Character association and direct/indirect effects of component traits on grain yield in finger millet [Eleusine coracana (L.) Gaertn.]

Introduction

Finger millet is an annual grass species of the Poaceae family that holds a versatile place in agriculture as a primary source of food, fodder, and livestock feed. Characterized by its remarkable capacity to thrive on marginal, nutrient-poor lands and endure harsh environmental stresses, millet cultivation has been sustained across various regions for over 5,000 years (Gowda et al., 2006) [13] . A strictly self-pollinating species, finger millet is an allotetraploid possessing a chromosome count of (2n = 4X = 36). Among small millets, it carries the highest economic significance, although related species such as foxtail, proso, barnyard, kodo and little millet remain indispensable to the subsistence of tribal farming communities (Patil et al., 2018) [28] . In India, finger millet is cultivated predominantly under rainfed regimes. Frequently termed a "poor person's crop", it serves as a staple for resource-limited smallholders managing waterconstrained agro-ecosystems across the semi-arid regions of South Asia and Sub-Saharan Africa (Vasisth et al., 2022) [36] . Historical accounts place its introduction to the Indian subcontinent more than three millennia ago. Beyond its agronomic resilience, the grain exhibits exceptional post-harvest storability without vulnerability to storage pests, making it a critical security buffer during periods of famine. Nutritionally, its grain profile is exceptionally rich, containing roughly 8 to 10 times more calcium than traditional staple cereals such as wheat and rice (Sharma et al., 2022) [32] . While finger millet is primarily grown for human grain consumption, its crop residues serve as a crucial source of dry matter for livestock, particularly during dry seasons when green forage is scarce. Finger millet straw is highly valued as livestock fodder due to its superior nutritional quality, containing up to 61% digestible nutrients. Although enhancing grain yield is the primary objective of breeding programmes, yield is a complex quantitative trait governed by multiple genes and strongly influenced by environmental factors, making direct selection less effective. Correlation analysis is an important statistical approach for understanding the relationships between grain yield and its component traits, thereby facilitating the identification of promising genotypes. Furthermore, path coefficient analysis separates correlation coefficients into direct and indirect effects, providing a more precise understanding of the contribution of individual traits to grain yield. This information enables breeders to identify key traits for effective selection. Therefore, the present study was undertaken to investigate the association among yield and its related traits through correlation and path coefficient analyses in 35 finger millet genotypes.

Materials and Methods

During Kharif-2024, 35 finger millet genotypes were evaluated in a Randomized Block Design with three replications across four SDAU research stations: Centre for Millets Research, Deesa; Agronomy Instructional Farm, Sardarkrushinagar; Maize Research Station, Bhiloda; and Agricultural Research Station, Ladol. Entries were grown in three-meter single row plots at 30 cm × 10 cm spacing following standard package of practices. Days to 50% flowering and days to maturity were monitored on a plot basis, whereas quantitative traits were recorded from five randomly sampled, non-border plants per replication. The following estimates of covariances were worked out as reviewed by Panse and Sukhatme (1978) [25] . σ p i p j = σ g i g j + σ e i e j σ p i p j = Phenotypic covariance between i th and j th character σ g i g j = Genotypic covariance between i th and j th character σ e i e j = Environmental covariance between i th and j th character Genotypic correlations primarily reflect the effects of pleiotropy and genetic linkage, whereas phenotypic correlations represent the combined influence of genetic and environmental factors. Genotypic and phenotypic correlation coefficients were computed using the procedure described by Hazel (1943) [14] . The significance of the correlation coefficients was tested for n -2 degrees of freedom by calculating the t-value according to the method of Panse and Sukhatme (1985) [26] . Path coefficient analysis was performed following the approach proposed by Wright (1921) [39] and later adapted by Dewey and Lu (1959) [7] . Genotypic correlation coefficients of the 15 component traits with grain yield per plant were used to partition the correlations into direct and indirect effects, thereby assessing the contribution of each trait to grain yield.

Results and Discussion

Assessment of trait associations is an important approach in plant breeding, as it facilitates indirect selection for yield improvement by identifying traits closely related to grain yield. Correlation analysis helps determine the magnitude and direction of relationships among yield and its component traits, thereby supporting the selection of superior genotypes. In the present study, genotypic correlation coefficients were generally higher than the corresponding phenotypic coefficients, indicating that genetic factors contributed more strongly to trait associations than environmental effects. The genotypic (rg) and phenotypic (rp) correlation coefficients of sixteen characters were presented in Table 1.

Trait productive tillers per plant was found to be in positive and significant phenotypic correlation with harvest index (rp = 0.19) and negative and significant genotypic correlation with plant height (rg = -0.61), number of finger per earhead (rg = -0.36) and fodder yield per plant (rg = -0.46). Abhilash et al. (2020) reported negative and significant results for days to 50% flowering. At both the genotypic and phenotypic levels, the number of fingers per earhead was significantly positively correlated with the plant height (rg = 0.36 and rp = 0.27), 1000 grain weight (rg = 0.63 and rp = 0.68), grain yield per plant (rg = 0.79 and rp = 0.80), fodder yield per plant (rg = 0.55 and rp = 0.59), harvest index (rg = 0.46 and rp = 0.51), iron content (rg = 0.44 and rp = 0.26) and zinc content (rg = 0.40 and rp = 0.24). It showed negative and significant correlation with days to maturity (rg = -0.43, rp = -0.22) at both genotypic and phenotypic levels, indicates the contrasting association between these traits, while productive tiller per plant (rg = -0.36) showed negative and significant genotypic correlation with number of fingers per earhead. Sapkal et al. (2019) [31] , Abhilash et al. (2020) [1] , Chavan et al. (2020) [4] , Rajabhau et al. (2020), Vidhate et al. (2020) [37] , Gayathri et al. (2022) [10] , Hema et al. (2022) [185] , Nikunj et al. (2023) [24] and Sul et al. (2025) [33] posted same results with grain yield per plant. Test weight or 1000 grain weight showed positive significant genotypic and phenotypic correlation with traits finger width (rg = 0.41 and rp = 0.31), number of fingers per earhead (rg = 0.63 and rp = 0.68), grain yield per plant (rg = 0.70 and rp = 0.71), fodder yield per plant (rg = 0.60 and rp = 0.50), harvest index (rg = 0.40 and rp = 0.35), iron content (rg = 0.44 and rp = 0.43) and zinc content (rg = 0.43 and rp = 0.42), while it showed positive significant phenotypic correlation with plant height (rp = 0.22). The favorable association among these traits proved advantageous in formulating effective genetic improvement programme. In contrast, days to maturity (rg = -0.39 and rp = -0.29) had a significant negative correlation with 1000 grain weight at both levels and significant negative phenotypic correlation with main earhead length (rp = -0.21). Lule et al. (2012) [20] , Suman et al. (2018) [34] , Chavan et al. (2020) [4] , Rajabhau et al. (2020), Aruna et al. (2023) [3] , Patel et al. (2023) [27] , Poola (2024) [29] and Ujjainkar et al. (2025) [35] reported similar results for grain yield per plant. Grain yield per plant, being the primary economic trait, exhibited significant positive genotypic and phenotypic correlations with several yield-contributing characters, including plant height (rg = 0.40 and rp = 0.31), finger length (rg = 0.51 and rp = 0.62), finger width (rg = 0.53 and rp = 0.62), number of fingers per earhead (rg = 0.79 and rp = 0.80), 1000 grain weight (rg = 0.70 and rp = 0.71), fodder yield per plant (rg = 0.58 and rp = 0.63), harvest index (rg = 0.45 and rp = 0.54), iron content (rg = 0.45 and rp = 0.29) and zinc content (rg = 0.36 and rp = 0.23). These results indicate a favorable association among these traits, which could be advantageous for crop improvement, particularly for increasing grain yield in the future. Trait main earhead length (rg = 0.32) had positive genotypic correlation with same trait. On the other hand, there was a notable inverse genotypic and phenotypic relationship between this trait and days to maturity (rg = -0.41 and rp = -0.24), while a negative genotypic correlation was also recorded with productive tillers per plant ((rg = -0.36). The positive association of grain yield with plant height observed in the present investigation is consistent with the findings of Wolie and Dessalegn (2011) [38] , Jayshree and Nagarajaiah (2013) [17] , Jadhav et al. (2015) [16] , Chavan et al. (2020) [4] , Giddaluri et al. (2022) [11] , Aruna et al. (2023) [3] , Patel et al. (2023) [27] , and Ujjainkar et al. (2025) [35] . Similar positive relationships for main earhead length were reported by Devaliya et al. (2017) [6] , Keerthana et al. (2019) [18] , Sapkal et al. (2019) [31] , and Gayathri et al. (2022) [10] . The positive correlation of grain yield with finger length agrees with the reports of Wolie and Dessalegn (2011) [38] , Ganapathy et al. (2011) [9] , Lule et al. (2012) [20] , Das et al. (2013) [5] , Jadhav et al. (2015) [16] , and Keerthana et al. (2019) [18] . Similar findings for finger width were documented by Das et al. (2013) [5] , Eric et al. (2016) [8] , Vidhate et al. (2020) [37] , Hema et al. (2022) [15] , Nikunj et al. (2023) [24] , Patel et al. (2023) [27] , and Poola (2024) [29] . The significant positive association between grain yield and 1000-grain weight corroborates the observations of Lule et al. (2012) [20] , Suman et al. (2018) [34] , Vidhate et al. (2020) [37] , Aruna et al. (2023) [3] , Patel et al. (2023) [27] , and Ujjainkar et al. (2025) [35] . Likewise, the positive correlation of grain yield with fodder yield per plant is in agreement with the findings of Wolie and Dessalegn (2011) [38] , Anuradha et al. (2013) [2] , Gohel and Chaudhari (2018) [12] , Suman et al. (2018) [34] , Keerthana et al. (2019) [18] , Madhavilatha et al. (2021) [21] , Gayathri et al. (2022) [10] , and Nikunj et al. (2023) [24] . A comparable positive association between grain yield and harvest index has also been reported by Wolie and Dessalegn (2011) [38] , Devaliya et al. (2017) [6] , Gohel and Chaudhari (2018) [12] , Suman et al. (2018) [34] , Abhilash et al. (2020) [1] , Gayathri et al. (2022) [10] , and Nikunj et al. (2023) [24] . Furthermore, the positive relationship between grain yield and iron content is in accordance with the findings of Devaliya et al. (2017) [6] . Fodder yield per plant had significant positive genotypic and phenotypic correlation with finger width (rg = 0.35 and rp = 0.26), number of fingers per earhead (rg = 0.55 and rp = 0.59), 1000 grain weight (rg = 0.60 and rp = 0.50), grain yield per plant (rg = 0.58 and rp = 0.63), iron content (rg = 0.64 and rp = 0.50) and zinc content (rg = 0.35 and rp = 0.28), while significant positive phenotypic correlation was also recorded with plant height (rp = 0.20). At the genotypic and phenotypic levels, there was a significant negative correlation found between the days to maturity (rg = -0.50 and rp = -0.34) and harvest index (rg = -0.45 and rp = -0.29), while significant negative genotypic correlation was also recorded with productive tillers per plant (rg = -0.46). Wolie and Dessalegn (2011) [38] , Anuradha et al. (2017) [2] , Gohel and Chaudhari (2018) [12] , Suman et al. (2018) [34] , Keerthana et al. (2019) [18] , Madhavilatha et al. (2021) [21] , Gayathri et al. (2022) [10] , Nikunj et al. (2023) [24] , Poola (2024) [29] and Sul et al. (2025) [33] reported significant positive correlation of grain yield with fodder yield per plant.

In the present study, grain yield per plant exhibited significant positive correlations with plant height, finger length, finger width, number of fingers per earhead, 1000grain weight, fodder yield per plant, harvest index, and iron and zinc contents at both the genotypic and phenotypic levels. These strong positive associations indicate that these traits play a major role in determining grain yield in finger millet. Therefore, improvement of these component traits through indirect selection is expected to result in a corresponding increase in grain yield, even without direct selection for yield itself. Consequently, these characters may serve as reliable selection criteria for the development of high-yielding finger millet genotypes. Path coefficient analysis provides a more precise evaluation of the relationships among traits by partitioning correlation coefficients into their direct and indirect effects on grain yield. In the present study, grain yield per plant was considered the dependent variable, while the remaining fifteen traits were treated as independent variables to assess their individual contributions. The direct and indirect effects of these traits on grain yield, estimated from genotypic correlation coefficients, are presented in Table 2 and illustrated in Figure 1. Path coefficient analysis indicated that fodder yield per plant exerted the highest positive direct effect on grain yield per plant (0.646), followed by harvest index (0.626), number of fingers per earhead (0.299), finger length (0.128), main earhead length (0.099), productive tillers per plant (0.090), finger width (0.072), iron content (0.046), protein content (0.023), plant height (0.014), and calcium content (0.004). The positive direct effects of these traits suggest that their association with grain yield is primarily due to their inherent contribution rather than indirect influences through other characters. Therefore, these traits can be considered reliable selection criteria and should be given priority in breeding programmes aimed at improving grain yield in finger millet. The present findings are consistent with the reports of Priyadharshini et al. (2011) [30] and Negi et al. (2017) [23] , who also observed a high positive direct effect of harvest index on grain yield. A similar positive direct contribution of fodder yield per plant was documented by Devaliya et al. (2017) [6] and Negi et al. (2017) [23] . The positive direct influence of number of fingers per earhead on grain yield corroborates the findings of Sapkal et al. (2019) [31] , Abhilash et al. (2020) [1] , and Lad et al. (2020) [19] . Likewise, the positive direct effect of finger length agrees with the observations of Priyadharshini et al. (2011) [30] , Mahanthesha et al. (2018) [22] , and Vidhate et al. (2020) [37] . Comparable positive direct effects for iron and calcium contents were also reported by Vidhate et al. (2020) [37] . Several traits contributed positively to grain yield through their indirect effects via harvest index. The highest positive indirect effect through harvest index was observed for number of fingers per earhead (0.139), followed by main earhead length (0.027), days to 50% flowering (0.015), 1000-grain weight (0.011), plant height (0.003), productive tillers per plant (0.003), protein content (0.001), and zinc content (0.0001). Likewise, fodder yield per plant exerted favourable indirect effects on grain yield through number of fingers per earhead (0.166), days to maturity (0.041), iron content (0.029), finger width (0.026), 1000-grain weight (0.016), finger length (0.010), plant height (0.004), and calcium content (0.001). These positive indirect contributions indicate that the influence of these traits on grain yield is also mediated through harvest index and fodder yield per plant, emphasizing their importance in selection strategies for yield improvement. The indirect contribution of 1000-grain weight to grain yield through fodder yield per plant (0.281) was the highest, followed by number of fingers per earhead (0.208), harvest index (0.150), days to maturity (0.032), iron content (0.020), finger width (0.018), finger length (0.009), plant height (0.003), and protein content (0.0005). Similarly, number of fingers per earhead exerted favourable indirect effects on grain yield through fodder yield per plant (0.254), harvest index (0.191), days to maturity (0.035), 1000-grain weight (0.028), iron content (0.020), finger width (0.008), plant height (0.005), and main earhead length (0.004). These findings indicate that the positive influence of 1000-grain weight and number of fingers per earhead on grain yield is also mediated through other important yield-contributing traits, particularly fodder yield per plant and harvest index. Productive tillers per plant exerted positive indirect effects on grain yield through finger length (0.099), harvest index (0.093), zinc content (0.008), and protein content (0.0001). In contrast, days to maturity (-0.082) and zinc content (-0.045) exhibited negative direct effects on grain yield per plant. These negative direct effects are likely attributable to their unfavourable indirect influences through other associated traits. Therefore, a multi-trait selection strategy that emphasizes traits with favourable direct and indirect effects would be more effective for improving grain yield than selection based on a single character. Such an approach enables breeders to exploit beneficial trait interactions while minimizing the adverse influence of traits with negative direct effects. The negative direct effect of days to maturity on grain yield observed in the present study is in agreement with the findings of Patel et al. (2023) [27] . Zn 'rg' with GYP DF 0.076 -0.064 -0.003 -0.004 0.051 -0.059 0.004 0.012 -0.003 -0.094 0.126 0.001 0.001 -0.009 0.013 0.049 DM 0.059 -0.082 -0.004 0.008 0.026 -0.019 -0.008 -0.128 -0.011 -0.339 0.065 0.001 -0.0001 -0.020 0.019 -0.412* PH -0.014 0.022 0.014 -0.024 -0.015 0.007 0.005 0.108 0.007 0.172 0.126 -0.002 -0.001 -0.001 -0.004 0.398* PTP -0.008 -0.017 -0.008 0.090 -0.008 0.099 -0.021 -0.107 -0.003 -0.270 0.093 0.0001 -0.001 -0.007 0.008 -0.157 MEHL 0.040 -0.022 -0.002 -0.003 0.099 -0.027 -0.001 0.110 -0.006 -0.025 0.174 -0.001 -0.0004 -0.011 0.005 0.320* FL 0.035 -0.012 -0.001 -0.009 0.098 0.128 -0.006 0.095 -0.002 -0.050 0.237 -0.0001 -0.001 -0.006 0.001 0.507** FW 0.004 0.009 0.001 -0.012 -0.015 0.011 0.072 0.034 0.007 0.230 0.196 -0.003 -0.0001 0.009 -0.015 0.531** NFEH 0.003 0.035 0.005 -0.014 0.004 -0.028 0.008 0.299 0.028 0.254 0.191 -0.001 -0.0003 0.020 -0.018 0.786** TW -0.007 0.032 0.003 -0.005 -0.022 0.009 0.018 0.208 0.027 0.281 0.150 0.001 -0.0002 0.020 -0.019 0.696** FYP -0.011 0.041 0.004 -0.018 -0.019 0.010 0.026 0.166 0.016 0.646 -0.243 -0.002 0.001 0.029 -0.016 0.630** HI 0.015 -0.009 0.003 0.003 0.027 -0.040 -0.023 0.139 0.011 -0.199 0.626 0.001 -0.001 -0.011 0.0001 0.541** PC 0.007 -0.008 -0.002 0.0003 -0.010 0.001 -0.019 -0.017 0.001 -0.111 0.055 0.023 0.001 0.024 0.003 -0.051 Ca 0.016 0.001 -0.002 -0.012 -0.008 0.014 -0.003 -0.020 -0.001 0.130 -0.150 0.003 0.004 0.021 -0.006 -0.013 Fe -0.015 0.036 -0.0002 -0.006 -0.023 0.018 0.014 0.079 0.012 0.410 -0.243 0.001 0.001 0.046 -0.031 0.290** Zn -0.022 0.035 0.001 -0.007 -0.011 0.003 0.023 0.080 0.012 0.125 -0.001 -0.001 0.001 0.031 -0.

Conclusion

Overall, the combined evaluation of trait correlations and path coefficients provides a clear selection framework for finger millet enhancement. Grain yield per plant showed significant positive associations at both the genotypic and phenotypic levels with plant height, finger length, finger width, test weight, fodder yield, harvest index, iron, and zinc content, establishing their utility for indirect selection. Path analysis resolved these associations into direct causal paths, identifying fodder yield per plant and harvest index as the primary direct contributors to productivity, alongside substantial positive contributions from main earhead length, productive tillers per plant, iron, protein, and calcium content. Because these attributes display both high direct effects and strong mutual associations with yield, incorporating them into selection indices offers an efficient strategy to achieve concurrent improvements in yield potential and grain quality.

Acknowledgement

AcknowledgementsThe authors express their sincere gratitude to Sardarkrushinagar Dantiwada Agricultural University (SDAU) for providing the necessary funding and research facilities. We are also deeply grateful to the Hill Millet Research Station, NAU, Waghai for generously sharing the research materials.

References

1. Anuradha N, Udaya BK, Patro TSSK, Sarma NDRK. Character association and path analysis in finger millet (Eleusine coracana (L.) Gaertn) accessions belongs to late maturity group. Int J Food Agric Vet Sci. 2013;3(3):113-115.
2. Aruna B, Ganapathy KN, Kumar CSM, Ronda V. Character association and path analysis for grain yield and yield components in white finger millet (Eleusine coracana L. Gaertn.) germplasm. Biol Forum An Int J. 2023;15(7):267-271.
3. Chavan BR, Jawale LN, Shinde AV. Correlation and path analysis studies in finger millet for yield and yield contributing traits (Eleusine coracana L. Gaertn.). Int J Chem Stud. 2020;8(1):2911-2914.
4. Das R, Sujatha M, Pandravada SR, Sivasankar A. Trait relationship and path coefficient analysis in finger millet (Eleusine coracana (L.) Gaertn.). J Prog Agric. 2013;4(1):81-84.
5. Devaliya SD, Manju S, Visat ML. Character association and path analysis in finger millet (Eleusine coracana (L.) Gaertn.). BioSci Trends. 2017;10(31):6690-6694.
6. Dewey DR, Lu KH. Correlation and path coefficient analysis of components of crested wheat grass seed production. Agron J. 1959;51(9):515-518.
7. Eric MO, Pangirayi T, Paul S, Mwangi G, Abhishek R. Correlations, path coefficient analysis and heritability for quantitative traits in finger millet landraces. Philipp J Sci. 2016;145(2):197-208.
8. Ganapathy S, Nirmalakumari A, Muthiah AR. Genetic variability and interrelationship analyses for economic traits in finger millet germplasm. World J Agric Sci. 2011;7(2):185-188.
9. Gayathri CU, Lal GM, Elangovan M, Anwar S. Estimation of genetic variability direct and indirect effects of yield contributing traits on rain yield in finger millet [Eleusine coracana (L)]. Int J Environ Clim Change. 2022;12(11):3355-3360.
10. Giddaluri C, Lal GM, Supriya P. Assessing genetic variability, correlation, direct and indirect effects of yield contributing characters on seed yield in finger millet (Eleusine coracana (L.) Gaertn). Int J Environ Clim Change. 2022;12(11):1552-1558.
11. Gohel DS, Chaudhari SB. Study of correlation and path analysis of finger millet genotypes (Eleusine coracana L. Gaertn.). J Pharmacogn Phytochem. 2018;7(6):12831288.
12. Gowda KTK, Gowda J, Ashok EG, Nagaraja A, Jagadish PS, Sashidhar VR, Seenappa K, Krishnappa M, Vasanth KR, Somu G, Jagadeesh BR, Bharathi S. Technology for increasing finger millet and other small millets production in India. Bangalore: Project Coordinating Cell, ICAR, University of Agricultural Sciences, Gandhi Krishi Vigyan Kendra Campus; 2006. p. 1-41.
13. Hazel L. The genetic basis for constructing selection indices. Genetics. 1943;28(6):476-490.
14. Hema P, Elangovan, Pandey S, Yadav BN. Study of genetic variability and character association in elite lines of finger millet [Eleusine coracana (L.) Gaertn] suitable for eastern plain zone of Uttar Pradesh. Int J Environ Clim Change. 2022;12(12):438-447.
15. Jadhav R, Ratnababu D, Ahamad ML, Rao S. Character association and path coefficient analysis for grain yield and yield components in finger millet (Eleusine coracana (L.) Gaertn). Electron J Plant Breed. 2015;6(2):535-539.
16. Jayshree MK, Nagarajaiah C. Genetic variability and character association studies in African and Indian finger millet (Eleusine coracana (L.) Gaertn) accessions. J Ecol Environ. 2013;31(4):1950-1953.
17. Keerthana K, Chitra S, Subramanian A, Nithila S, Elangovan M. Studies on genetic variability in finger millet [Eleusine coracana (L.) Gaertn] genotypes under sodic conditions. Electron J Plant Breed. 2019;10(2):566-569.
18. Lad DB, Mahalle SP, Bade BA. Association of characters and path coefficient analysis for yield and its components in finger millet (Eleusine coracana (L.) Gaertn.) genotypes. J Pharmacogn Phytochem. 2020;9(5):599-602.
19. Lule D, Tesfaye K, Fetene M, De Villiers S. Inheritance and association of quantitative traits in finger millet (Eleusine coracana Subsp. coracana) landraces collected from eastern and south eastern Africa. Int J Genet. 2012;2(2):12-21.
20. Madhavilatha L, Sudhakar P, Latha P, Priya MS, Kumar MH. Studies on genetic variability, correlation and path analysis for quantitative traits in finger millet. Pharma Innov J. 2021;10(6):709-712.
21. Mahanthesha M, Sujatha M, Meena AK, Pandravada SR. Correlation and path coefficient analysis in finger millet (Eleusine coracana (L.) Geartn). J Pharmacogn Phytochem. 2018;7(4):3193-3196.
22. Negi S, Bhatt A, Kumar V. Genetic diversity among finger millet [Eleusine coracana (L.) Gaertn] genotypes for yield and its contributing traits. J Appl Nat Sci. 2017;9(3):1624-1629.
23. Nikunj UK, Kumar DP, Kanwar RR, Kar DS, Singh DD. Studies on correlation and path analysis in finger millet (Eleusine coracana (L.) Gaertn). Pharma Innov J. 2023;12(3):2720-2722.
24. Panse VG, Sukhatme PV. Statistical methods for agricultural workers. New Delhi: Indian Council of Agricultural Research (ICAR); 1978;34(9):357-358.
25. Panse VG, Sukhatme PV. Statistical methods for agricultural workers. 2nd ed. New Delhi: Indian Council of Agricultural Research (ICAR); 1985. p. 1361.
26. Patel U, Patil HE, Ray P. Correlation and path analysis for grain yield and its component traits in finger millet (Eleusine coracana L.). Biol Forum An Int J. 2023;15(9):660-664.
27. Patil HE, Patel BK, Vavdiya P, Pali V. Breeding for quality improvement in small millets, a review. Int J Genet. 2018;10(9):507-510.
28. Poola S. Characterization and genetic divergence studies in indigenous finger millet [Eleusine coracana (L.) Gaertn] germplasm [doctoral dissertation]. Guntur: Acharya N.G. Ranga Agricultural University; 2024. 9(12):456-498.
29. Priyadharshini C, Nirmalakumari A, Jhon JA, Raveendran M. Genetic variability and trait relationships in finger millet hybrids. Madras Agric J. 2011;98(1-3):18-21.
30. Sapkal SR, Bhavsar VV, Barhate KK, Sarika NK. Correlation and path analysis for different characteristics in germplasm of finger millet (Eleusine coracana (L.) Gaertn.). Int J Curr Microbiol Appl Sci. 2019;8(1):2319-7706.
31. Sharma N, Negi AS, Jaiswal S, Kumar A. Genetic diversity and principal component analysis for yield and its attributing traits in finger millet (Eleusine coracana (L.) Gaertn.) germplasm. Pharma Innov J. 2022;11(7):1121-1126.
32. Ujjainkar VV, Sonkamble PA, Rathod TH, Gomashe SS, Lanjewar RD. Assessment of genetic variability among finger-millet [Eleusine coracana (L.) Gaertn] genotypes through multivariate analysis. Int J Agric Nutr. 2025;7(12):20-26.
33. Vasisth P, Rangaiah S, Sharma M, Balar V, Chittora V. Induced mutagenesis for male sterility in finger millet (Eleusine coracana L. Gaertn.). Agri Mech Asia. 2022;53(3):6935-6943.
34. Wolie A, Dessalegn T. Correlation and path coefficient analyses of some yield related traits in finger millet (Eleusine coracana (L.) Gaertn.) germplasms in northwest Ethiopia. Afr J Agric Res. 2011;6(22):50995105.
35. Wright S. Correlation and causation. J Agric Res. 1921;20:557-580.