In three groups (n = 12 each) of male controls (22--43 years), patients with recurring calcium urolithiasis (21--36 years) and hyperparathyroidism (HPT; 17--71 years) proven by surgery renal cyclic adenosine monophosphate (RcAMP), fractional tubular phosphate reabsorption and serum parathyroid hormone (PTH) were measured during endogenous creatinine clearance. RcAMP (muMol/g creatinine) was: controls 1.48 +/- SEM 0.27; stone formers 2.037 +/- 0.343 (not significantly different); HPT 6.234 +/- 0.454 (p less than 0.001). There is no overlap between HPT and controls. Phosphate reabsorption is least in HPT (0.84 +/- 0.015), higher in controls (0.924 +/- 0.004) and stone formers (0.941 +/- 0.007). All differences are statistically significant. Under the conditions selected (moderate hydration of individuals) Serum PHT (pg-equiv/ml) is lowest in stome formers (less than 100--339), higher in controls (less than 100--933) and HPT (400--1150). there is no overlap in PHT between the former and the latter group but a marked one between controls and HPT. For clinical purposes the resulting diagnostic uncertainty in a given patient can be overcome by additional determinations of RcAMP and ionised serum calcium: when referring to serum PTH HPT patients fall outside, RCU patients within 2 standard deviations of either parameter in control subjects. This procedure presently appears superior to those proposed in the past (urinary cAMP etc.) but requires confirmation in larger patient populations. Moreover, since HPT prevails in middle and upper age decades, their RcAMP values and those of RCU patients should be related to a range seen in closely age- and sex-matched controls.