Analysis of variance
Location-wise (Table 2) and pooled (Table 3) analysis of variance indicated significant differences among all the pairs of NILs in each location for all the three traits under study. The performance of six pairs of NILs was comparable across locations, as indicated by the non-significant NIL × location interaction, suggesting predictability of the performance of NILs across test locations. These NILs differing for growth habit controlled by major genes
(Keerthi et al., 2014; Keerthi et al., 2016; Basanagouda et al., 2022) offer a common genetic and PIS background to assess unconfounded and unambiguous effect of growth habit on days to 50% flowering, fresh pod number plant
-1 and fresh pod yield plant
-1.
Effect of growth habit on non-target traits
Both determinate and indeterminate components of NILs varied widely for all the three traits (Table 4). However, the extent of range with respect to days to 50% flowering and fresh pod number plant
-1 in indeterminate component lines of NILs was at least two folds greater than that in determinate component of NILs. The short growth period of determinate NILs could be attributed to breakage of shoot apical meristem (SAM) dominance
(Krylova et al., 2020; Li et al., 2021). Our results are not surprising given that the duration of overlap between the vegetative and reproductive phases in indeterminate genotypes is longer than those in the determinate ones (
Bernard 1972;
Hartung et al., 1981; Cober and Morrison 2010). Considering that differences in days to flowering between pairs of NILs are likely to confound the differences between them for other traits
(Curtis et al., 2000) such as fresh pod number plant
-1 and fresh pod yield plant
-1, we used days to flowering as a covariate. The magnitude of range was three times greater in indeterminate component of NILs than that in determinate component of NILs for fresh pod yield plant
-1. On an average, indeterminate NILs were late to flower by seven days and produced greater fresh pods number plant
-1 (Table 4, Fig 1) and fresh pod yield plant
-1 than determinate NILs (Table 4, Fig 2). In soybean,
Ablett et al., (1989); Robinson and Wilcox (1998);
Kato et al., (2018) and
Kato et al., (2019) have also reported significantly greater economic product (grain) yield in indeterminate genotypes than that in determinate ones. Our results suggest significant influence of growth habit genes for which NILs were selected on non-target traits such as those considered in the study. We believe that, substantial differences in days to 50% flowering, fresh pod number plant
-1 and fresh pod yield plant
-1 between determinate and indeterminate component genotypes of six pairs of NILs could be due to pleiotropic effects of growth habit genes in dolichos bean. However, the results need to be validated using a large number of NIL pairs differing for growth habit.
Previous reports on pleiotropic effects of growth habit genes on grain yield and its component traits using NILs in a range of crops such as common bean
(Wallace et al., 1993) and soybean
(Curtis et al., 2000; Ellis et al., 2000; Cober and Morrison 2010) and Barley (
Kuczyńska et al., (2014), would lend adequate support to our belief.
Hodgkin (2002) opined that pleiotropic genes are widely believed to be evolutionarily conserved as they are subjected to natural selection pressure acting on multiple traits. We believe that
Hodgkin’s (2002) opinion is relevant to genes controlling growth habit in dolichos bean, as growth habit is one of the evolutionarily shaped traits, which is associated with increased fitness under natural selection pressure.
