Much has been learned in the last decade about the epidemiology of genital herpes infections, including new information about seroprevalence and the risk of transmission of genital herpes to sex partners and at delivery. Unfortunately, the type-specific serologic assays now routinely used in these studies are not widely available, and commercially available assays that claim to be type-specific are not. Thus, most clinicians still do not have access to reliable type-specific assays. In cross-sectional seroprevalence studies, detection of HSV-2 antibody is positively associated with increasing age, lower levels of income or education, increased numbers of sexual partners, black or Hispanic race, female gender, male homosexual activity, and HIV infection. In addition, studies cited in this review have clarified the clinical spectrum of genital herpes infection in persons who have transmitted genital herpes to a sex partner, have shed virus asymptomatically, or are found to have HSV-2 antibody. Ten percent to 40% of these individuals are aware that they have genital herpes, whereas the remaining 60% to 90% are not. Among the latter, at least half have a history of recurrent genital lesions typical of genital herpes or can be taught to recognize typical, symptomatic episodes within 6 months if examined promptly after the onset of any unexplained genital symptoms. The remainder, about a third of the total, have no history of genital herpes and remain asymptomatic despite a careful history and follow-up examinations; in women in this group, asymptomatic shedding of HSV can be identified. Atypical lesions appear to play an important but as yet incompletely defined role. Most persons who transmit genital herpes to a sex partner or at delivery do not have a history of lesions at the time of transmission of HSV infection, suggesting that asymptomatic shedding or atypical, unrecognized lesions are responsible for most cases of transmission. In heterosexual couples, the risk of acquisition of HSV-2 infection from a sex partner with genital herpes is lowest in men (less than 5%), higher in HSV-1 seropositive women (less than 10%), and highest (about 30%) in women without antibody to HSV-1 or HSV-2. The risk of transmission to infants exposed to asymptomatic shedding at delivery is low (about 3%) in women with or without a history of genital herpes if HSV antibody of the same type is present in cord blood.(ABSTRACT TRUNCATED AT 400 WORDS)