One of sociology's defining debates centers on explanations of the geographic patterning of suicide. this classic debate is revisited using techniques of spatial analysis and data for two geographies: late nineteenth-century French departments, and late twentieth-century U.S. counties. Results of the French analysis contradict Durkheim's claim that "imitation" plays no role in shaping the geographic patterning of suicide. Suicide rates for northern and southern French departments cluster geographically even when the clustering of multiple dimensions of social integration is controlled. These findings are replicated in a contemporary analysis of nonwestern U.S. counties. Results for the American West, however support the Durkheimian view that suicide clusters in geographic space only because important structural predictors of suicide, including measures,of social integration, do so as. well. These discrepant findings are reconciled and it is concluded that the geographic patterning of suicide is shaped by both social integration and imitation.