Non-modulation is a trait characterized by abnormal angiotensin-mediated control of aldosterone release and the renal blood supply. To determine whether non-modulation defines a specific subgroup of the hypertensive population and its utility as an intermediate phenotype, we have studied the distribution of this quantitative trait, whether its features are reproducible on repeated testing, and whether there is concordance of its multiple features. Essential hypertensive patients (224) and normotensive subjects (119) received an infusion of angiotensin II (Ang II) at 3 ng.kg-1.min-1 for 30-45 minutes. p-Aminohippurate (PAH) clearance was assessed as an index of renal plasma flow while the subjects were on a 200 meq sodium diet; plasma aldosterone levels were measured while the subjects were on a 10 meq sodium diet. In 54 subjects, diuretic-induced volume depletion superimposed on a low salt diet was substituted for the Ang II infusion. The results of each study were submitted to maximum likelihood analysis to assess bimodality. In response to both diuretic-induced volume depletion (p < 0.000023) and Ang II infusion (p < 0.0009), aldosterone responses were bimodally distributed in the essential hypertensive but not in the normotensive subjects, suggesting that this trait identifies a discrete subgroup. In the 59 subjects who had both an adrenal and renal study, 50 (85%) were concordant. Finally, in 27 subjects studied two to six times over a span of 1-60 months, the intraclass correlations of the adrenal, PAH, or both responses were highly significant (p values between 0.001 and 0.00007), indicating high reproducibility of results on repeated testing.(ABSTRACT TRUNCATED AT 250 WORDS)