The control theory of Kacser and Burns [in: Rate Control of Biological Processes (Davies, D.D. ed) pp. 65-104, Cambridge University Press, London, 1973] and Heinrich and Rapoport [Eur. J. Biochem. (1974) 42, 97-105] has been used to quantify the amount of control exerted by different steps on mitochondrial oxidative phosphorylation in rat-liver mitochondria. Inhibitors were used to manipulate the amount of active enzyme. The control strength of the adenine nucleotide translocator was measured by carrying out titrations with carboxyatractyloside. In state 4, the control strength of the translocator was found to be zero. As the rate of respiration was increased by adding hexokinase, the control strength of the translocator increased to a maximum value of approximately 30% at approximately 80% of state 3 respiration. In state 3, control of respiration is distributed between a number of steps, including the adenine nucleotide translocator, the dicarboxylate carrier and cytochrome c oxidase. The measured values for the distribution of control agree very well with those calculated with the aid of a model for mitochondrial oxidative phosphorylation developed by Bohnensack et al. [Biochim. Biophys. Acta (1982) 680, 271-280].