We have investigated alterations in Epstein-Barr virus antigen specific cytotoxic T-lymphocyte (EBV-CTL) activity during human pregnancy. EBV-CTL activity was determined by a modified EBV induced B-cell focus regression assay and was expressed in terms of a regression index (IR50), i.e. the initial cell concentration required to achieve a 50%-incidence of regression in EBV-infected cell culture. Increased values of IR50 indicate the suppression of EBV-CTL activity. In 113 human female T-cell leukemia type-I (HTLV-I) non-carriers, the IR50 values (mean +/- S.E.) in non-pregnant, pregnant (the first trimester, second trimester and third trimester of pregnancy) and puerperal women were 10.6 +/- 1.4, 16.1 +/- 1.1 (20.1 +/- 2.0, 14.8 +/- 2.0, 14.6 +/- 1.6), and 12.1 +/- 1.9 respectively. Among HTLV-I carriers, the IR50 values (mean +/- S.E.) were likewise 34.6 +/- 8.0, 87.4 +/- 5.2 (101.7 +/- 6.3, 88.3 +/- 8.4 and 79.5 +/- 9.2) and 39.2 +/- 7.1 respectively. This data demonstrate: 1) EBV-CTL activity was suppressed during pregnancy (P < 0.05), especially in the first trimester (P = 0.0003). 2). In HTLV-I carriers, this suppression was shown in the first trimester (P = 0.0002), in the second trimester (P = 0.0002) and in the third trimester of pregnancy (P = 0.0014) and 3). One month after delivery, this suppression had returned to the non-pregnant level in both HTLV-I non-carriers and HTLV-I carriers. Pregnancy therefore has a suppressive effect on antigen specific cytotoxic T-lymphocyte activity and this effect is amplified in HTLV-I carriers.