The use of cell cultures to test the biocompatibility of dental materials is gaining in importance. Any cytotoxic effects that restorative materials may have will be on the dental pulp and for that reason cultured pulp cells should be the model of choice for biocompatibility testing. The aim of this investigation was to study the growth and morphologic characteristics and toxic response of human pulp lines and to compare these parameters to those of human buccal mucosa fibroblasts. Twenty-one specimens of pulp tissue and six from buccal mucosa were cultured using standard techniques. Six pulp cell lines were cultured successfully as were all six from the buccal mucosa specimens. From these specimens, 12 growth curves were computed. To study the morphology of the cultured cells, they were observed microscopically and classified into three morphological types: slender elongated cells (type I), epithelioid shaped cells (type II) and large stellate cells (type III). Their numbers and proportions were determined for each cell line and compared statistically. To gauge sensitivity to toxic materials, cells were exposed to concentrations of arecoline. An analysis of the growth curves showed no statistical difference between pulp cells and buccal mucosa cells; the slopes of the curves, however, differed significantly between individual cell lines, and these individual differences were greater among pulp cell lines. The morphology of the pulp and mucosa fibroblasts was similar microscopically. There was no significant difference between the number and proportion of the cell types in the two groups, but there were significant differences between the individual cell lines. Pulp cells showed a greater inhibition of growth when exposed to arecoline. Because pulp fibroblasts are difficult to culture, their reported survival rate is poor; due to the differences that exist between individual cell lines, we conclude that pulp cells when used as single cell lines or even pooled may not be ideal for testing biocompatibility, especially if reproducibility is a prerequisite. Any evaluation will require tests on not one, but several cell lines in order to minimize the effect of inter-cell-line differences. Their greater sensitivity to toxic substances, on the other hand, may show that pulp cells could be more sensitive indicators of cytotoxicity.