A study was undertaken in a group of patients undergoing 199 thoracic surgical procedures to assess the value of certain preoperative cardiopulmonary function tests in predicting fatal postoperative complications. Despite the limited number of tests studied, each of 28 patients who developed fatal complications after surgery had evidence of pre-existing disease involving the cardiovascular or respiratory system. Surgery in patients with normal test results was associated with negligible risk. The preoperative tests employed in this study were of unequal value in predicting the risk of surgery. The results of the electrocardiographic and the maximal breathing capacity tests, readily available to any physician, were of greatest value in predicting the outcome after surgery. Test results indicating the presence of significant chronic obstructive emphysema were most frequently associated with fatal postoperative complications. By serial application of two or more tests, groups of patients with high risk could be defined. The value of such multiple tests in predicting operative risk was greater than any single test. Althpugh it was possible to identify all cardiorespiratory fatalities on the basis of the preoperative tests, there was no way to differentiate between the 28 patients who died and 69 patients with the same findings who did not die. Further study may make such differentiation possible. Currently, preoperative tests indicate the presence of cardiac or pulmonary disease and given an estimate of the risk of surgery in the face of such disease.