Surgery an essentially manual exercise in its early days, has progressively grown richer in complementary techniques helping the surgeon to perfect his movements and increase efficiency. This is particularly the case with digestive surgery, which has been thoroughly transformed by radiology, endoscopy and extemporaneous examinations effected during surgical intervention. Such methods make it possible for surgeons to develop subtle nuances in operative techniques and to specify indi cations more and more precisely. Ultrasonography must now be included among such techniques. It supplements them and can sometimes even replace them. For these reasons, this work devoted to the use of ultrasound by the surgeon during surgical intervention is of great interest. Two general conditions had to be met before ultrasonography could be em ployed in abdominal surgery: the method and apparatus had to be adapted to its surgical utilization, and the surgeon had to adapt to a nonsurgical technique. The first condition has been fulfilled for all purposes. Intraoperative sonograms can be generated at a very high rate and are proving to be more and more useful. The miniaturization of probes permits their application almost everywhere with opti mal results. Of course, the costs are still high, but one can expect them to decrease.

For a long time, approximately since Oberlin and Guerin described the multifocal origin of pancreatic cancers and precancerous pancreatic lesions, no important study dealing with the entire subject of pancreatic cancer has been published in France and probably in the international literature. For some decades the knowl- edge acquired 40years or more ago was not improved appreciably, though the fre- quency ofthe disease started to increase in occidental countries. This has recently changed, and the progress ofthe medical sciences has spread to the pancreas. Although the surgical or medical prognosis of the most frequent form of pancreatic cancer, exocrine adenocarcinoma, remains very bad, recent studies have shown the multiplicityofits pathological forms, some being less severe so that curative surgery is possible. New experimental models, particularly in the hamster, and the use of carcinogenic drugs allow experimental studies on lesions similar to those in man. Oncologic immunology is still at its beginnings but shows promise for diagnosis and treatment. Though modem techniques of imaging - sonography, aspirative cytology, CT scan, endoscopic catheterism, arteriography, and maybe in the future nuclear magnetic resonance - have not yet significantly in- fluenced prognosis,they have made the diagnosis easierand more precocious. Yet in a diseasethat diffuses so rapidly to deep lymph nodes, it has not been proved whether early diagnosis can improve prognosis.