The determination of salivary drug concentrations is one of the major domains for the application of saliva in laboratory medicine. Its usefulness, however, has been criticized for many drugs because of the variability of the saliva/plasma concentration ratio (saliva/plasma ratio).Considering saliva as a model for transmembrane transport, drugs can be divided into 4 groups. The first group is characterized by a saliva/plasma ratio less than 1.0. In extreme cases a drug with negligible transport is either not detectable in saliva or is found in very low concentrations which are hard to measure or difficult to interpret.A second group leads to saliva/plasma ratios which are approximately constant and about 1.0 under most of the conditions studied. This group is ideal for monitoring salivary drug concentrations.A third group is sufficiently transferred into saliva, but the saliva/plasma ratio varies under different conditions. The reason for this variation is that the transport is influenced by several factors including an active transport mechanism. The varying influence of these factors on the saliva/plasma ratio largely depends on the physicochemical characteristics of the particular substance.In a fourth group very high saliva/plasma ratios are observed primarily due to the degree of ionization of weak bases.A representative of each group is presented with its saliva/plasma ratio and its physico-chemical properties: ceftazidim, ethanol, digoxin and prilocain. In all cases the salivary concentration probably reflects the intracellular concentration in target tissues. All examples confirm saliva as an ideal in vivo model for the study of transmembrane transport in the human organism.