It has been assumed in the past that pancreatic acinar cells represent an irreversible end stage in development. Consequently, when there was an increase in structures that had the morphology of ductules, the interpretation was that they were derived from the proliferation of stem cells and/or pre-existing ductular cells. Pancreatitis, however, is regressive in nature [Bockman (1984) In: Pancreatitis: Concepts and Classification. Gyr, K.E., Singer, M.V., Sarles, H., eds. Elsevier, Amsterdam, pp. 11-15]. That is, it is characterized by parenchymal destruction and loss, rather than by expansion of parenchyma. Furthermore, it was assumed that the organization of the pancreatic parenchyma is like bunches of grapes, with spheroidal acini representing the grapes, and the ductules representing the stems. Given this organization, it would be difficult to understand how regressive changes could lead to clusters of ductular structures. Investigations using three-dimensional reconstruction and retrograde injections have altered our idea of pancreatic organization. In addition to spheroidal acini, there also are other shapes, including tubular acini. Moreover, ductules do not necessarily stop when they encounter an acinus. They may emerge on the other side. Combined ductular and acinar lumina may anastomose with each other. It is now clear that pancreatic acini may undergo redifferentiation, taking on the morphology of ductules and forming tubular complexes during pancreatitis, as well as in response to pancreatic cancer, cystic fibrosis, or blockage of the ductal system. With this understanding of pancreatic architecture and morphological plasticity, it is easier to understand the changes one sees with pancreatic diseases.