A line of Buffalo rat liver cells (BRL 3A) that multiplies in the absence of serum produces a family of polypeptides termed MSA that can partially satisfy the serum requirement for growth of chick embryo fibroblasts. Temin, Pierson and Dulak (1972) proposed that BRL cells multiply in serum-free medium because they produce MSA. This does not appear to be the case. We have studied three BRL cell lines: 3A2 and 3A have diverged from the same original isolate from normal liver; 61t is a spontaneous transformant of a different isolate. All three cell lines showed a 10 fold increase in cell number during 5 days in serum-free medium. However, 3A-conditioned medium stimulated 3H-thymidine incorporation into DNA in chick embryo fibroblasts and human skin fibroblasts; 3A2- and 61t-conditioned media did not. After ion-exchange chromatography or gel filtration of the conditioned media and measurement of MSA by 3H-thymidine incorporation or radioreceptor assay, MSA again was found in the 3A medium but not in the 3A2 or 61t media. The absence of MSA in the 3A2 and 61t media was not due to inactivation of MSA by these two cell lines. Addition of partially purified MSA to 3A2 cells did not increase their multiplication rate in serum-free medium. We conclude that the ability of the BRL cells to multiply in serum-free medium is independent of the level of MSA in the medium.