The main cellular elements of atherosclerotic plaques are smooth muscle cells. Because these plaques differ from their precursors in the underlying artery wall in several ways, we have asked the question: Are human atherosclerotic plaques polyclonal or monoclonal in their origin? The X-linked glucose-6-phosphate dehydrogenase (EC 1.1.1.49) in heterozygotic females has been used to obtain an answer. 30 Plaques of different degrees of complexity and 59 samples of normal aorta and iliac artery walls from four females, 25-79 years old, were investigated. The data show that fibrous caps even of relatively large atheromatous plaques, 0.5 cm or greater in diameter, are composed of cells that produce solely or predominantly one enzyme type, whereas samples of artery wall media and intima as small as 0.1 mm(3) are regularly composed of a mixture of cell types. If plaques were a response to injury akin to a healing wound, a reaction to a growth stimulant, or formed due to an organization of a mural thrombus, they would be expected to be polyclonal. Hence, the results imply that atherosclerotic plaques in human beings arise by another mechanism. The mechanism compatible with the monoclonal nature of atherosclerotic plaques is mutation, and the likely causes are chemical mutagens or viruses.