Pneumocystis carinii infections were developed in animals immunosuppressed by dexamethasone treatment either from activation of latent infection (ferret) or by transtracheal inoculation with P. carinii-infected lung tissue from the homologous species (rat or mouse). Convalescent-phase antisera were obtained by stopping dexamethasone treatment after 2 to 4 weeks and allowing animals 5 to 8 weeks for recovery. P. carinii harvests from infected lungs were purified by differential filtration, solubilized in buffer containing urea, sodium dodecyl sulfate (SDS), and 2-mercaptoethanol, subjected to SDS-polyacrylamide gel electrophoresis, and blotted to polyvinylidene difluoride sheets for Western immunoblot analysis. These lung preparations are hereafter referred to as P. carinii antigens. Convalescent-phase antisera from each animal species were reacted on Western blots of P. carinii antigens prepared from organisms isolated from rats, ferrets, or mice. Each combination of P. carinii antigens and antisera from the same species of animal reacted with three or more P. carinii antigen proteins. Convalescent-phase mouse antisera reacted with P. carinii antigens from mice but not rats or ferrets. Convalescent-phase rat antisera reacted with P. carinii antigens from rats and mice but not ferrets, and convalescent-phase ferret antisera showed reactions with ferret and mouse P. carinii antigens but not rat antigens. These findings indicate antigenic differences among P. carinii strains infecting these animals. Pneumocystis carinii is an opportunistic organism thought to be ubiquitous in the environment. P. carinii infects many animal species besides humans, including rats, ferrets, and mice. Several lines of evidence suggest P. cannii strains infecting each species are different. Intraspecies P. carinii strain variations may also exist (6). Many investigators have studied the antigenic characteristics of P. carinii by using hyperimmune antisera or monoclonal antibodies with various immunologic techniques. Gigliotti et al. (9) determined that P. carinii isolates from different host species are not antigenically identical by using monoclonal antibodies that were reactive to rat P. cannii. Kovacs et al. (11, 12) also used monoclonal antibodies to rat and human P. carinii to demonstrate antigenic differences among P. carinii isolates. Linke et al. (14) analyzed P. cannii antigens for their susceptibilities to treatment with enzymes by immunoblotting and determined that rat and human P. cannii have both shared and species-specific antigenic determinants. Roth and Janitschke (15) used immunofluorescence to demonstrate differences among species with P. carinii by using hyperimmune sera. Convalescent-phase or infection-derived antisera developed in both rats and mice (5, 21, 22) have made it possible to compare naturally occurring antibodies among different species of animals infected with P. carinii (3). Walzer and Linke (19) showed that rat and human P. carinii isolates have both shared and unique antigenic determinants by