Purpose: Although no currently available technique for the measurement of osseointegration is entirely satisfactory, 3 clinical variables can be reasonably associated with the process: probing depth, micromobility, and crestal bone height. Micromobility can be quantified to some extent with the use of the Periotest, a commercially available instrument. In this investigation, the influence of surface characteristics and geometry upon Periotest value (PTV) and probing depth measurements was studied. Materials and Methods: In a multicenter trial, 120 healthy edentulous patients received 5 or 6 implants in the anterior mandible and were followed for 3 years. A total of 634 implants were placed. Every patient received at least 1 implant of each of 3 types: threaded titanium plasma-sprayed (TPS), threaded hydroxyapatite-coated (HA), and cylindric HA-coated. A randomization schedule assured that approximately equal numbers of each type of implant were placed and that they were uniformly distributed over the arch. Results: Of the 4 tested combinations of dependent and independent variables, the only statistically significant (P .05) effect was that of coating on PTV. At 1 year after prosthetic restoration, the mean PTV for HA-coated threaded implants was -5.36 ± 1.24, compared with -4.86 ± 1.70 for TPS implants. This difference steadily declined in magnitude and significance, until, after 3 years, the groups were indistinguishable. Discussion: This study agrees with the previous observations that HA coating tends to accelerate the initial rate of osseointegration. The absence of a difference between threaded and cylindric implants confirms that the PTV responds to micromobility near the surface, on a scale much smaller than such gross geometric features. Conclusion: On the basis of these results, one may conclude that HA-coated implants exhibit a more rapid decrease in micromobility than do TPS implants of identical geometry.