The magnetization of hydrogen nuclei in H2O constitutes the basis of most applications of magnetic resonance imaging (MRI.) Only ortho-water, where the two proton spins are in states that are symmetric with respect to permutation, features NMR-allowed transitions. Para-water is analogous to para-hydrogen, where the two proton spins are anti-symmetric with respect to permutation. The objective of this proposal is to render para-H2O accessible to observation. Several strategies will be developed for its preparation and observation in solids, liquids and gas phase, with yields up to 33%. When diluted in acetonitrile at room temperature, we found that Tortho(H2O) = 6 s. Based on experiments on H2C groups where Tpara/Tortho > 37, we conservatively estimate that Tpara/Tortho > 10 for H2O, so that we expect Tpara = 60 s. Dilution in aprotic solvents inhibits the exchange of protons and extends the lifetimes t(H2O) of water molecules from ca. 1 ms in pure water to 10 s and beyond, so that proton exchange does not hamper the use para-water. The ratio Tpara/Tortho of H2O depends on temperature, viscosity, paramagnetic agents, etc., which affect intra- and inter-molecular dipole-dipole interactions, chemical shift anisotropy, and spin rotation. In cases where proton exchange significantly shortens the lifetime of para-H2O, we shall prepare and observe para-ethanol and aqueous solutions of para-glycine, which cannot suffer from proton exchange, and allow similar perspectives as para-water. In conventional MRI, contrast stems mostly from spatial variations of T1 and T2. By monitoring the ratio Tpara/Tortho as a function of spatial coordinates, it will be possible to obtain a novel type of contrast. In suitable phantoms and porous media, para-water will allow us to characterize slow transport phenomena such as flow, diffusion, and electrophoretic mobility. The study of transport phenomena will become possible over longer time intervals, lower velocities or greater distances.