The role of white-tailed deer (Odocoileus virginianus) in the epizootiology of anaplasmosis in the southeastern United States was examined through retrospective and prospective serosurveys and by experimental infection studies. No serum antibody reactive to Anaplasma marginale was detected with an indirect fluorescent antibody (IFA) assay from any of 1,376 free-ranging deer sampled from 1968 through 1990 from 13 states and Puerto Rico. Thirty-one additional deer from three bovine anaplasmosis enzootic premises also were negative by IFA and Giemsa-stained blood films. Three captive deer given A. marginale intravenously developed antibodies 38 to 41 days post-inoculation (DPI) and remained seropositive for the duration of the study (161 to 287 DPI). At 42 DPI, rickettsemias of approximately 0.0001% infected erythrocytes were observed in all three deer using a DNA probe; low rickettsemias (maximum 0.01%) persisted through 56, 63, and 87 DPI, respectively. One deer had a recrudescent infection from 126 to 146 DPI (maximum rickettsemia 0.001%). We believe that white-tailed deer in the southeastern United States, even though susceptible to A. marginale infection, are not exposed naturally, even at enzootic sites. Furthermore, white-tailed deer did not develop rickettsemias sufficient to support mechanical transmission by biting flies, which is believed to be the primary means of anaplasmosis transmission in this region.