Towards the Development of a Finger Millet (Elesine Coracana) l. Garten) Breeding Programme in Zambia: A Multivariate Analysis of Agronomic Traits of Mutation Derived Lines

Abstract

Finger millet (Eleusine coracana (L.) Gaertn) is one of the climate-resilient cereal crops with the potential to mitigate the effect of climate change among the smallholder farmers who are the major growers of finger millet. It can adapt to extreme environmental conditions and grow under a wide range of diverse agroecologies, which is an advantage to breeding high-yielding stress-tolerant genotypes. However, the strength of breeding high-yielding, climate-resistant genotypes is highly or largely dependent on the genetic information of the base breeding population. The lack of knowledge on the genetic information of finger millet in Zambia has hampered the breeding of climate-resilient genotypes. Therefore, a study assessed the diversity of M6mutant lines. The mutants were selected from a pure line selection derived from a landrace finger millet variety called Mutubila.The Mutubila pure line was irradiated with a dose of 100gy gamma rays. The irradiated seed was planted to raise the M1 generation for eventual advancement to the sixth generation. From the M2 to M4 generations, mutants were selected based on the number of productive tillers and finger length using the independent culling selection approach at a selection intensity of 10%. In the M5 and M6 generations, however, the selection of advanced lines was expanded to include the number of fingers per ear head and other important agronomic traits. This resulted in 40 mutant lines being selected in the M6 generation. The 40 mutant lines and eight checks were planted using an Alpha Lattice design with two replications and eight blocks. The 48 genotypes were evaluated for agronomic performance, and the data collected was analysed using R software. The results revealed that the mutant lines were diverse in the number of productive tillers per plant (3 to 15), main ear length (4.4 to 14.5cm), number of fingers per ear head (3 to 17) and grain yield (0.58 to 2.95tons/ha). The yield components (number of productive tillers, Finger length, number of fingers, strawweight, and thousand seed weight) were strongly and positively correlated to grain yield (R2 > 0.67) as well as strongly and positively inter-correlated. Days to flowering were weakly and negatively correlated to grain yield (R2 > -0.17.The Mahalabonis analysis grouped the genotypes into ten (10) clusters, with cluster I having the highest number of sub-clusters (11) and cluster IX and X having the lowest sub-clusters (1). The highest inter-cluster distance (> 572,618) was observed between clusters V and IV; V and VII; V and IX; and VII and XI, indicating wider divergence among these clusters. The least inter-cluster distance (<6,043) was observed between II and III; VIII and X; and I and VIII. Distant clusters provide opportunities for breeding by crossing genotypes from two distinct clusters to generate superior cultivars. The study showed that selecting yield components as the mutants is advanced results in divergent lines, which can be recombined to breed high-yielding genotypes with multiple stress tolerance. We conclude, therefore, that mutation breeding for finger millet can be used to generate diversity and the identified divergent lines that can be recombined to develop improved finger millet varieties that can be commercialized.