Abstract

For two eggplant hybrids grown under greenhouse conditions the researchers calculated the correlations among ten variables: leaf width (LW, in cm), number of spines per leaf (NSL), plant height (PH, in cm), fruit length (FL, in cm), fruit diameter (FD, in cm), number of spines per calyx (NSC), number of seeds per fruit (NSF), fruit weight (FW, in g), number of fruits per plant (NFP), and yield (Y, in kg/plant). Data for variables PH, NSC and NSF did not show a normal distribution; the opposite was the case for all other variables. A significant Spearman correlation was found between NSC and NSL (r = 0.96*). Seven significant Pearson correlations were found between the following variables: FL and FW (r = 0.98**), FL and FD (r = 0.95**), FW and Y (r = 0.94**), NSL and FD (r = -0.93**), FL and Y (r = 0.93**), FD and FW (r = 0.92**), and FD and Y (r = 0.85*); in these cases, linear regression equations and determination coefficients (R2) were obtained. FW, FL and FD were the only variables that showed a significant correlation with Y, and so they should be considered key traits for breeding processes in eggplant.

Keywords Pearson correlation; Spearman correlation; linear regression; spines; plant breeding