1Natural Resources Institute, University of Greenwich, Chatham, Kent ME4 4TB, UK; 2Unidad de Fitoproteccion, Centro Agronomico Tropical de Investigacion y Ensenanza (CATIE), Costa Rica

Background and objectives
Black Sigatoka leaf spot disease of banana and plantain, caused by ascospores and conidia of the fungal pathogen Mycosphaerella fijiensis, severely affects fruit quality and plant yield. The world-wide spread of Black Sigatoka cannot easily be explained by windborne dispersal, even though the disease has appeared in areas that have not exchanged fruit or plants [1]. The arrival of Black Sigatoka in disease-free areas of the Caribbean is likely to prove catastrophic for the subsistence-based cultivation there and there is a strong need to understand and quantify the role of environmental factors in the spread of ascospores and conidia. In order to determine aspects of the windborne dispersal of these pathogens, NRI and CATIE have been investigating spore release from infected leaf material and their subsequent viability within the atmosphere, spore dispersal and disease development within infected and uninfected sites, and the dispersal of spores over tens of kilometres. The results of the studies over different distances should improve assessment of viable spore dispersal, contributing to the development of methods of disease control.

Materials and methods
Fieldwork was carried out at CATIE from October 1993 to February 1994 and from April to September 1995. Spore populations were assessed by counting spores released from rewetted leaf material; ascospore viability following exposure to ultraviolet radiation in simulated natural sunlight was also investigated. Spores were sampled daily in a naturally infected and an initially uninfected plot using shielded rotorod spore traps, for 20-min periods, with detailed measurement of weather conditions. Weekly or twice-weekly surveys of disease development were also conducted at both sites. Long-distance dispersal along a 5 km profile across the floor and up one side of a valley, to an elevation of 1000 m, was investigated using continuously-running volumetric spore traps.

Results and conclusions
Ascospores and conidia were found to be airborne within, above and outside infected plantations, but trap catches of ascospores were always low compared with the potential numbers available from spore release from necrotic tissue. No statistical relationship could be established between spore catches within 2 m of infected plants and disease levels on them, although more spores were caught in the vicinity of plants showing greater levels of necrosis: disease levels within a plantation could not be estimated using disease assessments on a random selection of plants [2]. Periods of alternate drying and wetting of diseased leaves, and changes in temperature, were more conducive to spore release than continuous wetting and exposure to constant temperatures, although shorter dry periods did tend to promote release. There is evidence that ascospores may be transported at least 5 km, and possibly over 40 km, from disease sources, but spores were only viable for 6 h under simulated natural conditions of tropical sunlight. Although the viability of spores undergoing long-distance transport is not yet known, the results of these investigations are supported by the disease incidence pattern within the Caribbean.

1. Burt PJA, Rutter J, Gonzales H, 1997. Plant Pathology 46, 451-458.
2. Smith M, Rutter J, Burt PJA, Gonzales H, Ramirez F, 1997. Annals of Applied Biology 131, 63-77.