Evaluating the Cold Tolerance of Stenotaphrum Trin Plants by Integrating Their Performance at Both Fall Dormancy and Spring Green-Up

: Owing to the poor cold tolerance of Stenotaphrum Trin and the urgent need for shade-tolerant grass species in temperate regions of East China, this study evaluated the cold tolerance of 55 Stenotaphrum accessions, aiming to provide shade-tolerant materials for temperate regions. A fine cold-tolerant turfgrass should have both the advantages of delayed fall dormancy and early spring green-up. However, previous research on the cold resistance of turfgrass has mainly focused on the performance of the spring green-up, with less attention paid to the fall dormancy, which has affected the ornamental and application value of turfgrass. This study first dynamically investigated the leaf colour of each accession during the fall dormancy and the coverage during the spring green-up and evaluated the cold resistance of the accession through membership functions and cluster analysis. Significant differences in the cold resistance were found with the assignment of breeding lines to four categories. The weak correlation (R 2 = 0.1682) between leaf colour during the fall dormancy and coverage during the spring green-up indicates that using the performance of a single period to represent the cold resistance of accessions is not appropriate. To test whether using the laboratory-based LT50 and stolon regrowth rating analysis can replace the above-improved method, we conducted a related analysis and found that the fit between these two methods is very poor. This phenomenon is attributed to the poor correlation between the laboratory-based parameters and the pot-investigated data. Therefore, this study presents a cold resistance evaluation method for Stenotaphrum that integrates performance in both the fall dormancy and spring green-up periods. This improved evaluation method cannot be simplified by the growth performance of a single period or replaced by using laboratory-based LT50 and stolon regrowth tests. With the help of this improved method, several excellent cold tolerance accessions (ST003, S13, and S12) were identified for temperate regions of East China.


Introduction
The tropical turfgrass Stenotaphrum Trin has both feed and ornamental value [1].Due to its much greater shade tolerance than that of other warm-season turfgrasses, Stenotaphrum species are preferred for use in shaded landscapes in tropical and subtropical regions [2,3].and evaluated their cold resistance by integrating the two-index using a membership function.By analysing the correlation between autumn/winter leaf colour and next year's green-up coverage, as well as comparing the growth performance of a single growth period with the ranking of a comprehensive evaluation, whether this method can be simplified by the growth performance of the fall dormancy or spring green-up is clarified.Furthermore, the cold resistance of these resources was evaluated through the laboratory-based cold resistance evaluation methods (leaf LT50 and stolon regrowth analysis) and was ranked and clustered through membership functions and cluster analysis.Whether field evaluation methods can be replaced by laboratory evaluation methods was determined by comparing the differences between laboratory cold resistance evaluation data and field evaluation data.By conducting a correlation analysis between field evaluation data and laboratory evaluation data, the reasons for the differences between field evaluation and laboratory evaluation were clarified.Through the above series of experiments, this study ultimately presented a method to compensate for the shortcomings of the current evaluation methods for the cold resistance of turfgrass and clarified whether this improved method can be simplified or replaced by other methods.

Plant Growth Conditions
The 50 accessions for the potted experiment were planted in flower pots with a bottom diameter of 18 cm.The flower pot was filled with half soil and half sand.Each accession contained three replicates.All the accessions were cultivated in the turfgrass nursery of Nanjing Botanical Garden Mem.Sun Yat-Sen, China.Compound fertilizer was applied once a month [31].The dosage used was 0.5 g/pot for the pot experiments.The lowest temperature in winter was −10 • C. The detailed climate information is listed in Supplementary Table S2.

Investigation of Leaf Colour during the Fall Dormancy and Coverage during the Spring Green-Up
From the end of October 2022 to the middle of December 2022, photos of potted accessions were taken during the fall dormancy.The interval at which the images were taken was 7 days.The leaf colour was scored according to the GBT30395-2013 standard as previous reported [32][33][34][35].
From the end of April 2023 to the beginning of June 2023, photos of potted accessions were taken during the green-up.The rate of green-up was obtained through visual inspection [32][33][34]36].

LT50
The experiment was conducted in September 2023.According to previous methods [37], the lethal temperature killing 50% of the plants (LT50) was determined by the electrolyte leakage method, and each treatment included 4 replicates.Briefly, the leaves of healthy plants were removed and rinsed 3 times with deionized water, after which the water that clung to the leaf surface was removed with filter paper.The leaves were cut to a length of approximately 0.5 cm and divided into 5 parts (2 g each).The samples were treated in a cryogenic circulator (Polyscience Company, Warrington, PA, USA), followed by an overnight pre-culture at 4 • C. The low-temperature gradients were set to 4, −1, −6, −11, and −16 • C.After thawing at 4 • C, the electrolyte was extracted by adding 20 mL of deionized water.The conductivity was measured with a conductivity meter (Shanghai Leici Instrument, Shanghai, China) before and after the samples were boiled in a water bath for 15 min.electrolyte leakage = conductivity before boiling water bath conductivity after boiling water bath * 100% The LT50 was obtained by fitting the logistic growth equation: where Y is the electrolyte leakage, x is the temperature, and YM and Y0 are the maximal and initial electrolyte leakage.K is the rate constant.The LT50 is the x value where Y equals 50%.

Stolon Regrowth Experiments
The experiment was conducted in September 2023.Using the methods in [38], stolon regrowth was evaluated via the number of stolons regenerated after low-temperature treatment.The stolons with 5 nodes were treated with 5 temperature gradients (8, 3, −2, −7, and −12 • C).Each treatment included 3 replicates (10 stolons per replicate).After thawing at 4 • C for 24 h, the stolons were planted in plugs filled with half soil and half sand.After 7 days of cultivation in the laboratory, the number of regrowing stolons was determined.

Membership Function Analysis
The comprehensive cold tolerance evaluation was carried out using the membership function method [39][40][41].Briefly, the subordination function value of the average greenness, average coverage, and total relative regrowth rates were calculated by the following equation: Uij = Xij − Ximin Ximax − Ximin whereas the subordination function value of the LT50 was calculated by the following equation: Uij = 1 − Xij − Ximin Ximax − Ximin i: a certain material, j: the index, U: the membership grade, Xij: the measured value of the index in a certain material, Ximin: the minimum value of the index in material i, and Ximax: the maximum value of the index in material i.

Data Statistics and Graphing
The cluster analysis was performed by using squared Euclidean distance coefficient and linkage between groups cluster method in SPSS 19.0 software.Correlation analysis was conducted using Graphpad Prism 9.5 software.For the tables, the data presented are the means ± standard error of at least three replicates.The differences among accessions in a same column was assessed using SPSS 19.0 software and a significant difference is indicated by different letters.

Investigation of Leaf Greenness during the Fall Dormancy and the Coverage during the Spring Green-Up Using the Pot Experiment
Since only 55 accessions survived a winter among the 80 accessions (Supplementary Table S1), detailed cold tolerance evaluations were conducted on the surviving 55 accessions.The investigation of leaf greenness began at the end of October and lasted until late December.As autumn and winter continued, the leaf greenness of Stenotaphrum gradually decreased and all the accessions progressed to a withered yellow state at the end of the monitoring period (Table 1).To objectively reflect the trend of changes in greenness, we averaged the data from 8 time points and obtained the average greenness.This indicator was subsequently used to rank the greenness of the 55 accessions in autumn and winter.The greater the average greenness was, the longer the green period was during the fall dormancy.The average greenness of these accessions varied greatly, ranging from 4.625 to 1.375.The three accessions with the highest average greenness were ST003, S12, and 674925-1.The three accessions with the lowest average greenness were S01, S27, and S28.
The investigation of coverage began at the end of April and lasted until early June.As the green-up period progressed, the grass gradually turned green, and by the end of monitoring, most accessions had approached a complete green-up state (Table 2).To objectively reflect the trend of this green-up change, we averaged the greening coverage at four-time points and obtained the average coverage.By using this indicator, the average coverage of the 55 accessions during the green-up was ranked.A higher average coverage indicates the faster green-up ability of an accession.The average coverage varied greatly among the accessions, ranging from 85% to 9%.The three accessions with the highest average coverage were 291594, S62, and S13.The three accessions with the lowest average coverage were S02, S10, and S28.Note: the word "Average" was abbreviated as Av.The significant differences among accessions in a same column is indicated by different letters.2.000 ± 0.000 op 6.000 ± 0.006 pq 15.000 ± 0.000 q 30.000 ± 0.000 j 13.250 ± 0.001 x S10 0.000 ± 0.000 p 5.000 ± 0.000 q 15.000 ± 0.000 q 25.000 ± 0.000 k 11.250 ± 0.000 y S28 1.000 ± 0.000 p 5.000 ± 0.000 q 10.000 ± 0.000 r 20.000 ± 0.000 l 9.000 ± 0.000 y Note: the word "Average" was abbreviated as Av.The significant differences among accessions in a same column is indicated by different letters.

Membership Function Analysis and Cluster Analysis Based on Pot Experiment Results
The data for the leaf colour during the fall dormancy and the turfgrass coverage during the spring green-up was analysed using membership function analysis.The score obtained was used to rank the cold tolerance.The cold tolerance varied greatly among the accessions, ranging from 0.87 to 0. The three accessions with the highest cold tolerance were ST003, S13, and S12.The three accessions with the lowest cold tolerance were S27, S02, and S28 (Table 3).The picture of these six accessions at fall dormancy and spring green-up was presented in Supplementary Figure S1.Cluster analysis revealed that these accessions were divided into two categories: cold tolerant and cold sensitive (Figure 1).The cold tolerance category can be further subdivided into two subcategories: super cold-tolerance and middle cold-tolerance.The cold sensitive category can be subdivided into two subcategories: super cold-sensitive and middle cold-sensitive (Figure 1).The accessions ranking 1-13 are mainly concentrated in the super cold-tolerant subcategory.The accessions ranking 14-46 are mainly concentrated in middle cold-tolerant subcategory.Those that ranked 47-51 are gathered in the middle cold-sensitive subcategory, and those that ranked 52-55 are gathered in the super cold-sensitive subcategory (Figure 1).The correlation analysis revealed that a positive correlation existed between autumn/ winter leaf greenness and the turfgrass coverage during the spring green-up (Figure 2).Accessions with delayed fall dormancy will have an early spring green-up rate.However, the poor R 2 (0.1682) indicates a weak correlation, suggesting that it is inappropriate to simplify the cold resistance evaluation by investigating the growth performance of a single period.This result was further confirmed by the discrepancy between the ranking of autumn leaf colour or spring coverage rate and the ranking of comprehensive evaluation (Table 3).

Laboratory-Based Leaf LT50 and Stolon Regrowth Rate Analysis
To investigate whether the pot evaluation can be replaced by the laboratory-based cold resistance evaluation methods, we conducted leaf LT50 and stolon regrowth rate analysis.The cold tolerance of the leaves was measured through an electrolyte leakage test (Table 4).Logistic equation fitting was performed on the relative conductivity of each accession, and the fitting degree (R 2 ) of each equation was greater than 0.86, indicating that the equation fit was good and that the obtained LT50 was highly reliable.A more negative LT50 value indicates a greater cold tolerance ability for leaves.A significant difference in leaf LT50 was detected among the accessions, ranging from −9.32 to 3.45.The three accessions with the most favourable LT50 values were S39, S005, and 410364.The three accessions with the lowest LT50 values were S11, S006, and S27.

Laboratory-Based Leaf LT50 and Stolon Regrowth Rate Analysis
To investigate whether the pot evaluation can be replaced by the laboratory-based cold resistance evaluation methods, we conducted leaf LT50 and stolon regrowth rate analysis.The cold tolerance of the leaves was measured through an electrolyte leakage test (Table 4).Logistic equation fitting was performed on the relative conductivity of each accession, and the fitting degree (R 2 ) of each equation was greater than 0.86, indicating that the equation fit was good and that the obtained LT50 was highly reliable.A more negative LT50 value indicates a greater cold tolerance ability for leaves.A significant difference in leaf LT50 was detected among the accessions, ranging from −9.32 to 3.45.The three accessions with the most favourable LT50 values were S39, S005, and 410364.The three accessions with the lowest LT50 values were S11, S006, and S27.
Since the spring green-up of Stenotaphrum depends on the budding of the stolons, the regrowth ability of the stolons was evaluated.For all the accessions, as the temperature during the low-temperature pre-treatment decreased, the ability of the stolons to regrow significantly decreased.At −7 • C, most accessions lost their ability to regrow, while at −12 • C, all accessions lost their ability to regrow (Table 5).Given the significant differences in the ability of the accessions to regrow at relatively normal temperatures (8 • C), the regrowth ability of stolons at 8 • C was used as a control and the regrowth ability of stolons under other temperature treatments was standardized.The total relative regrowth rate of stolons at low temperatures was subsequently obtained by averaging the total relative regrowth rate of stolons at −3 • C, −2 • C, and −7 • C. A higher total relative regrowth rate indicates a greater ability for stolons to regrow in the next year.The total relative regrowth rate varied greatly among the accessions, ranging from 52% to 13%.The top three accessions in terms of total relative regrowth rate were S30, S39, and S25.The three accessions with the lowest total relative regrowth rates were S27, S10, and 300129.Note: The significant differences among accessions in a same column is indicated by different letters.

Membership Function Analysis and Cluster Analysis Based on Laboratory Results
The score and rank based on the laboratory results were obtained by membership function analysis (Table 6).A significant difference between the laboratory-based rank and the pot-evaluated rank was detected, and the coincidence degree between them was poor (Tables 3 and 6).Cluster analysis revealed that these accessions could be divided into two categories and three subcategories based on the laboratory results (Figure 3).Specifically, the accessions ranking 1-9 are clustered in one subcategory (cold-tolerant).The accessions ranking 10-50 are clustered in one subcategory (middle cold-sensitive), while those rankings 51-55 are clustered in one subcategory (super cold-sensitive) (Figure 3).A very poor consistency with the clustering results between the pot evaluation and the laboratory-based evaluation was found (Figures 1 and 3).

Correlation Analysis between the Laboratory-Based Data and the Pot Analysis Data
A correlation analysis was conducted between LT50 and other data.It was found that LT50 was not correlated with the average greenness during the autumn/winter nor average coverage during the spring green-up but was correlated with the total relative regrowth rate (Figure 4).

Correlation Analysis between the Laboratory-Based Data and the Pot Analysis Data
A correlation analysis was conducted between LT50 and other data.It was found that LT50 was not correlated with the average greenness during the autumn/winter nor average coverage during the spring green-up but was correlated with the total relative regrowth rate (Figure 4).A correlation analysis was conducted between the total relative regrowth rate and other data.It was found that the total relative regrowth rate was not correlated with average greenness but was positively correlated with average coverage during the spring green-up (Figure 5).

Establishment of a Method for Evaluating the Cold Tolerance of Stenotaphrum by Integrating Its Performance at Both the Fall Dormancy and the Spring Green-Up
It is assumed that the cold resistance of the leaves of perennial turfgrass is crucial when it first encounters autumn/winter cold, while the regrowth ability of stolons is crucial for rapid spring green-up.Briefly, when plants encounter chilling injury, their leaves strive to maintain photosynthesis by prolonging the green period and synthesizing energy to resist chilling injury.Secondly, the leaves transport photosynthate to the stolon to store energy for spring green-up.Leaves with a strong cold resistance are more conducive to the implementation of the above process.The bud from the stolon was the initial site for spring A correlation analysis was conducted between the total relative regrowth rate and other data.It was found that the total relative regrowth rate was not correlated with average greenness but was positively correlated with average coverage during the spring green-up (Figure 5).

Correlation Analysis between the Laboratory-Based Data and the Pot Analysis Data
A correlation analysis was conducted between LT50 and other data.It was found that LT50 was not correlated with the average greenness during the autumn/winter nor average coverage during the spring green-up but was correlated with the total relative regrowth rate (Figure 4).A correlation analysis was conducted between the total relative regrowth rate and other data.It was found that the total relative regrowth rate was not correlated with average greenness but was positively correlated with average coverage during the spring green-up (Figure 5).

Establishment of a Method for Evaluating the Cold Tolerance of Stenotaphrum by Integrating Its Performance at Both the Fall Dormancy and the Spring Green-Up
It is assumed that the cold resistance of the leaves of perennial turfgrass is crucial when it first encounters autumn/winter cold, while the regrowth ability of stolons is crucial for rapid spring green-up.Briefly, when plants encounter chilling injury, their leaves strive to maintain photosynthesis by prolonging the green period and synthesizing energy to resist chilling injury.Secondly, the leaves transport photosynthate to the stolon to store energy for spring green-up.Leaves with a strong cold resistance are more conducive to the implementation of the above process.The bud from the stolon was the initial site for spring It is assumed that the cold resistance of the leaves of perennial turfgrass is crucial when it first encounters autumn/winter cold, while the regrowth ability of stolons is crucial for rapid spring green-up.Briefly, when plants encounter chilling injury, their leaves strive to maintain photosynthesis by prolonging the green period and synthesizing energy to resist chilling injury.Secondly, the leaves transport photosynthate to the stolon to store energy for spring green-up.Leaves with a strong cold resistance are more conducive to the implementation of the above process.The bud from the stolon was the initial site for spring green-up.Therefore, the cold tolerance of perennial turfgrass relies on both the leaves and stolons.This mechanism is completely different from the mechanism of cold resistance in annual plants, which do not experience fall dormancy and spring green-up [42][43][44].Moreover, the use of an evaluation system for the cold resistance of annual plants is inappropriate for evaluating the cold resistance of perennials [45][46][47][48].A successful cold-tolerant perennial grass plant needs to be characterized by both delayed fall dormancy and early spring green-up.In this study, through the dynamic investigation of changes in leaf colour in autumn/winter and green coverage in the following year and the membership function analysis and cluster analysis, the growth performance of 55 Stenotaphrum accessions in these two stages was evaluated (Tables 1-3; Figure 1).This improved evaluation method compensates for the limitation of the previous cold tolerance evaluation method, which mainly focused on the spring green-up stage and paid less attention to fall dormancy [10,11].The cold-tolerant accessions selected by this improved method have more ornamental value and are easier for the public to accept.The results showing that the autumn/winter leaf colour and the spring green-up coverage has a weak correlation and that the rank fit between a single period and these two periods is poor indicates that the growth performance of a single period cannot replace the overall evaluation (Figure 2; Table 3).The use of technical methods can only regulate the growth performance of one period (e.g., fall dormancy) and does not affect the growth performance of another period (e.g., spring green-up), supporting the above results [49,50].

This Improved Evaluation Method Cannot Be Replaced by Laboratory Evaluation
The laboratory-based cold tolerance results and the pot cold tolerance results are inconsistent (Tables 3 and 6; Figures 1 and 3).There was no correlation between LT50 and the autumn leaf colour or the spring green-up coverage.The total relative regrowth was not correlated with the autumn/winter leaf colour and only correlated with the next year's green coverage (Figures 4 and 5).This explains why LT50 and stolon regrowth data cannot fully reflect the cold resistance of Stenotaphrum accessions.The inconsistency of the LT50 or the stolon regrowth data to the cold tolerance found in other perennial grass might be also partially attributed to this reason [16,17].
The stolon regrowth rate is positively correlated with the spring green-up coverage (Figure 5), which is in agreement with reports on other turfgrasses [5,51].These data can to some extent reflect the situation of spring green-up of turfgrass.However, caution should be taken when using this method for cold tolerance evaluation in the future due to the weak correlation.

Screening of Several Excellent Cold-Tolerant Accessions That Can Be Directly Used in Temperate Regions of China
Through the comprehensive evaluation method, we selected excellent-cold-resistance accessions, which were represented by ST003, S13, and S12 (Table 3).These excellent accessions were obtained from Sydney, Australia, Wenchang, Hainan (China), and Tengchong, Yunnan (China) (Supplementary Table S1).Although several studies have evaluated the cold resistance of Stenotaphrum plants, most have focused on accessions from Europe and North America [19][20][21][22].This study evaluated the cold resistance of Stenotaphrum plants mainly using Chinese accessions.Due to cold tolerance being an important factor limiting plant geographical distribution [52] and cold tolerance being evaluated in temperate regions in China, the selected excellent-cold-resistance accessions can be directly applied locally.In the future, we plan to further evaluate the shade tolerance of these cold-tolerant accessions and screen for excellent shade-tolerant grass accessions that can be applied in temperate regions.

Conclusions
A turfgrass with good cold resistance should have both the advantage of delayed fall dormancy and early spring green-up.Based on the situation that the previous cold tolerance evaluation of turfgrass mainly focused on the spring green-up and paid less attention to its performance in the fall dormancy, this study integrates the performance of these two stages by dynamically investigating the autumn/winter leaf colour and next year's coverage and uses membership functions and cluster analysis to comprehensively evaluate the cold resistance of 55 Stenotaphrum accessions.This method cannot be simplified by the performance of one period, nor can it be replaced by indicator measurements conducted in the laboratory.The establishment of this method compensates for the shortcomings of previous methods for evaluating the cold tolerance of turfgrass.With the help of this improved method, we have screened several excellent-cold-tolerance accessions (ST003, S13, and S12) for the temperate regions of East China.

Supplementary Materials:
The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/horticulturae10070761/s1,Table S1: Aaccession information; Table S2: Extreme temperature and rainfall in Nanjing during the experimental process; Figure S1: The performance of six accessions with extreme cold tolerance and cold sensitivity.

Figure 1 .
Figure 1.Cluster analysis based on pot experiment results.

Figure 1 .
Figure 1.Cluster analysis based on pot experiment results.

Figure 2 .
Figure 2. Correlation between the average greenness during the fall dormancy and average coverage during the spring green-up.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 2 .
Figure 2. Correlation between the average greenness during the fall dormancy and average coverage during the spring green-up.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 3 .
Figure 3. Cluster analysis based on laboratory results.

Figure 3 .
Figure 3. Cluster analysis based on laboratory results.

Figure 4 .
Figure 4. Correlation between LT50 and other parameters.(A) Correlation between LT50 and average greenness.(B) Correlation between LT50 and average coverage.(C) Correlation between LT50 and total relative regrowth rate.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 5 .
Figure 5. Correlation between total relative regrowth rate and other parameters.(A) Correlation between total relative regrowth rate and average greenness.(B) Correlation between total relative regrowth rate and average coverage.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 4 .
Figure 4. Correlation between LT50 and other parameters.(A) Correlation between LT50 and average greenness.(B) Correlation between LT50 and average coverage.(C) Correlation between LT50 and total relative regrowth rate.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 4 .
Figure 4. Correlation between LT50 and other parameters.(A) Correlation between LT50 and average greenness.(B) Correlation between LT50 and average coverage.(C) Correlation between LT50 and total relative regrowth rate.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 5 .
Figure 5. Correlation between total relative regrowth rate and other parameters.(A) Correlation between total relative regrowth rate and average greenness.(B) Correlation between total relative regrowth rate and average coverage.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.

Figure 5 . 4 . 1 .
Figure 5. Correlation between total relative regrowth rate and other parameters.(A) Correlation between total relative regrowth rate and average greenness.(B) Correlation between total relative regrowth rate and average coverage.The P and R 2 are the fitted parameters.The symbol * indicates a correlation.4.Discussion 4.1.Establishment of a Method for Evaluating the Cold Tolerance of Stenotaphrum by Integrating Its Performance at Both the Fall Dormancy and the Spring Green-Up Author Contributions: J.-Q.Z. and Z.-Y.W. designed the work.J.Q. and D.-L.H. carried out the experiments.D.-L.H., J.-Q.Z. and Z.-Y.W. wrote the paper.J.-Y.Z., J.-B.C., D.-J.S. and J.-X.L. revised the paper.All authors contributed to the article.All authors have read and agreed to the published version of the manuscript.Funding: The authors are grateful for the financial support provided by the Forestry Science and Technology Innovation and Promotion Project of Jiangsu Province (LYKJ[2023]17), the Jiangsu Provincial Double-Innovation Doctor Program (Grant No. JSSCBS20221643), the Jiangsu Institute of Botany Talent Fund (Grant No. JIBTF202210), and the Program for the Young Innovative Talents of Jiangsu Vocational College of Agriculture and Forest (Grant No. 2021kj26).

Table 1 .
Leaf greenness during the fall dormancy.

Table 2 .
Coverage survey at the spring green-up.

Table 3 .
Membership function analysis based on pot experiment results.Note: the word "Average" was abbreviated as Av.The significant differences among accessions in a same column is indicated by different letters.

Table 6 .
Membership function analysis based on laboratory results.The significant differences among accessions in a same column is indicated by different letters.