Captive great egret (Ardea albus) nestlings were maintained as controls or were dosed with methylmercury chloride at low (0.5), and high doses (5 mg/kg, wet weight) in fish. Low dosed birds were given methylmercury at concentrations comparable to current exposure of wild birds in the Everglades (Florida, USA). When compared with controls, low dosed birds had lower packed cell volumes, dingy feathers, increased lymphocytic cuffing in a skin test, increased bone marrow cellularity, decreased bursal wall thickness, decreased thymic lobule size, fewer lymphoid aggregates in lung, increased perivascular edema in lung, and decreased phagocytized carbon in lung. High dosed birds became severely ataxic and had severe hematologic, neurologic, and histologic changes. The most severe lesions were in immune and nervous system tissues. By comparing responses in captive and wild birds, we found that sublethal effects of mercury were detected at lower levels in captive than in wild birds, probably due to the reduced sources of variation characteristic of the highly controlled laboratory study. Conversely, thresholds for more severe changes (death, disease) occurred at lower concentrations in wild birds than in captive birds, probably because wild birds were exposed to multiple stressors. Thus caution should be used in applying lowest observed effect levels between captive and wild studies.