No abstract

Although bacterial wilt remains a major plant disease throughout South America and the Caribbean, the diversity of prevalent Ralstonia solanacearum populations is largely unknown. The genetic and phenotypic diversity of R. solanacearum strains in French Guiana was assessed using diagnostic polymerase chain reactions and sequence-based (egl and mutS) genotyping on a 239-strain collection sampled on the families Solanaceae and Cucurbitaceae, revealing an unexpectedly high diversity. Strains were distributed within phylotypes I (46.9%), IIA (26.8%), and IIB (26.3%), with one new endoglucanase sequence type (egl ST) found within each group. Phylotype IIB strains consisted mostly (97%) of strains with the emerging ecotype (IIB/sequevar 4NPB). Host range of IIB/4NPB strains from French Guiana matched the original emerging reference strain from Martinique. They were virulent on cucumber; virulent and highly aggressive on tomato, including the resistant reference Hawaii 7996; and only controlled by eggplant SM6 and Surya accessions. The emerging ecotype IIB/4NPB is fully established in French Guiana in both cultivated fields and uncultivated forest, rendering the hypothesis of introduction via ornamental or banana cuttings unlikely. Thus, this ecotype may have originated from the Amazonian region and spread throughout the Caribbean region.

Cacao trees are affected by diseases that attack either their vegetative parts, their fruits or both. In cacao pod diseases, several factors are involved in disease susceptibility, such as the fruiting cycle, fruit size, age, position on the tree and cacao genotype. To gain a clearer understanding of how these characteristics influence cacao pod diseases, four models describing pod growth in several cacao genotypes were evaluated. Three models used to estimate pod volume or surface area were also compared. Observed pod growth was of a sigmoid form and fitted best to the Richards model, well to the Logistic and Beta growth models, and least to the Gompertz model. Pod volume and probably pod surface area were best estimated using a prolate spheroid model. Pod growth models can help improve pod disease management and thereby cacao production. They can help to predict yield, as well as provide information for the timing and frequency of control operations. Information on pod size, surface area and susceptibility will help to improve dose transfer and spray deposit studies intended to optimise control efficiency. (Résumé d'auteur