Correlation and path coefficient analysis of seed yield and its component traits in micronutrient-sprayed safflower (Carthamus Tinctorius L.)

Introduction

Safflower (Carthamus tinctorius L.), an important rainfed oilseed crop of the family Asteraceae is valued for its drought tolerance, wide adaptability and the high proportion of unsaturated fatty acids in its seed oil (Singh et al., 2006; Hussain et al., 2016) [2,1] . Seed yield being a complex quantitative trait resulting from the combined expression of several component characters has comparatively low heritability and direct selection for yield alone is therefore often inefficient. Correlation coefficient analysis quantifies the degree and direction of association between seed yield and its component characters indicating which traits could serve as indirect selection criteria. However a component character may appear highly correlated with yield merely through its association with other more influential characters without exerting any real influence of its own. Path coefficient analysis, first proposed by Wright (1921) [3] and later adapted for crop plants by Dewey et al. (1959) [4] overcomes this limitation by partitioning the correlation of each component character with yield into direct and indirect effects thereby permitting more precise identification of the true contributors to yield. Correlation and path analyses have been extensively used to identify effective selection criteria in safflower (Mokhtassi Bidgoli et al., 2006 [5] ; Mahasi et al., 2006 [6] ; Roopa and Ravikumar, 2008 [7] ; Arslan, 2007 [8] ; Golkar et al., 2011 [9] ; Topal et al., 2010 [10] ; Katar, 2013 [11] ; Dambal and Patil, 2016 [12] ; Perveen et al., 2026) [13] but information on the pattern of character association under micronutrients spray specifically remains limited. The present investigation was therefore undertaken to estimate correlation coefficients and to partition them through path coefficient analysis for thirteen morpho-physiological, yield-attributing and seed quality characters in relation to seed yield per plant in micronutrient-sprayed safflower.

Materials and Methods

Experimental site

The field experiment was conducted at the experimental farm of the Quality Seed Production (QSP) Unit, Khanapur-A, Vasantrao Naik Marathwada Krishi Vidyapeeth (VNMKV), Parbhani, Maharashtra during the Rabi season of 2025-26. Parbhani is situated at 19°26' N latitude and 76°47' E longitude at an elevation of 357 m above mean sea level (MSL). The region has a tropical climate with a mean annual rainfall of 888.5 mm, maximum temperature ranging from 28.6 to 42.4°c and minimum temperature from 13.4 to 32.2 °C. The experimental site consisted of uniform, well-ploughed and levelled black cotton soil (vertisol) of uniform fertility status.

Experimental design and treatments

The experiment was laid out in a Randomized Block Design (RBD) with nine treatments and three replications. The safflower variety PBNS-154 was used as the seed material and sown by dibbling. A recommended fertilizer dose of 60:30:00 kg NPK/ha was applied as basal dose. The nine treatments were: T 1 -Control; T 2 -Fe (FeSO 4 ) @ 4 mg/lit; T 3 -Zn (ZnSO 4 ) @ 3 mg/lit; T 4 -B (H 3 BO 3 ) @ 3 mg/lit; T 5 -Mn (MnSO 4 ) @ 2 mg/lit; T 6 -Fe + B (4 mg + 3 mg/lit); T 7 -Fe + Mn (4 mg + 2 mg/lit); T 8 -Fe + Zn (4 mg + 3 mg/lit) and T 9 -Zn + B (3 mg + 3 mg/lit each). Each net plot measured 1.8 x 3.0 m with row-to-row spacing of 45 cm and plant-to-plant spacing of 20 cm.

Characters studied

Observations were recorded on thirteen morpho-physiological, yieldattributing and seed quality characters viz., days to 50 per cent flowering (DFF), days to maturity (DM), plant height (PH), number of primary branches per plant (NPB), number of capitulum per plant (NCPP), number of seeds per capitulum (NSPC), 1000-seed weight / Test Weight (TW), biological yield per plant (BY), harvest index (HI), germination percentage (GP), seedling vigour index-I (SVI-I), seedling vigour index-II (SVI-II) and oil content (OC) in relation to seed yield per plant (SY). Seedling vigour index-I and II were computed as germination% x seedling length (cm) and germination% x seedling dry weight (g) respectively following Baki et al. (1973) [14] while harvest index as (seed yield / biological yield) x 100 and oil content (%) was estimated from a 25 g clean seed sample per plot using the Nuclear Magnetic Resonance (NMR) method.

Statistical analysis

The data recorded on the thirteen component characters and seed yield per plant were subjected to correlation and path coefficient analysis following the standard procedures of Panse et al. (1985) [15] , Wright (1921) [3] and Dewey et al. (1959) [4] as detailed below.

Character association

The correlation coefficients were worked out to determine the degree of association of each character with seed yield and also among the component characters themselves. The treatment mean values were used to compute phenotypic correlation coefficients as per the standard procedure of Panse et al. (1985) [15] :

Where, r = phenotypic correlation coefficient Cov(xy) = covariance between characters x and y Var(x) and Var(y) = variance of characters x and y respectively. The correlation coefficients so obtained were tested for significance at the error degrees of freedom at the 5 and 1 per cent probability levels.

Path analysis

Path analysis was carried out using the phenotypic correlation coefficients to determine the direct and indirect effects of the component characters on seed yield as suggested by Wright (1921) [3] and illustrated by Dewey et al. (1959) [4] . Path coefficients were obtained by solving the simultaneous equations expressing the basic relationship between correlations and path coefficients:

Where, r 1 y to r n y = correlation coefficients between causal factors 1 to n and seed yield (y); r 12 to r n-1 , n = correlation coefficients among causal factors a, b, c … n = direct effects of the respective characters on seed yield. This procedure partitions the phenotypic correlation coefficient of each component character with seed yield into a direct effect and indirect effects transmitted through the other component characters.

Calculation of residual effect (R)

The residual effect measures the contribution of other possible components of yield not included in the present study together with unexplained error and was computed as follows (Dewey et al., 1959) [4] :

The residual effect so obtained is depicted at the base of the phenotypical path diagram (Figure 1).

Results and Discussion

Correlation coefficients were worked out for thirteen morpho-physiological, yield-attributing and seed quality characters in relation to seed yield per plant, as well as amongst themselves, and are presented in Table 1. Path coefficient analysis was carried out at the phenotypic level for the thirteen characters, keeping seed yield per plant as the dependent variable, and the direct and indirect effects are presented in Table 2.

The residual effect was estimated at 0.364, indicating that the thirteen characters together accounted for nearly 87 per cent of the total variation in seed yield per plant. Diagonal values represent direct effects of each character on seed yield; off-diagonal values represent indirect effects transmitted via the respective character. Character abbreviations as in Table 1.

Correlation among seed yield and component traits

Days to 50 per cent flowering and days to maturity exhibited a negative, non-significant correlation with seed yield per plant (r = -0.375 and -0.218 respectively) indicating that an early-flowering and earlymaturing habit tended to favour higher seed yield although the association was not statistically significant. Plant height, number of primary branches per plant, number of capitulum per plant and number of seeds per capitulum each recorded a highly significant positive correlation with seed yield (r = 0.721**, 0.608**, 0.699** and 0.723** respectively) and were also significantly inter-correlated with one another and with 1000-seed weight, biological yield, harvest index and oil content. Among all the characters studied harvest index recorded the strongest positive correlation with seed yield (r = 0.835**) followed by seedling vigour index-II (r = 0.781**) and number of seeds per capitulum (r = 0.723**). Germination percentage alone showed a positive but non-significant association with seed yield (r = 0.328).

These trends are broadly consistent with the correlation studies reported for safflower by Mokhtassi Bidgoli et al. (2006) [5] , Mahasi et al. (2006) [6] , Roopa et al. (2008) [7] , Golkar et al. (2011) [9] , Dambal et al. (2016) [12] , Katar (2013) [11] and Perveen et al. (2026) [13] and are further supported by the reports of Bagheri et al. (2001) [16] and Omidi et al. (2009) [17] who also recorded significant positive association of seed yield with plant height, capitula per plant and seed weight in safflower and of Moghaddasi et al. (2010) [18] who observed similar trait relationships among exotic and native safflower genotypes evaluated through multivariate analysis.

Direct and indirect effects (path analysis)

Seedling vigour index-I recorded the highest positive direct effect on seed yield per plant (0.259) followed by number of seeds per capitulum (0.250) and harvest index (0.219) and in each case the high direct effect was reinforced by favourable positive indirect effects resulting in high and highly significant correlations with seed yield (r = 0.492**, 0.723** and 0.835** respectively). These three characters can therefore be considered reliable for direct selection. In contrast plant height, biological yield and oil content despite recording high and significant correlations with seed yield exerted comparatively low direct effects (0.087, 0.080 and 0.049 respectively) and influenced yield largely indirectly through their association with harvest index and the seedling vigour indices. Simple correlation coefficients for these traits would therefore have overestimated their direct contribution. Number of primary branches per plant showed a negative direct effect (-0.155) that was more than compensated by strong positive indirect effects again resulting in a high positive correlation with yield (r = 0.608**) driven mainly by indirect pathways while days to 50 per cent flowering and days to maturity exerted direct effects (-0.143 and 0.220) that were largely cancelled out by their indirect effects leaving neither as a dependable direct selection criterion. This pattern of direct and indirect effects is in agreement with the path-analysis framework of Wright (1921) [3] and Dewey et al. (1959) [4] and its application to safflower by Mokhtassi Bidgoli et al. (2006) [5] , Mahasi et al. (2006) [6] , Arslan (2007) [8] , Topal et al. (2010) [10] , Katar (2013) [11] and Perveen et al. (2026) [13] and is further corroborated by Behnam et al. (2012) [19] and Golparvar (2011) [20] who also reported a predominance of indirect over direct effects for several yield-contributing traits in safflower grown under contrasting moisture regimes as well as by Lucy-Kumari et al. (2010) [21] who recorded comparable direct-effect patterns for yield components in segregating safflower populations.

Implications for selection

Harvest index, number of seeds per capitulum and the seedling vigour indices combined a high direct effect with a high and significant correlation with seed yield and therefore emerged as the most reliable component traits for indirect selection of superior micronutrient treatments and by inference of superior safflower genotypes aimed at improving seed yield. The residual effect of 0.364 (Figure 1) indicates that the thirteen characters studied together accounted for nearly 87 per cent of the total variation in seed yield per plant, the remaining variation being attributable to characters not included in the present study or to error.

Conclusion

Correlation studies revealed that plant height, number of primary branches per plant, number of capitulum per plant, number of seeds per capitulum, 1000-seed weight, biological yield per plant, harvest index, seedling vigour index-I, seedling vigour index-II and oil content were positively and significantly associated with seed yield per plant in micronutrient-sprayed safflower whereas days to 50 per cent flowering and days to maturity showed a negative, non-significant association and germination percentage a positive, non-significant association with seed yield. Path coefficient analysis indicated that seedling vigour index-I, number of seeds per capitulum and harvest index exerted the highest positive direct effects on seed yield with a residual effect of 0.364. Harvest index, number of seeds per capitulum and the seedling vigour indices which combined a high direct effect with a high and significant correlation with seed yield may therefore be recommended as the most reliable component characters for indirect selection of superior micronutrient treatments for improving seed yield and seed quality of safflower under the agro-climatic conditions of Marathwada. The central node represents seed yield; the square node denotes the residual effect (0.364).

References

1. Hussain MI, Lyra DA, Farooq M, Nikoloudakis N, Khalid N. Salt and drought stresses in safflower: A review. Agronomy for Sustainable Development. 2016;36(1):Article 4.
2. Singh V, Nimbkar N. Safflower (Carthamus tinctorius L.). In Genetic resources, chromosome engineering and crop improvement. 2006:168.
3. Wright S. Correlation and causation. Journal of Agricultural Research. 1921;20(7):557-585.
4. Dewey DR, Lu KH. A correlation and path-coefficient analysis of components of crested wheatgrass seed production. Agronomy Journal. 1959;51(9):515-518.
5. Mokhtassi Bidgoli A, Akbari GA, Mirhadi MJ, Zand E, Soufizadeh S. Path analysis of the relationships between seed yield and some morphological and phenological traits in safflower (Carthamus tinctorius L.). Euphytica. 2006;148(3):261268.
6. Mahasi MJ, Pathak RS, Wachira FN, Riungu TC, Kinyua MG. Correlations and path coefficient analysis in exotic safflower (Carthamus tinctorius L.) genotypes tested in the arid and semiarid lands (ASALs) of Kenya. Asian Journal of Plant Sciences. 2006;5(6):1035-1038. International Journal of Advanced Biochemistry Research https://www.biochemjournal.com 26174707, 2026, 10(10), Downloaded from https://www.doi.org/10.33545/26174693.2026.v10.i10Sj.10379 by International Journal of Advanced Biochemistry Research on [09/10/2026]. See the Terms and Conditions (https://www.biochemjournal.com) for rules of use; OA articles are governed by the applicable Creative Commons License
7. Roopa VK, Ravikumar RL. Correlation and path analysis studies in safflower (Carthamus tinctorius L.) germplasm. Karnataka Journal of Agricultural Sciences. 2008;21(3):436-437.
8. Arslan B. The path analysis of yield and its components in safflower (Carthamus tinctorius L.). Journal of Biological Sciences. 2007;7(4):668-672.
9. Golkar P, Arzani A, Rezaei AM. Determining relationships among seed yield, yield components and morpho-phenological traits using multivariate analyses in safflower (Carthamus tinctorius L.). Annals of Biological Research. 2011;2(2):162-169.
10. Topal M, Ozturk E, Polat T. Path analysis of seed yield components using different correlation coefficients in safflower (Carthamus tinctorius L.). Bulgarian Journal of Agricultural Science. 2010;16(4):429-434.
11. Research: Agriculture and Veterinary. 2013;13(6):11-20.
12. Dambal GI, Patil RS. Character association and path analysis in safflower germplasm (Carthamus tinctorius L.). Research Journal of Agricultural Sciences. 2016;7(1):155-158.
13. Perveen N, Kumar S, Ratnakar TJ, Ghanghas R, Chinir J, Santhini V, Imtiyaz Md. A study of the correlation and path coefficient analysis of morphological traits for seed yield and oil content in safflower (Carthamus tinctorius L.) under late sown condition. Journal of Advances in Biology and Biotechnology. 2026;29(4):491-498.
14. Baki AAA, Anderson JD. Vigor determination in soybean seed by multiple criteria. Crop Science. 1973;13:630-633.
15. Bagheri A, Yazdi-Samadi B, Taeb M, Ahmadi MR. Study of correlations and relations between plant yield and quantitative and qualitative traits in safflower. Iranian Journal of Agricultural Sciences. 2001;32(2):295-307.
16. Omidi AR, Khazaei H, Hongbo S. Variation for some important agronomic traits in 100 spring safflower (Carthamus tinctorius L.) genotypes. American-Eurasian Journal of Agriculture and Environmental Science. 2009;5(6):791-795.
17. Moghaddasi SM, Omidi AH. Study of some morphological and phonological traits in exotic and native safflower genotypes using multivariate statistical methods. Advances in Environmental Biology. 2010;4(3):350-352.
18. Behnam T, Aharizad S, Mohamadreza S, Babazad BA, Gafari G. Path analysis of seed and oil yield in safflower. Communications in Plant Sciences. 2012;2(2):15-20.
19. Golparvar AR. Genetic improvement of oil yield in spring safflower cultivars under drought and non-drought conditions. Electronic Journal of Biology. 2011;7(2):40-43.
20. Lucy-Kumari, Ravikumar RL. Character association and path analysis in early segregating generations of interspecific crosses of safflower. Karnataka Journal of Agricultural Sciences. 2010;23(3):487-490. How to cite this article: Waikar P, Misal A, Rathod A, Devkule S, Umate S, Wankhade M, et al. Correlation and path coefficient analysis of seed yield and its component traits in micronutrientsprayed safflower (Carthamus Tinctorius L.). International Journal of Advanced Biochemistry Research. 2026;SP10(10):779-783. doi: https://www.doi.org/10.33545/26174693.20