Recent studies have shown that immunization with Streptococcus bovis using Freund's complete adjuvant (FCA) may confer protection against lactic acidosis in sheep. The major objective of this study was to compare the antibody responses to S. bovis in a practically acceptable adjuvant (Freund's incomplete adjuvant (FIA); QuilA; dextran sulphate (Dex); Imject Alum; or Gerbu) and in FCA. Thirty-five sheep were randomly allocated to 7 treatment groups. Six groups were immunized with S. bovis in an adjuvant; the other group served as the non-immunization control. The primary immunization was administered intramuscularly on day 0. followed by a booster injection on day 28. Immunization with FCA induced the highest saliva and serum antibody responses. The saliva antibody concentrations in the FIA and QuilA groups were significantly higher than those in the Alum, Dex and Gerbu groups (p < 0.01). The serum antibody concentration in the FIA group was significantly higher than those in the QuilA, Alum. Dex and Gerbu groups (p < 0.01). Immunization enhanced the antibody level in faeces (p < 0.05), but there was no significant difference between the different adjuvant groups (p > 0.05). Seven and 14 days following booster immunization, the saliva antibody levels induced by QuilA and/or FIA were comparable with the level stimulated by FCA (p > 0.05). There was a strongly positive correlation (R2 = 0.770, p < 0.01) between the antibody concentrations in salival and serum. Compared with the controls, a higher faecal dry matter content was observed in the animals immunized with either FCA or QuilA. The change in faecal dry matter content was positively associated with the faecal antibody concentration (R2 = 0.441, p < 0.05). These results indicate that FIA and QuilA were effective at inducing high levels of antibody responses to S. bovis, and suggest that either Freund's incomplete adjuvant or QuilA may be useful for preparing a practically acceptable vaccine against lactic acidosis.