Can ‘relative culm wall thickness’ be used to evaluate the lodging resistance of rice?

ABSTRACT Many indexes have been demonstrated to evaluate the lodging resistance of rice, but most of them are complex or not representative. Thus, we propose a simple and low-cost index, called relative culm wall thickness (RCWT), which can be calculated as the ratio of the basal internode wall thickness to its diameter, to measure the lodging characteristics of rice. A field experiment with 20 indica hybrid rice varieties was conducted in Dayi, Sichuan, China, in 2017 and 2018. The results show that RCWT was significantly negatively correlated with the lodging index, visual lodging rate, plant inclination angle, and bending moment, while it was significantly positively correlated with culm wall thickness, breaking resistance, cellulose, and lignin. Compared with the middle and low RCWT types, the high RCWT type had a lower plant height and center of gravity. Meanwhile, the short basal internodes, thick culm wall thickness, and small pith diameter increased the fullness of the stem, which increased the content of cellulose and lignin in the stem, and thus enhanced the bending resistance and decreased lodging index of the stem. These results indicate that the RCWT can be used as an important index to evaluate the lodging resistance of rice.


Introduction
Global food security and economic growth are closely related to rice Devkota et al. 2020). Rice, one of the most important foods worldwide, is cultivated in an area of more than 163 million hectares, feeding more than half of the world's population (IRRI 2013;FAO 2014). In China especially, rice is the staple food for nearly 1 billion people (Ding et al. 2020). Therefore, improving rice production with limited cultivated land is required to feed the rapidly growing population (Liu et al. 2013;Wang et al. 2020).
However, high-yield rice production faces a series of risks. These include environmental pollution, soil quality, and soil structural damage caused by the excessive application of pesticides and fertilizers (Zhu et al. 2016;Tom 2018;Chen et al. 2018a). High-fertilizer and high-density cultivation methods are commonly accompanied by large-scale lodging, which severely limits the improvement of both the yield and quality of rice (Zhang et al. 2014;Wu and Ma 2019). Lodging has become one of the main factors limiting the growth of rice yield. It is estimated that for every 2% increase in the rice lodging rate, the yield will decrease by 1%, and serious lodging will even reduce the rice yield by 50%-80% (Setter et al. 1997;Niu et al. 2016). After lodging, the canopy structure of the population is destroyed; the photosynthetic rate of the plant is reduced; and assimilate production and transportation are hindered. This ultimately leads to a reduction in rice yield and quality (Islam et al. 2007;Kashiwagi et al. 2010;Wu et al. 2012) and also increases the difficulty and cost of harvesting.
The occurrence of rice lodging is a complex and comprehensive phenomenon. According to the location, it can be divided into root lodging and stem lodging. Root lodging mainly occurs in dry rice or direct-seeded rice planting methods but rarely occurs in transplanted paddy rice (Zhang et al. 2014). Stem lodging is caused by bending or breaking from the base internodes and is the main type of lodging (Berry et al. 2003). Irrespective of the type of lodging, lodging is affected by factors, such as external natural conditions (Niu et al. 2016;Kato et al. 2020), water and fertilizer management, and other cultivation methods (Pan et al. 2019;Zhang et al. 2021), along with the genetic characteristics of the rice (Shahidullah et al. 2009). Previous studies on rice lodging have focused on agronomic traits, anatomical structure, and physiological and biochemical mechanisms (Zhang et al. 2010(Zhang et al. , 2017a. They suggest that the morphological, mechanical, anatomical, and chemical characteristics of the basal two or three internodes, especially the basal second internode, play a key role in the lodging resistance of rice Zhong et al. 2020).
Previously, researchers have proposed many indicators to evaluate the lodging performance of rice, including morphological indexes such as culm diameter, culm wall thickness, and stem fullness (Zuber et al. 1999;Wu et al. 2017). The lodging index, bending moment, and lodging mechanics model are comprehensive indexes for measuring the mechanical properties of rice (Ookawa and Ishihara 1992;Shrestha et al. 2020). In addition, anatomical indicators such as the number and area of large and small vascular bundles can be used to estimate the lodging resistance of rice Wu et al. 2017). Regarding chemical composition, lodging ability has been evaluated based on the content of cellulose and lignin Wu et al. 2017). These indicators can adequately explain the lodging resistance of rice, especially the lodging index, which is the most recognized evaluation index at present (Ookawa and Ishihara 1992). However, conducting tedious in-field experiments with expensive instruments or laboratory determination limits the large-scale application of these indicators.
The lodging resistance of rice has been closely related to the genotype, and the lodging resistance of different varieties often varies greatly Long et al. 2020). Previous studies have demonstrated that rice varieties with large culm diameters and thick culm walls have better lodging resistance (Zhang et al. 2014;Fan et al. 2018). However, other studies have found that lodging resistance is not always strong with a large or small culm diameter, but that it is positively correlated with culm wall thickness Zhao et al. 2019). Shi (2010) believed that lodging resistance differed between varieties of rice with thicker culm wall thickness, which is closely related to culm diameter. Nevertheless, in practical production and field experiments, we have found that rice varieties with a small pith diameter and thick culm wall thickness often show higher resistance to lodging. Therefore, the ratio of the basal internode wall thickness to the diameter (relative culm wall thickness), which integrates both diameter and culm wall thickness, may be a suitable index for evaluating the lodging resistance of rice.
Therefore, we hypothesize that the relative culm wall thickness (RCWT) could be used as an index to evaluate the lodging resistance of rice. To verify its feasibility, a field experiment with 20 indica hybrid rice varieties was conducted in Sichuan, China in 2017 and 2018. The specific objectives of this study were: (1) To analyze the lodging resistance of rice with different RCWT by combining stem morphology, mechanical properties, and chemical composition. (2) To evaluate the feasibility of RCWT as an index of lodging resistance. If the hypothesis is established, it can provide a simple and quick method for determining the lodging resistance of rice.

Study site and materials
Field experiments were conducted at Dayi County (30° 49′ N, 103° 57′ E), Sichuan province, China in 2017 and 2018 (He et al. 2019). Dayi has a subtropical, humid monsoon climate. The average temperatures from April to September in 2017 and 2018 were 23.25°C and 22.81°C, the average precipitation was 851.9 mm and 1570.0 mm, and the average sunshine hours were 670.00 h and 788.00 h, respectively. The former crop was Brassica juncea in both years. The experimental site had a medium loam soil. The primary soil properties of the 0-10 cm soil layer was measured as per Lu (2000) (Table S1). The tested varieties were 20 indica hybrid rice varieties bred or approved in Southwest China (Yunnan Province, Guizhou Province, Sichuan Province, and Chongqing City) in recent years. They are all established hybrid indica rice, which are widely planted in various regions in the southwest. Data for the 20 rice varieties used are shown in Table S2 (He et al. 2019).

Experimental design and management
A one-factor randomized block field experiment with 20 varieties and three replications was conducted in 2017 and 2018. The dimensions of each plot were 11 × 1.8 m in the two-year experiment. Seedlings were raised in a hard growing tray (58 cm × 28 cm × 2.5 cm) for 30 days. The row spacing used for transplanting by the rice transplanter VPD60 (YANMAR Agricultural Machinery Co., Ltd., Wuxi, China) was 30 cm × 20 cm, with three seedlings per hole. Transplanting was performed on 3 May 2017and 2018. For each plot, the total N application was 180 kg ha −1 , which included a basal fertilizer of 70% and a tiller fertilizer of 30%. One day before transplantation, 90 kg ha −1 of P 2 O 5 was applied as a single super-phosphate, and 180 kg ha −1 of K 2 O as potassium chloride was applied equally as a basal and tiller fertilizer. According to our previous research (Deng et al. 2015), highefficiency irrigation techniques were used to guide water resource management. Chemical pesticides were used to prevent yield loss caused by diseases, insects, and weeds.

Agronomic traits of stems
On the twenty-fifth day after the heading stage (Table S2), 10 representative main stems with uniform growth and the same number of internodes and leaves were selected from each plot to measure lodging-related physical traits. The plant height, gravity center height (the fresh plant is placed horizontally on the tip of the index finger and the fulcrum adjusted to keep it balanced; the length from the base to the fingertips is the height at the center of gravity), and relative gravity center height (gravity center height divided by plant height) of the stalk was calculated. Then, a ruler was used to measure the first, second, and third basal internode length (N1, N2, N3, respectively) and the distance from the internode base to the panicle top. After measurement, the internodes were cut, and Vernier calipers were used to measure the culm wall thickness and diameter of the N1-N3 internodes of the base. The rice stalks were simulated as approximately regular circles and the culm wall thickness and diameter of the basal internodes were measured twice, at positions 90 ° apart ( Figure S1).

Culmwallthickness mm
where D1 and D2 represent the two measurements of the diameter of the stem and T1 and T2 are the two measurements of the thickness of the stem wall. RCWT: relative culm wall thickness. CWT: culm wall thickness. CD: culm diameter.

Mechanical properties of stems
A Stalk Strength Tester (model YYDG-1, Zhejiang Top Instrument Co., Ltd., Hangzhou, China) was used to measure the basal internode (N1, N2, N3) breaking resistance (fulcrum spacing 5 cm) (Ookawa and Ishihara 1992). The mechanical properties were calculated following the methods of Ookawa and Ishihara (1992) with some modifications: where SL is the distance from the internode base to the panicle top (cm), and FW is the fresh plant weight from the internode base to the panicle top (g).

Breaking1mustrength gcm
where F is the breaking resistance (kg), and L is the distance between the two fulcrum points (cm).

Dry weight per unit length (DWUL) and chemical composition of basal internodes
After the above indexes were determined, the internodes were placed in an oven at 105°C for 30 min and then dried to a constant weight at 80°C to determine the internode dry mass and chemical composition.
The DWUL of N1, N2, and N3 was calculated by dividing the internode dry weight by its length. The dry internode samples of N1, N2, and N3 were crushed with a 100-type high-speed universal crusher and screened with 80 meshes to determine the content of cellulose and lignin in structural carbohydrates. The content of cellulose and lignin were determined following the methodologies of Zhang et al. (2017a) and Ahmad et al. (2018), respectively.

Investigation of visual lodging rate
Only stem lodging occurred in the mature stage of rice, and the lodging area was investigated in the field before harvesting and calculated the inclination angle of the plants in the plot. The plant inclination angle refers to the angle that the rice stalk deviates from the vertical direction ( Figure S2). The plant inclination angle can show the strength of rice stalks in withstanding external pressure. The greater the value, the weaker the pressure resistance of the stem. The lodging area refers to the area where the inclination angle of lodging plants in the field is greater than 15°.

Data analyses
SPSS 18.0 software (IBM, Inc., Chicago, IL, USA) was used for statistical analysis of the data. Relative culm wall thickness data for 2017 and 2018 were used for clustering, dividing them into three categories. The other lodging-related indicators were subsequently analyzed according to these three categories. Treatment means were compared using the least significant difference test at the 1% and 5% probability levels. Correlation was analyzed among groups (N = 3). The schematic diagram of the rice stalk section and other figures were drawn using AutoCAD 2010 (Autodesk Inc., San Rafael, CA) and GraphPad Prism 5.0 (GraphPad Software, Inc., California, USA), respectively.

The premise of choosing RCWT
Lodging index is one of the most effective indexes in evaluating lodging resistance of rice. The lodging index of the base internodes of N1-N3 had a strong correlation with the lodging-related indicators across years (Table 1). The lodging index of each internode was negatively associated (P < 0.05) with the culm wall thickness of N1 and N2, and was positively correlated (P < 0.05) with the culm diameter of the base N1-N3 internodes. However, the RCWT had a negatively association with lodging index (P < 0.01). Meanwhile, the broad-sense heritability (H 2 ) of the 20 varieties of RCWT was 82.62% in 2018, 76.27% in 2017, and 81.06% across years, which indicates that the characteristics of RCWT were mainly affected by genetic factors and the trait of RCWT was relatively stable (Table S3). Thus, RCWT can be used to evaluate the lodging resistance of rice.

Plant characteristics and internode length
No significant interactions between the study year and RCWT type on plant height, gravity center height, relative gravity center height, plant fresh weight, or internode length were observed (Table 2). However, except for relative gravity center height and the internode length of N3, study year and RCWT type had a significant (P < 0.01) effect on these indexes. The plant height and gravity center height, as well as the basal N1-N3 internode length, in 2018 were significantly greater than those in 2017. Compared with M-RCWT and L-RCWT, H-RCWT had the lowest plant height, gravity center height, relative gravity center height, and plant fresh weight. Meanwhile, the plant height, gravity center height, and plant fresh weight of the M-RCWT were greater than those of the L-RCWT. In addition, the basal internode lengths from N1 to N3 were ranked as L-RCWT > M-RCWT > H-RCWT. Overall, the H-RCWT rice varieties possessed a lower plant height, gravity center height, relative gravity center height, and plant fresh weight along with shorter basal internode lengths, which could improve the lodging resistance of the rice.

Morphological traits of culm
There were no significant interactions between study year and RCWT type on culm morphological traits (Table 3). Unlike the culm wall thickness of the N2 and N3 internodes at the base, year significantly (P < 0.05) affected the culm wall thickness of the N1 internode, and the culm diameter, pith diameter, and DWUL of all three basal internodes. These indicators were significantly lower in 2017 than in 2018. However, RCWT type had a significant (P < 0.05) effect on the culm wall thickness, culm diameter, and pith diameter. Compared with M-RCWT and L-RCWT, the H-RCWT had the H-RCWT, high relative culm wall thickness type; M-RCWT, medium relative culm wall thickness type; L-RCWT, low relative culm wall thickness type; T, relative culm wall thickness type; Y, study year. N1-N3 represents the first to third basal internodes of rice. Different lowercase letters in columns represent significant differences between treatments in the two years at the 5% level. ns: P > 0.05. *: P < 0.05. **: P < 0.01.
maximum culm wall thickness. The culm diameter and pith diameter of the three internodes at the base were all ranked as H-RCWT < M-RCWT < L-RCWT, and there were significant differences between H-RCWT and L-RCWT. In addition, the DWUL of the basal three internodes of the H-RCWT type was greater than that of the L-RCWT type in 2018. These phenomena indicate that the H-RCWT type had a large culm wall thickness, small culm diameter and pith diameter, and high stem fullness, which may lead to strong lodging resistance.

Mechanical properties of culm
Except for the lodging index of the N2 and N3 basal internodes, there were no significant interactions between the study years and RCWT types on the lodging index of the N1 internode, or the breaking resistance and bending moment of the three internodes at the base (Table 4). However, the year had a significant (P < 0.01) effect on the N1 internode breaking resistance, the bending moment of the three internodes, and the N2 and N3 lodging index at the base; the mechanical indexes in 2017 were larger than those in 2018. Furthermore, no significant differences in the mechanical properties of the culm were observed among the three RCWT types. The breaking resistance of the N1-N3 internodes was 10.00%, 9.09%, and 2.15% higher, respectively, in H-RCWT than in L-RCWT; and the breaking resistance of the N1-N3 internodes was 13.24%, 5.60%, and 4.40% larger, respectively, in M-RCWT than in L-RCWT in 2017. In 2018, the breaking resistance of the N1-N3 internodes was 22.22%, 26.27%, and 25.59% greater, respectively, in H-RCWT than L-RCWT; and the breaking resistance of the N1-N3 internodes was 16.88%, 19.20%, and 12.37% larger, respectively, in M-RCWT than L-RCWT. Regarding bending moment and lodging index, the three types showed H-RCWT < L-RCWT < M-RCWT, in which H-RCWT was significantly smaller than L-RCWT and M-RCWT. Thus, the large breaking resistance and small bending moment in H-RCWT decreased the lodging index.

Chemical composition of culm
There were no significant interactions between the study year and RCWT clusters on the cellulose and lignin content of basal N1-N3 internodes ( Table 5). The year had a significant (P < 0.01) impact on the cellulose and lignin content of each internode, which was significantly higher in 2018 than in 2017. The type significantly (P < 0.01) affected the cellulose content of the base N2 and N3 H-RCWT, high relative culm wall thickness type; M-RCWT, medium relative culm wall thickness type; L-RCWT, low relative culm wall thickness type; T, relative culm wall thickness type; Y, study year. N1-N3 represents the first to third basal internodes of rice. DWUL, dry weight of unit internode. Different lowercase letters in columns represent significant differences between treatments in the two years at the 5% level. ns: P > 0.05. *: P < 0.05. **: P < 0.01.
internodes and the N2 lignin content. Among the three types, the content of cellulose and lignin roughly presented the rule of H-RCWT > M-RCWT > L-RCWT, and most of the differences between H-RCWT and L-RCWT were significant.

Lodging conditions in the field
The interaction of the year and type had no significant influence on the plant inclination angle and visual lodging rate in rice fields, while the main effects of year and type had a significant (P < 0.01) influence on visual lodging in rice fields (Figure 1). The larger the lodging angle in the field, the more easily the plant will fall, increasing the visual lodging rate. Compared with 2017, the lodging angle and visual lodging rate in 2018 were increased by 170.33% and 213.41%, which may be related to the strong winds and heavy rainfall in the late rice filling stage in 2018. Among the different types, the plant inclination angle and visual lodging rate of rice were shown as L-RCWT > M-RCWT > H-RCWT. Compared with the H-RCWT, the lodging angles of M-RCWT and L-RCWT were greater than 2.8-fold and 5.8-fold, respectively; the visual lodging rate were greater than 4.3-fold and 6.5-fold, respectively.  1.67a 1.31a 0.98a 4446.74a 3750.18a 2939.43a 222.51a 239.39a 246.03a Analysis of variance Y ** ns ns ** ** ** ns ** * T ** ** * ** ** ** ** ** ** Y × T ns ns ns ns ns ns ns * * H-RCWT, high relative culm wall thickness type; M-RCWT, medium relative culm wall thickness type; L-RCWT, low relative culm wall thickness type; T, relative culm wall thickness type; Y, study year. N1-N3 represents the first to third basal internodes of rice. Different lowercase letters in columns represent significant differences between treatments in the two years at the 5% level. NS: P > 0.05. *: P < 0.05. **: P < 0.01. H-RCWT, high relative culm wall thickness type; M-RCWT, medium relative culm wall thickness type; L-RCWT, low relative culm wall thickness type; T, relative culm wall thickness type; Y, study year. N1-N3 represents the first to third basal internodes of rice. Different lowercase letters in columns represent significant differences between treatments in the two years at the 5% level. ns: P > 0.05. *: P < 0.05. **: P < 0.01.

Correlation analysis
Correlation analysis indicated a strong correlation among N1-N3 internode RCWT values and the main stem physical and mechanical characteristics in 2017 and 2018 (Tables 6 and S6). The RCWT of each internode was negatively correlated (P < 0.05) with the gravity center height and visual lodging rate, while it was significantly positively (P < 0.05) correlated with the culm wall thickness, cellulose, and lignin of the base N1-N3 internodes in 2017 and 2018. Meanwhile, the culm diameter, pith diameter, and lodging index were significantly negatively associated with the RCWT. Furthermore, the plant height of N1-N3 in 2018 and the plant inclination angle of N2 and N3 in 2017 were significantly negatively correlated with the RCWT. In addition, the RCWT values of N1 in 2017 and N2 in 2018 were significantly positively correlated with breaking resistance, while they were significantly negatively correlated with the bending moments of N1 and N2 in 2018. Therefore, the RCWT of the basal internode is closely related to lodging resistance traits.

Discussion
Lodging is one of the important factors that limit the production of rice (Lang et al. 2012;Khan et al. 2018). It increases the difficulty and cost of harvesting and reduces yield and quality Sher et al. 2018). To reduce the occurrence of rice lodging, many methods, such as the lodging index and the lodging mechanical model, have been proposed to evaluate and measure this phenomenon (Ookawa and Ishihara 1992;Shrestha et al. 2020). Most of these indicators require expensive instruments and complicated operation processes, which cannot be widely applied to production practice. Therefore, the RCWT was used to evaluate the lodging resistance of rice in this study. The RCWT characteristics of different rice varieties and their relationship with plant lodging were analyzed in detail; several interesting findings could provide a basis for the selection of rice varieties with high lodging resistance. The lodging resistance of rice is closely related to the morphological characteristics of stem, and the three basal internodes have a strong influence on the lodging of rice Zhong et al. 2020). Many studies have shown that plant height is one of the main causes of rice lodging Chen et al. 2018b). In our study, plant height was significantly negatively correlated with RCWT (Table 6). Compared with the medium and low RCWT types, high RCWT had the lowest plant height (Table 2), which may be the reason for the lower center of gravity height and fresh weight and the short internodes of the plants Wu and Ma 2019). Furthermore, internode length, culm diameter, culm wall thickness, pith diameter, and DWUL varied with RCWT type (Tables 2 and 3). The internode culm diameter and pith diameter of rice decreased significantly with an increase in RCWT, while internode culm wall thickness and DWUL increased significantly with the increase in RCWT (Table 6). This indicates that a short internode length, large culm diameter, and thick culm wall can increase stem fullness, thus enhance the lodging resistance of rice (Zhang et al. 2014(Zhang et al. , 2017bFan et al. 2018). Moreover, studies have shown that, in terms of morphological indicators, the breaking resistance can be doubled by the leaf sheath covering (Ookawa and Ishihara 1992). Therefore, the reason for the strong lodging resistance of H-RCWT rice was attributed to its short plant height, low gravity center height, thick culm wall thickness, small pith diameter, and good stem fullness.
In field production, the ability of rice to resist or buffer external forces mainly depends on the mechanical properties of the stem. Increasing the mechanical strength of the rice basal internode stem is one of the key factors in improving lodging resistance . In this study, RCWT was significantly positively correlated with the breaking resistance between the basal N1-N3 internodes of rice, but negatively correlated with the bending moment and lodging index (Table 6). It is common knowledge that under conditions of short internodes, large DWUL, thick culm diameter, and thick culm walls, the breaking resistance of the stem can be improved . Furthermore, Shrestha et al. (Shrestha et al. 2020) believe that the bending moment of the basal N2 internode can contribute approximately 60% to the incidence of lodging. The bending moment is the product of the weight and height. The above-ground biomass of high RCWT rice will also decrease as the plant height decreases, thereby reducing the bending moment of the rice (Tables 2 and 4), and ultimately reducing its lodging index (Wu and Ma 2019), which is also consistent with the smaller plant inclination angle and visual lodging rate of H-RCWT (Figure 1). Meanwhile, the lodging resistance of rice is also affected by the chemical composition of the stem. The accumulation of structural carbohydrates in the rice stem and the transport of appropriate amounts of non-structural carbohydrates to the grain play decisive roles in producing stem mechanical strength and high yield (Ishimaru et al. 2008;Zhang et al. 2016;Zhong et al. 2020). Excessive transport of non-structural carbohydrates affects the concentration of structural carbohydrates, which will cause early senescence of the stem (Kashiwagi et al. 2006) and reduce culm density , thereby reducing the mechanical strength of the stem and increasing the risk of lodging Zhong et al. 2020). In this study, the breaking strength of the stem was largest under H-RCWT (Table 4), which may be closely linked to the accumulation of cellulose and lignin in stems (Table 5). As structural carbohydrates, cellulose and lignin are the main components of plant cell walls and some researchers attribute stem strength to them (Zhang et al. 2014;Liu et al. 2016). Accumulation of cellulose and lignin in the stem will increase the thickness of the cell wall, which will increase the culm wall thickness, and ultimately enhance the breaking strength of the stem Zhang et al. 2017a;Fang et al. 2018). This was also confirmed by the increase in dry weight per unit internode in this study (Table 3). Thus, the augmentation of cellulose and lignin content of the H-RCWT type improves the culm wall thickness, which in turn increases the mechanical strength and enhances the lodging resistance of rice.
The lodging of rice is closely correlated with the stem morphology, chemical composition, and mechanical properties Zhang et al. 2016;Wang et al. 2020). Under the conditions of low plant height, short base internode length, thick culm walls, and high content of stem cellulose and lignin, the lodging index will be decreased and the lodging resistance of rice will be improved Tian et al. 2018). In this study, there were differences (P < 0.05) in lodging-related traits among the different genotypes and years (Table S5), indicating that the diversity of lodging-related traits varies with genotype. However, the broad-sense heritability (H 2 ) of the 20 varieties of RCWT was 82.62% in 2018, 76.27% in 2017, and 81.06% across years, indicating that the genetic diversity of RCWT was stable (Table S3). The rice plant height, gravity center height, plant inclination angle, visual lodging rate, culm diameter, pith diameter, bending moment, and lodging index all decreased significantly with the increase in RCWT after clustering (Table 6). Meanwhile, culm wall thickness, breaking resistance, cellulose, and lignin were significantly positively correlated with RCWT (Table 6). Therefore, with the increase in RCWT, the lodging resistance of the stem also increases. Moreover 'RCWT' refers to the ratio between the culm wall thickness and culm diameter, which means that the size of the RCWT is affected by the combination of the culm wall thickness and diameter. It is more appropriate than using culm wall thickness or culm diameter alone to evaluate the lodging resistance of rice (Shi 2010;Weng et al. 2017). In addition, RCWT determination needs only Vernier calipers to measure the culm diameter and culm wall thickness, which are convenient and quick to operate. Therefore, 'RCWT' is a highly feasible index for evaluating the lodging resistance of rice.

Conclusion
In this study, the twenty tested varieties were divided into three RCWT types. Compared with M-RCWT and L-RCWT (especially), the plant height, gravity center height, plant inclination angle, culm diameter, bending moment, and lodging index of H-RCWT rice were lower, while the culm wall thickness, breaking resistance, and both cellulose and lignin content were higher. Correlation analysis showed that plant height, gravity center height, plant inclination angle, culm diameter, bending moment, and lodging index were significantly negatively correlated with RCWT. However, the culm wall thickness, cellulose, lignin, and breaking resistance were significantly positively correlated with RCWT. These traits indicate that the physical characteristics and mechanical strength of the H-RCWT stalks are improved, thereby reducing the lodging rate of the rice. Therefore, in this study, RCWT was used as an index to evaluate the lodging resistance of rice. The proposed index provides a more convenient method for determining the lodging resistance of rice. In the next study, we will conduct further verification under field cultivation, to include factors, such as planting density and water and fertilizer management.