Purpose: Osseointegration has been defined in many ways, from both basic science and clinical perspectives, but generally represents the restoration of bony homeostasis following implant placement and is usually judged by some form of bone/implant disruption test. In this study, bone anchorage to two different implant surfaces, in tensile and shear modes, was compared to investigate the relation between implant surface topography and osseointegration over time. The purpose was to determine if mathematical parameters could be derived that would reflect the biologic relevance of the implant surface design. Materials and Methods: Rectangular titanium implants (n = 244) were placed in the distal femora of 122 male Wistar rats proximal to the knee joint. Implants were either microsurfaced (MS) or nanosurfaced (NS). Animals were euthanized at one of six time points ranging from 5 days to 6 months, and the force required to disrupt the bone implant interface, in either shear or tension, was measured using an Instron machine. Data were analyzed by fitting the function F = C (1-e-x/τ), where F is the measured disruption force, C is the predicted average maximum disruption force, x is the time postimplantation, and τ is a time constant defined as the time required for F to reach 63.2% of C. Results: Analysis showed that shear testing resulted in significantly larger values of C than seen in tension, but no significant difference was observed when comparing the values of C for NS and MS implants in shear (P = .7). Thus, in accord with clinical reports, both implants performed equivalently at longer implantation periods. The differences in C were significant in tension (P .05). Importantly, NS implants had a significantly smaller τ than the MS implants (P .01, in shear), but no significant differences were observed in τ due to mechanical testing vector. The disruption force values reached a plateau with time, representing bony homeostasis as a result of osseointegration. With time, both implant surfaces reached the same maximum (C) values, in shear. However, the value of τ was smaller in NS compared to MS implants, which represented a higher rate of osseointegration. Conclusion: Thus, τ emerges as a measureable and biologically relevant parameter that can be employed to compare the osseointegration potential of putative implant surfaces. Parole chiave: bone anchorage, homeostasis, implant surface, mathematical modeling, nanotopography, new parameter, osseointegration, surface roughness, tau