The present experiment was performed to study the peri-implant tissue response to non-submerged (1-stage) and initially submerged (2-stage) implant installation procedures. 6 beagle dogs were used. All mandibular premolars and the 1st, 2nd and 3rd maxillary premolars were extracted. After 3 months of healing, 3 fixtures of the Astra Tech System were installed and submerged in the right (or the left) edentulous, premolar region in each of the 6 dogs. Radiographs were obtained immediately after fixture installation. In the radiographs, the distance between the abutment-fixture junction and the most "coronal" bone in contact with the implant surface was determined. 3 months later, abutments were connected to the initially submerged fixtures and another 3 fixtures of the same system were installed in the contralateral, edentulous premolar region. Abutments were, however, immediately connected to the newly-installed fixtures (non-submerged side; test side). The mucosal flaps were replaced, adjusted and sutured in such a way that the coronal portion of the abutments remained exposed in the oral cavity. A new set of radiographs were obtained from all 6 implant sites in each animal. A period of plaque control was initiated. Clinical examinations were performed and radiographs obtained from all implant sites after another 3 months and at the termination of the experiment. 9 months after the 1st fixture installation procedure, the animals were sacrificed, the mandibles were removed, and each implant region dissected. The most mesially-located implant sites were processed for ground sectioning. The remaining biopsies were processed and embedded in EPON. The histometric analysis included assessment of the vertical dimension of the marginal soft and mineralized peri-implant tissues. The ground sections were used for measurements describing (i) "bone to implant contact" and (ii) "bone density". It was observed that the mucosa and bone tissue that formed at implants placed in a non-submerged or a submerged procedure had many features in common. Thus, figures describing (i) the height of the mucosa, (ii) the length of the junctional epithelium and the height and quality of the zone of "connective tissue integration", (iii) the % of bone to implant contact as well as (iv) the density of the peri-implant bone, were similar in the submerged and the non-submerged groups. It is therefore suggested that a non-submerged (1-stage) installation technique may provide conditions for tissue integration that are similar to those obtained using a submerged (2-stage) approach.