Abstract
A narrow genetic diversity limits the scope for obtaining new recombinants. In general, Bangladesh lentils have narrow genetic base with respect to morphological, phenological and agronomic traits as well as biotic and abiotic stresses. Therefore, efforts were undertaken to study 110 accessions of lentil from home and abroad to identify diverse group of genotypes. Diversity was studied for both morphological and molecular level. For morphological characterization both qualitative and quantitative characters showed significant diversity among the accessions. Variability was found for stem pigmentation, tendril formation, flower colour, cotyledon colour, testa colour and testa pattern. Genetic diversity was estimated using unweighted pair group with arithmetic means (UPGMA). Dendrogram of quantitative characters was constructed for both individual and combined environments. The accessions were grouped into five, four, three and five major clusters for Env.1, Env.2, Env.3 and Env.4, respectively. For combined environment, it was six. The site regression (SERG) stability analysis using biplot technique was also done for both individual and combined environments. Stability analysis for first flowering (37.39 days) and maturity (97.53 days) showed that BLX98006(3) was the earliest but unstable for these characters. The accessions BLX98002(3), ILLX87040 and N11-424 were early maturing genotypes and stable for this character. Accessions DPL-44, ILL5113, ILL 2532 and ILL8147 were the most stable for number of pods per plant. Accessions ILL7656, ILL 2532, and ILL 2581 produced higher seed yield and stable for seed yield per plant. Stemphylium blight disease incidence under natural epiphytotic conditions was observed in 110 lentil accessions using 0-5 scoring method. Under natural management conditions six accessions seemed to be highly resistant and 10 were resistant, 43 were moderately resistant and the rest were moderately susceptible, susceptible and highly susceptible. Molecular analysis for SSR markers was carried out with twenty random primers. Of them, nine primers produced different levels of amplifications. A total of 104 alleles were determined ranging from 4 to 22 per locus. Estimated gene diversity value for 33 loci was 0.69%. The frequency of the most common allele at each locus ranged from 12.50% (SSR33) to 86.49% (SSR340). The polymorphic Information Content (PIC) values ranged from 0.23 to 0.93 with an average of 0.66. Cluster analysis using the Unweighted Pair Group Method with Arithmetic Mean classified accessions into eleven distinct groups. The groups were A (A1, A2, and A3), B (B1, B2, B3, B4 and B5) and C (C1, C2, and C3). There was a little synchronization of clusters membership in morphological and molecular data. In majority of the cases the grouping of accessions based on morphological characters did not match with the grouping based on SSR markers. Two years experimentations both at Ishurdi and Magura, two accessions, ILL 4605 and 8605-2 were identified as promising for trial in Ishurdi, Pabna, Kushtia and Meherpur region and another two accessions, ILL 2581, and ILL 8006 were identified as promising for further testing in Magura, Jessore and Faridpur region. The other four promising accessions, ILL 5150, ILL 2581, ILL 7656 and ILL 2532 identified from the combined data of Ishurdi and Magura over two years experimentations can be proposed for further trial in wide range of environments across Bangladesh. Preliminary yield trials conducted at Ishurdi and Magura showed high yield potentiality of four accessions, ILL 7656, ILL 2532, ILL 5150 and ILL 2581.