Uncertainty about the origin of anomalous fluid pressures in certain geologic settings has caused researchers to take a second look at osmosis, or flow driven by chemical potential differences, as a pressure‐generating process in the subsurface. Interest in geological osmosis has also increased because of an in situ experiment by Neuzil (2000) suggesting that Pierre Shale could generate large osmotic pressures when highly compacted. In the last few years, additional laboratory and in situ experiments have greatly increased the number of data on osmotic properties of argillaceous formations, but they have not been systematically examined. In this paper we compile these data and explore their implications for osmotic pressure generation in subsurface systems. Rather than base our analysis on osmotic efficiencies, which depend strongly on concentration, we calculated values of a quantity we term osmotic specific surface area (A_so) that, in principle, is a property of the porous medium only. The A_so values are consistent with a surprisingly broad spectrum of osmotic behavior in argillaceous formations, and all the formations tested exhibited at least a modest ability to generate osmotic pressure. It appears possible that under appropriate conditions some formations can be highly effective osmotic membranes able to generate osmotic pressures exceeding 30 MPa (3 km of head) at porosities as high as ∼0.1 and pressures exceeding 10 MPa at porosities as high as ∼0.2. These findings are difficult to reconcile with the lack of compelling field evidence for osmotic pressures, and we propose three explanations for the disparity: (1) Our analysis is flawed and argillaceous formations are less effective osmotic membranes than it suggests; (2) the necessary subsurface conditions, significant salinity differences within intact argillaceous formations, are rare; or (3) osmotic pressures are unlikely to be detected and are not recognized when encountered. The last possibility, that osmotic pressures routinely escape detection or are attributed to other mechanisms, has important implications for understanding subsurface flow regimes.