in each plot.Lodging events occurred in the late growth period(close to harvest)in En1(August 14)and En2(August 9–12).

Fig.1.Correlations among yield,yield components,and lodging resistance among environments and cultivars.KN,kernel number per ear;TKW,thousand-kernel weight;RPS,rind penetration strength;SBS stalk bending strength;DWUL,dry matter weight per internode length.

2.2.3.Plant and internode morphology

Three representative plants per plot were cut close to the ground.PH and EH were measured with a ruler.The third internodes were cut from the basal stems to determine mean diameter(SD,in mm),internode length(IL,in cm),and cross-sectional area(CSA,in mm2).The mean diameter and CSA at the middle point of the third internode were calculated as.

where a and b are the respective diameters of the long and short axes of the stem[44,48,49].

2.2.4.Assessment of stalk lodging resistance

The measurement procedure for stalk lodging resistance was as described in our previous studies[44,49].At silking and harvest(detailed measurement days are presented in Table 1),stem bending resistance(SBS)was measured with a Stalk Strength Tester(YYD-1;Zhejiang Top Instrument Co.,Ltd.,Hangzhou,Zhejiang,China).The U-shaped probe of the tester was vertically lowered against the middle of the internode at approximately 200 mm min-1until it broke,and the maximum force(Fmax)was recorded.The bending strength of the stem(SBS,in N m)was calculated as.

where L is the distance between the two fulcra[14].

The RPS of the third internode was measured with the same stalk-strength tester.A test probe was vertically inserted into the middle of the third internode at a slow and uniform speed.The maximum force(FRmax)to penetrate the stalk epidermis was recorded and the RPS was calculated as.

where 0.01 is the area(cm2)of the test probe[9,10].

Finally,the third stem internode was oven dried at 80°C to determine the dry weight per unit internode length(DWUL,g cm-1).

2.2.5.Dry matter accumulation and partitioning

At silking and harvest stages,three representative plants per plot were sampled to determine dry matter accumulation and partitioning.Each plant was separated into leaf,ear,and stem and each organ was oven dried at 80°C to constant weight to measure the dry matter of each organ(DM-x).Dry matter allocation to each organ at the silking stage(x%),and dry matter accumulation after silking of each organ(ΔDM-x)were calculated as follows:

where x indexes the total plant,stem,leaf,and ear[50,51].

2.3.Statistical analyses

Analyses of variance of yield traits,lodging parameters,physical characteristics,and dry matter accumulation and partitioning traits were performed with PROC MIXED in SAS 9.3(SAS Institute,Cary,NC,USA).Effects of environments and cultivars and their interactions were taken as fixed and effects of replicates as random.Significance of differences between environment and cultivars treatment means were determined by a t-test(α＜0.05)with Satterthwaite’s approximation.

Principal component analysis(PCA)was conducted in R 4.0.0[52]to(a)identify relationships among yield traits,lodgingresistance traits,plant and basal internode physical characteristics,and dry matter accumulation and partitioning traits of the four cultivars and(b)characterize the dissimilarity among the four cultivars(Fig.2).A collinearity test among these variables was performed before PCA analysis using SPSS 20.0(IBM Corp.,Armonk,NY,USA).Variables were removed if the tolerance was smaller or equal to 0.1.According to tolerance values,SBS and CSA showed collinearity and were removed before further analysis.The maize growth period was divided into two parts:dry matter accumulation and partitioning for KN and lodging resistance formation before silking,and dry matter accumulation and partitioning for TKW increases and lodging resistance maintenance after silking.Taking yield and lodging proportion into consideration,all parameters were classified into three groups(A)yield traits and lodging proportion,(B)KN,lodgingresistance traits,plant and basal internode physical characteristics,and dry matter accumulation and partitioning traits at silking stage,and(C)TKW,lodging-resistance traits at harvest stage,plant and basal internode physical characteristics,and dry matter accumulation and partitioning traits after silking stage.

A structural equation model(SEM)was fitted in R using the lavaan package to connect the direct effects of the environment on both yield and lodging resistance parameters(Fig.3).This analysis can also identify the relationship of yield components and lodging resistance with morphological variables and dry-matter accumulation and partitioning.Only significant explanatory and necessary variables that were strongly associated with both yield components and lodging resistance were considered in SEM analysis.

Fig.2.Principal component analysis(PCA)of(A)yield traits and lodging proportion,(B)kernel number,lodging-resistance traits at silking stage,plant and basal internode physical characteristics,and dry matter accumulation and partitioning traits at silking stage,and(C)thousand-kernel weight,lodging-resistance traits at harvest stage,plant and basal internode physical characteristics,and dry matter accumulation and partitioning traits after silking.Collinearity test among variables was conducted before PCA analysis and stalk bending strength(SBS)and cross-section area(CSA)showed collinearity and were removed.TKW,thousand-kernel weight;KN,kernel number per ear;RPS,rind penetration strength;DWUL,dry matter per internode length;PH,plant height;EH,ear height;IL,length of the third aboveground internode;ID,diameter of the third aboveground internode;DM-stem,dry matter of stem;DM-leaf,dry matter of leaf;DM-ear,dry matter of ear;stem%,dry matter allocation to stem at silking stage;leaf%,dry matter allocation to leaf at silking stage;Ear%,dry matter allocation to ear at silking stage;ΔDM-stem,dry matter accumulation of stem after silking;ΔDM-leaf,dry matter accumulation of leaf after silking;ΔDM-ear,dry matter accumulation of ear after silking.

Fig.3.Structural equation model(SEM)for environmental factors,yield and yield components,lodging-resistance traits,physical characteristics of plant and third aboveground internode and traits of dry matter accumulation and partitioning at silking and harvest stage.KN,kernel number per ear;TKW,thousand kernel weight;GDD,growing degree days;SR,solar radiation;DWUL,dry matter per internode length of third aboveground internode;RPS,rind penetration strength;EH,ear height;PH,plant height;DM-stem,dry matter of stem at silking stage;Ear%,dry matter allocation to ear at silking stage;ΔDM-leaf,dry matter accumulation of leaf after silking;ΔDM-ear,dry matter accumulation of ear after silking.Only significant explanatory and necessary variables that were both highly associated with yield components and lodging resistance were considered in SEM analysis.For example,dry matter of each organ at harvest(DM-stem,DM-leaf,DM-ear),ΔDM-leaf andΔDM-ear were both highly correlated with yield components(TKW)and lodging-resistance traits(DWUL)(Fig.2C),but DM-stem did not differ among cultivars(Table S3)indicating DM-stem was not a determinant of yield and lodging among cultivars.Given that dry matter accumulation after silking of organs plays a more important role in kernel weight increases and lodging resistance maintenance,ΔDM-leaf andΔDM-ear were selected in the SEM model to explain TKW and lodging resistance.The number adjacent to each arrowed line is a factor loading(R2)that shows the variance explained by the variable.*and**indicate significant differences at the 0.05 and 0.01 probability levels,and the dotted line indicates no significant differences.

3.Results

3.1.Relationships among yield,yield components,and lodging resistance

3.1.1.Grain yield,yield components,and harvest index

Environment(En)significantly affected grain yield,yield components,and harvest index(HI),cultivar(C)had significant effects on KN,TKW and grain yield,and their interaction(En×C)had significant effects on grain yield,yield components(except for ear number),and HI(Table 2).

On average across cultivars,ear number of En2 was larger than that of other environments(Table 2).En2 and En4 showed similar values for KN and KN per m2,which were larger than that of En1 and En5.TKW among environments ranked in the order of En5(376.8 g)＞En3 and En4＞En1＞En2(288.3 g).The yield varied from 9408.1 kg ha-1in En1 to 12,138.1 kg ha-1in En4,and yield in En4 was 29.03% and 7.55% larger than that in En1 and En3,respectively.The HI varied from 0.55(En1)to 0.67(En2),and En3 had a value similar to En4(0.66 vs.0.62),greater than that in En5(0.58).

Yield differences among cultivars were significant in En1,3,and 5.LS cultivars XY335 had the largest yield values in En1 and En3,and LR cultivars LS1 had the lowest yield values.KN was significantly different among cultivars in En3–En5,with LS cultivar XD20 on average having the highest KN(496.8 No.ear-1)and LR cultivar LS1 having the lowest KN(435.4 No.ear-1).TKW differences among cultivars were significant in five environments,with XY335 having the largest TKW and LS1 having the lowest values(except for En4).

3.1.2.Lodging percent,RPS,SBS and DWUL

Lodging occurred at the late grain filling stage after typhoon Yagi(August 14)in 2018 and after Lekima(August 9–12)in 2019,and no lodging events were observed in 2020(Table 3).Lodging percent of XY335 and XD20 was larger than that of LS1 and FM985 in 2018 and 2019.

At silking and harvest stage,En,C,and their interactions significantly affected RPS,SBS,and DWUL of the third aboveground internode,except for the effect of En on RPS at harvest(Table 3).

At the silking stage,on average across cultivars,the RPS was largest in En2 compared to other environments(Table 3).The SBS and DWUL were both larger in En1 and En5 than in En3 and En4.RPS differences among cultivars were significant in five environments,with LS cultivars XD20 having the lowest RPS in En1,En3,and En4.LS1 and XY335 have the largest RPS in En1 and En5,respectively.SBS was significantly different among cultivars in En1-4,LS cultivars XY335 and XD20 had the lowest RPS in En2 and En4,respectively.LS1 and XY335 had the largest SBS in En3,and FM985 had the lowest SBS in En1.DWUL differences among cultivars were significant in En1-3,and the DWUL of XD20 was larger than that of other cultivars in En2.The DWUL of FM985 was 17.51% and 23.25% lower than that of other cultivars in En1 and En3,respectively.On average across environments,the SBS of LS1(5.2 N m)was similar to that of XY335(4.9 N m),which was larger than that of FM985(4.5 N m)and XD20(4.7 N m).FM985 and XD20 had the lowest values in DWUL and RPS,respectively.

Table 2 Grain yield,ear number,kernel number per ear,kernel number per m2,thousand kernel weight(TKW),and harvest index(HI)of four cultivars in five environments in 2018–2020.

Table 3 Rind penetration strength(RPS),stalk bending strength(SBS)and dry matter per internode length(DWUL)of the third aboveground internode at silking and harvest stage and lodging percent at harvest stage as affected by environments and cultivars.

At the harvest stage,on average across cultivars,SBS and DWUL among environments both ranked in the order En1＞En4–5＞En2–3(Table 3).RPS was significantly different among cultivars in environments except for En2.LS1 and FM985 had the largest RPS in En1 and En5,respectively;XD20 had the lowest RPS.SBS was significantly different among cultivars in five environments.The SBS of LR cultivars LS1 was larger than that of other cultivars in En1 and En5.The SBS of LS cultivars XY335 was 39.27%,40.52%,and 17.22% lower than that of other cultivars in En2,En3,and En4,respectively.DWUL was significantly different among cultivars in different environments except for En4.The DWUL of LS1 was 29.17%,54.98%,and 63.33% larger than that of other cultivars in En2,En3,and En5,respectively.LR cultivars LS1 had the largest SBS,RPS,and DWUL in all the environments.The LR cultivars FM985 had lower RPS,SBS,and DWUL than LS cultivars.

3.1.3.Relationships among yield,yield components and lodging resistance

Yield was significantly negatively correlated with RPS,SBS,and DWUL(r=0.262,0.498,and 0.520,respectively;Fig.1).KN was significantly negatively correlated with SBS and DWUL(r=0.432 and 0.436).TKW was significantly negatively correlated with RPS(r=0.392).

3.2.Traits affecting yield formation and lodging resistance

3.2.1.Plant and stem internode morphology

Environments(En),cultivars(C),and their interactions(En×C)had significant effects on all plant morphological parameters and the third internode above ground(Table S1).

On average across cultivars,PH of En5 was 9.0% and 11.4%higher than that of En1 and En3,respectively,and was similar to that of En2 and En4(Table S1).En2 displayed the lowest EH(86.3 cm)among five different environments.Compared to the other four environments,En1 on average had a longer and thicker third aboveground internode.

PH varied from 202.1 cm of LR cultivar FM985 in En2 to 279.5 cm of LS cultivar XY335 in En5(Table S1).In most cases(En2,En3,and En4),LS cultivars XY335 and XD20 showed higher PH than LR cultivars LS1 and FM985.Likewise,compared to LS1 and FM985,XY335 and XD20 had higher EH in all five different environments(Table S1;Fig.S2).LS cultivars XY335 had the longest third aboveground internode in En3,En4,and En5,and XD20 displayed the longest third aboveground internode in En1.The averaged internode diameter(ID)varied from 15.1 mm of FM985 in En2 to 22.0 mm of XY335 in En1.

3.2.2.Dry matter accumulation and partitioning

At silking,En,C,and their interactions significantly affected the dry matter weight of each organ and dry matter allocation to each organ,except for the effect of C on dry matter weight of leaf(DMleaf)(Table S2).

Total plant(DM-plant),stem(DM-stem)and leaf(DM-leaf)had the largest dry matter weight in En1 compared to the other environments,and En2 produced the lowest DM-plant and DM-stem(Table S2).En2 had a similar ear dry matter weight(DM-ear)with En1,which was significantly higher than other environments.En1 on average allocated 6.67% lower dry matter to stem(Stem%)and 13.53% higher dry matter to leaf(Leaf%)than other environments.

The DM-plant at silking was significantly different among cultivars in all environments but En4(Table S2).The DM-plant,DMstem,and DM-ear of LS cultivars XY335 and XD20 on average were respectively 26.43%,41.38%,and 116.44% larger than those of LR cultivars LS1 and FM985 in En1.XD20 had the largest values in DM-plant and different organs in En2 and LS1 had the lowest values.XY335 had the largest DM-plant and different organ dry matter weight in En3 and En5.DM-plant and DM-leaf were not significantly different among cultivars in En4.DM-ear of LS cultivars on average was 69.33% higher than that of LR cultivars,and XY335 showed the lowest DM-stem in En4.LS1 and XY335 showed the lowest values in Stem%in En1 and En4,and XD20 had the lowest values in En3 and En5.Leaf% was not significantly different among cultivars in En4,with XD20 having the largest values in En3 and En5.Dry matter allocation to ear(Ear%)of LS cultivars was higher compared to LR cultivars in En1,En3,and En4(on average 13.22% vs.7.85%),and was lower in En2 and En5(on average 11.33% vs.8.85%).

After silking,En,C,and their interactions significantly affected dry matter weight of each organ at harvest and dry matter accumulation of each organ,except for the effect of C on DM-stem and post-silking dry matter accumulation of stem(ΔDM-stem)(Table S3).

Post-silking dry matter accumulation of total plant(ΔDMplant)in En1 and En2 was on average 24.42% lower than that of En4 and En5(Table S3).En1 had the largest value in leaf dry matter weight reduction(ΔDM-leaf)from silking to maturity.The dry matter accumulation of ear after silking(ΔDM-ear)in En1 was 25.11%lower than that of other environments.At the harvest stage,En2 had the lowest values in DM-plant,DM-stem,DM-leaf,En1 had the lowest values in DM-ear,and En5 had the largest dry matter weight in each organ.

TheΔDM-plant,ΔDM-leaf,ΔDM-ear of XY335 was significantly larger than that of other cultivars in En1,En4 and En5(Table S3).XY335 had the largest DM-plant and different organs at harvest in these environments.FM985 displayed the largest values in dry matter weight of each organ at harvest and post-silking dry matter accumulation of each organ in En2.TheΔDM-stem of LS1 was larger than that of other cultivars in En1-2,and FM985 had the lowest values in En4-5.

The PCA analysis of yield traits,lodging-resistance traits,physical characteristics,and dry matter accumulation and partitioning parameters showed clear differences among the four cultivars(Fig.2).The principal component(PC1,Fig.2A)accounted for 60.3% variance and was predominant for yield and KN.XY335,XD20,and FM985 were located in the same direction as yield and had positive factor loadings on PC1,indicating that these cultivars had higher yield.LS1 was far from and opposite to yield,indicating that LS1 had a lower yield.XY335 and XD20 were located close to lodging percent,indicating a high LS.Yield was positively correlated with yield components(KN and TKW)and lodging percent.Likewise,Fig.2B shows that KN was positively correlated with DM-stem,Stem%,DM-ear,Ear%,and PH,and was negatively correlated with DM-leaf,Leaf%,ID,and lodgingresistance traits(RPS and DWUL at silking stage).TKW was positively correlated with PH,EH,and dry matter accumulation and partitioning of each organ after silking(except forΔDM-stem),but was negatively correlated withΔDM-stem and lodgingresistance traits(RPS and DWUL at harvest stage)(Fig.2C).Lodging percent was positively correlated with PH,EH,and RPS(silking and harvest stage)and was negatively correlated withΔDM-stem(Fig.2B,C).

3.3.Environmental impacts on lodging resistance and yield

The SEM showed that KN and TKW both had positive effects on yield(R2=0.50 and R2=0.53,P≤0.01;Fig.3).Pre-silking DM-stem had positive effects on the diameter of third aboveground internode(ID),DWUL,and EH with factor loadings of 0.48(P≤0.01),0.29(P＜0.05)and 0.37(P≤0.01),respectively.Higher EH increased KN with a factor loading of 0.41(P≤0.01).The GDD had a negative effect on pre-silking DM-stem and a positive effect on EH(R2=-0.37,P≤0.01 and R2=0.37,P≤0.01).Likewise,GDD during post-silking phase had negative effects on TKW and postsilkingΔDM-leaf with factor loadings of-0.71(P≤0.01)and-0.64(P≤0.01),respectively.SR during post-silking phase had a positive effect on TKW(R2=0.50,P＜0.05).The largerΔDM-leaf directly decreased RPS(R2=-0.263,P＜0.05).EH had negative effects on DWUL and RPS with factor loading of-0.27(P≤0.01)and-0.64(P≤0.01),respectively.ID had positive effects on DWUL and RPS(R2=0.76 and R2=0.46,P≤0.01),respectively.

4.Discussion

4.1.Relationships between yield components(KN and KW)and lodging resistance in maize

This study observed that LS cultivars XY335 and XD20 yielded more than the LR cultivars LS1(Tables 2,3)and lodging susceptibility had a positive relationship with yield(Fig.2A),in agreement with findings that high yielding crop cultivars face high lodging risk[25,27].The negative correlations between yield,KN,TKW and stalk lodging resistance further confirmed the tradeoffs between yield and lodging resistance[27,53](Fig.1).

The LR cultivars LS1 and FM985 had higher lodging resistance than the LS cultivars XY335 and XD20 mainly because of the stronger basal stems,lower PH,and EH[9,44](Tables 3,S1).A stronger basal stem demands more biomass investment for support structures[54]and competes more vigorously for assimilates with ear dry matter growth[26].LS1 transported more assimilates to basal stem,resulting in a thicker basal stem(Tables S1,S2),which decreased dry matter allocation to ear(Ear%)and reduced KN[29](Tables 2,S2).The lower KN of LS1 was also due partly to the lower PH and EH,which increased lodging resistance but also limited plant dry matter accumulation because a lower PH affects light distribution in the canopy[55]and reduces the capacity for storage of soluble carbohydrates[56].In contrast,the higher KN of LS cultivars XY335 and XD20 needed larger pre-silking dry matter accumulation and required more photoassimilates allocated to the ear,limiting the dry matter allocation to stem[31,32,57](Table S2).The lower stalk lodging resistance of LS cultivars(Table 3)also indicated that biomass partitioning to stem increased mainly SL and PH and did not increase stem strength.Thus,plants that combine higher stalk lodging resistance with larger KN should have a lower EH,and produce and allocate more photoassimilates to the ear.Similar results were also found in studies[20,21,58]of plant growth regulators,in which the relationships between plant morphology,lodging resistance,and yield were detected.Applying ethephon(2-chloroethyl phosphonic acid)after the 13-leaf stage can avoid damaging floret development and allocate more photosynthate to the ear by reducing internode length,which increased both yield and lodging resistance[59].

Tradeoffs between kernel weight and lodging resistance were also detected in this study,in agreement with those reported in wheat[25,26].During the grain filling stage,LS cultivars XY335 produced more assimilates because of the greater leaf photosynthetic productivity compared with LR cultivars LS1[49](Table S3).The higher TKW achieved by XY335 resulted partly from higher assimilate availability for increasing kernel weight compared to LS1[60–62](Tables 2,S3).XY335 transported more assimilates from stalk to ear during the period from silking to maturity(as in En2 and En3)to increase ear weight(Table S3),greatly reducing stalk quality and increasing lodging risk[12](Table 3).The height of the plant’s center of gravity rose with gradually increased ear weight.Stalk lodging frequently occurs during the grain-filling stage[9]and LS cultivars lodged severely at late growth stage in En1 and En2(Table 3).The higher lodging resistance of LS1 was due mainly to the larger assimilate partitioning to stem after silking(Tables 3,S3)and lower PH and EH(Table S1).Thus,during grain filling,lodging resistance and grain yield are affected by plant architecture,assimilate production,and assimilate partitioning to maize organs.

FM985 was located more closely to yield and KN than was LS1(Fig.2A),indicating the FM985 has a higher obtainable yield among LR cultivars.Averaged across five environments,the yield of FM985(11,377.16 kg ha-1)was lower than that of XY355(12,392.56 kg ha-1)but higher than that of XD20(10,948.77 kg ha-1)(Table 2),implying that FM985 was a relatively high-yielding cultivar.Compared to LS cultivars XY335 and XD20,FM985 was located opposite to lodging percent but close to stalk lodging resistance(RPS and DWUL at harvest)(Fig.2),confirming a higher lodging resistance.Thus,LR cultivar FM985 showed both high-yielding(Table 2)and high-LR traits,successfully combing high yield and high lodging resistance in one maize cultivar.The higher yield of FM985 was due mainly to the larger KN compared to LS1(Table 2;Fig.2A,B).FM985 showed a similar dry matter allocation to ear(Ear%)at the silking stage but attained a larger KN than LS1(Tables 2,S2),implying that FM985 required a smaller assimilate flux for an individual flower primordium to develop a kernel[32,33].The high leaf productivity of FM985 provided enough photo-assimilates for lodging resistance and KN formation before silking as well as for lodging resistance maintenance and kernel weight increase after silking(Table S3).In addition,compared to LS1,the EH was reduced by 10.46% in FM985(Table S1;Fig.S2).The optimized EH enabled stem to compete less for assimilates with reproductive organs,promoting kernel formation and growth[59](Table S2;Fig.3).The optimized EH also gave FM985 a smaller lodging risk with a thinner basal stem internode and a lower basal stem DWUL(Tables 3,S2).Although FM985 had some stem morphological deficiencies,our previous study[49]revealed that the larger sclerenchyma and more compact distribution of epidermal cells in the rind resulted in a greater stalk lodging resistance compared to LS cultivars.As discussed above,highyielding and high-LR traits of FM985 were related to optimized EH and stem anatomical structure,higher leaf productivity,low assimilate demand for kernel formation,and fluent assimilate partitioning to the ear.

4.2.Evaluation of the effects of growth environments(GDD and SR)on the formation of yield and lodging resistance.

Growing environment can influence both yield and lodging resistance.Higher mean temperature shortened the vegetative phase,accelerated GDD accumulation,but decreased cumulative incident radiation[38](compare En5;Table 1).The SEM analysis revealed that a higher pre-silking GDD can accelerate stem extension but decreased stem dry matter accumulation and weakened basal stem(Fig.3;Table 3,S2).For the high stalk lodging resistance in En1(Table 3),the higher SR was an explanation(Table 1),as the mechanical strength of the maize stalk increased with SR[10,14].No significant effects of pre-silking GDD and SR on KN were detected(Fig.3);the KN difference between environments is probably due to the abiotic stresses(high temperature and rain)at the flowering stage(Table 2).The mean maximum temperature of En1 around silking(5 days before and after silking)was approximately 34.0 °C,higher than the optimum temperature(31.8 °C)for maize anthesis[63].As for En5,rainy stress was the pivotal limiting factor reducing KN[64]as the cumulative rainfall was 107.6 mm during the flowering period(August 1–20).Thus,optimizing sowing dates to avoid environmental restraints on kernel formation and to promote dry matter allocation to the basal stem should be taken into account.

Environmental factors during grain filling strongly affect kernel weight[65,66]and lodging resistance(Fig.3).The mean temperature during grain filling stage in En1 was approximately 28.4 °C,higher than in other environments(23.6–27.0 °C).The higher mean temperature accelerated leaf aging,reduced postflowering photosynthetic production,and resulted in a low kernel weight[67](Table 2;Fig.3).High mean temperature during grain filling stage promoted dry matter allocation from leaf to stem,and increased stem dry matter accumulation and stalk lodging resistance(En1,Figs.S2 and 3;Table S3).The higher temperature during grain filling stage may reduce grain kernel weight but partly increase stalk lodging resistance.En5 had the largest dry matter accumulation of vegetable organs after silking(Table S3),partly because the lower mean temperature(23.63°C)reduced plant respiration[68](Table 1).The higher assimilate availability in En5 promoted kernel weight increases and resulted in a higher yield(Table 2).SEM revealed that higher SR increased dry matter accumulation and partly compensated low kernel weight induced by higher GDD(Fig.3).These results indicated that post-silking temperature is a more important influence on lodging resistance and yield than are other environmental factors[40](Fig.3).

5.Conclusions

Tradeoffs between yield formation and lodging resistance are present during the entire maize growth period.Before silking,the LR cultivars LS1 transported more assimilates to the basal stem,resulting in a thicker basal stem,which reduced dry matter allocation to the ear and in turn KN.The lower KN of LS1 was also due partly to the lower PH,which increased lodging resistance but limited plant dry matter production.In contrast,the LS cultivars XY335 and XD20 produced and allocated more photoassimilates to ears,but limited dry matter allocation to stems.After silking,LS cultivars showed higher TKW than LR cultivars as a function of high photoassimilate productivity and high assimilate allocation to the ear.High-yielding and high-LR traits of FM985 were related to optimized EH and stem anatomical structure,higher leaf productivity,low assimilate demand for kernel formation,and assimilate partitioning to the ear.Optimizing sowing dates to avoid environmental restraints on kernel formation and to promote dry matter allocation to the basal stem should be taken into account.Post-silking temperature has greater influence on lodging resistance and yield than other environmental factors.

CRediT authorship contribution statement

Ping Zhang:Conceptualization,Investigation,Writing–review& editing,Formal analysis,Writing–original draft.Shuangcheng Gu:Investigation,Writing–review & editing,Formal analysis,Writing–original draft,Data curation.Yuanyuan Wang:Investigation,Formal analysis.Chenchen Xu:Data curation.Yating Zhao:Investigation.Xiaoli Liu:Formal analysis.Pu Wang:Conceptualization,Funding acquisition,Resources.Shoubing Huang:Conceptualization,Funding acquisition,Resources,Writing–review &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

The authors want to thank the staff of Wuqiao Experimental Station of China Agricultural University for the work.This study was supported by the project of National Key Research and Development P rogram o f C hina ( 2016YFD0300301 a nd 2017YFD0300603)a ndT he2 115T alentD evelopmentP rogramo f ChinaA griculturalU niversity.

AppendixA.Supplementarydata

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2022.04.020.

Referencesyield potential of maize under medium and high plant density,Plant Biol.23(2021)485–496.

[24]Y.S.Yang,X.X.Guo,G.Z.Liu,W.M.Liu,J.Xue,B.Ming,R.Z.Xie,K.R.Wang,P.Hou,S.K.Li,Solar radiation effects on dry matter accumulations and transfer in maize,Front.Plant Sci.12(2021)727134.

[25]M.J.Foulkes,G.A.Slafer,W.J.Davies,P.M.Berry,R.Sylvester-Bradley,P.Martre,D.F.Calderini,S.Griffiths,M.P.Reynolds,Raising yield potential of wheat.III.optimizing partitioning to grain while maintaining lodging resistance,J.Exp.Bot.62(2011)469–486.

[26]F.J.Pi?era-Chavez,P.M.Berry,M.J.Foulkes,G.Molero,M.P.Reynolds,Avoiding lodging in irrigated spring wheat.II.genetic variation of stem and root structural properties,Field Crops Res.196(2016)64–74.

[27]W.Wu,B.L.Ma,J.J.Fan,M.Sun,Y.Yi,W.S.Guo,H.D.Voldeng,Management of nitrogen fertilization to balance reducing lodging risk and increasing yield and protein content in spring wheat,Field Crops Res.241(2019)107584.

[28]F.H.Andrade,C.Vega,S.Uhart,A.Cirilo,M.Cantarero,O.Valentinuz,Kernel number determination in maize,Crop Sci.39(1999)453–459.

[29]L.Borrás,L.N.Vitantonio-Mazzini,Maize reproductive development and kernel set under limited plant growth environments,J.Exp.Bot.69(2018)3235–3243.

[30]V.H.Gonzalez,E.A.Lee,L.L.Lukens,C.J.Swanton,The relationship between floret number and plant dry matter accumulation varies with early season stress in maize(Zea mays L.),Field Crops Res.238(2019)129–138.

[31]L.Echarte,F.H.Andrade,C.R.C.Vega,M.Tollenaar,Kernel number determination in Argentinean maize hybrids released between 1965 and 1993,Crop Sci.44(2004)1654–1661.

[32]C.R.C.Vega,F.H.Andrade,V.O.Sadras,S.A.Uhart,O.R.Valentinuz,Seed number as a function of growth.A comparative study in soybean,sunflower,and maize,Crop Sci.41(2001)748–754.

[33]G.A.Slafer,M.Elia,R.Savin,G.A.García,I.I.Terrile,A.Ferrante,D.J.Miralles,F.G.González,Fruiting efficiency:an alternative trait to further rise wheat yield,Food Energy Secur.4(2015)92–109.

[34]J.Cárcova,B.Andrieu,M.E.Otegui,Silk elongation in maize:relationship with flower development and pollination,Crop Sci.43(2003)914–920.

[35]H.G.Jung,M.D.Casler,Maize stem tissues:cell wall concentration and composition during development,Crop Sci.46(2006)1793–1800.

[36]V.Mahalakshmi,S.Sivaramakrishnan,F.R.Bidinger,Contribution of preanthesis and postanthesis photosynthates to grain in pearl-millet under water deficit,J.Agron.Crop Sci.170(1993)91–96.

[37]K.E.D’Andrea,C.V.Piedra,C.I.Mandolino,R.Bender,A.M.Cerri,A.G.Cirilo,M.E.Otegui,Contribution of reserves to kernel weight and grain yield determination in maize:phenotypic and genotypic variation,Crop Sci.56(2016)697–706.

[38]I.Rajcan,M.Tollenaar,Source:sink ratio and leaf senescence in maize:I.dry matter accumulation and partitioning during grain filling,Field Crops Res.60(1999)245–253.

[39]M.E.Otegui,S.Melón,Kernel set and flower synchrony within the ear of maize:I.sowing date effects,Crop Sci.37(1997)441–447.

[40]B.Zhou,Y.Yue,X.Sun,Z.Ding,W.Ma,M.Zhao,Maize kernel weight responses to sowing date-associated variation in weather conditions,Crop J.5(2017)43–51.

[41]P.Hou,Y.E.Liu,W.M.Liu,H.S.Yang,R.Z.Xie,K.R.Wang,B.Ming,G.Z.Liu,J.Xue,Y.H.Wang,R.L.Zhao,W.J.Zhang,Y.H.Wang,S.F.Bian,H.Ren,X.Y.Zhao,P.Liu,J.Z.Chang,G.H.Zhang,J.Y.Liu,L.Z.Yuan,H.Y.Zhao,L.Shi,L.L.Zhang,L.Yu,J.L.Gao,X.F.Yu,Z.G.Wang,L.G.Shen,P.Ji,S.Z.Yang,Z.D.Zhang,J.Q.Xue,X.F.Ma,X.Q.Wang,T.Q.Lu,B.C.Dong,G.Li,B.X.Ma,J.Q.Li,X.F.Deng,Y.H.Liu,Q.Yang,C.L.Jia,X.P.Chen,H.Fu,S.K.Li,Quantifying maize grain yield losses caused by climate change based on extensive field data across China,Resour.Conserv.Recycl.174(2021)105811.

[42]S.Asseng,F.Ewert,P.Martre,R.P.R?tter,D.B.Lobell,D.Cammarano,B.A.Kimball,M.J.Ottman,G.W.Wall,J.W.White,M.P.Reynolds,P.D.Alderman,P.V.V.Prasad,P.K.Aggarwal,J.Anothai,B.Basso,C.Biernath,A.J.Challinor,G.D.Sanctis,J.Doltra,E.Fereres,M.Garcia-Vila,S.Gayler,G.Hoogenboom,L.A.Hunt,R.C.Izaurralde,M.Jabloun,C.D.Jones,K.C.Kersebaum,A.K.Koehler,C.Müller,S.Naresh Kumar,C.Nendel,G.O’Leary,J.E.Olesen,T.Palosuo,E.Priesack,E.E.Rezaei,A.C.Ruane,M.A.Semenov,I.Shcherbak,C.St?ckle,P.Stratonovitch,T.Streck,I.Supit,F.Tao,P.J.Thorburn,K.Waha,E.Wang,D.Wallach,J.Wolf,Z.Zhao,Y.Zhu,Rising temperatures reduce global wheat production,Nat.Clim.Change 5(2014)143–147.

[43]W.Wu,B.L.Ma,Assessment of canola crop lodging under elevated temperatures for adaptation to climate change,Agric.For.Meteorol.248(2018)329–338.

[44]P.Zhang,S.C.Gu,Y.Y.Wang,R.M.Yang,Y.Yan,S.Zhang,D.C.Sheng,P.Wang,S.B.Huang,Morphological and mechanical variables associated with lodging in maize(Zea mays L.),Field Crops Res.269(2021)108178.

[45]B.Tian,J.Zhu,Y.Nie,C.Xu,Q.Meng,P.U.Wang,Mitigating heat and chilling stress by adjusting the sowing date of maize in the north china plain,J.Agron.Crop Sci.205(2019)77–87.

[46]X.Dong,L.Guan,P.H.Zhang,X.L.Liu,S.J.Li,Z.J.Fu,L.Tang,Z.Y.Qi,Z.G.Qiu,C.Jin,S.B.Huang,H.Yang,Responses of maize with different growth periods to heat stress around flowering and early grain filling,Agric.For.Meteorol.303(2021)108378.

[47]M.J.Novacek,S.C.Mason,T.D.Galusha,M.Yaseen,Twin rows minimally impact irrigated maize yield,morphology,and lodging,Agron.J.105(2013)268–276.

[48]M.A.O.Oladokun,A.R.Ennos,Structural development and stability of rice Oryza sativa L.var.Nerica 1,J.Exp.Bot.57(2006)3123–3130.

[49]P.Zhang,Y.Yan,S.C.Gu,Y.Y.Wang,C.L.Xu,D.C.Sheng,Y.B.Li,P.Wang,S.B.Huang,Lodging resistance in maize:a function of root-shoot interactions,Eur.J.Agron.132(2022)126393.

[50]P.Ning,S.Li,P.Yu,Y.Zhang,C.J.Li,Post-silking accumulation and partitioning of dry matter,nitrogen,phosphorus and potassium in maize varieties differing in leaf longevity,Field Crops Res.144(2013)19–27.

[51]C.Rivera-Amado,E.Trujillo-Negrellos,G.Molero,P.M.Reynolds,R.Sylvester-Bradley,M.J.Foulkes,Optimizing dry-matter partitioning for increased spike growth,grain number and harvest index in spring wheat,Field Crops Res.240(2019)154–167.

[52]R Core Team,R:A Language and Environment for Statistical Computing,R Foundation for Statistical Computing,Vienna,Austria,2020.

[53]C.Z.Li,C.J.Li,Ridge-furrow with plastic film mulching system decreases the lodging risk for summer maize plants under different nitrogen fertilization rates and varieties in dry semi-humid areas,Field Crops Res.263(2021)108056.

[54]P.M.Berry,R.Sylvester-Bradley,S.Berry,Ideotype design for lodging-resistant wheat,Euphytica 154(2007)165–179.

[55]D.J.Miralles,G.A.Slafer,Yield,biomass and yield components in dwarf,semidwarf and tall isogenic lines of spring wheat under recommended and late sowing dates,Plant Breed.114(1995)392–396.

[56]F.D.Beed,N.D.Paveley,R.Sylvester-bradley,Predictability of wheat growth and yield in light-limited conditions,J.Agric.Sci.145(2007)63–79.

[57]M.Tollenaar,L.M.Dwyer,D.W.Stewart,Ear and kernel formation in maize hybrids representing three decades of grain yield improvement in Ontario,Crop Sci.32(1992)432–438.

[58]C.L.Xu,Y.B.Gao,B.J.Tian,J.H.Ren,Q.F.Meng,P.Wang,Effects of EDAH,a novel plant growth regulator,on mechanical strength,stalk vascular bundles and grain yield of summer maize at high densities,Field Crops Res.200(2017)71–79.

[59]Y.Zhao,Y.Lv,S.Zhang,F.Ning,Y.Cao,S.Liao,P.U.Wang,S.Huang,Shortening internodes near ear:An alternative to raise maize yield,J.Plant Growth Regul.41(2022)628–638.

[60]G.A.Maddonni,M.E.Otegui,R.Bonhomme,Grain yield components in maize:II.postsilking growth and kernel weight,Field Crops Res.56(1998)257–264.

[61]L.Borrás,M.E.Otegui,Maize kernel weight response to postflowering sourcesink ratio,Crop Sci.41(2001)1816–1822.

[62]B.L.Gambín,L.Borrás,M.E.Otegui,Source-sink relations and kernel weight differences in maize temperate hybrids,Field Crops Res.95(2006)316–326.

[63]B.Sánchez,A.Rasmussen,J.R.Porter,Temperatures and the growth and development of maize and rice:a review,Glob.Change Biol.20(2014)408–417.

[64]Z.Gao,H.Y.Feng,X.G.Liang,S.Lin,S.L.Zhou,Adjusting the sowing date of spring maize did not mitigate against heat stress in the north china plain,Agric.For.Meteorol.298–299(2021)108274.

[65]L.E.Bonelli,J.P.Monzon,A.Cerrudo,R.H.Rizzalli,F.H.Andrade,Maize grain yield components and source-sink relationship as affected by the delay in sowing date,Field Crops Res.198(2016)215–225.

[66]I.R.Hisse,K.E.D’Andrea,M.E.Otegui,Kernel weight responses to the photothermal environment in maize dent×flint and flint×flint hybrids,Crop Sci.61(2021)1996–2011.

[67]L.Borrás,G.A.Slafer,M.E.Otegui,Seed dry weight response to source-sink manipulations in wheat,maize and soybean:a quantitative reappraisal,Field Crops Res.86(2004)131–146.

[68]D.J.Allen,D.R.Ort,Impacts of chilling temperatures on photosynthesis in warm-climate plants,Trends Plant Sci.6(2001)36–42.