This paper describes the role of venous ultrasonography in the diagnosis of suspected deep venous thrombosis and pulmonary embolism. Inability to compress the common femoral or popliteal vein is usually diagnostic of a first episode of deep venous thrombosis in symptomatic patients (positive predictive value of about 97%). Full compressibility of both of these sites excludes proximal deep venous thrombosis in symptomatic patients (negative predictive value of about 98%). In patients with suspected deep venous thrombosis or in those who present with suspected pulmonary embolism but have a nondiagnostic lung scan, the subsequent risk for symptomatic venous thromboembolism is very low (<2% during 6 months of follow-up) provided that ultrasonography of the proximal veins remains normal in the course of 1 week (suspected deep venous thrombosis) or 2 weeks (suspected pulmonary embolism). Anticoagulation and further diagnostic testing can usually be safely withheld in these situations. Venous ultrasonography is much less reliable for the diagnosis of asymptomatic, isolated distal, and recurrent deep venous thrombosis than for the diagnosis of a first episode of proximal deep venous thrombosis in symptomatic patients. Clinical evaluation of the probability of deep venous thrombosis or pulmonary embolism, preferably by using a validated clinical model, complements venous ultrasonographic findings and helps to identify patients who would benefit from additional (often invasive) diagnostic testing. Thus, venous ultrasonography is thought to be a very valuable test for the diagnosis and management of patients with suspected deep venous thrombosis or pulmonary embolism.