Our primary objective was to test the hypothesis that a defect in acidification is more common in patients who have idiopathic calcium phosphate kidney stones than in those whose stones are formed mainly of calcium oxalate. Additionally, other risk factors might differ for these 2 stone types. Urine pH was measured serially over 24 hours, and along with ammonium and titratable acid, it was measured before and serially after ingestion of ammonium chloride in 3 groups of subjects: 24 patients with predominantly calcium phosphate stones, 30 patients with calcium oxalate stones, and 15 health non-stone-formers. Twenty-six parameters potentially related to stone formation and acidification were assayed on urines collected over 24 hours, and 15 parameters on blood. The data base was a computerized list of 5900 analyses of stones from patients living in Newfoundland. Patients not known by their physician to have had urinary tract infection, anatomical abnormality, hyperparathyroidism, or renal tubular acidosis were asked to participate in the study. Differences between means were considered significant if p values were less than 0.05 for F by analysis of variance and also less than 0.01 by t-test. In all patients with calcium oxalate stones and all non-stone-formers, urine acidified to pH less than 5.25, but in 8 of the 23 phosphate stone formers who completed the ammonium chloride study urine failed to acidify to pH less than 5.25. As all 8 had normal values for venous pH, total CO2, and chloride, they were considered to have incomplete renal tubular acidosis (IRTA). The 8 phosphate stone formers with IRTA had greater mean values for urine pH on all 9 specimens collected serially over 24 hours (all means greater than 6.2), and after administration of ammonium chloride (p less than 0.01), as well as lower mean values for urine titratable acid excretion (p less than 0.01), both after administration of ammonium chloride and in 24-hour urine samples, compared with the remaining phosphate stone formers whose urine acidified and the oxalate and non-stone-forming control groups. Nearly all the phosphate stone formers had 1 or more risk factors for stone formation, but with frequencies not significantly higher than those found in the oxalate group. Hypercalciuria and hypocitruria were the commonest, but increased oxalate or urate also occurred. Thus, idiopathic calcium phosphate stone formation can be associated with 1 or more of several risk factors, and, with the possible exception of those with IRTA, treatment should be similar to that given to patients with calcium oxalate stones.