Multivariate analysis of genetic divergence in Lathyrus (Lathyrus sativus L.) germplasm

Introduction

Lathyrus (Lathyrus sativus L.) is a member of the Leguminosae family, it is a significant crop rich in protein (28%), next to soybeans (43%). Papilionaceae is a subfamily of lathyrus with chromosomal number 2n=14. Lathyrus has extensive tolerance of drought, water logging soil. Analysis of genetic diversity in relation to geographical distribution may contribute to the identification of centers of genetic diversity and elucidate the origins and domestication history of cultivated crop species. The edible Lathyrus (Lathyrus sativus L.) originated in the Mediterranean region of West Central Asia. It was domesticated in North India, where remains dating back 3600 years have been found. The predominant cultivation practice for Lathyrus (Lathyrus sativus L.) in India is the relay cropping system, commonly referred to as the "utera" system. The crop is well adapted to diverse environmental conditions, exhibiting considerable tolerance to drought as well as resilience under conditions of excessive rainfall. Lathyrus (Lathyrus sativus L.) is relatively resistant to many diseases and insect pests. The effectiveness of a crop improvement programme is largely determined by the extent of genetic variability available within the breeding material. Genetic diversity serves as the foundation for plant breeding by providing opportunities to select diverse parents with desirable traits for hybridization. Systematic evaluation of germplasm enables the identification of genetic differences among accessions for important agronomic characters, thereby supporting the development of improved cultivars. Furthermore, characterization of germplasm contributes to the efficient conservation and utilization of valuable genetic resources in future breeding programmes.

Multivariate statistical methods, particularly Mahalanobis D² analysis, are widely employed to assess genetic divergence among crop genotypes. This approach groups genotypes into different clusters by considering variation across several quantitative traits simultaneously, thereby providing a comprehensive measure of genetic diversity. The clustering pattern helps breeders to identify genetically diverse parents for hybridization. Crosses involving genotypes from distantly related clusters are expected to produce greater genetic variation and increase the probability of obtaining superior recombinants in subsequent generations. Estimation of genetic diversity not only supports the selection of superior genotypes but also contributes to the development of high-yielding, climateresilient, and low-ODAP cultivars with improved adaptability. Therefore, systematic evaluation of available germplasm for genetic divergence is an important step toward the effective genetic improvement of Lathyrus

Materials and Methods

The present research work was undertaken during Rabi 2024-25 at Research farm of Agricultural Botany Section, College of Agriculture, Nagpur to estimate the genetic diversity amaong 64 germplasm of lathyrus. These 64 germplasm were grown in RBD (Randomized Block Design) for ten characters and data was recorded on five competitive plants. This characters are days to 50% flowering (plot wise), days to maturity (plot wise), plant height (cm), number of primary branches plant -1 , number of pods plant -1 , number of seeds pod -1 , 100 seed weight (g), seed yield plant -1 (g), harvest index (%), ODAP content (%).The recorded observations were statistically analyzed using analysis of variance (ANOVA) to evaluate the existence of significant variation among the genotypes for the studied traits. Genetic divergence was assessed through Mahalanobis D 2 analysis, and the genotypes were classified into distinct clusters using Tocher's clustering technique. Mean intra-cluster and inter-cluster distances were computed to measure the degree of genetic diversity within and among clusters. The percentage contribution of each trait to the overall genetic divergence was also determined, while canonical vector analysis was employed to identify the characters responsible for the observed variability. The findings of these analyses were utilized to identify genetically diverse and agronomically promising genotypes that can serve as potential parents in future hybridization and crop improvement programmes.

Result and Discussion

The mean squares for genotypes were highly significant for all ten characters which indicated significant variation among all the genotypes for for the characters under study. Among 64 genotypes, the genotype Mahateora (57.66 days) attained earliest flowering followed by IC-120472 (58.33), NLK-49 (58.66 days) and NLK-63 (58.66 days). The genotype which attained 50% flowering late were NLK-74 (65.66 days) followed by AKL-12 (65.00 days), NLK-73 (65.00 days) and EC-209026 (65.00 days). The genotype BIOR-231 (115.66 days) matured earliest followed by NLK-36 (116 days), NLK-102 (116 days), NLK-53 (116 days) and the genotype NLK-63 (127.33 days) attained late maturity followed by NLK-94 (124.33 days). The genotype NLK-3 (90.29 cm) was tallest followed by NLK-55 (89.56 cm) and the genotype RLK-1045 (67.17 cm) was shortest followed by IC-120491 (68.57 cm) and IC-120472 (68.77 cm). The genotype NLK-55 (5.20) and NLK-70 (5.20) recorded maximum number of primary branches plant -1 while minimum number of primary branches plant -1 was recorded by the genotype NLK-63 (3.60) followed by NLK-94 (3.70). The genotype AKL-19 (91.51) produced maximum number of pods plant -1 followed by AKL-12 (90.47) and NLK-37 (88.80). The minimum number of pods plant -1 was produced by the genotype NLK-54 (45.69) followed by NLK-3 (55.47) and BIOR-222 (57.58). The genotype NLK-37 (3.82) followed by NLK-3 (3.71), NLK-48 (3.70) and NLK-6 (3.70) had maximum number of seed pod -1 . The genotype IC-120469 (1.99) exhibited minimum number of seeds pod -1 . The genotype Mahateora (11.48 g) had maximum seed weight followed by RLK-1045 (11.47g) and the genotype NLK-63 (5.66 g) had minimum seed weight followed by PUSA-90-2 (6.23 g). The genotype NLK-49 (59.56 g) had maximum yield and the genotype JRL-115 (10.81g) had minimum yield followed by KH-1 (11.64 g) and IC-120491 (11.79 g). The genotype IC-120491 (51.63%) had maximum harvest index followed by PUSA-56 (51.61%) and the genotype NLK-47 (22.42%) had minimum harvest index. The genotype Ratan (0.236 %) contain less ODAP followed by NLK-73 (0.250 %), Prateek (0.279 %) and Mahateora (0.276 %). Although per se performance provides a preliminary indication of the breeding value of a genotype, it does not adequately reveal its potential for producing desirable progenies over successive generations. Based on the evaluation of yield and its associated traits among sixty-four genotypes of lathyrus, NLK-49, JRL-115, IC-120472, NLK-63, BIOR-231 NLK-36, NLK-102, NLK-53, NLK-55, NLK-70, RLK-1045, NLK-37, NLK-3, NLK-6, NLK-48, AKL-12, AKL-19, NLK-37 exhibited superior performance compared with the check variety Ratan. These genotypes recorded higher values for important yield-attributing characters such as number of pods per plant, number of seeds per pod and 100seed weight. The grouping of 64 genotype into different clusters was done by Tocher's method. The entire genotypes (64) were grouped on the basis of D 2 statistics into 14 clusters. Cluster I was largest comprising of 41 genotypes, followed by cluster II comprising of 5 genotypes, cluster III, IV, V, VI, VII, and VIII, comprising of single genotype and cluster IX comprising of 4 genotypes again cluster X, XI, XII comprising of single genotype and cluster XIII comprising of 4 genotypes at last cluster XIV comprising of single genotype. Studied by Yadav et al. (2026) and Das et al. (2026) The promising check Ratan, Prateek and Mahateora grouped into cluster IX. The coefficients of the first three canonical vectors revealed the relative contribution of different traits to genetic divergence among the genotypes. In vector I, which explained 19.87% of the total variation, major contributions were observed from days to 50% flowering, ODAP content (%), number of pods plant -1 . Vector II accounted for 16.23% of the total variation, where number of seed pod -1 , number of primary branches plant -1 , number of pods plant -1 , days to 50% flowering, 100 seed weight, plant height. In vector III, contributing 14.63% of the total variation, 100 seed weight, harvest index, number of primary branches plant -1 , days to 50% flowering were the key characters influencing divergence. The results indicate that genotypes selected based on traits such as number of seeds pod -1 , days to 50% flowering, seed yield plant -1 , number of pods plant -1 , and 100 seed weight are likely to exhibit substantial genetic diversity and may serve as promising parents for hybridization programs. Başaran and Sarıtaş (2026) The intra cluster distance range from 0.00 to 28.61. Cluster XIII possessed highest intra cluster distance (D² = 28.61) followed by cluster II (D² = 16.92) and cluster I (D² = 13.70) and cluster IX (D 2 = 12.66). The average inter cluster distance was maximum between XI and cluster XIV (D² = 224.69), followed by cluster V and cluster XIV (D 2 = 204.77), cluster X and cluster XIV (D 2 = 211.95). The inter cluster distance was found to be minimum between cluster VIII and cluster XII (D² =9.66). From the data it can be observed that the average intra cluster distance was maximum in cluster XIII, cluster II, cluster I and cluster IX and the average inter cluster distance was maximum in cluster XI and XIV and also in between cluster X and cluster XIV. Widely diverged clusters remain distinct in different environment Therefore, genotypes originating from widely separated clusters can be utilized in hybridization programs to generate a broader range of variability among the segregating progenies. The comparison of cluster means for ten characters under study marked considerable genetic difference between groups. Highest cluster mean for days to 50% flowering was recorded by cluster X (64.33) followed by cluster XII (64.00) and cluster VI (63.67) while cluster XIV (58.67) followed by cluster V (59.00) and cluster IX (60.67) represented the lowest mean for days to 50% flowering. Highest cluster means for days to maturity was recorded by cluster XIV (126.00) followed by cluster XIII (122.67) and cluster IX (121.42) while cluster II (118.13) followed by cluster X (118.33) and cluster I (119.60) represented the lowest mean for days to maturity. Highest cluster mean for plant height was recorded by cluster IV (89.56) followed by cluster XIII (84.76) and cluster XII (82.75) while cluster II (74.85) followed by cluster XI (68.57) and cluster V (68.73) represented the lowest mean for plant height. Highest cluster mean for number of primary branches plant -1 was recorded by cluster IV (5.21) followed by cluster V (4.60) and cluster IX (4.55) while cluster III and cluster VII (3.80) represented the lowest mean for number of primary branches plant -1 . Highest cluster mean for number of pods plant -1 was recorded by cluster X (88.59) followed by cluster VI (79.26) and cluster IX (78.72) while cluster VII (60.92) and cluster III (61.96) represented the lowest mean for number of pods plant -1 . Highest cluster mean for number of seeds per pod was recorded by cluster VI (3.70) followed by cluster VIII (3.31) and cluster IV (3.30) while cluster III (2.00) followed by cluster XIV (2.10) and cluster V (2.10) represented the lowest mean for number of seeds per pod. Highest cluster mean for 100 seed weight, cluster IX (10.77) followed by cluster II (10.75) while cluster XIII (7.10) and cluster IV and cluster XII (7.25) represented the lowest mean for 100 seed weight. Highest cluster mean for Seed Seed yield plant -1 was recorded by cluster XIV (59.56) while cluster V (10.82) followed by cluster XI (11.80) and cluster III (14.99) represented the lowest mean for Seed yield plant -1 . Highest cluster mean for harvest index was recorded by cluster XI (51.64) followed by IX (45.50) and cluster (45.36) while cluster X (31.46) followed by XIII (32.52) represented the lowest mean for harvest index. Highest cluster mean for ODAP was recorded by cluster XI (0.61) followed by cluster X (0.58) and cluster II (0.55) while cluster IX (0.26) followed by cluster VIII and cluster XII (0.44). The 64 genotypes exhibited significant differences for the traits under study. The characters like days to number of seeds pod -1 , number of pods plant -1 , and Seed yield plant -1 , Number of primary branches plant -1 , number of seeds pod -1 were identified to contribute maximum towards genetic divergence from canonical analysis studied, cluster mean estimate and contribution of individual characters to divergence. Sixty four genotypes were grouped into fourteen clusters. Ten inter cluster combination were identified for selecting promising diverse genotypes to be included in crossing programme to get potential variety. Twenty seven genotypes viz. AKL-19, NLK-94, NLK-67, NLK-62, NLK-51, EC-209081, NLK-39, NLK-54, NLK-74, NLK-48, EC-2609071, NLK-6, AKL-7, JRL-16, BIOR-231, NLK-49, IC-120472, NLK-37, IC-120491, NLK-40, KH-1, NLK-102, NLK-5, PUSA-24, AKL-12, NLK-44, BIOR-222 were identified as potential parents and we can use these genotype in crossing programme.

Conclusion

Significant differences were observed for sixty-four germplasm. Data were recorded on five competitive plants for 10 quantitative characters viz. days to 50% flowering (on plot basis), days to maturity (on plot basis), plant height (cm), number of primary branches plant -1 , number of pods plant -1 , number of seeds pods -1 , 100 seed weight (g), yield plant -1 (g), harvest index (%) and ODAP content (%). Mahalanobis generalized distance (Mahalanobis 1928 andRao 1952) was used in this investigation for computing genetic divergence for quantitative characters. The sixtyfour germplasm exhibited significant differences for the traits under study. The characters like days to number of seeds pod -1 , number of pods plant -1 , and yield plant -1 , Number of primary branches plant -1 , number of seeds pod

References

1. Başaran U, Sarıtaş Y. Evaluating some local grass pea (Lathyrus sativus L.) varieties under warm humid continental climate conditions for agricultural sustainability with a special reference to seed yield and β-ODAP content. Frontiers in Plant Science. 2026;17:1750659.
2. Das S, Sawarkar A, Saha S, Raman RB, Dasgupta T. Principal component and cluster analysis in mungbean (Vigna radiata (L.) Wilczek). Legume Research -An International Journal. 2026;49(1):15-24.
3. Gaur VS, Maloo SR. Genetic divergence in grass pea (Lathyrus sativus). Indian Journal of Agricultural Sciences. 2011;81(2):172-173.
4. Khillari AV, Patil AE, Gawande NG, Jayde VS. Genetic divergence analysis among explored germplasm and selected varieties in lathyrus (Lathyrus sativus L.). Journal of Food Legumes. 2016;29(1):1417.
5. Madke VS, Patil SA, Deotale RD, Patil SR, Shende PV, Patil SS, et al. Genetic divergence for selection of parents for hybridization in lathyrus. South Asia. 2019;5(7).
6. Mahalanobis PC. On the generalized distance in statistics. Proceedings of the Indian National Science Academy. 1936;2:49-55.
7. Mahapatra NS, Das A, Bhattacharyya PK, Bhattacharya S, Pal S, Barpete S. Studies on genetic variability, divergence and association of characters in grasspea. Journal of Crop and Weed. 2020;16:155-161.
8. Mishra A, Sahu M, Pandey HK. Assessment of variability and association among green forage yield and contributing traits in indigenous genotypes of grass pea [Lathyrus sativus (L.)]. Journal of Advances in Biology & Biotechnology. 2025;28(1):503-512.
9. Ranjithkumar G, Debnath S, Rajasekhar D. Genetic divergence analysis in grasspea (Lathyrus sativus L.). International Journal of Current Microbiology and Applied Sciences. 2019;8(12):562-567.
10. Rao CR. Advanced statistical methods in biometric research. New York: John Wiley & Sons; 1952.
11. Sharma D, Padmanabham. Food science and technology. In: Liener IE, editor. New York: Academic Press; 1969. p. 209-233.
12. Yadav A, Nath S, Chourasiya VK, Boro J, Kumar D, Srikanth B, et al. Multivariate genetic diversity assessment in chickpea (Cicer arietinum L.) genotypes under different environmental conditions. Legume Research. 2026;49(6):967-973.