It is well established that glucocorticoids (GCs) are beneficial for the treatment of many inflammatory and allergic diseases, including asthma 1. , in which tissue infiltration by eosinophil leukocytes plays a major role in pathogenesis. Eosinophils are produced and stored in the bone marrow, to be subsequently released into the circulation and recruited, under the influence of chemotactic factors, into inflammatory sites, where they can contribute to tissue damage by secreting toxic proteins, inflammatory mediators and cytokines 2. . Many studies have shown that GCs inhibit the production of a subset of cytokines [interleukin 5 (IL-5), granulocyte-macrophage colony-stimulating factor (GM-CSF) and IL-3] that can contribute to eosinophilic inflammation by stimulating the production of eosinophils in the bone marrow (eosinopoiesis). These same cytokines increase the survival of mature eosinophils that are recruited to the tissues and promote their activation in a variety of ways. The ability to suppress synthesis of the eosinopoietic cytokines is thought to underlie the ability of GCs to reduce blood and tissue eosinophilia 2. , 3. . In addition, because GCs have been shown to inhibit many of the effects of these cytokines on mature eosinophils, it is often assumed that GCs will always antagonize the effects of IL-5, GM-CSF and IL-3, regardless of the maturation stage of the target cell. However, as we argue below, the relationship between GCs and eosinopoietic cytokines includes synergistic as well as antagonistic aspects, and might depend on the maturation stage of the target cells. In this context, recent evidence from our own 4. and other laboratories 5. that GCs can actually potentiate the eosinopoietic actions of GM-CSF and IL-5 in the bone marrow is likely to raise important questions. Here, we suggest that these findings have implications for therapy with GCs and discuss how these paradoxical findings relate to differences between eosinophils in bone marrow and other sites.