Abstract
Disease severity data from grids of contiguous quadrats were used to investigate spatial pattern of alfalfa leaf spot, which is caused primarily by Leptosphaerulina briosiana. Three methods of analysis were used: discrete probability distributions, indices of dispersion, and a blocked quadrat variance method. A range from regular to moderately aggregated disease pattern was found among the samples. Distributional techniques generally did not allow differentiation of spatial pattern among samples. Indices of dispersion generally indicated a regular to random pattern in the data at the quadrat level, although slight clustering was indicated in a few cases. Taylor''s b was estimated on both a field and plant-to-plant (within quadrat) basis and indicated a greater degree of aggregation at the plant-to-plant level than at the field level. The blocked quadrat variance method showed multiple peaks in the mean square versus block size plots which may indicate clustering at several scales.