Three families with polycythemia inherited through apparently different modes are described. Secondary causes of polycythemia were ruled out. Erythropoietin (EPO) levels were normal or low, even after phlebotomy. In vitro erythroid colony growth in standard assay cultures containing EPO was normal; however, in the absence of added EPO, a few progenitors from most of the affected individuals were able to generate recognizable colonies of mature erythroblasts, although these were smaller and proportionately less numerous than seen in polycythemia vera (PV). To search for EPO-receptor changes as a possible pathophysiologic mechanism, we examined, by Southern blot analysis, genomic DNA samples from affected and nonaffected family members, as well as three patients with PV. Two different probes, derived from the human EPO-receptor, were used. We found no evidence for chromosomal rearrangements or gene amplification in hereditary polycythemia or PV patients. Further, no nucleotide sequences were found that were homologous to the Friend spleen focus-forming virus glycoprotein gp55, which has been shown to bind to and activate the murine EPO-receptor. Functional studies examining number and binding affinity of the EPO- receptor on erythroid progenitors from three hereditary polycythemia patients demonstrated no abnormalities. We conclude that the mechanism(s) for the erythrocytosis in familial and congenital polycythemia and in PV may not involve the EPO-receptor and, therefore, may result from alterations of postreceptor responses.