STATEMENT OF PROBLEM: The in vivo release of Pd from palladium alloys into the oral environment and sensitivity reactions by patients has been of concern. However, little information is available about the variation in elemental release from different palladium alloys. PURPOSE: The purpose of this in vitro study was to compare the elemental release into a corrosion-testing medium from a high-palladium alloy (Freedom Plus, 78Pd-8Cu-5Ga-6In-2Au) and a Pd-Ag alloy (Super Star, 60Pd-28Ag-6In-5Sn) under different conditions. MATERIAL AND METHODS: Alloys were cast into Ø12×1-mm-thick disks, subjected to simulated porcelain-firing heat treatment, polished, and ultrasonically cleaned in ethanol. Three specimens of each alloy were immersed for 700 hours in a solution for in vitro corrosion testing (ISO Standard 10271) that was maintained at 37 °C. Two solution volumes (125 mL and 250 mL) were used, and the solutions were subjected to either no agitation or agitation. Elemental compositions of the solutions were analyzed by using inductively coupled plasma-mass spectroscopy (ICP-MS). Concentrations of released elements from each alloy for the 2 solution volumes and agitation conditions were compared by using the restricted maximum likelihood estimation method with a 4-way repeated-measures ANOVA, the Satterwhite degrees of freedom method, a lognormal response distribution, and the covariance structure of compound symmetry. RESULTS: For the 4 combinations of solution volume and agitation conditions, the mean amount of palladium released was 3 orders of magnitude less for the Pd-Ag alloy (0.009 to 0.017 µg/cm2 of alloy surface) compared with the Pd-Cu-Ga alloy (17.9 to 28.7 µg/cm2). Larger mean amounts of Sn, Ga, Ag, and In (0.29 to 0.39, 0.57 to 0.83, 0.71 to 1.08, and 0.91 to 1.25 µg/cm2, respectively) compared with Pd were released from the Pd-Ag alloy. Smaller amounts of Cu, Ga, and In (4.8 to 9.9, 5.9 to 12.8, and 4.2 to 9.5 µg/cm2, respectively) compared with Pd were released from the Pd-Cu-Ga alloy. The Ru released was much lower for the Pd-Ag alloy (0.002 µg/cm2) than the Pd-Cu-Ga alloy (0.032 to 0.053 µg/cm2). Statistically significant differences (P<.001) in elemental release were found for the factors of alloy and element and the alloy×element interaction. Significant differences were found for the solution volume (P=.022), solution volume×element interaction (P=.022), and alloy×solution volume×element interaction (P=.004). No significant effect was found for agitation condition. CONCLUSIONS: The relative amounts of released elements from each alloy were not proportional to the relative amounts in the composition. The amounts of Pd and Ga released from the Pd-Cu-Ga alloy were consistent with the breakdown of a Pd2Ga microstructural phase and perhaps some dissolution of the palladium solid solution matrix. Precipitates, rather than the palladium solid solution matrix, appeared to undergo greater dissolution in the Pd-Ag alloy. The Pd-Ag alloy should have lower risk of adverse biological reactions than the Pd-Cu-Ga alloy.