Effects of pH and co-solvents on the bio-mimetic artificial membrane permeation assay were investigated to determine the optimal conditions for the prediction of oral absorption. The permeability (P(am)) of 33 structurally diverse drugs to the PC/PE/PS/PI/CHO/1,7-octadiene membrane system (bio-mimetic lipid (BML) membrane) was measured at pH 5.5, 6.5, and 7.4. The pH dependence of P(am) was in accordance with the pH partition theory. The better prediction of oral absorption (fraction of a dose absorbed) was shown under the pH 5.5 condition (r=0.866, n=25) and/or pH 6.5 (r=0.865, n=28), rather than pH 7.4 (r=0.767, n=24). Then, the appropriate conditions for determining the permeability of poorly soluble compounds were examined. Dimethysulfoxide (DMSO), ethanol (EtOH) and polyoxyethyleneglycol 400 (PEG 400) were added up to 30% to the transport medium as solubilizers. DMSO, EtOH and PEG 400 decreased P(am) of hydrocortisone and propranolol. For example, DMSO (30%) decreased P(am) of hydrocortisone by 60% and by 70% in the case of propranolol. DMSO and PEG 400 also decreased P(am) of ketoprofen. In contrast, EtOH produced an opposite effect on permeability, i.e. an increased P(am) of ketoprofen. Therefore, the high concentration of these co-solvents could lead to the under- or overestimation of drug permeability.