SummaryFour self-curing acrylic hone cements were surveyed by infrared, solubility, viscometry, quantitative metallography, microscopy, and physical testing techniques: CMW, Palacos R, Sulfix-6, and Surgical Simplex P. Results show that these hone cements were primarily composed of poly(methy1 methacrylate) and that no crosslinking was evident. Solubility analysis confirmed this latter observation, as the hone cements dissolved completely except for a small insoluble fraction, ,which was identified as the radiopaque filler. For each bone cement, the viscosity-average molecular weights of both the powdered phase and the cured two-phase product remained unchanged, varying overall from 1 to 5 X 105. Using standard quantitative metallography, porosity ranged from 1 to 8% and the dispersed powder phase decreased 11-46%. Microscopy revealed the nature of the porosity, radiopaque fillers, the powder size and shape, and the fracture morphology. From tensile and fracture toughness tests, five physical properties were determined at ambient conditions and at 37°C after conditioning in distilled water a t 37°C for 10 months: the modulus of elasticity, the ultimate tensile strength, the elongation at break, the fracture energy, and the mean inherent flaw size. At ambient conditions, the ultimate tensile strength decreased 33-55% when compared with commercial unmodified poly(methy1 methacrylate), Plexiglas G. While the fracture energy remained rather invariant, the mean inherent flaw size increased fivefold over the commercial acrylic tested. This marked increase in the mean inherent flaw size could lower the fatigue resistance of a material, since more and/or larger fracture initiation sites are available. When tested a t 37°C after protracted conditioning, the deleterious trends observed a t ambient temperature continued. To some degree, porosity, particle-matrix interfaces, residual stresses, low molecular weight products, inorganic and/or other organic additions, and water contributed to the inherent flaw size a t the expense of the working stress. Several modifications are suggested by which the importance of these factors might he minimized.