To investigate whether the efficacy of live vaccines is influenced by the mode of vaccine delivery, a widely-used UK live commercial Salmonella Enteritidis vaccine was delivered to pullet chicks either by spray, in drinking water, or in combination with a bivalent vaccine containing inactivated Salmonella Enteritidis and Salmonella Typhimurium. The birds were subsequently challenged with 10(2) or 10(8) colony-forming units (cfu) of Salmonella Enteritidis through drinking water at either six or 20 weeks of age. Ten days after the challenge, the birds were euthanased and their caecal contents cultured for Salmonella. All of the vaccinated groups contained fewer Salmonella Enteritidis-positive birds than the unvaccinated groups. The 'spray-vaccinated' group contained significantly fewer Salmonella Enteritidis-positive birds than the 'water-vaccinated' group after challenge with 10(8) cfu at 20 weeks. However, there was little or no difference at the other challenge time points between the groups that received vaccine through different modes of delivery.