Commercial root end filling materials, namely two zinc oxide eugenol-based cements [intermediate restorative material (IRM), Superseal], a glass ionomer cement (Vitrebond) and three calcium-silicate mineral trioxide aggregate (MTA)-based cements (ProRoot MTA, MTA Angelus, and Tech Biosealer root end), were examined for their ability to: (a) release calcium (Ca(2+) ) and hydroxyl (OH(-) ) ions (biointeractivity) and (b) form apatite (Ap) and/or calcium phosphate (CaP) precursors. Materials were immersed in Hank's balanced salt solution (HBSS) for 1-28 days. Ca(2+) and OH(-) release were measured by ion selective probes, surface analysis was performed by environmental scanning electron microscopy/energy dispersive X-ray analysis, micro-Raman, and Fourier transform infrared spectroscopy. IRM and Superseal released small quantities of Ca(2+) and no OH(-) ions. Uneven sparse nonapatitic Ca-poor amorphous CaP (ACP) deposits were observed after 24 h soaking. Vitrebond did not release either Ca(2+) or OH(-) ions, but uneven nonapatitic Ca-poor CaP deposits were detected after 7 days soaking. ProRoot MTA, MTA Angelus, and Tech Biosealer root end released significant amounts of Ca(2+) and OH(-) ions throughout the experiment. After 1 day soaking, nanospherulites of CaP deposits formed by amorphous calcium/magnesium phosphate (ACP) Ap precursors were detected. A more mature ACP phase was present on ProRoot MTA and on Tech Biosealer root end at all times. In conclusion, zinc oxide and glass ionomer cements had little or no ability to release mineralizing ions: they simply act as substrates for the possible chemical bonding/adsorption of environmental ions and precipitation of nonapatitic Ca-poor ACP deposits. On the contrary, calcium-silicate cements showed a high calcium release and basifying effect and generally a pronounced formation of more mature ACP apatitic precursors correlated with their higher ion-releasing ability.