The Coulter VCS is a tridimensional flow cytometer designed for differential leukocyte counting on the basis of cell volume, light scatter and conductivity. The VCS differential performed on 192 non-flagged samples was compared with the blood smear, taken as the reference method. A good correspondence was found between the two methods. Two hundred and seventy pathological samples were studied by both the VCS and the reference method: 219 samples with no anomalies other than quantitative changes and 51 with abnormal cells. 41% of the quantitative changes presented without any flag; 59% were flagged: either isolated X5 flags (30%) or other flags needing blood smear review (29%). Samples with abnormal cells (51) were detected with one or more flags: X3 + X6 in myelocytic disease X2 + X7 in lymphocytic disease. Only two (1 hairy cell leukemia, 1 lymphoma) were not flagged; however in these cases the abnormalities were revealed by the presence of either a striking neutropenia or a lymphocytosis. This corresponded to a false-negative rate of 0.4% in the group studied. Further studies were performed on patients with X5 flagging only. The VCS differential was compared to the optical differential as a reference method. There was a good relationship between the two methods; thus X5 flagging could be established without further slide review provided the numerical values were within the range valid for the laboratory. Thus, in our system of working, both VCS flags (except isolated X5) and laboratory performance limits were taken into account as regards the need for blood smear review. This resulted in a review rate of 15%.