Abstract
The spatial pattern of microsclerotia of Macrophomina phaseolina was determined from the assay of 2,250 soil samples taken from three blocks of 750 quadrats (1.0 .times. 0.3 m) in a cultivated field at Marana, AZ [Arizona, USA]. This spatial pattern was aggregated as measured by the mean-to-variance ratio and Morisita''s index of dispersion. Aggregate size was determined for each block using Morisita''s index of clumping. Based on Moran''s I, a statistic of spatial autocorrelation, significant positive autocorrelation was found among quadrats in all three blocks, indicating that similar levels of inoculum density tended to occur in quadrats of close proximity. Because the spatial pattern of microsclerotial population was found to be nonrandom, systematic sampling strategies were found to be most useful for mean estimation. Comparisons probabilistic and systematic sampling strategies are discussed.