Epidemiology of Botrytis spp. associated with neck rot of onion (Allium cepa L.) in northern Tasmania, Australia

Neck rot of onion, caused by a complex of Botrytis spp., is an important fungal disease of onion worldwide. In Tasmania, it has caused considerable losses to the onion industry in some seasons. Botrytis infects the onion plant in the field but usually the infection remains symptomless, with the fungus growing into the bulb during curing and producing a rot of the bulb in storage. The taxonomy of Botrytis species causing onion neck rot is currently under review. Recent studies of the ribosomal internal transcribed spacer (ITS) region of the genome of Botrytis spp. associated with neck rot have confirmed the existence of three distinct groups. A preliminary study of 24 isolates' from Australia identified one isolate as B. aclada type Al, and 22 isolates as.B. aclada type All (B. allii) with no isolates of B. byssoidea found. A survey of 16 commercial onion bulb crops in northern Tasmania from October to December of 1999 detected B. allii in leaf samples from six crops at an incidence of 0.1% to 0.3%. In three crops that were re-sampled in January 2000, incidence had increased from 0.1% to 3.2%, 0% to 0.5% and 0.3% to 5.8%. Infected crops were found in both main onion-growing regions (northwest and the northeast) in Tasmania. After storage of bulb samples from eight of the surveyed crops, the incidence of Botrytis neck rot ranged from 0.4% to 16.3% with an average incidence of 5.3%. Two field trials were conducted to examine the spatio-temporal spread of B. allii during the season from a point source (2000) and line source (2001). Data of B. cinerea spatio-temporal spread was also collected. In 2000, 100 plots (1.60 by 4.48 m) ‚Äö were established in a 10-by-10 lattice. Leaf samples were taken six times during the growing season and incubated under high humidity to determine the incidence of B. allii in foliage. The incidence of B. allii increased from 0.0% at placement of inoculum (150 days after sowing) to a cumulative 3.8% and a non-cumulative 1.8% at 84 days after placement of inoculum (234 days after sowing). The cumulative sample method was based on always treating a plant as infected once it was found to be infected. The non-cumulative sampling method was the incidence found at each sample time. The disease progress curve was described best by exponential and logistic models (P = 0.001, R2 = 0.99), indicating secondary spread of the pathogen. The disease gradient was steep and best described by an exponential model, usually indicative of a rain splash-dispersed pathogen. Spatial analysis using ordinary runs analysis, beta-binomial distribution analysis, spatial analysis by distance indices and radial correlation analysis determined that the pattern of infection was aggregated.