The solvent composition dependence of the partial molal volume of the transition state of benzyl chloride solvolysing in water–ethanol mixtures is compared with that of anilinium chloride and a number of quaternary alkyl ammonium chlorides. It is concluded that anilinium chloride is not a good electrostatic model for the benzyl chloride transition state even though it may be a fair structural model. The partial molal volume of tetraethylammonium chloride is very similar to that of the transition state suggesting that charge development in the activation process is far from being fully ionic.