Akhtar et al., J. Anim. Plant Sci. 21(4):2011 660 ESTIMATION OF HERITABILITY, CORRELATION AND PATH COEFFICIENT ANALYSIS IN FINE GRAIN RICE (Oryza sativa L.) N. Akhtar, M. F. Nazir, A. Rabnawaz * , T. Mahmood ** , M. E. Safdar, M. Asif and A. Rehman University College of Agriculture, University of Sargodha, Sargodha, Pakistan * Rice Research Institute Kala Shah Kaku, Punjab, Pakistan ** PMAS Arid Agriculture University, Rawalpindi, Pakistan Corresponding author: naeem.uca@gmail.com ABSTRACT Yield contributing traits in ten rice genotypes were studied for variances, heritability, correlation (genotypic and phenotypic) and path coefficient analysis. The heritability was found to be highest for number of grains panicle -1 , days to maturity, plant height and paddy yield while lowest for number of tillers plant -1 . Paddy yield had strong genetic correlation with number of grains panicle -1 , days to maturity and 1000-grain weight. Genotype PK4831-6-1 produced highest paddy yield (5.1 t ha -1 ), number of grains panicle -1 (144) as well as higher 1000-grain weight (25.56 g) whereas PK7909-27-1 and PK7378-4-1-1-1 achieved higher plant heights (162.5 cm and 148.6 cm, respectively) and spent longer growth duration (112 days) to reach maturity. Super Basmati produced highest number of tillers plant -1 (20) but lowest paddy yield (3.1 t ha -1 ). Regression analysis revealed that paddy yield had significant positive correlation with number of grains panicle -1 (b = 0.0164) and 1000-grain weight (b = 0.1356). It can be concluded that number of grains panicle -1 , 1000-grain weight and days to maturity are important plant traits which should be considered when any breeding program for higher paddy yield in rice is to be planned. Key words: Correlation, fine rice, heritability, yield components INTRODUCTION Grain yield, being a quantitative trait is a complex character of any crop. Various morphological and physiological plant characters contribute to yield. These yield contributing components are interrelated with each other showing a complex chain of relationship and also highly influenced by the environmental conditions (Prasad et al., 2001). Breeding strategy in rice mainly depends upon the degree of associated characters as well as its magnitude and nature of variation (Zahid et al., 2006; Prasad et al., 2001). For selection in rice, information on correlation coefficient always has been helpful as a basis for selection in a breeding programme. Path coefficient analysis partitions into direct and indirect matrix presenting correlation in a more meaningful way (Mohsin et al., 2009). The present research study was conducted to find out the genetic variability among different plant traits, direct and indirect contribution of these parameters towards paddy yield and to identify better combinations as selection criteria for developing high yielding fine rice genotypes. MATERIALS AND METHODS The research study was conducted at Rice Research Institute Kala Shah Kaku, Lahore, Pakistan for two years i.e. during summer seasons of 2009 to 2010. The nursery of eight uniform fine grain rice genotypes viz., PK4831-6-1, PK8644-1-1, PK8535-4-2, 99417, PK7378-4-1-1-1, PK7705-4-1-1-1, PK7705-4-1-1-1-1, PK7909-27-1 along with two standard varieties (Super basmati and Basmati 2000) were sown during the month of June each year. The nursery was raised by conventional method which included 24 h seed soaking followed by 48 h incubation for seed sproutening then broadcasting of sprouted seed in the puddled soil. The seed rate was kept 10 kg ha -1 . Thirty days old nursery was transplanted in the field with a plant spacing of 22.5 cm x 22.5 cm during both the years. The experiment was laid out In Randomized Complete Block Design (RCBD) with three replications. At maturity, data on 10 random plants from each replication were recorded for number of grains panicle -1 , number of tillers plant -1 , plant height, 1000-grain weight, days to maturity and paddy yield. The data were subjected to analysis of variance and covariance through the procedure outlined by Steel et al., (1997) in randomized complete block design (RCBD). Genetic and phenotypic correlation coefficients were calculated as outlined by Known and Torrie (1964) and means of both years (2009 & 2010) of each character were compared by Duncan’s Multiple Range (DMR) test. The present study with eight promising lines including two check varieties of Oryza sativa L. was carried out to find the interrelationship between different yield contributing factors for developing new varieties with better combinations of these traits. The Journal of Animal & Plant Sciences, 21(4): 2011, Page: 660-664 ISSN: 1018-7081