INTRODUCTION Pearl millet (Pennisetum glaucum) is a diploid (2n=2x=14) cereal widely known as "Bajra" in the Poaceae family. It was first found in West Africa and then spread to India. Pearl millet is a common coarse grain cereal crop grown in dry and semi-arid climates. It may be cultivated in light textured soil with little rainfall. Among the millets, it is the most significant and perhaps most promising crop. It is widely farmed as a dual-purpose crop in Africa, Asia, and Australia, but it is exclusively planted as a feed crop in the subtropics of the United States. Pearl millet grains are high in nutrients and are consumed by around 10% of India's population as a staple food. In Zone 1C (Hyper Arid Partially Irrigated Western Plains), which includes Bikaner, Jaisalmer, and three tehsils in Churu district, pearl millet production is exceptionally low (325 kg/ha). Because arable land is a valuable and limited resource, increasing cropping intensity and maximizing the use of current resources appears to be a more valid option than expanding the area under cultivation. As a result, there is an urgent need to increase grain yield per unit area. Its inflorescence is a panicle, which is a complex terminal spike. Because of its protogynous nature, pearl millet is a highly cross-pollinated species. A caryopsis is a type of millet seed. Its root system is typical of monocotyledonous plants, with seminal or main roots, adventitious roots, and crown or collar roots. The correlation coefficient into the direct and indirect effects of various traits on a dependent trait (El-Din et al., 2012). Direct effects are where a trait directly affects another without being influenced by other traits whereas indirect effects occur when the relationship between two traits is mediated by one or more traits (Tyagi and Lal 2007). Knowledge of the associations between yield and its component traits and among the component traits themselves would allow for more effective selection for yield. In finger millet, grain yield has been reported to be highly directly associated with: panicle mass and straw yield per plant (Sonnad et al., 2008); productive tillers and 1000 grain mass (Bezaweletaw et al. 2006); biomass yield, finger length and number of fingers per panicle (Ganapathy et al. 2011; Wolie and Dessalegn 2011); and basal tillers, flag leaf blade length, and panicle length and width (Bharathi 2011). Studies which have generated such information on finger millet in east Africa are limited. MATERIALS AND METHODS The study was conducted at Agricultural Research Station, SKRAU, Bikaner, during Kharif season which is falls under Agro-Climate Zone IC of Rajasthan where average rainfall is about 260 mm, mostly received during July-September. The experimental material for the studies consisting 45 hybrids developed by three male sterile lines (A line) and the fifteen restorers lines (R line) through Line × Tester approach. The experimental material was laid out in a Randomized Block Design with three replications during Kharif-2018 with 60 cm × 15 cm plant geometry. Normal and uniform cultural operations were followed during the crop season to raise a good crop. Individual plant observations were made on ten randomly selected plants from each replication for ten traits: plant height, number of effective tillers per plant, ear head diameter, ear head length, 1000 grain weight, biological yield per plant, harvest index, and seed yield per plant, while whole plot observations were made on two characters: days to 50% flowering and days to 50% maturity. Panse and Sukhtame performed analysis of variance on all treatments (1961). Burton's suggested genetic variability parameters were calculated (1958). According to Searle, the phenotypic and genotypic correlation coefficients were calculated from the phenotypic and genotypic variance and covariance (1961). Path coefficient analysis was used to estimate the direct and indirect effects, as suggested by Wright (1921) and elaborated by Dewey and Lu (1959). RESULTS AND DISCUSSION The experimental material, which consisted of 45 pearl millet hybrids, was laid out in a three-replication randomized block design. For the purpose of experimental design, the values of all characters were subjected to analysis of variance. Table 1 shows the mean square values for the characters analyzed. The analysis of variance found substantial variations between the hybrids for all traits except the number of effective tillers per plant, showing that pearl millet hybrids have a lot of variability. A wide range of variability for various traits has been observed earlier by Kumari et al. (2018a); Kumari et al. (2018b); Kumar et al. (2020a); Kumar et al. (2020b). Table 2 shows the mean, range, coefficients of variation, heritability, and genetic advance (as a percentage of the mean). The estimates of genotypic coefficient of variation (GCV) for biological yield per plant were high (>20%), while harvest index and number of effective tillers per plant were moderate (10-20%). For days to 50% flowering, days to 50% maturity, plant height, ear head length, 1000 grain weight, and ear head diameter, it was low (less than 10%). For number of effective tillers per plant, ear head length, biological yield per plant, and seed yield per plant, the phenotypic coefficient of variation (PCV) was high (>20%), but moderate (10-20%) for plant height, 1000 grain weight, and harvest index. For days to 50% flowering, days to 50% maturity, and ear head diameter, it was low (less than 10%). For plant height, heritability estimates were moderate (50-70 percent). Days to 50% flowering, days to 50% maturity, number of effective tillers per plant, plant height, ear head length, 1000 grain weight, ear head diameter, biological yield per plant, and seed yield per plant were all low (below 50%). For the characters of biological yield per plant and seed yield per plant, genetic advance (as a percentage of mean) was high (>20 percent), whereas for plant height, 1000 grain weight, and harvest index, it was moderate (10-20 percent). Days to 50% flowering, low genetic advance (as a percentage of the mean) (less than 10%) was observed. At both the genotypic and phenotypic levels, character association was calculated for various pairs of agronomic attributes, including grain yield per plant (Table 3). For all of the characters, the genotypic correlation estimates were lower than the phenotypic condition coefficient. Days to maturity were positively and significantly correlated with days to 50% flowering (0.999**), plant height (0.198**), and 1000 grain weight (0.346**), while harvest index (-0.405**), seed yield per plant (-0.484**), and biological yield per plant (-0.314**) were significant and negative phenotypic correlations. The number of effective tillers per plant (0.622**), plant height (0.330**), harvest index (0.382**), and biological yield per plant at the phenotypic level (0.869**) all had positive and significant correlations with seed yield per plant. At the phenotypic level, this character was negatively and significantly correlated with days to 50% flowering (-0.484**) and days to 50% maturity (-0.483**). Grain yield per plant was positively and significantly correlated with characters such as number of effective tillers per plant (0.945**), biological yield per plant (0.851**), plant height (0.374*), and ear head length (0.767**) at the genotypic level, but was negatively and significantly correlated with days to 50% flowering (-0.393**) and days to 50% maturity (-0.402**) at the genotypic level. These characters need due consideration during any selection programmes. Similar findings of positive and significant correlation had been reported by number of workers for grain yield per plant with biological yield per plant, number of effective tillers per plant by Izge et al. (2006); Bikash et al. (2013); Kumar et al. (2020b). Negative and significantly correlation of grain yield per plant with days to 50 per cent flowering and days to 50 per cent maturity reported by Nehra et al. (2017); Kumar et al. (2020a). At the genotypic level, harvest index (0.838) had the greatest direct positive effect on seed yield per plant, followed by ear head diameter (0.622), plant height (1.528), days to 50% flowering (24.202), and biological yield per plant (0.949), while days to 50% maturity (-25.062), ear head length (-0.800), number of tillers per plant (-0.508), and 1000 grain weight had the greatest direct negative effect (-0.023). The study's low residual effect at the genotypic (0.018) level indicated that all of the characters contributed significantly to seed yield per plant. This research backs up the findings of and Bikash et al. (2013); Kumari et al. (2018a); Kumari et al. (2018b); Kumar et al. (2020a).