Phenotypic evaluation of variability among dry bean cultivars landraces from breeding collections of Institute of Field and Vegetable Crops Novi Sad

Abstract

Dry bean (Phaseolus vulgaris L.) is one of the most important food legumes and source of proteins and dietary fibers. Two major gene pools exist in cultivated dry bean, one being Middle American, and second Andean. Dry bean accessions and cultivars belonging to each of these two gene pools are distinguished based of phenotypic traits, as well as on molecular and biochemical level. Most of the dry bean germplasm is maintained ex situ in gene banks and collections of the research institutes around the world which is characterized with high level of genetic diversity. Possibility to identify this genetic variation is of greatest importance for conservation and utilization of dry bean germplasm. Phenotypic evaluation is the first step in discrimination and classification of these accessions. The aim of this work was to assess phenotypic diversity of 37 dry bean genotypes (domestic and foreign cultivars and landraces) held at dry bean breeding collection of Institute of Field and Vegetable Crops, Novi Sad. We analyzed eight qualitative traits (growth type, color of flower standard, color of flower wings, pod color, seed pattern, color of seed coat, seed brilliance, seed shape) following the Bioversity International Key descriptor for common bean and six quantitative traits (number of pods per plant, number of seeds per plant, seed weight per plant, 1000 seed mass, number of days to flowering, number of days to maturity). Multivariate analyses, consisting of principal component analysis and cluster analysis were applied. In a principal component analysis first two axis explained 46.9% of variation, and the most important traits in discriminating between genotypes were separated. The genotypes differed mostly for number of pods per plant, number of seeds per plant, color of flower standard, color of flower wings, seed coat color and 1000 seed mass. Positive correlations were observed between number of pods per plant, number of seeds per plant, seed weight per plant and growth habit, as well as between growth habit and number of day to flowering and number of days to maturity. Negative correlation was observed between 1000 seed mass and other yield components and growth habit. Both principal component analysis (PCA) and cluster analysis clustered the genotypes into two major groups presenting Mesoamerican and Andean gene pools. Only two foreign cultivars (KP-12 and KP-13) were distinguished as third, separate group. Possible hybrids between these two genepools were identified, meaning genotypes with a combination of traits representing Andean or Mesoamerican gene pool.