Various procedures of vaccination for pseudorabies were compared for their effects on shedding, latency, and reactivation of attenuated and virulent pseudorabies virus. The study included 6 groups: group 1 (10 swine neither vaccinated nor challenge-exposed), group 2 (20 swine not vaccinated, but challenge-exposed), and groups 3 through 6 (10 swine/group, all vaccinated and challenge-exposed). Swine were vaccinated with killed virus IM (group 3) or intranasally (group 4), or with live virus IM (group 5) or intranasally (group 6). The chronologic order of treatments was as follows: vaccination (week 0), challenge of immunity by oronasal exposure to virulent virus (week 4), biopsy of tonsillar tissue (week 12), treatment with dexamethasone in an attempt to reactivate latent virus (week 15), and necropsy (week 21). Vaccination IM with killed or live virus and vaccination intranasally with live virus mitigated clinical signs and markedly reduced the magnitude and duration of virus shedding after challenge exposure. Abatement of signs and shedding was most pronounced for swine that had been vaccinated intranasally with live virus. All swine, except 4 from group 2 and 1 from group 4, survived challenge exposure. Only vaccination intranasally with live virus was effective in reducing the magnitude and duration of virus shedding after virus reactivation. Vaccination intranasally with killed virus was without measurable effect on immunity. Of the 55 swine that survived challenge exposure, 54 were shown subsequently to have latent infections by use of dexamethasone-induced virus reactivation, and 53 were shown to have latent infections by use of polymerase chain reaction (PCR) with trigeminal ganglia specimens collected at necropsy. Fewer swine were identified by PCR as having latent infections when other tissues were examined; 20 were identified by testing specimens of olfactory bulbs, 4 by testing tonsil specimens collected at necropsy, and 4 by testing tonsillar biopsy specimens. Eighteen of the 20 specimens of olfactory bulbs and 3 of the 4 tonsil specimens collected at necropsy in which virus was detected by PCR were from swine without detectable virus-neutralizing antibody at the time of challenge exposure. One pig that had been vaccinated intranasally with live virus shed vaccine virus from the nose and virulent virus from the pharynx concurrently after dexamethasone treatment. Evaluation of both viral populations for unique strain characteristics failed to provide evidence of virus recombination.