Materials and methods: A wear method was developed that mainly simulates attrition wear using a commercially available chewing simulator (Willytec, Germany). In this test, a standardized stylus made of pressable ceramic (Empress) hits flat specimens 120,000 times with a 5 kg weight, a descent speed of 60 mm/s and a lateral movement of 0.7 mm with a speed of 40 mm/s under constant exchange of water at different temperatures (325 x 5 degrees C/55 degrees C). The volume loss was measured on plaster replicas with the Laserscan 3D (Willytec) and the Match 3D software. Twenty-four experimental and commercial composites (n = 8) were tested with a volumetric wear range of between 5.5 and 147 x 10(-2) mm(3). On standardized specimens made of the same composites, the Vickers hardness (H), elastic modulus (E) and fracture toughness (K-1c) were measured. The mean particle size (d) and volume content (v(f)) of the inorganic filler were evaluated. Furthermore, a differentiation was made between the main filler with the largest mean size (d(1), v(f.1)) and the total filler content (v(f.tot)). Results: The best linear regression curve fit with an adjusted R-2 of 0.908 was found for wear index = d(1)(0.6) /K(1c)v(f.1)vf.(tot)(E/H) Conclusions: The good mathematical fit of the formula may be an indication that the wear method is based on physical properties and that it provides a highly reproducible standard. (C) 2007 Elsevier B.V. All rights reserved.