Johne's disease is a well recognized problem in dairy herds. Relatively little information is available on either the prevalence or the control of Johne's disease in commercial cow-calf operations. In the fall of 1999, blood samples were collected during pregnancy testing from cows on community pastures in Saskatchewan. Sera from these cows were analyzed using a commercial ELISA for antibodies to Mycoplasma avium subspecies paratuberculosis. All cows from each herd examined at the community pastures were sampled. Of the 1799 samples tested, 15 had sample to positive (S/P) ratios greater than 0.25 and were considered positive (apparent sample prevalence, 0.8%; 95% CI, 0.4% to 1.5%). If we assume test sensitivity of 25% and specificity of 98% as recommended by the National Johne's Working Group, the true sample prevalence is not significantly different from 0.0%. The ELISA S/P results for the antibody test-positive animals ranged from 0.27 to 2.5. If a herd was classified as positive based on one test-positive animal, the average herd apparent prevalence was 15.2% (95% CI, 7.1% to 28.6%). If the potential for false-positive results was considered with 2 or more positive animals being required for positive herd status, the herd prevalence was 3.0% (95% CI, 0.4% to 13.4%). Because of the very low prevalence in cow-calf herds, future research to identify risk factors and control points should target problem herds and utilize a case-control study design.