Implicit in attempts to characterize and purify biologically active factors is the premise that the bioassay system employed will show a progressive increase in the response as the concentration of the responsible factor increases. We employed the hamster cheek pouch to assay the neovascularization potential of growth factors, including endothelial cell growth supplement (ECGS), epidermal growth factor (EGF), fibroblast growth factor (FGF) and platelet-derived growth factor (PDGF). Each growth factor was applied to the system in graded concentration and two approaches used to assess neovascularization: first, direct serial inspection of the cheek pouches at 40 power; second, tritiated thymidine incorporation into endothelial cells, assessed by radioautography. PDGF induced a dose-related increase in neovascularization, with a threshold dose of 17.5 micrograms/ml and a peak, 56% response, at a PDGF concentration of 175 micrograms/ml. Progressive increases in PDGF concentration, thereafter, induced progressive reductions in the neovascularization rate. Under some conditions optimal bioassay requires serial dilutions of the assay material over a wide range.