Myocardial cells are conventionally maintained in a healthy state in tissue culture by including animal serum in the incubation medium. However, serum-containing media are complex, undefined, and variable mixtures of hormones, nutrients, binding proteins, growth factors and other constituents that may hinder biochemically defined experiments. In addition, serum-associated mitogens stimulate proliferation of non-myogenic cells that usually contaminate these cultures. Thus, long-term metabolic experiments on cultured cardiac cells usually involve a mixture of cell types. We demonstrate here that rat myocardial cells can be maintained in a viable, contractile state for many weeks in a simple serum-free medium that contains certain chemically defined supplements. We screened various serum constituents that are known to be anabolic for myocardium, but that are not mitogenic for mesenchymal cells (e.g. fibroblasts). We found that a mixture of insulin, thyroid hormone, testosterone, and the iron-binding protein transferrin maintained levels of protein synthesis in myocardial cell cultures near the levels achieved with fetal bovine serum supplement, and reduced catabolism of proteins measured directly and by monitoring net changes in protein levels in the cultures. Although insulin alone accounted for the amelioration of protein balance, the other hormone supplements contributed to long-term preservation of contraction and morphologic integrity. In contrast to serum, these supplements alone or in combination did not support the proliferation of non-contractile fibroblastoid cells isolated from rat myocardium. Application of this novel approach should facilitate many types of heretofore difficult studies that require biochemical definition and cellular homogeneity.