A n unknown environmental agent has been suspected to induce systemic lupus erythematosus (lupus) in man. Prompted by our recent immunochemical findings, we sought evidence for an association between Epstein-Barr virus infection and lupus. Because the vast majority of adults have been infected with Epstein-Barr virus, we chose to study children and young adults. Virtually all (116 of 117, or 99%) of these young patients had seroconverted against Epstein-Barr virus, as compared with only 70% (107 of 153) of their controls (odds ratio 49.9, 95% confidence interval 9.3-1025, P < 0. 00000000001). The difference in the rate of Epstein-Barr virus seroconversion could not be explained by serum IgG level or by cross-reacting anti-Sm/nRNP autoantibodies. No similar difference was found in the seroconversion rates against four other herpes viruses. An assay for Epstein-Barr viral DNA in peripheral blood lymphocytes established Epstein-Barr virus infection in the peripheral blood of all 32 of the lupus patients tested, while only 23 of the 32 matched controls were infected (odds ratio > 10, 95% confidence interval 2.53-infinity, P < 0.002). When considered with other evidence supporting a relationship between Epstein-Barr virus and lupus, these data are consistent with, but do not in themselves establish, Epstein-Barr virus infection as an etiologic factor in lupus.