Syncope is a common medical problem and is caused by a wide variety of diseases ranging from physiologic derangements with few consequences to conditions that may be immediately life-threatening. Because of the large differential diagnosis, many diagnostic tests are available for its evaluation. However, a cause of syncope is not established in 38% to 47% of patients despite these tests. In those patients in whom a diagnosis can be assigned, the history and physical examination identify a potential cause in 49% to 85%. Furthermore, in 8% of additional patients, history and physical examination are suggestive of causes that need confirmation by specific tests. Routine blood tests rarely yield diagnostically helpful information. In those patients in whom a potential cause for syncope is identified, arrhythmias are diagnosed by electrocardiogram in 2% to 11% of patients, cardiac monitoring in 3% to 27% (telemetry or Holter), stress test in less than 1%, carotid massage in less than 1%, and electrophysiologic studies in less than 3%. Diagnosis of arrhythmias as a cause of syncope is problematic because symptomatic correlation during electrocardiographic monitoring is rarely found (approximately 4%), and as a result, there is no uniform agreement on diagnostic criteria for abnormalities. Similar problems exist in the use of electrophysiologic studies. Upright tilt testing and psychiatric examination may be useful in evaluation of recurrent syncope of unknown cause in patients without organic heart disease. Based on the results of recent studies, strategies for evaluation of patients with syncope are possible that utilize selective and goal-directed diagnostic testing.