Character Association and Path Coefficient Analysis of Certain Biometrical Traits among Fenugreek Genotypes under Different Environmental Conditions

Spices are good appetizers and are also considered essential in the culinary art all over the world. Spices are used for adding aroma and flavor in foods. Some of spices also used as a preservative in some food products like pickles and chutneys etc. while others some spices also possess strong anti-microbial, anti-oxidant properties and antibiotic activities. Many of them possess medicinal properties and have a profound effect on human health.
       
Fenugreek (Trigonella foenum-graecum L.) is extensively used as fresh leaves (green leafy vegetable), chopped leaves (flavoring agent), seeds (spice, condiment or medicines), extracts and powders (medicines). India is the largest producer, consumer and exporter of fenugreek in the world with an annual production of 2, 56,000 MT from 2,10,000 hectares (Anonymous, 2017). In India, it is largely cultivated in Rajasthan, Gujarat, Punjab, Uttar Pradesh, Madhya Pradesh and Maharashtra. Rajasthan and Gujarat are contributes more than 80 per cent of total seed spices produced in India. In Rajasthan area and production of fenugreek in are 1,57,000 hectare and 1,90,362 MT, respectively. In Rajasthan it is mainly grown in Chittorgarh, Sikar, Nagaur, Jaipur, Jhalawar, Kota and Jhunjhunu districts.
       
Fenugreek can be grown under a wide range of climatic conditions and is particularly sensitive to water stress from one week before to two weeks after flowering. Therefore, high yielding fenugreek could be achieved through full irrigation at the flowering stage, even if the soil water content is sub-optimal during the vegetative growth and pod filling stages.  A long dry spell and scared water has been reported limiting factor in production of fenugreek during the season.
       
The productivity of any crop is mainly contributed to the yield attributing traits. Yield depends on various characters and environmental conditions that exist during crop growth period. Therefore, it is essential to study the character association among themselves and with seed yield of the crop. Correlation provides a measure of association between two or more characters and help to identify more useful relationship between characters. Indirect associations become complex and important when a number of variables are included in the study of correlation. A study on correlation alone is not enough to give an exact picture of relative importance of direct and indirect influence of each of the component traits on seed yield. In some cases more refined technique such as path coefficient analysis helps to find out direct and indirect causes of character association. Every component character has a direct effect on the yield. The effects of an independent on dependent traits via other independent trait are known as indirect effects. If correlation is due to direct effect, it reflects true relationship and selection is practiced for such a character for improving the yield. In case, the effect is indirect through another component trait, the breeder has to select the later trait through which indirect effect is exerted. The low productivity of fenugreek in India is mainly due to non-availability of suitable high yielding varieties for various agro climatic regions and poor crop husbandry. The performance of locally available cultivars of fenugreek is poor in the Hadoti region of Rajasthan. Hence, there is an urgent need for genetic improvement to develop high yielding varieties suitable for adverse situations.

The experiment was laid out with thirty genotypes of fenugreek at Vegetable Research Farm, College of Horticulture and Forestry, Jhalawar (Rajasthan) during Rabi season 2019-20. Seeds of thirty diverse genotypes were collected from different part of the country and detailed of genotypes with source and morphological characters presented in Table 1.
 

       
These genotypes were sown in a RBD with three replications in two different environments, namely, (1) normal (E₁) and (2) limited (E₂) irrigation condition. Each   genotype was sown in a double row plot of 4 m length with row to row and plant to plant distance at 30 cm and 10 cm, respectively. All standard recommended agronomic practices were carried out during entire cropping season except irrigation in limited irrigation condition (E₂). Irrigations were given at the time of seed sowing for establishing the crop in both the environments. Normal irrigation (E₁) condition was created by providing required irrigations sowing to maturity of the crop as per recommendation and limited irrigation (E₂) condition was created by providing half irrigation of normal condition in limited manner. Mean weekly meteorological data for the period of experiment are presented in Graph 1.
 

 
All the observations i.e. days to 50 per cent flowering, days to maturity, plant height at maturity (cm), number of branches per plant, number of pods per plant, number of seeds per pod, test weight(g), seed yield per plant (g) and proline content in leaves (mg/g) (Bates et al., 1973) at 60 days after sowing (DAS) and maturity were recorded on five randomly selected plants avoiding border plant from each replication, each genotype and each environment for eleven traits except days to 50 percent flowering and days to maturity. Days to 50 percent flowering and days to maturity were recorded on the plot basis in both the environments.
       
Correlation (Searle, 1961) and path analysis (Wright, 1921 and Dewey and Lu, 1959) were calculated as per statistical method using web based statistical package OPSTAT from CCS HAU, Hisar, Haryana.

In general, the magnitudes of genotypic correlation coefficient were higher than their corresponding phenotypic correlation coefficient for all the characters thereby indicating a strong inherent association between various traits. The association is almost same for all the characters except small variations in values at both genotypic and phenotypic level.
       
Genotypic and phenotypic correlation in both the environment i.e. E₁ and E₂: Days to 50 percent flowering was observed highly positive significant for days to maturity (0.463, 0.320 and 0.410, 0.366)  and negatively significant for number of branches per plant (-0.523, -0.469 and -0.708 -0.559), number of pods per plant (-0.516, -0.423 and -0.666 -0.588), pod length (-0.476,-0.383and -0.432,-0.374), number of seeds per pod (-0.404,-0.339 and -0.649, -0.508) and seed yield per plant (-0.477,-0.474 and -0.404,-0.410). Days to maturity had highly positive significant for test weight (0.283, 0.266 and 0.277, 0.258) and negatively significant for number of pods per plant (-0.324, -0.320 and -0.354, -0.315). Number of branches per plant was observed highly positive significant with number of pods per plant (0.442, 0.383 and 0.616, 0.533), pod length (0.686, 0.586 and 0.421, 0.338) and seed yield per plant (0.282, 0.268 and 0.313, 0.282). Number of pods per plant had highly positive significant for pod length (0.544, 0.484 and 0.426, 0.374), number of seeds per pod (0.314, 0.272 and 0.482, 0.440) and seed yield per plant (0.765, 0.725 and 0.670, 0.651). Pod length and number of seeds per pod were found highly positive significant with seed yield per plant (0.376, 0.348 and 0.351, 0.327) and (0.705, 0.650 and 0.666, 0.640). Test weight was observed negatively significant with proline content in leaves at 60 DAS and maturity (-0.227,-0.226 and -0.305,-0.304) and (-0.222,-0.220 and -0.311,-0.309). Proline content in leaves at 60 DAS was observed highly positive significant with proline content in leaves at maturity (0.999, 0.989 and 1.000, 0.995) (Table 2). Significant correlation of  characters suggested that there is much scope for direct and indirect selection for further improvement  which  means  by  improving of  these traits we can achieve the improvement with respect to seed yield per plant and positive selection should be employed for these traits to improve seed yield. However, there were generally negative and significant genotypic and phenotypic relationships between seed yield per plant with days to 50 percent flowering. The trait days to 50 per cent flowering  can also be  used  as  a  selection  criterion for  increasing  yield  only  when negative selection is practiced. This finding is in agreement with the findings of Pushpa et al., (2010), Jain et al., (2013), Kole and Saha (2013), Singh (2014) and Gurjar et al., (2016). 
 

 
Path coefficient analysis
 
In the present investigation path coefficient analysis was carried out for characters under study using genotypic and phenotypic correlation coefficient and taking seed yield per plant as dependable variable, in order to see the causal factor and also to identify the components which are responsible for producing seed yield per plant. The genotypic and phenotypic path coefficient analysis revealed that number of pods per plant (0.693, 0.602 and 1.030, 0.754), number of seeds per pod (0.396, 0.378 and 0.490, 0.421) and test weight (0.338, 0.304 and 0.405, 0.368) gave the highest direct effect on seed yield per plant in both the environments i.e. normal (E₁) and limited (E₂) irrigation condition (Table 3). This suggests that these traits directly contribute towards seed yield per plant. So, importance is given for these traits during crop improvement programme to improve seed yield per plant. The several studies have shown importance by various scientists reported by Dashora et al., (2011), Fikreselassie et al., (2012) and Yadav et al., (2013). Negligible positive direct effects were exerted by trait like, plant height at maturity in both the environments i.e. normal (E₁) and limited (E₂) irrigation condition showing that this trait contribute very less to seed yield per plant. This result is in agreement with the findings of Jain et al., (2013). The value of residual effect (0.086, 0.155 and 0.238, 0.241) in both the environments i.e. normal (E₁) and limited (E₂) irrigation condition at genotypic and phenotypic levels indicates that there may be some other secondary components that should not be ignored.
 

The biometrical traits number of pods per plant, number of seeds per pods and test weight has been identified as selection criteria for obtaining good genotypes in fenugreek breeding programme  in both the environments i.e. normal (E₁) and limited (E₂) irrigation condition.