PURPOSE: To study the effects of surface roughness and composition of titanium on proliferation and differentiation of osteoblasts. METHODS: Osteoblasts were cultured on 5 commercially pure titanium (cp Ti) substrates of ground (S0), blasted with 108~130 |xm(S1), 216~301 m(S2), 356~411 m (S3) TiO2 particles and titanium-sprayed plasma (TPS) surfaces. Surfaces prepared by hand grinding with SiC paper of 600 grits served as control (S0). Electron microprobe was used to evaluate the TiO2 film structure of the 5 titanium surfaces. For proliferation and differentiation measurement, osteoblasts were cultured for 1, 3, 5 and 7 days,evaluated by MTT Assay, ALP activity and OC level. SPSS12.0 software package was used for one-way ANOVA. RESULTS: The number of cells was the highest on S3 and TPS surfaces after 1 day culture. The same results were observed after 5 days. On day 3 and 7, S3 surface was the highest (P<0.05). The number of cells on all experimental groups were higher than S0 surfaces at each time point (P<0.05). Increase of ALP activity were detected on S0, S1 and S2 surfaces after 1, 3, 5 days. However, there was no difference between S3 and TPS surfaces (Pg0.05). After 7 days, ALP activity increased significantly on TPS surface than on S1 or S2 surfaces. ALP activity on S3 surfaces was the highest ( P<0.05) . The ALP activity on all experimental groups were higher than S0 surfaces at each time point ( P<0.05) . An increase in OC production was detected on S0, S1, S2 and S3 surfaces after 1, 3, 5 days. The highest OC production was on S3 surface was on day 7 (P<0.05). CONCLUSIONS: It was concluded that for TiO2 blasted surfaces, Ra ranging from 1.260 m to 3.530 m would optimize proliferation and differentiation of osteoblasts. Blasting was an effective treatment method for osteointegration in vitro. [ABSTRACT FROM AUTHOR]