When red cell mean corpuscular volume (MCV) is determined by use of the Coulter counter, the blood sample is first diluted in a solution with fixed osmolality (Isoton). Therefore we studied the interference of blood osmolality with MCV measured by this method. In 14 patients with hyposmolality, the correction of osmolality was accompanied by an increase in MCV of 4.7% (p < 0.001) and a decrease in mean corpuscular hemoglobin concentration (MCHC) of 4.8% (p < 0.001). In 9 patients with hyperosmolality the decrease in MCV was of 5% (p < 0.001) and the rise in MCHC of 4% (p < 0.001) after osmolality correction. Before correction of hyposmolality, 1 patient had false microcytosis and 3 had masked macrocytosis. In the hypertonic group 3 patients had initially false macrocytosis. Red blood cells (RBC) from hypotonic patients probably shrink when they are acutely placed in the Isoton which is a hypertonic solution (330 mosm/kg H2O). Conversely RBC from patients with severe hyperosmolality swell in the same conditions. The patients osmolality must be considered to interpret the MCV measured by the Coulter counter correctly, as a 10-mosm/kg H2O change in serum osmolality is responsible for an artefactual change in MCV of 1 f1.