The purpose of this investigation was to determine the ability of human gingival fibroblasts, in vitro, to migrate along a chemotactic gradient over 3 different guided tissue regeneration barrier materials; i.e., polytetrafluoroethylene, polylactic acid, and sterile calcium sulfate. Forty petri dishes were divided into 4 equal groups. In each group of 10 dishes, a different barrier material served as the fibroblast substrate with the polystyrene floor of one group of Petri dishes serving as the control. The under agarose technique of measuring cell migration was employed using platelet derived growth factor-BB homodimer as the chemoattractant and Hanks balanced salt solution to test random migration. In addition, fibroblasts were directly cultured on triplicate sets of barrier materials and the control surface for 24 hours and examined by scanning electron microscopy. Comparative analysis of the fibroblast migration data showed the mean migration distance (adjusted for random migration) for controls to be significantly greater than any of the three barrier materials. Further, mean migration distance over calcium sulfate was significantly greater when compared to that of the polylactic acid barrier group. All other comparisons between groups were not statistically significant. Scanning electron microscopic examination fibroblasts cultured directly on barrier membranes and compared to controls indicated that the calcium sulfate substrate appeared to facilitate cell attachment and spreading whereas cells on polytetrafluoroethylene and polylactic acid barriers exhibited a morphology not conducive to migration or, in many cases, cell health. Based on these limited in vitro results and, given the 3 barrier materials considered, it would appear that calcium sulfate offers the greater potential for guided tissue regeneration in surgical sites where primary wound closure cannot be obtained.