An assessment of the three-population differential count on the Coulter Counter Model S Plus IV.

The accuracy and precision of a three-population differential count on a Coulter Counter S Plus IV were assessed. In general a good correlation was found between manual and automated lymphocyte and granulocyte percentages although various abnormal samples caused discrepancies, and it was found that some eosinophils were categorized as mononuclear cells rather than granulocytes. The automated mononuclear cell percentage was found to be imprecise and often inaccurate as a reflection of monocyte numbers. The instrument sometimes failed to either reject or 'flag' samples containing an appreciable percentage of lymphoma cells, nucleated red blood cells, blast cells, or eosinophils but was found useful in reducing the number of differential counts required.