Isotope effects have been measured with Mg2+ as the activator and L-malate labeled with deuterium or tritium at carbon 2 as the substrate over the pH range 4-10. Comparison of the nearly pH-independent deuterium-isotope effect on V/Kmalate of 1.5 with the tritium effect of 2.0 by the method of Northrop (Northrop, D.B. (1975), Biochemistry 14, 2644) gives limits on the true effect of deuterium substitution on the bond-breaking step of 5-8 in the forward reaction and 4-6.5 in the reverse direction. Comparison of the deuterium effect on V/K with the 13C-isotope effect of 1.031 reported by Schimerlik et al. (Schimerlik, M.I., Rife, J.E., and Cleland, W.W. (1975), Biochemistry 14, 5347) allows the deduction that at pH 8 reverse hydride transfer is six to eight times faster than decarboxylation, which is thus largely rate limiting for the catalytic reaction. The absence of a deuterium-isotope effect on V at pH 7-8 and comparison of the Ki of pyruvate as an uncompetitive inhibitor of the forward reaction and a substrate for the reverse reaction indicate that at neutral pH the release of TPNH from enzyme-reduced triphosphopyridine nucleotide (E-TPNH) is the rate-limiting step in the forward direction. The observation of a deuterium effect on V that approaches 3 at pH 4 and 10 shows, however, that, at very low and very high pH, hydride transfer may become partly rate limiting. In the reverse reaction, the probable rate-limiting step at pH 7 is the isomerization of E-TPNH, while at pH 8.5 and above V becomes too large to measure and appears infinite. Substitution of Co2+, Ni2+, or low levels of Mn2+ for Mg2+ gives similar deuterium-isotope effects, although the values of V and Kmalate vary considerably with metal. The kinetics of Mn2+ show pronounced negative cooperativity, with Km values of 7 muM and 5 mM for concentration ranges from 4 to 100 muM and over 1 mM.