The diffusion process through a non-porous barrier membrane depends on the properties of the drug, vehicle and membrane. The aim of the current study was to investigate whether a series of oily vehicles might have the potential to interact to varying degrees with synthetic membranes and to determine whether any such interaction might affect the permeation of co-formulated permeants: methylparaben (MP); butylparaben (BP) or caffeine (CF). The oils (isopropyl myristate (IPM), isohexadecane (IHD), hexadecane (HD), oleic acid (OA) and liquid paraffin (LP)) and membranes (silicone, high density polyethylene and polyurethane) employed in the study were selected such that they displayed a range of different structural, and physicochemical properties. Diffusion studies showed that many of the vehicles were not inert and did interact with the membranes resulting in a modification of the permeants' flux when corrected for membrane thickness (e.g. normalized flux of MP increased from 1.25 ± 0.13 μg cm−1 h−1 in LP to 17.94 ± 0.25 μg cm−1 h−1in IPM). The oils were sorbed differently to membranes (range of weight gain: 2.2 ± 0.2% for polyurethane with LP to 105.6 ± 1.1% for silicone with IHD). Membrane interaction was apparently dependent upon the physicochemical properties including; size, shape, flexibility and the Hansen solubility parameter values of both the membranes and oils. Sorbed oils resulted in modified permeant diffusion through the membranes. No simple correlation was found to exist between the Hansen solubility parameters of the oils or swelling of the membrane and the normalized fluxes of the three compounds investigated. More sophisticated modelling would appear to be required to delineate and quantify the key molecular parameters of membrane, permeant and vehicle compatibility and their interactions of relevance to membrane permeation.