Carbohydrate metabolism has been extensively studied in micro-organisms, animals and plants. The major metabolic routes are well established and many of the enzymes have been extensively purified and studied in vitro. Experiments with whole animals, perfused organs and isolated cells have established an overall picture of the regulation of carbohydrate metabolism, and the study of enzymes in vitro has provided further information on the fine controls that are available to the cell. However, it is still not clear how the regulatory properties of enzymes studied in vitro correlates with the controls observed in animal cells in vivo. Considerable information about the regulation of carbohydrate metabolism in bacteria has been obtained by isolating mutants with alterations in their ability to utilize carbohydrates. The development of techniques to culture animal cells has now advanced to the stage where single-cell clones of both wild-type and mutant phenotype may be isolated and studied. Animal cells with alterations in carbohydrate metabolism would allow us to exploit an enormous body of information to ask specific questions about the regulation of carbohydrate metabolism in animal cells in culture. We have initiated a programme to study carbohydrate metabolism in an established line of Chinese-hamster ovary cells (CHO-K1 ; CC1-61) and to isolate such mutants. Cells were grown on Ham’s F12 medium, with the appropriate carbon source added, supplemented with the macromolecular fraction of foetal bovine serum (macro-serum; 8 %) and with the addition of antibiotics. Medium containing macro-serum but lacking an added carbon source contains less than 10pwglucose and is unable to support the growth of single-cell clones. CHO-K1 cells grow on a number of hexoses, but are unable to grow on pentoses, C3 compounds such as pyruvate, or intermediates of the tricarboxylic acid cycle (Table 1). Growth on disaccharides is also limited, and with maltose it was discovered that growth was permitted only because of the presence of an active maltase in the serum. These results are similar to those reported by Eagle et ul. (1958) for HeLa cells. They also reported that HeLa cells would grow on ribose in the presence, but not in the absence, of added pyruvate; we have been unable to confirm this for CHO-K1 cells. Auxotrophic mutants of CHO-Kl cells have been obtained after treatment with various mutagenic agents and negative selection with 5’-bromodeoxyuridinecontaining medium (Puck & Kao, 1967). It has been reported that the bromodeoxyuridine technique requires modification outside of the United States of America (Thirion etul., 1976). We have subjected CHO-Kl cells to mutagenesis with ethyl methanesulphonate and N-methyl-N’-nitro-N-nitrosoguanidine and have applied the bromodeoxyuridine selection technique to isolate successfully glycineand thymidine-requiring auxotrophs. These mutants will be used to facilitate genetic studies of our carbohydrate mutants. After mutagenesis with N-methyl-N-nitro-N-nitrosoguanidine, various numbers of cells were plated in media containing different carbohydrate sources unable to support the growth of wild-type cells in an attempt to isolate cells that could now utilize such a carbohydrate. After several weeks incubation, small colonies were seen in plates containing lactose, sucrose and ribose, but none in plates containing lactate, pyruvate, citrate, succinate, fumarate or malate. A selection of the colonies were picked to grow up for further studies, but so far only the colonies growing in ribose medium have been successfully transferred. These clones have been purified by recloning and their growth