1. An alternative explanation for the kinetic data obtained by Bachelard (1971) for the brain hexokinase reaction is presented. 2. Apparently sigmoidal saturation curves for MgATP2− based upon Bachelard's (1971) studies can be corrected to hyperbolic curves by use of a stability constant for MgATP2− complex formation. 3. A number of other effects related to the concentration-dependent stability of the MgATP2− complex and to the presence of the inhibitory free uncomplexed ATP4− concentration are also explained in terms of a non-allosteric role for either Mg2+ or MgATP2− fully consistent with a number of previous reports on this enzyme. 4. A brief discussion of the validity of Hill plots in studies of multisubstrate co-operative enzymes is presented. 5. A simple model is presented that demonstrates how enzymes obeying Michaelis–Menten kinetics can demonstrate sigmoidal velocity responses if the true substrate of the reaction is the metal–substrate complex.