The specific aim of the study was to determine the effect that food-stimulating solvents have on varyingly constituted resin composites and glass ionomer. Samples were stored in food-simulating liquids (FSL) (in increasing order of solubility parameter): Heptane, 100%, 75%, 50%, 25%, and 0% aqueous ethanol solutions for 0, 1, 3, 7, 18, and 30 days at 37 degrees C prior to Knoop Hardness measurements. Measurements were made on both the unpolished matrix-rich surface and the polished filler-rich surface. Multifactorial analysis of variance revealed significant differences in hardness among materials, surface finish, solution, storage time, and all of their interactions. The average hardness of the polished surface was higher than that of the unpolished surface. Resin composites with similar matrix chemistry tended to behave similarly in FSL. Resin with Urethane DMA matrix were found to be significantly more susceptible to FSL than were those with the BisGMA matrix. Glass ionomer behaved differently in FSL compared with composites, due to differences in matrix chemistry, the nature of the bond between filler and matrix, and its susceptibility to dehydration.