Is avian humoral immunocompetence suppressed by testosterone?

A key issue in sexual selection theory is how a correlation between male se condary sexual characters and male genetic quality can be maintained. The i mmunocompetence-handicap hypothesis proposes that testosterone-dependent ma le characters remain honest signals because of the immunosuppressive effect of elevated steroid hormone levels. The hypothesis requires that physiolog ical levels of testosterone depress immune system function. We quantified t estosterone titers and humoral immunocompetence of captive male and :female red-winged blackbirds (Agelaius phoeniceus) at three points in the annual cycle (autumn, prebreeding, and breeding). We also conducted an implant exp eriment on the males to assess the effects of prolonged, above-normal testo sterone titers on humoral immune responses. Humoral immunocompetence was me asured as secondary antibody production to a non-pathogenic protein antigen , keyhole limpet hemocyanin, using an enzyme-linked immunosorbent assay we developed for A. phoeniceus. Secondary antibody responses of individuals we re highly repeatable between sampling periods. Neither physiological nor ab ove-normal levels of plasma testosterone suppressed secondary antibody prod uction. In paired tests of the same individuals between prebreeding and bre eding, and between breeding and implant, plasma testosterone increased sign ificantly but secondary antibody responses were unaffected. We are confiden t in these results because with 80% power, an 11-14% difference in antibody titers would have been detected. There was no relationship between plasma testosterone levels and humoral immunocompetence in free-ranging males test ed at the peak of breeding. These results cast doubt on a key assumption of the immunocompetence-handicap hypothesis.