VIEW ARTICLE

Techniques

Illusions in Visual Assessment of Stagonospora Leaf Spot of Orchardgrass. R. T. Sherwood, Research plant pathologist, U.S. Regional Pasture Research Laboratory, USDA-ARS, University Park, PA 16802 (also adjunct professor of plant pathology, Pennsylvania State University, University Park); C. C. Berg(2), M. R. Hoover(3), and K. E. Zeiders(4). (2)(3)(4)Research agronomist, research associate, and research plant pathologist, respectively, U.S. Regional Pasture Research Laboratory, USDA-ARS, University Park, PA 16802 (also adjunct assistant professor of plant breeding, support scientist, and adjunct instructor, respectively, Pennsylvania State University, University Park). Phytopathology 73:173-177. Accepted for publication 6 July 1982. This article is in the public domain and not copyrightable. It may be freely reprinted with customary crediting of the source. The American Phytopathological Society, 1983. DOI: 10.1094/Phyto-73-173.

Two groups of five experienced scorers estimated percent leaf area spotted by matching orchardgrass leaves infected by Stagonospora arenaria against published area diagrams. Actual area of spotting was determined by weighing paper replicas of photographs. Nine of the scorers usually overestimated spotted areas. Overestimation was greatest when infected area was smallest, often being two to three times the actual area, and it decreased as the infected area increased. When two leaves had equal total spotted areas, the leaf having substantially more (but smaller) spots was usually scored higher. Regression analysis showed that overestimation was inversely proportional to the natural logarithm of the disease area for all scorers and also directly proportional to the number of spots for five scorers. There were significant group, scorer within group, leaf, and scorer × leaf effects. If visual area assessments overestimate disease more seriously at low disease incidence, their use in equations for predicting disease increase or in equations for yield reduction will result in underestimation of the true rate or amount of loss. The coefficients of variation (x*=21.5%) indicated that visual estimates were not highly precise.