A high-throughput method for analyzing cellular response to crystallinity in a polymer material is presented. Variations in crystallinity lead to changes in surface roughness on nanometer length scales, and it is shown that cells are exquisitely sensitive to these changes. Gradients of polymer crystallinity were fabricated on films of poly(L-lactic acid) using a gradient in annealing temperature. The resultant morphologies were characterized using an atomic force microscope. Root-mean-square (rms) roughness values ranging from 0.5 to 13 nm were created on a single sample. MC3T3-E1 osteoblastic cells were cultured for 1, 3 and 5 d, and the number of cells was measured using automated fluorescence microscopy. It is shown that the rate of proliferation on the smooth regions of the films is much greater than that on the rough regions, and a monotonic variation in rate is observed as a function of roughness. The critical rms roughness, above which a statistically significant reduction in rate of proliferation occurs, was approximately 1.1 nm. Fluorescence microscopy measurements on immunostained cells indicate there is no significant change in cell area, the number or type of adhesions formed, or the degree of actin polymerization. Results from enzyme-linked immunofluorescence assays indicated that there was no detectable change in adhesion protein accessibility, suggesting the cells directly respond to substrate topography. The use of the gradient library approach yielded the functional dependence of cell proliferation on nanometer-scale roughness and gave a sensitive estimate of the critical roughness for which a decrease in proliferation is observed.