Purpose: To evaluate the bone contact percentage around a proprietary high-temperature dual-etched (DE) implant surface (Osseotite) versus implants with machined, hydroxyapatite (HA), and titanium plasma-sprayed (TPS) surfaces. Materials and Methods: Each implant type was placed in rabbit tibiae of the same animal and assessed at 1 to 8 weeks. Histologic sections were prepared and analyzed histomorphometrically. Results: The DE implant surface achieved higher levels of bone contact percentage than the other surfaces. This enhanced contact level was apparent by 3 weeks and seen at all time intervals except 2 weeks, at which machined exceeded the DE mean. In evaluating which surface outscored the others in each individual rabbit, there was a statistically significant confidence for the DE surface (P .001). The other 3 surfaces failed to show significance, although the numeric scores for the TPS surfaces were below random expectations and the machined scores were slightly above. There was no correlation between degree of roughness and bone contact percentage. Discussion: Arbitrarily roughening the implant surface may not result in a large change in bone conductivity. The specific texture of the DE process yielded more contact, possibly as the result of better fibrin clot retention and growth factor enhancement. Conclusions: There was no advantage demonstrated in this model to an HA surface over titanium. The bone contact to the rough HA surface scored similarly to that for the TPS surface of similar roughness, and well below that for the DE titanium surface. The DE surface appeared to have an advantage in bone contact percentage, particularly in early healing in a rabbit tibia model.