In an effort to better understand the supporting anatomy for unloaded endosseous dental implants, this study focused on the histomorphometric analysis of 3 different types of implants placed into non-human primate jaws and allowed to heal for 6 months. This report describes data from 24 screw-type dental implants placed in edentulated (2 months healing time) posterior arches of 4 adult female baboons. Three different implants were placed and allowed to heal for 6 months prior to processing for evaluation: commercially pure titanium (n = 8), titanium alloy (n = 8), and titanium plasma-sprayed (n = 8). Circumferential bone-implant interface sampling from 6 regions along the entire length of each implant was obtained for evaluation of percent bone-implant contact (%BIC) and percent bone area (%BA), within 3 mm of the implant. Data were collected (reliability of 1.6% for both parameters) and analyzed by an observer blinded to implant material using IMAGE analysis software for differences between jaws, implant biomaterials, and jaw/biomaterial (analysis of variance, pairwise comparison using least squares method with Bonferroni adjustment). The results indicated that the overall mean %BIC was 55.8 and mean %BA was 48.1. Maxillary and mandibular differences for both parameters were statistically significantly different: %BIC in maxilla 50.8, in mandible 60.8; %BA in maxilla 43.6, in mandible 52.6 (both significant at the P < .05 level). The biomaterial analyses revealed no significant differences between the different implants for %BIC or %BA. The trend observed--that mandibular values were greater than maxillary values for the overall jaw comparisons--was found to be consistent at the jaw/biomaterial level, although the small sample size limited statistical power. These data, along with data from a previous 3-month study, provide insight into baseline supporting anatomy for dental implants.