Test doses (0.5 ml) of the perfluorochemical blood substitute Fluosol-DA causes pulmonary hypertension in some patients. To investigate this phenomenon we infused Fluosol-DA into unanesthetized sheep with chronic vascular catheters on ten occasions. In six of these experiments lung lymph flow was measured. Since other fluorochemical blood substitutes alter the coagulation cascade we also sampled blood to perform coagulation screening tests and look for evidence of thrombin generation as assessed by the survival of 125I labelled sheep fibrinogen. Test doses (0.5 ml) of Fluosol-DA resulted in transient but marked pulmonary hypertension in nine of ten experiments (17 +/- 1.2 SE to 43 +/- 3.9 SE torr). Cardiac output decreased by an average of 30%, and lung lymph flow transiently increased without a change in the lymph to plasma protein concentration (L/P) ratio in five of six experiments. When 50 ml of Fluosol-DA was administered one hour later the pulmonary hypertension was more prolonged but less severe (18 +/- 1.0 SE to 34 +/- 3.2 SE torr). Lymph flow again increased without a change in the L/P ratio for protein. The activated partial thromboplastin and prothrombin times, and survival in plasma of 125I labelled sheep fibrinogen were unaffected by administration of Fluosol-DA. We conclude that Fluosol-DA causes pulmonary hypertension in sheep and increases lung lymph flow. This results from vasoconstriction of the pulmonary vessels and subsequent recruitment of the pulmonary microvasculature and not from thromboembolic phenomena.