In order to asses accuracy and consistency of diagnostic opinion in suspected pancreatic disease, using available diagnostic procedures 74 patients, subjected to selective angiography, pancreatic function test (Lundh test), and scintigraphy, were studied retrospectively. The patients were divided into four diagnostic subgroups, three of which represented defined pancreatic disease, the fourth serving as a control group since such disease could be excluded. The primary diagnostic results were compared. After a stipulated minimum of six months the angiographic films and scintigraphic recordings were re-appraised in a randomized manner without access to clinical data. The evaluations thus achieved were compared to those delivered primarily. The results of the present study seem to indicate that the Lundh test and pancreatic scintigraphy are screening tests with a high level of diagnostic reliability and consistency; when normal, practically ruling out pancreatic disease. Angiography is recommendable only if one or both of the above mentioned screening methods yield pathologic results, supported by clinical evidence. The principal aims of selective angiography in suscepted pancreatic disease, are to map vascular anatomy, to differentiate non-neoplastic disease from cancer and to assist estimation of operability. The results of the reappraisal indicate that the diagnostic accuracy of angiography, more than that of scintigraphy, is dependent on the experience of the examiner.