Tricalcium phosphate (TCP) has been historically a well-accepted material for bone augmentation. We examined the use of a porous beta-TCP (100%) in a split mouth model for sinus floor augmentation. Five patients were treated bilaterally, receiving 1-2 mm-sized beta-TCP particles (Cerasorb) on one side (test side) and autologous chin bone particles on the other (control) side. Four other patients were treated with a unilateral sinus floor augmentation using 100% beta-TCP (no controls). Biopsies of the augmented sites were taken at 6 months. Histomorphometry measurements were carried out in order to quantify bone augmentation at test and control sides. The average bone volume (BV) formed in the augmented sinus at the control side was 41% (32-56%) and 17% (9-27%) in the test side when all nine patients were included (statistically significant, P=0.04). When only the five bilateral patients were included, mean BV of the test side was 19% (13-27%), which was also significantly different from the control side (P=0.009). Osteoid formation tended to be higher in the test side biopsies (1.3%) than in the controls (0.3%) (marginally significant, P=0.1), indicating ongoing bone formation in the TCP material. The amount of lamellar bone at the test side was less than half the amount in the control side, indicating that remodelling had only recently started in the TCP-augmented side. The resorption surface, however, did not differ significantly between the two sides. These histological results indicate that Cerasorb is an acceptable bone substitute material for augmentation of the maxillary sinus. Due to the osteoconductive, but not osteoinductive properties of this material, the rate of bone formation is somewhat delayed in comparison to autologous bone.