Four lines of evidence indicate that the Rrebs TCA cycle exists in vivo as a complex of enzymes—a metabolon—in the mitochondrial matrix. The evidence includes specific interaction of sequential enzymes, specific binding of the enzymes to the inner surface of inner mitochondrial membrane, increased rates of portions of the Krebs TCA cycle in a partially intact system compared to a soluble system, and phenotype changes in yeast cells lacking citrate synthase. Further studies of the yeast system using 13C metabolites yielded results that could be interpreted as channeling of substrates and surprisingly indicated that the symmetrical substrates, succinate and fumarate, might be transferred with their rotation restricted. Several experiments were carried out to establish the validity of this interpretation. Several of these experiments disproved the hypothesis, but other explanations of the data are still possible. Such experiments are being carried out to characterize this process.