OUR knowledge as to the constitution of the coloured compound formed by the interaction of phenol with freshly prepared ferric chloride solution is very meagre. Brode and Wesp1 suggested that the violet-red colour is probably due to an ion of the type [Fe(OPh)6]-3; but they have not presented any evidence in its favour. We have studied the reaction by four different methods: thermometric, conductometric, absorption and transport measurements. Thermometric and conductometric titration curves of phenol with ferric chloride solutions show two breaks corresponding to Fe+3: phenol equal to 1 : 3 and 1 : 6; so these may be supposed to represent the formation of the compounds Fe(OPh)3 and H3[Fe(OPh) 6]. By applying Job's method of continued variation, it is found that, for equimolecular solutions, the maxima in the optical density curve against the volume of one of the reactants appear at the point Fe+3: phenol equal to 1:3. The maxima of the curves are found to be independent of the wave-length (490 μ, 550 μ) and the concentration of the reactants used, and so are due to the formation of the compound Fe(OPh)3. Now the compound Fe(OPh)3 may also be formulated as Fe[Fe(OPh)6], in which case also the ratio of Fe+3: phenol is equal to 1:3. When ferric chloride - phenol solutions are shaken with non-polar organic solvents, the coloured compound is not found to be partitioned between the two liquid phases, and this supports the formulation of Fe(OPh)3 as Fe[Fe(OPh)6]. Finally, transport experiments by Duval's apparatus with solutions containing ferric chloride and phenol in the ratio 1:3 show that iron goes to the anode and thus give additional and direct evidence in favour of the formulation Fe[Fe(OPh)6]. © 1950 Nature Publishing Group.