The present clinical taxonomy of distal renal tubular acidoses includes "gradient," "secretory," and "voltage" defects. These categories refer to presumed collecting duct defects in which the epithelium may be abnormally permeable and unable to sustain an ion gradient, in which luminal proton ATPases are defective, or in which electrogenic Na⁺ reabsorption is impaired and luminal electronegativity is reduced. Classification of affected individuals is based on urinary pH and ion concentrations during spontaneous acidosis and during SO infusion, as well as urinary PCO₂ during HCO loading. A model of rat CD has been developed that has been used to examine determinants of urinary acidification (Weinstein AM. Am J Physiol Renal Physiol 283: F1252-F1266, 2002) and the interplay of HCO and PO loads to generate a disequlibrium pH and equilibrium PCO₂. In this paper, pure forms of gradient, voltage, and secretory defects are simulated, with attention to variability in the locus of the defect in the cortical (CCD), outer medullary (OMCD), or inner medullary collecting duct (IMCD). The objective of these calculations is to discover whether the intuitive description of these defects is sustained quantitatively. The most important positive finding is that the locus of the transport defect along the CD plays a critical role in the apparent severity of the lesion, with more proximal defects being less severe and more easily correctable. In particular, model calculations suggest that for gradient or secretory defects to be clinically detectable they need to involve the OMCD and/or IMCD. Additionally, the calculations reveal a possible mechanism for CD K⁺ wasting, which does not involve failure of H⁺-K⁺-ATPase but derives from a paracellular anion leak and thereby a more electronegative lumen. The most important negative finding is the lack of support for the category of renal tubular acidosis associated with a voltage defect. Although CD lesions that present with both K⁺ and H⁺ secretory defects suggest mediation by transepithelial electrical potential difference (PD), both PD changes and proton pump PD sensitivity appear too small to account for the observed abnormalities.