The studies associating infections by herpes simplex virus type 2 (HSV-2) with carcinoma of the human uterine cervix are reviewed within the context of three possible interpretations. Extensive seroepidemiologic evidence indicates that the virus does not grow preferentially in neoplastic tissue, nor is the association of HSV-2 with cervical carcinoma a reflection of their independent relationship to promiscuity. While a number of infectious agents, including other viruses, are associated with cervical atypia, only HSV-2 is a significant risk factor for CIS. In vitro transformation data supporting the oncogenic potential of the virus are summarized, and evidence is presented that an antigen designated ICP 10/AG-4 is a valid candidate for the role of a virus-encoded protein involved in the maintenance of a transformed phenotype. Antibody to AG-4 is IgM, and it is detected by microquantitative complement fixation. With this assay, it is demonstrated that conversion to anti-AG-4 occurs during primary infection with HSV-2; however, it is transient. In testing 1325 patients, a correlation was observed between antibody to AG-4 and cervical carcinoma. Thus, whereas only 11.7% of controls and 7.7% of patients with cancer at other sites are AG-4 seropositive, as many as 49.6% of patients with dysplasia, 63% of those with CIS, and 72.7% of those with invasive cancer are positive for the antibody. Antibody to AG-4 is related to tumor growth. This is evidence by 1) retrospective analyses demonstrating that the proportion of AG-4 seropositive individuals is directly correlated to the state of the disease and 2) prospective study of 209 patients demonstrating loss of antibody in patients with a successfully removed tumor mass and reappearance of AG-4 antibody in cancer recurrence. The possible use of AG-4 (or its antibody) in the diagnosis and monitoring of cervical carcinoma and its treatment is discussed.