The mechanisms of retention of two recent stationary phases of interest in lipophilicity measurements, namely of the silica-based Discovery-RP-Amide-C16 phase and the polymer-based ODP-50-4B phase, were assessed and compared. A set of 41 model solutes and drugs with well-defined solvatochromic parameters were selected to allow a broad distribution of property spaces. The chromatographic results showed that, under the conditions of study, the lipophilicity index log k w obtained with the former stationary phase was more closely related to experimental log P oct values than was log k w obtained with the ODP-50-4B phase. Linear solvation/ free-energy relationship (LSER) analyses showed that the retention mechanisms of the two stationary phases are different, retention on the Discovery-RP-Amide-C16 phase and partitioning in octan-1-ol/H 2 O being controlled by the same balance of intermolecular forces (Van der Waals volume V w , H-bond acceptor basicity b, and dipolarity/polarizability p*).Introduction. ± The lipophilicity of solutes, traditionally expressed by their partition coefficients in the octan-1-ol/H 2 O system (noted log P oct ), is of high significance from both a physicochemical and a pharmacological viewpoint [1] [2]. The partitioning process expresses the combined effects of a number of intermolecular forces between a solute and its environment, here the solvents between which the solute partitions. These intermolecular forces are of particular importance in pharmacology since they also control the partitioning of solutes into biomembranes. Numerous studies have reported a relationship between log P oct and absorption or permeability in cell cultures and tissue preparations used as models of, e.g., the gastrointestinal tract or the bloodbrain barrier [3 ± 7].The reference procedure to measure log P oct is the shake-flask method which, however, is time-consuming and limited in range (ca. À 3 < log P < 4). Beyond these limits, log P oct values measured by the shake-flask method become unreliable.The reversed-phase HPLC method is a promising alternative to the shake-flask method, having such advantages as a higher throughput, an insensitivity to impurities or degradation products, and a broader lipophilicity range. In reversed-phase HPLC, lipophilicity indices are derived from the capacity factor log k, which is calculated by Eqn. 1, where t R and t 0 are the retention times of the solute and of an unretained compound, respectively. Some workers have used isocratic log k values measured in an appropriate mobile phase as a lipophilicity parameter [8 ± 10]. However, many more investigators use capacity factors extrapolated to 100% H 2 O (log k w ) to eliminate organic-solvent effects [11 ± 15], and they have indeed demonstrated the usefulness of