The objective of this study was to investigate how the connection of superstructures to implants with different surface properties affects the surrounding bone. The right and left mandibular premolars and molars of 5 dogs were extracted. After 12 weeks, a machined implant was placed mesially and an anodized implant was placed distally on one side of the edentulous jaw, with the positions reversed on the opposite side. Twelve weeks after implantation, splinted superstructures were set to the implants. At 24 weeks after implantation, the implant stability quotient (ISQ) was measured, radiographs were obtained. Removal torque values were measured and histologic observation was performed. The ISQ values at 24 weeks after implantation were not significantly different between the groups. The removal torque values were significantly different between the distal anodized and distal machined implants (p < 0.05). From 12 to 24 weeks, marginal bone losses were not significantly different between the groups. Fluorescent observation of tissue samples revealed bone-remodeling activity around all of the implants. The results of this study suggest that when implants with different surface properties are connected, machined implants at the most distal sites might be a potential risk factor for implant-bone binding.