Polygenic variations and cause effect relationship in some photo-insensitive recombinant inbred lines ( RIL ’ s ) of Basmati derivative

The present investigation consisted 18 photo-insensitive recombinant inbred lines (RILs) of Basmati derivative developed through disruptive seasonal selection and two check varieties raised during Kharif 2010 to study the polygenic variations in yield and yield attributing characters and their cause and effect relationship. Seven RILs viz., Bidhan Moti 6, 10, 11, 15, 23 and 30 surpassed the check varieties that is, Satabdi (IET 4786) and Khitish (IET 4094) in respect of grain yield and some other yield related traits. The maximum range of variation in mean was observed for floret number/panicle followed by number of grains/panicle, plant height and fertility percentage. The highest estimate of phenotypic and genotypic variances was observed for floret number /panicle followed by number of grains/ panicle and plant height. High estimates of genotypic coefficients of variation (GCV) and phenotypic coefficients of variation (PCV) were obtained for grain yield/plant, number of grains/panicle, panicle weight and floret number/panicle. High heritability was observed for floret number/panicle, fertility percentage, kernel length, panicle weight, grain length, number of panicle/plant, kernel length (L)/breadth (B) (L/B) ratio, grain breadth and days to maturity. Grain yield showed positive significant correlation with number of panicles/ plant, number of grains/ panicle and fertility percentage. Path coefficient analysis revealed that seven characters viz. plant height, days to 50% flowering, number of panicles/plant, panicle length, number of grains/ panicle, grain length and kernel breadth had positive direct effect on grain yield while nine characters namely days to maturity, panicle weight, floret number/panicle, fertility percentage, grain breadth, grain L/B ratio, kernel length, kernel L/B ratio and 1000 grain weight incurred negative direct effect on grain yield.


INTRODUCTION
Generally, the export quality traditional basmati varieties are tall in nature but poor yielders.In order to improve the yield potential without sacrificing the special quality features, knowledge on the correlation between yield and its component characters can help improve the efficiency of selection.Correlation studies permit only a measure of relationship between two traits.Hence, path coefficient analysis becomes necessary as it indicate separation of direct and indirect effects via other related characters by partitioning through correlation coefficients that helps in designing appropriate breeding procedure for evolving high yielding genotypes.In the present investigation, an

RESULTS AND DISCUSSION
Estimated mean performance for different traits of RILs and check varieties are presented in Table 1.All the lines exhibited considerable variability for all the characters studied.A comparatively dwarf stature was observed for Bidhan Moti 2,6,16,24,Satabdi (IET 4786) and rest of the lines belonged to the semi dwarf group.Bidhan Moti 2,3,6,10,26,Satabdi and Khitish (IET 4094) were Sawarkar and Senapati 113 flowering earlier in comparison to other genotypes like Bidhan Moti 11,22,17,30. Bidhan Moti 8,10,14,17,26 accounted for more number of days to mature than the rest of the genotypes.Bidhan Moti 5,6,11,17,  panicle, plant height and fertility percentage.Phenotypic coefficient of variation was found to be higher than respective genotypic coefficient of variation for all the characters under study indicating the influence of environment on the expression of characters (Dutta et al., 2013).High estimates of genotypic coefficient of variation and phenotypic coefficient of variation were obtained for grain yield/ plant, number of grains/ panicle, panicle weight and floret number/ panicle which is corroborated by the results of Singh et al. (2000) and Chand et al. (2004).
High heritability was observed for most of the characters studied viz floret number/panicle, fertility percentage, kernel length, panicle weight, grain length, number of panicles/ plant, kernel L/B ratio, grain breadth and days to maturity (Bisney et al., 2009;Sao et al., 2002;Roy et al., 2001).
Plant height also recorded high heritability in this regard.
Florets number/ plant recorded the maximum value for genetic advance followed by number of grains/panicle, plant height and fertility percentage.This finding corroborates earlier findings of Bisney et al. (2009) and Singh et al. (2002).Genetic advance as a percentage of mean was highest for grain yield/ plant followed by Magnitude of genotypic correlation coefficients was in general higher than that of the corresponding phenotypic ones (Table 3).This findings corroborates with those of Mamun et al. (2012).Grain yield showed positive and significant correlation with fertility percentage at genotypic level and number of panicles/plant, number of grains/panicle both at genotypic and phenotypic levels.Similar findings were reported earlier by Swain and Reddy (2006), for number of panicles/plant and Senapati et al. (2009) for number of panicles/plant, number of grains/panicle.
Therefore, these were the main yield determining characters in basmati derivatives.Plant height showed positive and highly significant correlation with grain length, kernel length, kernel L/B ratio and 1000 grain weight suggesting tall plants might be expected to produce large slender grain with more test weight.Zahid et al. (2006) also reports similar association for grain length and Sabesan et al. (2009) for 1000 grain weight.
Days to 50% flowering had positive and highly significant correlation with florets/panicle whereas significant negative correlation was observed with grain breadth.Positive significant correlation coefficients were observed for panicle weight with number of grains/panicle and fertility percentage while it had positive significant correlation coefficients with L/B ratio at genotypic level only.Panicle length incurred positive significant  correlation with kernel L/B ratio while it had significant negative correlation coefficient with kernel breadth.Number of grains/ panicle had significant negative correlation coefficients with kernel L/B ratio and plant height.Grain length had positive correlationcoefficients with grain breadth, kernel length (Sabesan et al., 2009) and plant height.The kernel length had positive and highly significant correlation coefficient with kernel L/B ratio.This is similar to the result of Singh et al. (2013) and Subudhi et al. (2011).Significant negative correlation were noted for plant height with floret number per panicle and number of grain per panicle; similar relationship was also observed between days to 50% flowering and grain breadth; panicle length and kernel breadth; floret number/ panicle and 1000 grain weight; number of grains/panicle and kernel L/B ratio, kernel breadth and kernel L/B ratio.The results are in agreement with Sood and Siddiq (1980), Deosarkar and Nerkar (1994) and Christopher et al. (1999) for kernel breadth and kernel L/B ratio.
Path coefficient analysis (Table 4) revealed that the number of character chosen for the study were very much appropriate as evident from low value of residual effect (0.1860).Seven characters viz.plant height, days to 50% flowering, number of panicle/plant, panicle length, number of grains/panicle, grain length and kernel breadth had positive direct effect while nine characters namely days to maturity, panicle weight, floret number/panicle, fertility percentage, grain breadth, grain L/B ratio, kernel length, kernel L/B ratio and 1000 grain weight imparted negative direct effect on grain yield.Grain length imparted the highest positive direct effects on yield followed by number of grains/panicle, plant height, number of panicles/ plant and panicle length.It is similar with the findings of Nagesh et al. (2012), Khan et al. (2009), Mohadeseh et al. (2013) and Das et al. (2004).
Grain length and plant height incurred high positive direct effect on grain yield which was masked by their high negative indirect effects via number of grains/panicle, grain breadth, grain L/B ratio and kernel length that ultimately lead to non-significant negative correlation with yield.In this regard, a restricted simultaneous selection model is to be followed to make selection effective for plant height and grain length and restriction are to be imposed on number of grain/panicle, grain breadth, grain L/B ratio and kernel length to nullify their undesirable indirect effects on grain yield.Shanthala et al. (2004), Mahto and Yadav (2003) also reports similar results for this trait.Yield was positively correlated with number of panicle/plant, number of grain/panicle and fertility percentage.This is in agreement with the findings of Zahid et al. (2006), Asm et al. (2012) and Hasan et al. (2013).Correlation coefficient between number of panicle/plant and yield was highly significant but the direct effect was negative.This revealed that, indirect effects were mainly responsible for yielding of such correlation coefficient.Therefore, it would be effective to consider simultaneously the other characters that showed high indirect effect on grain yield, that is, number of grains/panicle, plant height and number of panicles/ plant would also be helpful for the improvement of yield in rice.Number of grains/panicle and panicle number/plant showed highly significant positive correlation simultaneous with high amount of positive direct effect on yield.These are in conformity to the findings of Mahto et al. (2003), Gazafrodi et al. (2006), Senapati et al. (2009).
Therefore, the present investigation highlighted the differential performance of the newly developed RILs.Some RILs like Bidhan Moti 6,10,11,15,23 and 30 showed very promising performance that may be useful for the development of variety in future for this areas of West Bengal.Panicle number/ plant and grain number/panicle were the prime yield contributing character in rice and their direct selection would be effective in yield improvement in rice.

Table 1 .
Mean performance of RILs of Basmati derivative.
Khitish (IET-4094) was the top scorer followed by Bidhan Moti 6, 14, 15 and 26.All of the RILs possessed extra long grain length.The minimum value (8.23 mm) was observed in Bidhan Moti 30 while the maximum value was registered by Bidhan Moti 26 (11.56 mm) followed by Bidhan Moti 21 (11.40 mm) and Bidhan Moti 1(11.33 mm), respectively.Grain breadth, one of the important features of rice which is one of the important criteria for the market price, showed lesser degree of variation in comparison to other characters.Bidhan Moti 14 recorded the minimum value (1.53 mm) for grain breadth followed by Bidhan Moti 30, 15 and 17, respectively.The L/B ratio that determines the grain shape was very high in all the genotypes.Highest L/B ratio was observed in Bidhan PH, Plant height; NP, number of panicles/plant; NF, number of florets/plant; GL, grain length; KL, kernel length; DFF, days to 50% flowering; PW, panicle weight; NG, number of grains/plant; GB, grain breadth; KB , kernel breadth; DM, days to maturity; PL, panicle length; F%, fertility percentage; GL:B, grain length: breadth ratio; K L:B, kernel length breadth ratio; BM, Bidhanmoti; GW, 1000 grain weight; GY, grain yield/plant;

Table 2 .
Variability and genetic parameters for some polygenic traits in Basmati derivatives.

Table 3 .
Genotypic (G) and phenotypic (P) correlation coefficient among different polygenic traits in Basmati derivatives.

Table 4 .
Path coefficient (genotypic) analysis showing direct (bold) and indirect effects of component traits in rice.