Seed Yield Component Identification and Analysis for Exploiting Recombinative Heterosis in Bottle Gourd

Abstract

Bottle gourd [Lagenaria siceraria (Molina) Standley] is an important African vegetable, but its production is low due to lack of high-yield varieties. This study was conducted to select parameters on which efforts should be concentrated to improve its performance. Path analysis was performed on 52 open-pollinated accessions of bottle gourd grown in a field experiment to determine how per plant seed yield was directly and indirectly affected by: number of days from sowing to emergence (TE) and to first fruit maturity (FM); number of branches (NB); plant length (PL); number of fruit per plant (NF); fruit weight (FW); number of seed per fruit (NS); 100-seed weight (100-SW); sex ratio (SR); and harvest index (HI). All traits except TE and FM were highly variable (CV ≥ 30%) and suitable for exploiting recombinative heterosis in L. siceraria. Five traits exhibiting high CV were correlated with seed yield and NS, FW, and 100-SW had relatively strong direct effects on seed yield enhancement. Stepwise multiple regression analysis indicated that 89.4% of total variation in per plant seed yield could be explained by variation in NS, FW, and 100-SW; 70.5% could be attributed to variation in NS. About 19% of total variation in seed yield could be explained by variation in FW and 100-SW. Residual effects were relatively high (mean combined value = 0.302), indicating that substantial variation was accounted for by components of SY not analyzed. Path analysis identified fruit weight number of seed per fruit, and 100-seed weight as having the greatest effect on seed yield. The accessions showing the highest values for these parameters (NI106; NI266; NI273; NI283; NI354; NI421) and lowest values (NI180; NI202; NI212; NI215; NI248) should be used as parental genotypes to investigate recombinative heterosis for use in breeding.