The interpretation of labile Al, as defined operationally by a 15‐s reaction with 8‐hydroxyquinoline, remains problematic. In solutions free of complexing ligands other than OH⁻, one question is whether the chemical species (Al(OH)⁺₂ should be included. To resolve this issue, labile Al was measured in Al(ClO₄)₃ solutions at total Al (Al_T) concentrations in the range 4 to 60 µmol kg⁻¹ and pH in the range 4 to 6. In solutions equilibrated for 4 h without shaking, no effect of pH in the range 4 ≤ pH ≤ 5 on the values of labile Al was observed. Thus, Al³⁺ and all significant mononuclear hydrolytic species of Al that formed under the experimental conditions were reactive with 8‐hydroxyquinoline within 15 s. Shaking the sample solutions at pH 5 throughout equilibration, however, produced a gradual decline in labile Al with concomitant changes in light scattering that indicated the formation of colloidal or macromolecular Al species that were unreactive with 8‐hydroxyquinoline during a 15‐s period. Even in unshaken solutions, a decline in labile Al (and the inferred formation of colloidal Al species) was observed at pH 5 if solutions with Al_T > 1 mmol kg⁻¹ were used to prepare standard solutions for calibration. These results are important not only for the application of the 15‐s 8‐hydroxyquinoline method, but also for any experimental study of aqueous Al at pH values where Al hydrolysis is significant.