Aerosol vaccination is used effectively to immunize poultry against Newcastle disease, but to the authors' knowledge, this vaccination procedure is not well studied in other species. The efficacy of IM and aerosol vaccination of pigs against Mycoplasma hyopneumoniae infection was evaluated. Twenty-one pigs from a Mycoplasma-free herd were randomly allotted by litter and body weight into 3 groups. One group was given aerosolized phosphate-buffered saline solution (PBSS) by inhalation. The second group (AERO) was given aerosolized M hyopneumoniae vaccine by inhalation. The third group (IM) was given the same vaccine by IM injection. Vaccination by IM administration was repeated once, and aerosol vaccination was repeated twice at 2-week intervals. Two weeks after the last vaccination, all pigs were intratracheally challenge-exposed with 3 ml of broth culture containing 10(7) color-changing units (CCU) of a low-passage strain of virulent M hyopneumoniae. Pigs were observed daily for coughing. Four weeks after challenge exposure, all pigs were necropsied. Percentage of lung affected by gross pneumonia was measured, bronchioalveolar lavage fluid (BALF) cells were counted, and quantitative culture for mycoplasmas was performed on lung sections. Additionally, M hyopneumoniae-specific antibodies were measured in prevaccination, postvaccination, and postchallenge-exposure serum and BALF by use of indirect ELISA. Mean prevalence of persistent coughing in pigs of the AERO group (4.6 d/pig) was not different from that in pigs of the PBSS group (3.7 d/pig). Prevalence of coughing in IM vaccinated pigs (1.0 d/ pig) was lower (P < 0.05) than that in pigs of the PBSS group. Mean gross lung lesion scores and BALF cell counts were not different between the AERO (15% pneumonia, 5,233 cells/microliter) and PBSS (11% pneumonia, 3,022 cells/microliter) groups, but were lower (P < 0.05) in the IM group (1.5% pneumonia, 400 cells/microliter) than in the PBSS group. Mean lung mycoplasmal counts were not significantly (P < 0.05) different among the PBSS (10(5.6) CCU/g), AERO (10(5.3) CCU/g), and IM (10(3.3) CCU/g) groups. Postvaccination M hyopneumoniae-specific IgG or IgA was not detectable in BALF after either vaccination procedure. Postvaccination M hyopneumoniae-specific serum IgG concentration was not different among the 3 groups. Postchallenge exposure M hyopneumoniae-specific IgG and IgA were detectable in BALF of all pigs, but were not different among the 3 treatment groups. Postchallenge exposure-specific serum IgG concentration was not different between the PBSS (mean OD, 0.739) and AERO (mean OD, 0.672) groups, but was higher (P < 0.05) in the IM group (mean OD, 1.185) than in the PBSS group. Aerosol vaccination failed to induce local and systemic antibody responses detectable by ELISA, and failed to protect pigs against mycoplasmal pneumonia. Intramuscular vaccination failed to induce local and systemic antibody responses detectable by ELISA, but substantially reduced the clinical signs and lesions caused by challenge exposure to virulent M hyopneumoniae.