L-Lactate dehydrogenase (L-LDH) from Saccharomyces cerevisiae and L-mandelate dehydrogenase (L-MDH) from Rhodotorula graminis are both flavocytochromes b2. The kinetic properties of these enzymes have been compared using steady-state kinetic methods. The most striking difference between the two enzymes is found by comparing their substrate specificities. L-LDH and L-MDH have mutually exclusive primary substrates, i.e. the substrate for one enzyme is a potent competitive inhibitor for the other. Molecular-modelling studies on the known three-dimensional structure of S. cerevisiae L-LDH suggest that this enzyme is unable to catalyse the oxidation of L-mandelate because productive binding is impeded by steric interference, particularly between the side chain of Leu-230 and the phenyl ring of mandelate. Another major difference between L-LDH and L-MDH lies in the rate-determining step. For S. cerevisiae L-LDH, the major rate-determining step is proton abstraction at C-2 of lactate, as previously shown by the 2H kinetic-isotope effect. However, in R. graminis L-MDH the kinetic-isotope effect seen with DL-[2-2H]mandelate is only 1.1 +/- 0.1, clearly showing that proton abstraction at C-2 of mandelate is not rate-limiting. The fact that the rate-determining step is different indicates that the transition states in each of these enzymes must also be different.