The epidemiologic evidence establishing a low HDL cholesterol level as an independent risk factor for atherosclerosis and coronary heart disease is convincing beyond all possible doubt. However, the mechanisms by which a low HDL level predisposes to and a high HDL level protects against atherosclerosis are not firmly established. By all odds the most widely accepted hypothesis is that put forward by Glomset almost 20 years ago, namely, the hypothesis that HDL is involved in “reverse cholesterol transport” (1). Based on his studies of lecithin-cholesterol acyltransferase, Glomset saw that HDL could accept free cholesterol from tissues and that this free cholesterol could be converted to the ester form by the action of LCAT. This would then allow the surface of the HDL molecule to accept another molecule of free cholesterol from the cells, and so on. This hypothesis is supported by many cell culture studies showing that HDL can act as an acceptor of cholesterol from cells overloaded with cholesterol [reviewed by Reichl and Miller (2)]. On the other hand, lipoprotein-deficient serum or even serum albumin alone can also facilitate the removal of cholesterol from cholesterol-loaded cells. Studies by Oram and colleagues, demonstrating a set of saturable binding sites on cells in culture that are up-regulated when the cells are loaded with cholesterol, also support the hypothesis (3,4) and several other groups have either observed specific, saturable binding of HDL or have actually partially purified a putative HDL receptor (see ref. 2). However, the results of these in vitro studies cannot be fully evaluated until there is a translation to the in vivo situation.