English French

Dr. Doumbia I. Zan



Several techniques, such as chemical and bio-insecticides have been applied by farmers to control thrips (Megalurothrips sjostedti) damage on cowpea. Many factors, such as improper amount of insecticides to dilute in water, cost and unavailability of insecticides and equipment limit the effective control of these thrips. Incorporating thrips resistant genes into local adapted landraces could increase farmers‘ income. These new lines could be grown with minimal insecticide application. Participatory Rural Appraisal (PRA) was conducted within two districts through focus group discussions and questionnaire administration to identify farmers‘ perception about thrips and also their preferences for varietal development in Mali. Most of the farmers were aware of the thrips and different types of damage associated with their attack. Extracts from plants, such as Neem (Azadirachta indica), Pepper (Piper guineense) and Papaya (Carica papaya) were applied by most of the farmers on cowpea fields for thrips control. High yielding varieties with medium maturing, semi-prostrate growth forms and white seeds were the major preferences of famers across PRA locations. Amary shô, Kalifala and M‘barawa were farmers‘ preferred varieties suggested for improvement for thrips resistance. A study was conducted to assess the variability among 120 genotypes under natural and artificial thrips infestation within Cinzana and N‘Tarla locations. Comparison of infested and control experiments identified CIPEA82672 and Suivita2 genotypes as highly tolerant having lower damage scoring than the resistant check Sanzisabinli. Positive correlations were detected between damage indices and number of adults thrips from Cinzana (R2= 0.264) and N‘Tarla (R2= 0.603) locations. Two thrips species, Frankliniella schultzei and Sericathrips occipitalis, were identified from the two screening sites and from thrips alternative host (Pterocarpus santalinoides). Discriminant analysis revealed some qualitative traits (plant pigmentation, mature pod colour and seed coat color) and quantitative traits (days to 50% flowering and maturing) that highly differentiated genotypes. Cluster analysis from qualitative and quantitative data grouped the genotypes into three major clusters while Simple Sequence Repeat (SSR) markers partitioned them within seven groups with the two resistance checks being in different clusters with varieties suggested for improvement. CIPEA82672 was closely related to Sanzisabinli. The Polymorphic Information Content of the SSR markers used to differentiate varieties was between 0.37 (VM22) and 0.82 (SSR1). Allele size ranged from 100 to 1, 000 bp with the mean 0.71, 5.5, 0.59 and 0.34, respectively, for allele frequency, number of alleles, gene diversity and heterozygosity. Some duplicate accessions were identified based on molecular markers and they were removed from the collection. Variability was observed among the progenies of 4 crosses and their parents for all traits across three study locations. The F1 generations performed better than the mid-parent values for some crosses with some transgressive segregations observed in F2 and backcross populations with damage scoring and other traits skewing towards resistant parents (Sanzisabinli or TVu 1509). Maternal effects were detected since more yields and less damage were noticed with most of the generations when the resistant parent was used as female in crosses. Additive (n), additive x dominance (j) and dominance x dominance (l) effects were the modes of gene action predominantly involved with thrips resistance control in cowpea. Number of effective factors for thrips damage control varied from 3.4 (number of peduncle per plant) to 6.17 (total number of pods per plant). Polygenic resistance was involved to control thrips resistance among segregating and non-segregating populations that suggest ways to enhance landrace resistance through classical breeding.