Vaccination of experimental animals can provide efficient protection against ocular herpes simplex virus type 1 (HSV-1) challenge. Although it is suspected that local immune responses are important in protection against ocular HSV-1 infection, no definitive studies have been done to determine if local ocular vaccination would produce more efficacious protection against HSV-1 ocular challenge than systemic vaccination. To address this question, we vaccinated groups of rabbits either systemically or periocularly with recombinant HSV-2 glycoproteins B (gB2) and D (gD2) in MF59 emulsion or with live KOS (a nonneurovirulent strain of HSV-1). Three weeks after the final vaccination, all eyes were challenged with McKrae (a virulent, eye disease-producing strain of HSV-1). Systemic vaccination with either HSV-1 KOS or gB2/gD2 in MF59 did not provide significant protection against any of the four eye disease parameters measured (conjunctivitis, iritis, epithelial keratitis, and corneal clouding). In contrast, periocular vaccination with gB2/gD2 in MF59 provided significant protection against conjunctivitis and iritis, while ocular vaccination with live HSV-1 KOS provided significant protection against all four parameters. Thus, local ocular vaccination provided better protection than systemic vaccination against eye disease following ocular HSV-1 infection. Since local vaccination should produce a stronger local immune response than systemic vaccination, these results suggest that the local ocular immune response is very important in protecting against eye disease due to primary HSV-1 infection. Thus, for clinical protection against primary HSV-1-induced corneal disease, a local ocular vaccine may prove more effective than systemic vaccination.