Epidemiological data were recorded and blood samples were obtained from a group of 297 individuals from Luxembourg who had been in contact with rabies-infected or -suspected animals from 1979 to 1985, and who had been vaccinated with a complete course of HDCS rabies vaccine. The majority of the subjects were farmers aged between 30 and 59. The most frequent source of infection was cattle, followed by foxes, sheep and cats. Dogs were in the fifth position only, probably as a direct result of obligatory vaccination of dogs in Luxembourg. Animal rabies was proven in 253 of the 297 cases; in 35 cases, rabies diagnosis was not confirmed, and in 9 cases the animals escaped. Concerning the mode of contact, most people (86%) reported having had contact which caused no skin lesions. Contact with skin lesion, i.e. by biting, scratching or scraping, was only reported in connection with dogs, cats and wild animals, whereas with cattle and other farm animals, contact was mostly without skin lesion. The hand was the part of the body (85%) most frequently involved. Sera were tested by an enzyme immunoassay for rabies antibodies. The results were unsatisfactory: 21.5% of the cases had no demonstrable antibodies and 15.2% had antibodies in low titre. No decline in antibody titre was found according to the time interval after vaccination. Therefore, the enzyme immunoassay does not appear to be optimal to determine long-term immunity.