To estimate the prevalence of infection by hepatitis C virus (HCV) in general practice, and to investigate attitudes to screening for HCV, a specific questionnaire was sent to the sentinel general practitioners (SGPs) of the French Sentinel Network. From the figures reported by SGPs, the number of anti-HCV tests performed in general practice during the six months before the survey (July December 1994) was extrapolated to about 350,000 tests for the whole of France during all of 1994 (95% CI: 314,000 382.000). Of the 312 participating SGPs, 224 (72%) had prescribed at least one anti-HCV test during these six months. and 217 (70%) knew of at least one anti-HCV-positive patient in their practice. With regard to their screening attitudes. 244 SGPs (78%) said they had prescribed anti-HCV tests for patients with no signs of previous or recent hepatitis infection; SGPs who had at least two HCV-positive patients in their practice were more likely to screen for anti-HCV than SGPs who did not (94% vs. 72%, p < 0.001). The main reason for prescription were that patients had an anti-HIV-positive status, were intravenous drug users and/or were blood transfusion recipients. For the 88 SGPs who had not prescribed any anti-HCV test during the six months before the study, the main reasons given were that few patients were at risk in their practice (81 SGPs) and that the risk factors for HCV infection were not well known (12 SGPs). Only 6 SGPs gave the absence of effective treatment as a reason for non-prescription. From the figures reported by SGPs, we extrapolated to 100,000 (95% CI: 90,000-110.000) the number of anti-HCV-positive patients known by the 52,000 French general practitioners in 1994. These results indicate that the Sentinel network may be a useful tool for assessing the evolution of attitudes to HCV screening in general practice.