Purpose: The purpose of this study was to evaluate differences in the healing of cortical and cancellous bone around titanium implants with two different surfaces (machined or anodized) in rabbit tibiae. Materials and Methods: Screw-shaped commercially pure titanium implants of two different surface types were fabricated: machined implants (control) and anodized implants; each had two transverse canals that served as sites for bone ingrowth. In the tibiae of six New Zealand white rabbits, a total of 24 implants (12 implants of each surface) were surgically placed in a randomized arrangement. The upper transverse canal was positioned in the cortical bone region, and the lower transverse canal was positioned in the cancellous bone region. After a 1-month healing period, undecalcified ground sections were subjected to histologic and histomorphometric analyses. A mixed-model analysis of variance was used to compare the two types of implants and to control for the random effect of animals. Results: The percentage of bone-to-implant contact (BIC) inside the upper canals was 16.45% ± 4.05% for controls (machined surface) and 24.85% ± 4.86% for anodized implants. BIC inside the lower canals was 7.01% ± 2.34% for controls and 11.35% ± 2.67% for anodized implants. The percentage of bone area inside the upper canals was 10.94% ± 4.06% for controls and 27.95% ± 12.38% for anodized implants. The percentage of bone area inside the lower canals was 3.16% ± 1.08% for controls and 4.66% ± 1.53% for anodized implants. For all measures, the anodized implants had higher values than the controls (P .0001). Conclusion: The results suggest that anodized surface modification of titanium implants is beneficial to both cortical and cancellous bone healing. Palabras clave: anodic oxidation, cancellous bone, cortical bone, healing, histomorphometric analysis, titanium implant