Assessment of purity of yellow common bean seeds in Tanzanian markets

Abstract

In Eastern Africa, common beans are a major source of food and revenue for smallholder farmers who make up the bulk of the population. However, yellow bean trade is affected by limited information and unstructured system on production which has negative implications on seed supplies and other possible investments. The presence of improved varieties in seed markets shelves is a positive sign of a vibrant seed system. The aim of this study was to assess the genetic and phenotypic purity of yellow bean varieties traded in the major markets in Tanzania using morphological traits (basing on phenotypic traits of common beans) and genotypic traits (Single Nucleotide Polymorphisms (SNPs)). A total of 300 seed samples were collected from Tanzanian markets and compared with 14 reference seed samples maintained at the International Centre for Tropical Agriculture (CIAT) gene bank in Uganda and the Tanzania Agricultural Research Institute (TARI). The seed samples were planted in out of the screen house at CIAT-Uganda. For the phenotypic purity study, the samples were planted in augmented design and the data collected from the growing plants included days to flowering and to physiological maturity, number of pods per plant and number of seeds per pod, flower colour and mature pod colour. The ANOVA, principal component analysis (PCA) and cluster analysis were conducted to characterize the seed samples. For reference seeds, the results of the ANOVA revealed significant (P<0.05) differences in five phenotypic traits out of eight traits that were studied. These were growth habit (GH), days to flowering (DF), flower colour (FCOL), mature pod colour (MPDCOL) and seeds per pod (SDPD). For market seeds, significant differences were observed only in two traits, namely Flower colour (FCOL) and Mature pod colour (MPDCOL). For reference seeds, the PCA distinguished traits seed weight (SW) and pods per plant (PDPL) as having greater variability, while for market seeds the number of days to flowering (DF), days to physiological maturity (DPM), seeds per pod (SDPD) and seed weight (SW) had greater variability. For the genotypic trait study, a total of 236 yellow bean seed samples from Tanzanian markets and 14 reference seed samples from TARI and CIAT were used. They were planted out of the screen house at Kawanda Research Station in Uganda. Two weeks after germination, leaf samples were taken and genotyped with 50 quality control SNP markers at the Intertek laboratory in Australia. The analysis of molecular variance (AMOVA) showed that reference seeds had lower average heterozygosity (51.0%) than market seeds (75.2%). The genetic distance that was obtained from pairwise population from SNPs and for reference seeds derived populations was moderate (0.040), while that of market seed samples was higher at 0.055. Nevertheless, the study showed that there is phenotypic and genotypic variability in yellow common beans in Tanzanian markets. Therefore, it was recommended that the Seed Quality Control and Certification Department should be strengthened so that it can effectively monitor and ensure that the quality of yellow bean seeds is maintained as they move from breeders to other players along the seed value chain. In addition, there is need to increase the supply and accessibility of better quality seeds in the markets, and to make seed laws functional so that those found selling fake seeds are punished.