The distinction between 'medical' and 'surgical' jaundice is often difficult due to the lack of specificity of biochemical liver function tests, and the difficulty in assessing hepatocellular function and biliary tract patency in the jaundiced patient. We present a noninvasive isotopic technique which gives reliable information on these parameters, resulting in a high degree of accuracy in the classification of jaundice. 131I Rose Bengal, which is handled by the liver like bromsulphthalein, is administered intravenously and count rates during hepatic uptake and biliary excretion are recorded over the liver, biliary tract and duodenum using a gamma camera/computer system, into which is built an electronic light pen, so that particular areas of interest can be studied. Blood clearance of the iostope is simultaneously measured, and from these data the T 1/2 of uptake, the plasma retention and the excretion quotient are determined. The results of the first 80 examinations show that by qualitative assessment of the uptake and excretory phases of the 'hepatogram' and by determination of the T 1/2 and excretion quotient, jaundice has been classified with an 86 per cent accuracy in cases where the diagnosis is not apparent from clinical and biochemical data. The T 1/2 of uptake correlates well with bromsulphthalein retention in the non-jaundiced patient, and when applied to patients with obstructive jaundice enables the progress of hepatocellular impairment with time to be studied, and its recovery after relief of the obstruction. Preliminary results suggest that the technique may also have applications in the study of biliary tract dynamics in patients with post-cholecystectomy syndromes and relapsing acute pancreatitis.