The safety and efficacy of a remotely delivered multivalent Pasteurella haemolytica supernatant vaccine (serotypes A2 and T10) were examined in captive Rocky Mountain bighorn sheep (Ovis canadensis canadensis). Twenty bighorn sheep were grouped according to baseline leukotoxin neutralizing antibody titers (≤2 or .2 log₂⁻¹) and vaccination history (previously vaccinated or unvaccinated). Within these groups, animals were randomly assigned to one of two delivery treatments: hand injection (control) or biobullet implantation. All bighorns received a single dose from the same lot of vaccine (n = 10/treatment); four additional animals were injected intramuscularly with 0.9% saline as unvaccinated sentinels. Mild, transient lameness one day after hand injection or biobullet implantation was the only adverse effect. Serum neutralizing antibody titers to P. haemolytica leukotoxin differed between delivery treatments (P = 0.009) and among baseline titer/vaccination history groups (P = 0.013). Neutralizing titers were higher among hand-injected bighorns. Although neutralizing titers were lower among implanted bighorns than hand-injected controls at 1 wk (P = 0.002) and 2 wk (P = 0.021) after vaccination, seroconversion rates in response to implantation (6/10) and hand injection (9/10) did not differ (P = 0.303). Agglutinating antibody titers to T10 were high and did not vary over time or between delivery treatments. Agglutinating antibody titers to A2 in the hand-injected controls were not different (P ≥ 0.07) than those in bighorns vaccinated with biobullet implantation. These data demonstrate that although hand injection elicits higher absolute titers, biobullet implantation may also stimulate effective antibody responses to P. haemolytica supernatant vaccine. Further evaluation of biobullet vaccination against pneumonic pasteurellosis in free-ranging populations of wild bighorn sheep is warranted.