Spatial Distribution of Coffee Wilt Disease Under Roguing and Replanting Conditions: A Case Study from Kaweri Estate in Uganda.

Based on time and spatial dynamic considerations, this study evaluates the potential role of short- and long-distance dispersal in the spread of coffee wilt disease (CWD) in a large commercial Robusta coffee estate in Uganda (Kaweri, 1,755 ha) over a 4-year period (2008 to 2012). In monthly surveys, total disease incidence, expansion of infection foci, and the occurrence of isolated infected trees were recorded and submitted to spatial analysis. Incidence was higher and disease progression faster in old coffee plantings compared with young plantings, indicating a lack of efficiency of roguing for reducing disease development in old plantings. At large spatial scale (approximately 1 km), Moran indices (both global and local) revealed the existence of clusters characterized by contrasting disease incidences. This suggested that local environmental conditions were heterogeneous or there were spatial interactions among blocks. At finer spatial scale (approximately 200 m), O-ring statistics revealed positive correlation between distant infection sites across distances as great as 60 m. Although these observations indicate the role of short-distance dispersal in foci expansion, dispersal at greater distances (>20 m) appeared to also contribute to both initiation of new foci and disease progression at coarser spatial scales. Therefore, our results suggested the role of aerial dispersal in CWD progression.

Molecular mapping of genes conferring field resistance to South American Leaf Blight ( Microcyclus ulei) in rubber tree.

The South American Leaf Blight (SALB), caused by the fungus Microcyclus ulei, is the major rubber tree disease in all Central and South America. A population of 192 progeny individuals derived from a cross between a resistant clone and a susceptible cultivated clone was planted in a field trial in French Guiana in order to evaluate the resistance parameters under real infestation conditions. The resistance type (RT), the presence of stromata (ST) and the level of attack (AT) were observed 20-times on a 22-months period, and semi-quantitative evaluation of stromata was registered only once. The search for QTLs was performed using the Kruskal-Wallis test, Interval Mapping and the Composite Interval Mapping method. One major QTL located on linkage group g13 was detected on the RO 38 map, responsible for 36 to 89% of the phenotypic variance of resistance. This resistance QTL corresponds to one that had previously been detected under controlled conditions of infestation and we called it M13-1bn. Surprisingly, the effect of this QTL was larger under natural conditions of infestation than under controlled inoculation. Other minor QTLs (four on the RO38 map and one on the PB 260 map) were also detected. The type of resistance brought by M13-1bn, as well as its durability, are discussed. Applications for rubber tree breeding programs are considered.