VIEW ARTICLE

Disease Detection and Losses

A Sampling Procedure To Detect Grapevine Downy Mildew. R. C. Seem, Associate professor, Department of Plant Pathology, New York State Agricultural Experiment Station, Cornell University, Geneva 14456; P. A. Magarey(2), P. I. McCloud(3), and M. F. Wachtel(4). (2)(4)Plant pathologist and technical officer, Department of Agriculture, Loxton Research Centre, Loxton, South Australia 5333; (3)Biometrician, Department of Agriculture, Mathematics and Computing Services, Adelaide, South Australia 5000. Phytopathology 75:1252-1257. Accepted for publication 8 June 1985. Copyright 1985 The American Phytopathological Society. DOI: 10.1094/Phyto-75-1252.

A sampling procedure was developed that would detect primary infections by Plasmopara viticola in commercial vineyards with 99% confidence when the incidence of infected leaves was as little as 0.01%. The probability of detecting disease, if present, is the product of the probability of finding at least one diseased leaf (given that disease is present) and the probability of selecting a diseased sampling unit. Disease was simulated with artificial lesions and four to six incidence levels were assigned to half-vine sampling units consisting of 7.3 meters of vine row. Based on experimental results, probabilities of detecting at least one lesion within a sampling unit (disease incidence of 0.01- 0.07%) were determined for different sampling period durations, times of season, and observers on cultivars Rhine Riesling, Muscat Gordo, Sultana (synonym Thompson Seedless), and Shiraz. An analysis of variance with a linear logistic model indicated that only disease incidence and cultivar accounted for significant variation in the probability of detecting disease. In vineyards with known distribution of disease, the minimum numbers of sampling units per vineyard, based on 2-min sampling periods per sampling unit and a 99% confidence level for detection were 13 for Riesling, 31 for Gordo, 39 for Sultana, and 73 for Shiraz; when the likely distribution of disease in a vineyard is not known, more sampling units are required.