A B S T R A C T Rats were made acutely hyper-or hyponatremic by infusion of hypertonic saline or water, respectively. Other rats were maintained in these states from 1 to 7 days to observe the effects of time. Brain tissue water, Na, Cl, and K were compared with serum Na and Cl concentration (NaE and ClE). The following observations are noted: Brain Cl content varies directly with CIE and brain Na content in the Cl space (Nae) varies directly with NaE, indicating little or no restraint on the inward or outward movement of Na or Cl from the Cl space of brain. The intracellular volume of brain fluid (Vi) derived as the difference between total water and Cl space, decreases with hypernatremia and increases with hyponatremia. The changes in V, in the acute studies are not accompanied by any change in brain K content, or calculated intracellular Na content, and are approximately 0.6 the changes predicted from osmotic behavior of cells, which apply four assumptions: (a) NaE is proportional to osmolality; (b) Experimental studies (3, 6-1 1) have documented the changes in brain fluid and electrolyte that occur with changes in osmolality that were specific for each study. The results have usually shown a change in brain Cl and Na content in the direction of the change in concentration of C1 and Na induced in plasma. Changes in K content in the brain have been noted in some experiments (6-9) but not in others (3,10,11). Except for the early study by Yannet (6), there has been little effort to generalize a quantitative relation between changes in plasma osmolality to changes in brain tissue water and electrolytes.The effects of changes in plasma osmolality upon brain fluid and electrolyte depend upon the rate with which brain water equilibrates with plasma water when a gradient in water potential develops, and the degree to which brain solute content is altered, since the latter influences the final distribution of water. Because rates of exchange for water and solute vary, time is an important element in characterizing changes in brainfluid volume. This is evident in clinical experience: hypertonic urea injections decrease brain volume, because urea diffusion into brain water is relatively slow and osmotic redistribution of water continues until urea equilibrium is reestablished; rapid reduction of plasma urea concentration in uremia by hemodialysis leads to transitory signs of brain swelling (12) for the same reason. It is also evident that those patients who develop hyponatremia slowly have fewer symptoms of central nervous system dysfunction than do those who develop it rapidly (13, 14).The studies reported here are designed to compare the effects of acute changes in plasma osmolality with those of sustained changes upon the Na and Cl content of the brain, the fluid phases of the brain, the K content, and the ratio of cation to water in brain fluid. In particular, we have compared the effects of changing osmolality upon observed cell volume with that predicted from the assumption that cell volume is a reciprocal func...