Alveolar hypoxia (FIO₂ 0.10) rapidly produces inflammation in rat muscle, brain, and mesentery microcirculations. Dissociation between tissue PO₂ values and inflammation, plus the observation that plasma from hypoxic rats activates mast cells and elicits inflammation in normoxic tissues, suggest that the response to hypoxia is initiated via mast cell activation by an agent released from a distant site and carried by the circulation. These experiments tested the hypothesis that this agent originates in alveolar macrophages (AM). Male rats were depleted of AM by tracheal instillation of clodronate-containing liposomes. Four days after treatment, AM recovered by bronchoalveolar lavage were less than 10% of control. Controls received buffer-containing liposomes. Alveolar hypoxia (FIO₂ 0.10) in controls rats increased leukocyte-endothelial adherence, produced degranulation of perivascular mast cells, and increased fluorescent albumin extravasation in the cremaster microcirculation. None of these effects were seen in AM-depleted rats exposed to hypoxia. Plasma obtained from control rats after 5 min of breathing 10% O₂ elicited inflammation when applied to normoxic cremasters. In contrast, normoxic cremasters did not develop inflammation after application of plasma from hypoxic AM-depleted rats. Supernatant from AM cultured in 10% O₂ produced increased leukocyte-endothelial adherence, vasoconstriction, and albumin extravasation when applied to normoxic cremasters. Normoxic AM supernatant did not produce any of these responses. The effects of hypoxic supernatant were attenuated by pretreatment of the cremaster with the mast cell stabilizer cromolyn. These data support the hypothesis that AM are the source of the agent that initiates hypoxia-induced systemic inflammation by activating mast cells.