Cassava is a staple food that plays a very important role for the Ivorian population. In the Central, Central-Western and Western regions of Côte d’Ivoire, where cassava is a staple food, growers cultivate several varieties whose diversity remains poorly understood. The aim of the present study was to analyze the agromorphological variability of cultivars collected in these growing basins, in order to make better use of this varietal diversity. Agromorphological characterization was carried out on a collection of 70 accessions collected from growers, using ten quantitative variables. Analysis of variance (ANOVA) showed significant phenotypic differences for all the traits studied. Principal component analysis (PCA) confirmed this polymorphism at 83.49% for the first three axes. Accessions were structured into 5 groups of morphological variability following Hierarchical Ascending Classification (HAC). These groups are distinguished by plant height, height of first branching, central lobe length, central lobe width, number of lobes, leaf length to width ratio, petiole length, tuberous root weight per plant and total yield. The first three groups have cultivars with yields in excess of 20 t/ha and constitute the best genetic resources for increasing cassava production in these growing basins and in Côte d’Ivoire.

This study was conducted to assess the molecular diversity of eight NJM x GVT coconut palm (Cocos nucifera L) hybrid progenies and a BP121+ hybrid control using 15 microsatellite loci. The NJM x GVT hybrids are tolerant to the Lethal yellowing coconut disease when the control is very susceptible to the disease. As a result of this work, 86 alleles were identified at the 15 microsatellite loci with a variation of 2 to 11 alleles per locus. The intra-progeny diversity (HS = 0.50) is higher than the inter-progeny diversity (DST = 0.04). The allelic richness of the progeny ranges from 2.07 to 3.13. The NJM x GVT progenies are molecularly different from the control PB121+. In sum, 63.95% of the alleles present in the NJM x GVT progenies are absent in the BP121+ control. Furthermore, 8.14% of the alleles present in the control are absent in the NJM x GVT hybrids. Collecting samples for further work such as QTLs identification could take these results into account. Thus, it would be wise and advantageous to sample a high number of trees per progeny than to choose several families of progenies. The allelic richness of the progenies could guide the choice of progenies. The markers comprising the private alleles detected between the PB121+ and NJM x GVT progenies can be used in the varietal purity tests.