This paper describes evaluation and testing of the Savannah River Site (SRS) impactor-type alpha continuous air monitor (CAM) to determine the particle delivery and collection efficiency in the device using fluorescence-labeled, polystyrene latex spheres and liquid droplet aerosols. Particle sampling efficiency through the dome-shape sampling inlet for particles with aerodynamic diameters of 6.2 and 10 mu m was 98% at 20 cfm (566 L min), 90% at 32 cfm (906 L/min), and 87% at 40 cfm (1132 L/min) sampling flow rates. Internal delivery efficiency through the CAM was > 94% at 20 and 32 cfm for 0.5-, 1.1-, 2.2-, and 3.2- mu m particles and > 90% for 6.2- mu m particles. For 10- mu m particles, the internal delivery efficiency was 91% at 20 cfm and decreased to 83% at 32 cfm and 77% at 40 cfm. The 50% cutoff aerodynamic diameter for the impactor was 3.2 mu m at 20 cfm, 2.6 mu m at 32 cfm, and 2.3 mu m at 40 cfm. For a typical radioactive aerosol in the workplace (activity median aerodynamic diameter of 5 mu m with a geometric standard deviation of 2), these cutoff diameters provide collection efficiencies of 74% at 20 cfm, 83% at 32 cfm, and 87% at 40 cfm. The normal grease layer of 1.5 mg that is routinely applied to the planchet of the SRS CAM was adequate to quantitatively retain all collected particles with diameters of 3.2 mu m or less at flow rates of 20, 32, and 40 cfm. For particle sizes of 6.2 and 10 mu m, approximately 80-85% of the particles were retained on the impactor planchet and 15-20% were re-entrained into the exhaust airstream due to particle bounce. The delivery and collection efficiencies in the SRS CAM can be combined to give overall efficiencies for detection of airborne actinide aerosols as a function of particle size. For 6.2 and 10 mu m aerodynamic diameter particles, the total collection efficiency of the SRS CAM at 20, 32, and 40 cfm exceeds the conservative assumption of 50% efficiency for plutonium particle collection that has been traditionally used at SRS.