The ionic nature of ionic liquids (ILs) results in a unique combination of intrinsic properties that produces increasing interest in the research of these fluids as environmentally friendly "neoteric" solvents. One of the main research fields is their exploitation as solvents for liquid-liquid extractions, but although ILs cannot vaporize leading to air pollution, they present non-negligible miscibility with water that may be the cause of some environmental aquatic risks. It is thus important to know the mutual solubilities between ILs and water before their industrial applications. In this work, the mutual solubilities of hydrophobic yet hygroscopic imidazolium-, pyridinium-, pyrrolidinium-, and piperidinium-based ILs in combination with the anions bis-(trifluoromethylsulfonyl)imide, hexafluorophosphate, and tricyanomethane with water were measured between 288.15 and 318.15 K. The effect of the ILs structural combinations, as well as the influence of several factors, namely cation side alkyl chain length, the number of cation substitutions, the cation family, and the anion identity in these mutual solubilities are analyzed and discussed. The hydrophobicity of the anions increases in the order [C(CN) 3 ] < [PF 6 ] < [Tf 2 N] while the hydrophobicity of the cations increases from [C n mim] < [C n mpy] e [C n mpyr] < [C n mpip] and with the alkyl chain length increase. From experimental measurements of the temperature dependence of ionic liquid solubilities in water, the thermodynamic molar functions of solution, such as Gibbs energy, enthalpy, and entropy at infinite dilution were determined, showing that the solubility of these ILs in water is entropically driven and that the anion solvation at the IL-rich phase controls their solubilities in water. The COSMO-RS, a predictive method based on unimolecular quantum chemistry calculations, was also evaluated for the description of the water-IL binary systems studied, where it showed to be capable of providing an acceptable qualitative agreement with the experimental data.