First, second, third, and fourth authors: Department of Plant Pathology; second author: Department of Statistics; fourth author: Agricultural Research Service, U.S. Department of Agriculture, University of Wisconsin, Madison 53706
Go to article:
Accepted for publication 30 September 1996.
Snap bean plants within seven-row segments that ranged from 65 to 147 m were sampled, using a cyclic sampling plan. In the cyclic sampling plan, only 6 of every 31 plants were sampled, but sampled plants were spaced such that pairs of plants that were 1, 2, 3, 4,…, 1,525 plants apart could be identified within each sample. Every leaflet on every sampled plant was assessed for bacterial brown spot, and the proportion of disease leaflets per plant was determined. Arcsine square-root-transformed disease incidence values were analyzed for spatial patterns by autocorrelation and spectral analyses. Disease patterns were detected at several different scales within a single snap bean row, at distances that ranged from ˜20 to ˜100 m. Approximately 23 to 53% of the disease variability in the samples could be described by sine and cosine curves, indicating a substantial component of regularity in the disease patterns. Possible origins for these regular patterns, including cultural practices and seed infestation, are discussed.
The American Phytopathological Society, 1997