Eight puppies (group 1) were vaccinated once with a bivalent modified-live vaccine against infectious tracheobronchitis by the intranasal route and at the same time with an injectable trivalent vaccine against canine parvovirus, canine distemper virus and canine adenovirus; a second group of eight puppies (group 2) was vaccinated only with the intranasal bivalent vaccine, and a further eight puppies (group 3) were vaccinated only with the injectable trivalent vaccine. Three weeks later they were all challenged with wildtype Bordetella bronchiseptica and canine parainfluenza virus by the aerosol route, and their antibody responses to the five vaccine organisms were determined. Oronasal swabs were taken regularly before and after the challenge for the isolation of bacteria and viruses, and the puppies were observed for clinical signs for three weeks after the challenge. There were no significant differences in the puppies' titres against canine parvovirus, canine distemper virus and canine adenovirus type 2 between the groups vaccinated with or without the bivalent intranasal vaccine. After the challenge the mean clinical scores of the two groups vaccinated with the intranasal vaccine were nearly 90 per cent lower (P=0.001) than the mean score of the group vaccinated with only the trivalent injectable vaccine, and the puppies in this group all became culture-positive for B bronchiseptica and canine parainfluenza virus. There were only small differences between the rates of isolation of B bronchiseptica from groups 1, 2 and 3, but significantly lower yields of canine parainfluenza virus were isolated from groups 1 and 2 than from group 3.