We applied neutron reflectivity measurements to hydrophilic and hydrophobic self-assembled monolayers (SAMs) to probe water density at these interfaces. The measurements were motivated by a previous theoretical study which reported a reduced water density at a hydrophobic surface (Lum, K.; Chandler, D.; Weeks, J. D. J. Phys. Chem. B 1999, 103, 4570) and by our own computer simulations on the hydration force and water density between two methoxy tri(ethylene glycol) terminated undecylthiolate SAMs adsorbed on gold substrates used to study protein adsorption. These simulations predicted that the surfaces are slightly hydrophobic and are characterized by a reduced water density at the interface (Pertsin, A. J.; Hayashi, T.; Grunze, M. J. Phys. Chem. B 2002,106, 12274). In disagreement with the marginal reduction in water density derived in the simulations, the neutron reflectivity measurements reported here indicate an unexpectedly extended (similar to4 nm) water layer at the SAM surface with a noticeably reduced density (85-90% of the density of bulk water). The reproducibility of the experimental results with the methoxy tri(ethylene glycol) terminated undecylthiolate SAMs was confirmed with four different samples and by one measurement using a contrast-matched D2O/H2O water mixture. We also used neutron reflectivity measurements to study the water density at the water/SAM interface of hydroxy hexa(ethylene glycol) and hydroxy tri(ethylene glycol) terminated undecylthiolate SAMs. Except for one of the hydroxy hexa(ethylene glycol) samples studied, the results are consistent with the presence of bulk water in direct contact with the interface. We also discuss possible artifacts and problems in the analysis. Our results on nonfunctionalized hydrophobic octadecanethiolate and hydrophilic hydroxy-terminated undecylthiolate SAMs give physically unreasonable and nonconclusive models for the water interface on these surfaces, respectively. The best fit of the data for the hydrophobic surface gives an unreasonably low water density, possibly due to the presence of air inclusions in the film and/or adsorbed air "nanobubbles". The results obtained on the hydrophilic hydroxy-terminated surface can be fitted equally well with a model assuming an interface water density that is higher or lower than that of bulk water, demonstrating the ambiguities associated with describing the organic/liquid interface with a box model for the Q range accessible in our experiment.