To compare rickettsial infectivity and seropositivity rates against spotted fever group (SFG) rickettsiae, ticks and wild mammals were collected from three areas where Rickettsia rickettsii was thought to be enzootic in Connecticut during 1978-1979, and from four additional sites (with no reported human cases) between 1976 and 1979. Of the 1,001 Dermacentor variabilis examined by the hemolymph test, 59 (5.9%) contained rickettsia-like organisms; direct immunofluorescence tests verified the presence of SFT rickettsiae in 14 specimens. Prevalence of rickettsiae-infected ticks at Newtown, an area where human cases of Rocky Mountain spotted fever probably originated, was 2.2%. Rates for six other areas ranged between 0 and 6.3%. Isolations included Rickettsia montana from four ticks collected at Branford and Woodbridge, and R. rickettsii (R-like strain) from the blood of an acutely ill person. Microagglutination (MA) tests indicated that 15 (14.9%) of 101 Peromyscus leucopus (white-footed mice) from Newtown had agglutinins in titers greater than or equal to 1:8 against R. rickettsii, whereas five of 92 white-footed mice (5.4%) from Brandord, West Hartford, Woodbridge, and Sharon were considered MA-positive. Indirect microimmunofluorescence tests of Procyon lotor (raccoon) sera revealed antibodies to R. rickettsii in 33 of 69 (47.8%) samples from Newtown and in two of 60 (3.3%) from Guilford. Additionally, 17 raccoons had sera specific to R. montana (n = 8) or to the 369-C rickettsia strain (n = 9). Since rickettsia-positive ticks are high-titered seropositive mammals occurred at widely separated sites in Connecticut there are probably several foci of SFG rickettsiae distributed throughout the D. variabilis range.