Objectives: To determine the effect of alveolar ridge preservation (ARP) for molar sites without primary flap closure. Materials and Methods: Three groups were established: the sockets grafted with deproteinized bovine bone mineral containing 10% collagen (DBBM-C) and covered by a native bilayer collagen membrane (CM) (test group 1), the sockets grafted with DBBM-C only (test group 2), and the sockets healed naturally (control group). Primary flap closure was not attempted. Cone-beam computed tomography scans were obtained immediately after ARP and 4 months later. A biopsy was performed at the time of implant surgery. The change of marginal bone level was measured. Results: There was significantly less horizontal resorption in the test group 1 than the control group at 1 mm (-1.02±0.88 [mean±SD] vs. -4.44±3.71 mm) and 3 mm (-0.31±1.51 vs. -2.27±1.15 mm) levels below the crest, and significantly less vertical reduction in the midcrestal area in test group 1 than in test group 2 (-0.25±0.95 vs. -1.15±1.63 mm) (p0.05). There were no significant differences between the test groups in clinical and histomorphometric measurements. All groups exhibited stable marginal bone level after one year loading. Conclusion: ARP without primary flap closure in molar areas was effective in minimizing ridge resorption and favored implant treatment. Palabras clave: bone regeneration, bone substitute, randomized controlled trial, tooth extraction, wound healing