The ability to discriminate between herpes simplex virus types 1 and 2 (HSV-1 and HSV-2) infections by serological means is of increasing importance for clinical virological diagnostics. HSV type-specific serology has a number of applications, e.g. to guide duration and dosage of antiviral therapy, to allow for stringent epidemiological analyses, to evaluate efficacy of HSV vaccine candidates, and to help the clinician during counseling of couples where one has genital herpes. The genomes of the two HSV types appear to be sufficiently intratypically stable and intertypically discrepant to permit such a discrimination. Despite recent advances in methodology, the ideal HSV type-specific serological assays still remain to be developed. The viral antigens utilized for such an assay should evoke strong antibody responses, but only against unique determinants, in order to be able to combine high sensitivity and specificity. Of all the HSV envelope proteins, the most promising candidate antigens, i.e. HSV-1 glycoprotein G (gG-1) and its HSV-2 counterpart gG-2, contain relatively long type-unique stretches of amino acids. However, whether the type-specific determinants preferentially localize to the unique or homologous stretches of gG is still unknown. Paradoxically, type-specific monoclonal antibodies to the moderately type-specific antigen, HSV-1 glycoprotein C, have hitherto been mapped to homologous rather than to type-unique stretches. A definition of human type-specific epitopes on gG-1 and gG-2, as well as a broader search for new candidate HSV antigens, might be needed to fully discriminate dual infections from high titer single HSV-1 or HSV-2 infections.