LondonSUMMARY A technique has been developed which allows the position of the upper gastrointestinal tract, as determined radiographically, to be superimposed on the pancreatic photoscan, performed after an injection of 75Se-selenomethionine. Results of the technique applied to 27 'abnormal' and 23 'normal' subjects are described. The importance of the additional information gained by the combined technique is discussed in terms of normal variation and increased diagnostic yield, and the situations in which it was found most useful are described. The addition of correlated duodenography rendered scan reporting more objective in 50 % of cases.Comparisons of the usefulness of radiological and other techniques in solving specific diagnostic problems in pancreatic disease have been undertaken by a number of workers (Sullivan, Meaney, Rodriguez-Antunez, and : Kreel, 1967Eaton, Fleischli, Pollard, Nebesar, and Portsaid, 1968). The reports suggest that all the techniques have well defined 'blind spots'. Pancreatic scanning after injection of 75Se-selenomethionine (Rodriguez-Antu(nez, Filson, Sullivan, and Brown, 1966;Centi Colella and Pigorini, 1967;Brown, Sircus, Smith, Donaldson, Dymock, Falconer, and Small, 1968;Melmed, Agnew, and Bouchier, 1968) and hypotonic duodenography (Raia and Kreel, 1966;Bilbao, Frische, Dotter, and Rosch, 1967;Kreel, 1969;Martel, 1968) have proved to be simple, safe procedures particularly when 'tubeless' hypotonic duodenography is employed. Our experience is similar to that of Eaton et al (1968) and suggests that these are generally the most useful techniques employed in the diagnosis of pancreatitis and pancreatic carcinoma, though distinction between the two conditions is often difficult.A technique, described below, has been developed which allows the position of the duodenal loop, as determined by duodenography, to be superimposed on the photoscan. Using this technique a study was undertaken aimed primarily at answering the question, 'Does the addition of correlated hypotonic duodenography as a routine procedure in patients subjected to pancreatic scanning yield additional information helpful to the interpretation of the scan?'The study also allowed close examination of the scan in difficult interpretive situations. These arise when glandular uptake of the isotope is poor and not greatly in excess of background activity; when the isotope appears in abnormal sites or concentrations; or when the liver or bowel overlaps the pancreas. Under such conditions it may be difficult if not impossible to delineate precise boundaries of the gland. For such reasons one-tenth of scans are unreadable and as many more are interpreted with little confidence. The technique was therefore employed in the hope of reducing this uncertainty and it permitted close study of isotope distribution in the context of known anatomical relations of the pancreas. As no similar studies had been previously reported, interpretation of the results of the study was carried out at a communal reporting session for the first 3...