We describe a statistical method appropriate for the analysis of spatial autocorrelation in data varying in time as well as space. In particular, the technique was developed lo address the issue of geographic synchrony in ecological variables that may change markedly from year to year such as population density of animals or seed production of trees. The method yields ‘modified correlograms” that test for significant autocorrelation between sites located within any given range of distances apart. This technique facilitates detecting and understanding spatial processes m a variety of ecological phenomena, including testing the plausibility of causational hypotheses using cross-correlational analyses. Several examples are discussed, including population densities of squirrels in Finland, winter densities of two hawk species in California, and acorn production and radial growth by individual blue oak Quercus douglasii trees in central coastal California.