Stability analysis in Indian mustard

Introduction

Indian mustard (Brassica juncea L. Czern and Coss) is one of the most important oilseed crops of the Indian subcontinent, grown across a wide range of agro-climatic conditions during the rabi season. Because of this wide cultivation, the expression of seed yield and its component characters in a given genotype is governed not only by its genetic constitution but also by the environment in which it is grown, and by the interaction between the two. This genotype × environment interaction complicates the selection of genotypes for release, since a genotype performing well in one environment may not perform as well in another. Eberhart and Russell [1] proposed a regression-based model of phenotypic stability in which the regression coefficient (bi), obtained by regressing a genotype's performance in each environment on an environmental index, and the deviation from regression (S²di) are used together with the mean performance to characterise a genotype's stability. A genotype is considered stable under this model if it combines a high mean, a regression coefficient not differing significantly from unity, and a deviation from regression not differing significantly from zero. Genotype × environment interaction and stability of yield and its component characters in Indian mustard have been examined by several workers under diverse growing conditions (2 to 12), who have consistently reported significant interaction and identified genotypes with wider or specific adaptability. The present investigation was carried out to estimate the stability parameters for seed yield and its component characters and to identify stable genotypes among a set of Indian mustard genotypes evaluated at four locations under the first date of sowing.

Materials and Methods

The experimental material comprised ten genotypes of Indian mustard, seven test entries (ACN-226, ACN-237, ACN-250, ACN-308, PM-26, Radhika and DRMRJ-150-35) along with three check varieties (NRCHB 101, Kranti and TAM 108-1). The genotypes were evaluated at four locations, namely Nagpur, Sindewahi, Amravati and Sakoli, under the timely sown conditions, during Rabi 2025. At each location the trial was laid out in a Randomized Block Design with three replications. Data were recorded on 5 randomly selected plants in each replication for plant height(cm), number of branches plant -1 , number of siliquae plant -1 , siliqua density on the main branch and seed yield plant -1 (g), and on a plot basis for days to 50% flowering, days to maturity and 1000 seed weight(g). Analysis of variance was carried out separately for each location following the standard procedure for a Randomized Block Design. Pooled analysis of variance over the four locations partitioned the total variation into components due to locations, genotypes, genotype × location interaction and pooled error. Stability analysis was carried out following the regression model of Eberhart and Russell [1] , in which the mean performance, the regression coefficient (bi) of each genotype on an environmental index, and the deviation from regression (S²di) were estimated for every genotype and character. A genotype was considered stable if its mean was favourable, its regression coefficient did not differ significantly from unity, and its deviation from regression did not differ significantly from zero. Significance of the mean squares was tested against the pooled error by the F test at the 5% and 1% levels of probability, and significance of the regression coefficient against unity was tested by the t test.

Results and Discussion

Pooled analysis of variance (Table 1) revealed significant differences among locations for all eight characters, at the 1% level for seven characters and at the 5% level for days to maturity. Differences among genotypes were highly significant for all eight characters, confirming adequate genetic variability in the material evaluated. The genotype × location interaction was significant for all eight characters, highly significant for six and significant at the 5% level for plant height and number of siliquae plant -1 , indicating that the relative performance of the genotypes changed across the four locations. Significant genotype × environment interaction for yield and its component characters in Indian mustard has been reported earlier [2,3,5,6,8,9] . Since genotype × location interaction was significant for all characters studied, the data were further analysed for stability following the regression model of Eberhart and Russell [1] ; the results are presented in Table 2. The environment + genotype × environment component, when partitioned into linear and non-linear parts, showed the linear component of the environmental effect to be significant for seven of the eight characters, all except days to maturity, indicating that variation among the four locations followed a largely predictable, ordered pattern for these traits. The genotype × environment linear component was significant for two characters, days to maturity and thousand seed weight, showing that genotypes differed genuinely in their regression response to the environmental index for these two traits; for the remaining six characters this component was non-significant. The non-linear component, measured as the pooled deviation, was significant for two characters, days to 50% flowering and number of branches plant -1 , pointing to an unpredictable genotype response to the changing locations for these two traits, while it remained non-significant for the other six characters. For most characters the magnitude of the linear components was considerably larger than the non-linear pooled deviation, supporting the use of the regression model for predicting stability in this material, a pattern also noted in Indian mustard by Kalpande et al. [12] . Estimates of the stability parameters, mean performance, regression coefficient (bi) and deviation from regression (S²di), are presented in Tables 3 and4, and genotypes meeting the stability criteria for each character are summarised in Table 5. Applying the criteria described above, genotype ACN-237 was identified as stable for five of the eight characters studied, namely number of branches plant -1 , number of siliquae plant -1 , siliqua density on the main branch, 1000 seed weight and seed yield plant -1 , the largest number recorded among the genotypes evaluated. ACN-226, ACN-250, DRMRJ-150-35 and NRCHB 101 each met the stability criteria for three characters, and ACN-308, Radhika and TAM 108-1 for two characters each; PM-26 and Kranti did not meet the criteria for any of the eight characters. Similar identification of genotypes combining favourable mean performance with a stable regression response has been reported in Indian mustard [2,3,6,9,10,11,12] . For seed yield plant -1 , the character of greatest direct economic value, ACN-237 recorded a mean of 8.75 g plant -1 along with a regression coefficient not differing significantly from unity and a non-significant deviation from regression, confirming its stable performance across the four locations under the timely sown conditions. Identification of genotypes combining high seed yield with a stable regression response has similarly been reported in Indian mustard, including the recent report of Kalpande et al. [12] , who identified stable genotypes for seed yield among Indian mustard genotypes using the same Eberhart and Russell model.

Conclusion

Under the timely sown conditions across the four locations, seed yield and its component characters in Indian mustard were significantly influenced by location, by genotype and by their interaction. Genotype ACN-237 combined a favourable mean, a regression coefficient close to unity and a deviation from regression close to zero for five of the eight characters studied and is recommended as the most stable genotype among the material evaluated for cultivation across the locations represented in this study. Genotypes ACN-226, ACN-250, DRMRJ-150-35 and NRCHB 101, each stable for three characters, and ACN-308, Radhika and TAM 108-1, each stable for two characters, may also be considered for further evaluation and multi-location testing.

Acknowledgement

AcknowledgmentsI sincerely express my gratitude to all those whose encouragement, guidance, and support contributed to the successful completion of this research. I am especially thankful to my research guide for providing constant supervision, valuable suggestions, and constructive comments throughout the course of the study. I also extend my heartfelt appreciation to the All-India Co-ordinated Research Project on Linseed and Mustard, College of Agriculture, Nagpur, for making available the necessary facilities and resources that enabled the successful execution of this research work.

References

1. Eberhart SA, Russell WA. Stability parameters for comparing varieties. Crop Science. 1966;6:36-40.
2. Sah RP, Kumar A, Ghosh J, Prasad K. Stability study in Indian mustard (Brassica juncea L.). Journal of Hill Agriculture. 2015;6:40-44.
3. Priyamedha, Kumar A, Haider ZA. Stability for seed yield and component traits in Indian mustard (Brassica juncea L.) under Jharkhand condition. Journal of Oilseed Brassica. 2017;8(1):37-42.
4. Jambhulkar S, Kumari V, Chaudhary HK, Sharma BK, Sood P, Guleria SK, et al. Phenotypic stability for seed yield and related traits in Trombay mustard genotypes under North Western Himalayas. Journal of Oilseed Brassica. 2019;10(1):33-37.
5. Shekhawat N, Sharma H, Meena BL. Assessment genotype × environment interaction and stability analysis in Indian mustard (Brassica juncea L.). International Journal of Current Microbiology and Applied Sciences. 2020;9(4):1766-1776.
6. Srivastava A, Srivastava K. Stability analysis in Indian mustard (Brassica juncea L. Czern & Coss.) over normal and terminal heat stress environments. Plant Archives. 2020;20(1):3343-3347.
7. Chauhan A, Tewari S, Pant U, Sharma D. Assessing stability of Indian mustard (Brassica juncea L. Czern) varieties under different agroforestry systems in Tarai region of Uttarakhand. The Pharma Innovation Journal. 2021;10(Special Issue 12):117-1174.
8. Kamdi S, Ingole H, Bhure S, Meshram M, Tajane D, Patil P. Stability analysis in Indian mustard. The Pharma Innovation Journal. 2022;11(7):4351-4354.
9. Kumawat GL, Sharma S. Stability analysis in Indian mustard [Brassica juncea (L.)] varieties. The Pharma Innovation Journal. 2022;11(11):797-801.
10. Yadav AP, Lal GM. Stability analysis for seed yield trait in Indian mustard (Brassica juncea L. Czern. & Coss.). International Journal of Environment and Climate Change. 2023;13(10):3506-3513.
11. Sadhu S, Chakraborty M, Roy SK, Mondal A, Dey S. Stability analysis on elite genotypes of Indian mustard Brassica juncea L. in Terai Agro-Climatic Region. Electronic Journal of Plant Breeding. 2024;15(3):660670.
12. Kalpande PS, Kamdi SR, Bhure S, Tajne D, Meshram MP, Kadu PR, et al. Stability analysis of Indian mustard genotypes through Eberhart and Russell's model. International Journal of Plant & Soil Science. 2025;37(6):56-62.