On the basis of the hypothesis that sarcoidosis might be a transmissible disease, we thought a study of the familial contacts of known cases, as is done in tuberculosis, might reveal asymptomatic cases. In order to assess the efficiency of chest radiographs in detecting asymptomatic sarcoidosis in such contacts, we carried out a cross-sectional study. The population studied consisted of 66 index cases of sarcoidosis who were self-selected by their agreeing to take part in the study. Criteria were established with regard to the size of the family to be studied, along with three grades of intensity in the relationship between the index cases and their respective contacts. 976 familial contacts over 14 years of age, from grandparents to grandchildren of the index cases were identified, including nonconsanguineous contacts. 496 close contacts were asked to take part in the survey. Among these contacts there were 2 with a previous sarcoidosis, a prevalence of 4/1,000. 239 of these 496 contacts agreed to radiograph screening. Among those screened, no new cases of sarcoidosis were detected. We concluded that at present in our environment, with a low incidence and good prognosis, routine screening of relatives is not indicated. Nevertheless, longitudinal epidemiological studies designed to ascertain the pattern of transmission of the disease and including chest radiograph screening in a population with a high incidence of the disease could be of interest.