The Efficiency of Pedigree and Single Seed Descent Selections for Yield Improvement at Generation 4 (F4) of Two Yardlong Bean Populations

Abstract

The efficiency of the selection procedure during succeeding generations is the most important role of any breeding program. There are many selection methods, but no method is perfect for general use with all crop plants. Breeders must carefully consider which method will be the most effective for their purposes. This study was conducted to compare the effectiveness between the pedigree selection (PS) and single seed descent (SSD) methods in two yardlong bean crosses. Both procedures started from the same F2 population in each cross. In 2004, 15 F4 progenies were selected by each selection method from each population. Thirty F4 progenies and the parents of each population were tested in separate experiments with two check cultivars at the Songkhla Field Crop Research Station, Songkhla Province. Selected-PSU and VU135 were grown as check yardlong bean cultivars. Narrow sense heritability (h²) for the pod yield and yield components were estimated through regression analysis of
the F4 progenies on F3 parental plants. Correlations among yield and yield components were also estimated. The results showed that the mean pod yield and yield components of selected F4 progenies derived from both PS and SSD were not significantly different in both the 4501 and 4502 populations. Low heritability for pod yield per plant was found in both the 4501 and 4502 populations at 4.62 and 2.96%, respectively, which indicated why selection for yield improvement in early generations had been ineffective. The highest positive correlation was found between the number of pods per plant and pod yield in both the 4501 and 4502 populations, with correlation coefficients of 0.7540 and 0.9229 (p<0.01), respectively. This study shows that PS and SSD are equally effective for yield improvement in yardlong bean.