Study on relationships between yield and its components will improve the efficiency
of a breeding program with appropriate selection criteria. In this study,
interrelationships between yield and its components in cantaloupe were investigated
using sequential path coefficient analysis. To determine important components of yield
in cantaloupe, 49 genotypes generated from 7 × 7 diallel, were evaluated in α-lattice
design with 3 replications. The highest coefficient correlation was obtained between
fruit weight and fruit number with yield. Sequential stepwise multiple regression
analysis was performed to organize the predictor variables into first, second and third
order paths based on their respective contributions to the total variation of yield and
minimum collinearity. Based on the variance inflation factor and magnitude of direct
effects, average of fruits weight and number of fruit per plant were considered as first
order variables and accounted for 96 percent of total variation of yield. The t-test of
significance, using standard error values, obtained through bootstrap resampling,
indicating that all direct effects were significant. The results also indicated that both two
traits of average of fruits weight and number of fruit per plant could be considered as
selection criterion in selecting for increased yield in cantaloupe.