Correlation and Path Analysis between Yield and Yield Components in Cassava (Manihot esculenta Crantz) in Southern Tanzania

An experiment to study the relations among root yield and some traits of cassava using correlation and path coefficient analysis was conducted in the Sothern Tanzania during the 2011 – 2012 cropping season. The design of the experiment was a Randomized Complete Block with three replications using twelve genotypes. Highly significant and positive correlations were found between plant height and stem girth (r = 0.5900***), plant height and roots per plant (r = 0.4463***), plant height and harvest index (r = 0.3005***), branches per plant and roots per plant (0.2441***), stem girth and roots per plant (0.5046***) and roots per plant and harvest index (0.2647***).  Also, highly significant and positive correlations were found between root yield and plant height (r =0.5436***), stem girth (r = 0.3874***), roots per plant (r = 0.7053***) and harvest index (r = 0.3025***). Compared to the simple correlation analysis, path analysis of roots per plant, plant height and root size evolved the highest direct influence 0.619, 0.290 and 0.153, respectively. Therefore, indirect selection for higher root yield may be effective for improving these characters. This study suggests the relative higher value of residual effect (0.51) indicated more yield components should be considered in the future to account for the variation in cassava root yield. Key words: Correlation, Path analysis, Cassava, Yield, Yield components.