Current implant placement utilizes both nonsubmerged and submerged techniques. However, the implications of the location of a rough/smooth implant interface as well as the location of a microgap between implant and abutment on crestal bone changes are not well understood. The purpose of this study was to radiographically evaluate crestal bone changes around unloaded nonsubmerged and submerged titanium implants in a side-by-side comparison. Fifty-nine (59) implants were placed at different levels to the alveolar crest in 5 foxhounds. Standardized radiographs were taken at baseline and at monthly intervals until sacrifice at 6 months. Radiographic assessment was carried out by measuring the distance between the top of the implant/abutment and the most coronal bone-to-implant contact (DIB), and by evaluation of bone density changes using computer-assisted densitometric image analysis (CADIA). DIB measurements revealed that in 1-part, nonsubmerged implants, the most coronal bone-to-implant contact followed at all time points the rough/smooth implant interface. In all 2-part implants, nonsubmerged and submerged, the most coronal bone-to-implant contact was consistently located approximately 2 mm below the microgap. In addition, CADIA values for all 2-part implants were decreased in the most coronal area-of-interest (AOI). All bone changes were statistically significant and detectable 1 month after implant placement in nonsubmerged implants or 1 month after abutment connection in submerged implants. Neither implant position nor individual dog effects were statistically significant. These results demonstrate that the rough/smooth implant interface as well as the location of the microgap have a significant effect on marginal bone formation as evaluated by standardized longitudinal radiography. Bone remodeling occurs rapidly during the early healing phase after implant placement for non-submerged implants and after abutment connection for submerged implants.