Inter-relationships among potato traits and their significance in determining tuber yield

Abstract

Research aimed at selecting potato varieties from Standard International Field Trials (SIFT) with high tuber was carried out at five locations during three seasons of 2001 - 2002 in Uganda. Data were collected on number of leaves, number of stems, plant height, percentage tuber dry matter, number of tubers, mean tuber weight and yield. Results from the simple phenotypic correlation and multiple regressions indicated that mean tuber weight (r = 0.61, P<0.001) and number of tubers per plant (r= 0.47, P<0.001) had the highest correlation values and were also retained in the regression equation across all locations, suggesting that these are the most important components of tuber yield. Path analysis was used to estimate the magnitude, significance and directions of causal relationships between these variables and yield; and thereafter an output path diagram developed. Path analysis identified number of leaves, number of stems, percentage tuber dry matter, number of tubers and mean tuber weight as the main yield components, with mean tuber weight (P = 0.62) and number of tubers per plant (P = 0.46), having the highest direct path coefficients. Although the direct effect of plant height on tuber yield was negative, plant height and number of stems per plant were important components of mean tuber weight. Path coefficient analysis also showed that direct and indirect effects of plant traits on yield explained only 47% of the total variation in fresh tuber yield probably due to environmental influence.