Most articles about restenosis proceed from the assumption that luminal narrowing in atherosclerosis and restenosis occurs as a result of migration of medial cells across the internal elastic lamina and proliferation of smooth muscle cells in the pathologic intima (1^ 4). To a large extent, this view is derived from two sources of conventional wisdom. First, for the last 20 years, the most popular concept of vascular response to injury has been that put forward by Ross (5, 6). In brief, this model emphasizes smooth muscle proliferation as a dominant component of atherosclerosis, both in terms oflesion formation and progression. It seems only reasonable to imagine that the same process (es) would play key roles in the accelerated narrowing of the lumen seen after angioplasty. Second, animal models of response to injury have shown a dramatic proliferative response to vascular injury with balloon catheters. Pharmacologic studies have identified the m olecules responsible for smooth muscle migration and subsequent intimal proliferation. It seems reasonable to proceed on the assumption that the molecules responsible for forming an intima following injury also play a role in human atherosclerosis or restenosis. This view could be erroneous. Attempts to find replication in human atherosclerotic lesions have generally been unsuccessful