Radioimmunoassays for erythropoietin are limited so far to a few specialized laboratories and this requires transport and storage of samples. We therefore tested the stability of immunoreactive erythropoietin in serum and plasma samples obtained from a uremic and a nonuremic anemic patient. No significant change in the concentration of immunoreactive erythropoietin was found in either serum or plasma samples for up to 14 days of storage. This type of stability was observed no matter whether the samples were stored at room temperature, 4 degrees C, or -20 degrees C. There was no difference between the estimates of erythropoietin in serum and heparinized plasma. Validity of the radioimmunoassay used in this study was demonstrated by parallelism of dilution curves of test specimens and the 2nd International Reference Preparation for erythropoietin and by a close correlation between the immunoreactivity and the bioactivity of the hormone, as assessed in the same samples by the exhypoxic polycythemic mouse bioassay. In conclusion the data obtained clearly indicate that the necessity of storage and transport of clinical samples does not limit the practicability of the radioimmunoassay for erythropoietin.