The immune system and several endocrine axes communicate with each other through a network of molecules which collectively produce a coordinated response to immune challenges. This phenomenon, necessary for the survival of the organism, is thought to involve the release, by activated cells in the periphery, of proteins, called cytokines, which inform the brain about immune activation. The brain then organizes a series of neuroendocrine responses which participate in the regulation of the host response. With regard to the influence of cytokines on the hypothalamic-pituitary-gonadal axis, we know that the injection of these proteins lowers gonadotropin-releasing hormone release, which in turn inhibits luteinizing hormone (LH) secretion. These changes would be expected to decrease sex steroid production and, indeed, estrogens and testosterone are low in female and male rats, respectively, following acute intracerebroventricular (i.c.v.) injection of interleukin (IL)-1 beta. There is, however, another possibility that central cytokines could alter ovarian and testicular function independently of changes in gonadotropin levels. Prolonged i.c.v. infusion of the cytokine into the female rat brain produced a dramatic rise in progesterone levels. The absence of a comparable change in the progesterone release rate of males infused with IL-1 beta, and the presence of marked surges of prolactin (PRL) in the females, suggests that IL-1 beta altered ovarian function, and that the persistence of large corpora lutea induced PRL release. The possibility that the cytokine might stimulate the brain circuits that regulate PRL release, while possible, appears remote, because male rats injected with IL-1 beta showed significantly blunted PRL levels. In intact adult male rats, i.c.v. IL-1 beta administration caused the expected decrease in LH and testosterone levels, but was also accompanied by a loss of testicular responsiveness to gonadotropins. Though elevated levels of corticosteroids are known to interfere with normal gonadal steroidogenesis, blockade of IL-1-induced corticosterone release did not reverse the inhibitory influence of the cytokine. One mechanism that deserves attention is the possibility that i.c.v. injection of IL-1 beta might increase circulating cytokine levels, and indeed plasma IL-6 concentrations were significantly elevated in rats treated with IL-1 beta. This humoral mechanism may disrupt testicular function through the documented inhibitory effects of blood-borne cytokines on Leydig cell function. In addition, brain cytokines might influence a variety of peripheral events through direct (neural?) connections. This brief review discusses the hypothesis that there are brain-to-gonad connections that bypass the pituitary, and presents results that might support the possibility that central injection of IL-1 beta decreases testosterone secretion independently of blunted LH levels.