Micronucleated blood cells--whether generated spontaneously or by clastogen treatment--are present in the blood and bone marrow as rare events. Historically they have been scored manually by microscopic inspection which is labor-intensive and tedious. It has been recognized by investigators that a need exists for an automated method which can accurately, objectively and quantitatively score rare micronucleated cells. In order to improve assay statistics more cells must be processed, making high-speed scoring an important objective of any automated procedure. Flow cytometry can provide the means to quantitatively analyze micronucleated cells at high speeds and with great accuracy once the chemical, biological and instrumentation conditions are optimized. Recent literature suggests that noise and fidelity of the data, as well as the sensitivity of present flow cytometers, are major obstacles that still must be overcome. Experiments are described herein which demonstrate that flow cytometry is able to score micronucleated cells under conditions where noise levels are low, and the fidelity and accuracy are high. In addition, the accuracy of scoring rare events is maintained at high speeds (e.g. 1,000,000 cells/min). A major emphasis of this manuscript is to demonstrate the means for evaluating the accuracy and sensitivity of the flow cytometer in scoring rare events. Both computer simulation and reconstruction experiments were used to gauge scoring accuracy and guided optimization experiments. These experiments demonstrate that when optimum conditions are used in conjunction with a suitable flow cytometer, it is possible to score micronucleated cells at high speeds with great precision.