It is generally thought that osmoregulation is aimed not at osmotic pressure, or extracellular Na concentration per se, but at the regulation of cell volume. Tacit assumptions often made in considering body fluid distribution include the following: cell membranes support no significant hydrostatic pressure gradient; cell membranes are much more permeable to water than to the major solutes; therefore intracellular and extracellular osmotic pressure are equal; cell solute content is stable and not related to cell volume or osmotic pressure; accordingly cell volume varies inversely with osmotic pressure. In fact, counter examples are known to each of these and the one generalisation that seems to hold is that there are no water pumps and that water rapidly attains thermodynamic equilibrium between interstitial and intracellular fluid. Three topics will be discussed that illustrate some of these points. Even in the simplest model cells are not expected to shrink in direct proportion to increasing osmotic pressure, since they contain 20 to 30% solids. However, after allowing for this, cells still shrink less than predicted. One way of interpreting this retention of water is in terms of an increasing negative intracellular fluid pressure that results from the gelled nature of cytoplasm (Hladky and Rink, 1978). Many cell types respond to osmotic swelling or shrinking by subsequent adjustment of their solute content to return the volume towards normal. Human lymphocytes can do this remarkably quickly, via large increases in permeability to K and Cl. The mechanisms that produce these changes have possible implications for the functioning of osmoreceptors.(ABSTRACT TRUNCATED AT 250 WORDS)