In November 1988, the Food and Drug Administration approved reagents for serologic screening for human T-lymphotropic virus type I (HTLV-I) infection in blood donors and patients suspected of having HTLV-I-related illnesses. These reagents are able to detect HTLV type II (HTLV-II), a close relative of HTLV-I with no known pathologic effect. The distinction between the two forms of HTLV is important to the donor and to any recipient of blood containing HTLV. The application to sera from 38 seropositive blood donors and 2 recipients (37 "confirmed" positive and 3 indeterminate by Western blot) of two methods (Western blot and peptide enzyme immunoassay) for serologic distinction between HTLV-I and -II is described. These results were compared to those from polymerase chain reaction (PCR) analysis of HTLV proviral DNA obtained from donor peripheral blood mononuclear cells. The peptide enzyme immunoassay was found to be less sensitive than the Western blot, but completely concordant with PCR results when differential reactivity could be established. The Western blot pattern showed complete diagnostic concordance with the samples with confirmed-positive serologic tests, but was incorrect in two HTLV-II-infected donors with indeterminate serologic tests. Thirty-three (89%) of the 37 individuals from this predominantly Native American and Hispanic group of blood donors were found to have HTLV-II. These findings confirm and extend previous reports that HTLV-I infection may be less common, and HTLV-II infection more common, than previously believed. The peptide enzyme immunoassay can provide most individuals who have positive results with the HTLV-I/II screening test with serologic distinction between HTLV-I and HTLV-II.