In 1933, M.H. Jacobs (J. Cell. Comp. Physiol. 4, 161-183) developed the theoretical basis for calculating permeability constants of nonelectryolytes passively penetrating erythrocytes from experimentally determined hemolysis times in isotonic solution of penetrating solute. This derivation had assumed that the reflection coefficient sigma=1, whereas, usually 0 less than sigma less than 1, By comparison of Jacobs' original derivation with the equations revised to include sigma, it is shown that: (see article) where r=(k1/k2), the apparent ratio of solute permeability constant (k1) to osmotic volume flow constant of water (k2) as determined by the Jacobs approach; and ro=(omega/Lpcs), the ratio of the true permeability constant (omega) to the osmotic flow calculated from the product of the pressure-filtration coefficient (Lp) and the concentration gradient (cs). The correct ratio may be expressed as a function of the apparent ratio: (see article) For large or small values of r, simpler approximations may be used: r greater than 1, ro approximately r r less than 1, ro approximately sigma4r These provide less than 20% error if sigma greater than 0.4 and r greater than 10 or r less than 0.1. Corrections for sigma less than 1 are applied to the classical (k1/k2) permeability constants of ethylene glycol, glycerol and propanol for bovine eryghtocytes. The sigma values for monoacetin and diacetin are predicted to be 0.7 and 0.6, respectively, on the basis of the deviation of their (k1/k2) constants from the expected relationship to partition coefficient and vapor pressure.