Antibody titers to Epstein-Barr virus were determined in 13 patients with iridocorneal endothelial syndrome and in 13 healthy race-, age-, and sex-matched controls. Both the geometric mean titer of IgG antibodies to the Epstein-Barr virus capsid antigen and the proportion with high titers of IgG antibodies to the Epstein-Barr virus capsid antigen (greater than or equal to 1:640) were significantly higher in 12 seropositive patients with iridocorneal endothelial syndrome than in 12 seropositive controls (1/761:1/202, P = .001; 83.3%:8.3%, P less than .001). Ten of 12 seropositive patients with iridocorneal endothelial syndrome and five of 12 seropositive controls had antibodies to Epstein-Barr virus-induced early antigens (greater than or equal to 1:10) (Fisher's Exact Test, P less than .05), while four seropositive patients with iridocorneal endothelial syndrome and one seropositive control had low to undetectable levels of antibodies to Epstein-Barr virus-associated nuclear antigen (less than or equal to 1:5) (P greater than .1). Antibody levels to cytomegalovirus or measles virus were not different between patients with iridocorneal endothelial syndrome and controls. Additional studies showed no evidence of humoral immune disorder or collagen vascular disease in the patients with iridocorneal endothelial syndrome. The serologic profiles suggest that the patients with iridocorneal endothelial syndrome examined had a cellular immune abnormality sufficient to permit reactivation of latent Epstein-Barr virus infection and imply, but do not establish, a role for Epstein-Barr virus infection in the pathogenesis of some cases of the iridocorneal endothelial syndrome.