At least two exogenous sources of agents able to control vascular smooth muscle proliferation can be identified. Platelets contain and release mitogens as well as a factor, TGF-beta, that inhibits cell growth on plastic surfaces while stimulating it when cells are grown in suspension in soft agar. Macrophages release mitogens, including PGDF, and macrophage invasion is characteristic of early experimental lesions in fat-fed animals. Finally, it is at least possible that endothelial cell production of mitogens may represent a response to some as yet undefined external injury. The vessel wall also offers sources of growth control endogenous to the smooth muscle cell layers. The vessel wall contains heparan sulfate able to inhibit cell growth of smooth muscle cells, which by themselves can synthesize PDGF. This provides possible positive and negative control of replication intrinsic to the smooth muscle cells themselves. The role of these intrinsic or extrinsic factors in the smooth muscle proliferation of hypertension and atherosclerosis remains hypothetical. It is intriguing to implicate platelets and/or macrophages in the denuding injuries seen in small hypertensive vessels and in advancing atherosclerotic plaques. At least for the latter case, however, there seem to be other critical factors. Simple denudation and thrombosis, for example, are not sufficient to stimulate smooth muscle growth, and the kinetics of proliferation after balloon denudation imply the presence of some other event required to initiate smooth muscle proliferation. Similarly, smooth muscle replication in large vessels of hypertensive animals occurs without loss of endothelial continuity. This implies that replication in response to hypertension depends on factors intrinsic to the vessel wall. Benditt's observation of monoclonality also implies some intrinsic mechanism allowing cells to grow in a focal manner. It is intriguing to consider the possibility that this commitment process could require the release of cells from the intrinsic inhibitory effects of heparan sulfate located around the cells or the synthesis of growth factors secreted by the smooth muscle cells themselves. If we add the hypothesis that only some cells are capable of such a response, we would expect the sort of oligodense phenomenon demonstrated by Benditt. Proof of such a hypothesis, however, will have to await development of methods to explore these mechanisms directly in the vessel wall responding to injury.