It is proposed that the immune/inflammatory system plays a yet unrecognized role in the mechanics of prostaglandin (PG) F2 alpha-induced luteal regression. Eosinophils are specifically attracted into luteal tissues and activated to degranulate (i.e. secrete cytotoxins) before symptoms of luteolysis are manifested (sheep). Further, because eosinophils are often associated with tissue reactions involving antigen-antibody binding, it is hypothesized that a luteal cell antigen could be expressed/unmasked as a result of the action of PGF2 alpha. Identification of the antigen by an appropriate autoantibody (e.g. complement-fixing) is an alternative mode by which cellular destruction can be mediated. Sialic acid residues that coat the surface of luteal membranes might act as a protective agent to autoimmune recognition. The hypothesis that luteolysis comprises an autoimmune reaction is extended to indicate that rescue of the corpus luteum from regression during early pregnancy involves a local immunosuppressive mechanism.