The effect of correction of chronic hyponatremia at different rates was studied in 91 rats maintained at a plasma [Na+] of 112 +/- 1 mmol/liter for 19 +/- 1 days. Hyponatremia was corrected into normal ranges (140 to 145 mmol/liter) using three different methods. Rats corrected by water restriction achieved normal plasma [Na+] by 2.1 +/- 0.2 day and had a maximal (4 hr) correction rate of 1.0 +/- 0.1 mmol/liter.hr; rats corrected by water diuresis achieved normal plasma [Na+] by 1.6 +/- 0.1 day and had a maximal correction rate of 2.8 +/- 0.2 mmol/liter.hr; rats corrected by hypertonic saline infusion achieved normal plasma [Na+] by 5.4 +/- 0.3 hr and had a maximal correction rate of 5.7 +/- 0.4 mmol/liter.hr. A fourth control group was not corrected. No demyelinative lesions were found in the brains from the uncorrected rats, whereas the occurrence of such lesions in the brains of the corrected rats was highly correlated with the maximal rate of increase in plasma [Na+] (r = 0.68, P less than 0.001), and to a lesser degree with the magnitude of the increase in plasma [Na+] over the first 24 hours of correction (r = 0.41, P less than 0.001). Brain myelinolysis was first observed in animals whose maximal (4 hr) rate of correction exceeded 1.75 mmol/liter.hr, and the incidence of demyelination increased progressively in rats with more rapid rates of correction. Similarly, myelinolysis was first observed in rats whose magnitude of correction at 24 hours exceeded 16 mmol/liter and also increased in rats with larger 24 hour magnitudes of correction.(ABSTRACT TRUNCATED AT 250 WORDS)