When bovine capillary endothelial (BCE) cells were treated with 10 ng/ml of basic fibroblast growth factor (bFGF) for 10 or 30 minutes at 37"C, washed extensively with phosphate-buffered saline (PBS) and incubated in bFGF-free medium, plasminogen activator (PA) production was stimulated to the same extent as in cells exposed continuously to bFGF. Three methods of removing bFGF from heparin-like binding sites in the extracellular matrix, but not from bFGF receptors, abolished this long-term effect of a brief exposure to bFGF. First, BCE cells exposed to bFGF for 30 minutes were washed with 2 M NaCl and incubated in bFGF-free medium. Second, BCE cells were incubated with bFGF for 10 minutes in the presence of heparin, and cells were washed with PBS and incubated in bFGF-free medium. Third, BCE cell cultures were treated with heparinase and exposed to bFGF. Each of these treatments abolished the long-term (24-48 hours) stimulation of PA production normally observed after brief exposure to bFGF. In each of these experiments, incubation of cells in bFGF-containing medium after the treatments resulted in normal stimulation of PA production, demcjiistrating that the treatments did not harm the cells. Stimulation of DNA synthesis was observed when cells were exposed to bFGF for 2 hours at 4"C, incubated in bFGF-free medium for 24 hours at 37"C, and assayed for 3H-thymidine incorporation. However, no stimulation was observed if the 2 hours incubation at 4ยฐC was carried out in the presence of heparin. Thus, long-term stimulation of PA activity and DNA synthesis after a brief exposure to bFGF seems to be a consequence of bFGF binding to the extracellular matrix. The extracellular matrix may act as a physiologic buffer, binding bFGF when concentrations are high and releasing it later for interaction with its receptor. This interaction with matrix may be required for the in vivo action of bFCF.