TUBERCULOUS pericarditis, chronic, insidiously progressive, frequently characterized in its early stages by an inflammatory effusion and culminating almost inevitably in great thickening and fibrous contraction of the diseased pericardium, presents real difficulties in its management.1 From its inception, it poses the twofold problem of controlling infection and of correcting the mechanical interference with the flow of blood imposed by constriction of the heart or its great vessels, most particularly the venae cavae, the compression of which is believed to be the primary reason for the persistent and recurring pleural and peritoneal effusions. The gravity of the disease is illustrated by the study of Harvey and Whitehill2: "Of 20 proven cases with effusion 16 died and of 17 proven cases without an effusion 15 died, a mortality of 83 per cent." Blalock and Levy,3 after a study of 22 proved cases in 1937, reported: "Every patient, with one