The reliability of cytology, colposcopy, and directed biopsy were assessed in pregnancy and compared with that observed in a matched non-pregnant control group in order to evaluate the influence of pregnancy. One hundred and seventeen gravidas in which the final pathology could be ascertained and 234 control patients were examined by the same colposcopist. Concordance between cytology and final diagnosis was complete or within one degree of severity in 55.6% and 77.8% of patients, respectively. Unsatisfactory colposcopy was significantly less frequent in pregnant patients than in controls (12.8% vs. 23.1%, P = 0.023). In pregnant patients, colposcopy provided concordance, overestimation and underestimation of the final diagnosis in 72.6%, 17.6% and 9.8%, respectively. In the same group, the concordance between the histologic findings of directed biopsy and final diagnosis was complete or within one degree of severity in 83.7% and 95.9%, respectively. The reliability of cytology, colposcopy and directed biopsy was not related to pregnancy. These data show that the physiological changes which occur in pregnancy do not significantly alter the reliability of colposcopy and directed biopsy, if the colposcopist is aware of the peculiar difficulties and does not overreact to the accentuated patterns that may occur during pregnancy.