Rate constants for the reaction catalyzed by alanine aminotransferase were measured by 13C saturation transfer and inversion transfer between the enriched methyl groups of pyruvate and alanine. At a high enzyme concentration four techniques provided a reasonable estimate of parameters. (1) saturation transfer, (2) nonlinear least-squares fitting of inversion-transfer data, (3) linear least-squares fitting of the initial slopes of inversion-transfer data, and (4) a combination of initial slope and saturation-transfer data. At a lower concentration of enzyme, the saturation-transfer method provided the most accurate determination of rate constants. An inversion-transfer experiment produces a series of measurements of magnetization at various delays after inversion from which the parameters of interest may be determined by nonlinear least-squares fitting. Error in the determined parameters is influenced by the delays at which data are acquired, the number of data points acquired, and error in the measurement of magnetization. In the presence of significant noise in the data, the complete inversion-transfer experiment (inversion of both sites, sequentially) must be analyzed for an accurate determination of the parameters. If the inversion-transfer experiment is performed, an experimental protocol that minimizes error in the parameters may be designed to efficiently acquire inversion-transfer data if an estimate of the parameters and error in measurement of magnetization is available. The most efficient approach is the combination of initial slope data from the inversion-transfer experiment with measurement of equilibrium magnetization during saturation.