A method of data reduction for the statistical examination of mean arterial pressure in essential hypertensive subjects to determine salt sensitivity was examined. A computerized method was used to estimate the best-fitting cosine curve for data collected every 30 minutes for 24 hours. The effect of sodium loading on the cosinor parameters (mesor, amplitude, and acrophase) in 45 subjects with essential hypertension and five normotensive control subjects was assessed. Twenty-five percent of the essential hypertensive subjects in the study were found to be salt-sensitive with a statistically significant increase in their mesors with sodium loading (p less than 0.05). The non-salt-sensitive group was found to contain a subpopulation with a statistically significant decrease in their mesor with sodium loading (p less than 0.05). Sodium loading appears to affect the lability of mean arterial pressure independently of mesor changes.