This column provides a ranking of presently available personal samplers designed to estimate the inhalable fraction of airborne aerosol. Specifically, the IOM, I CIPI0-I(2) and GSP(3) samplers exhibited the lowest bias. The inhalable convention specifies a particle size dependent efficiency with which an acceptable sampler is expected to capture aerosol particles. The inhalable, together with the respirable and thoracic, conventions have been adopted by the International Standards Organization, the Comite Europeen de Normalisation, the American Conference of Governmental Industrial Hygienists ACGIH (4) and by the American Society for Testing and Materials. The inhalability convention was developed (5-10) so as to mimic aerosol capture by the human head and is in line with many health effects studies reviewed(11) by ACGIH. The convention is specified in terms of a curve giving the sampling efficiency versus aerodynamic aerosol diameter defined so as to result in equivalent motion of a unit density sphere in air up to 100 Ám, representative of human exposure, for wind speeds less than 4 m/s, averaging over all wind directions. As such, the largest particles of significance to health are covered. Since large aerosol particles are difficult to handle experimentally, much less information is available for characterizing inhalable aerosol samplers than, for example, respirable samplers. Nevertheless, information exists which suggests choices among several specific samplers available at present. The most recent comparison of presently used inhalable aerosol samplers is that carried out under the auspices of the European Commission EC, (12,13) which generally corroborates earlier sampler studies.(1,14) This research was begun at the Warren Spring Laboratory, Stevenage, United Kingdom, and, upon the closure of this laboratory, was completed at the Atomic Energy Authority Harwell United Kingdom. Eight samplers in present use were tested: CIPlO-I France, 37-mm closed-face cassette Spain and United States, 37mm open-face cassette Sweden) PAS 6 Netherlands, PERSPEC Italy, GSP Germany, IOM United Kingdom, and the seven-hole sampler United Kingdom. Conditions of the experiment included measurements of sampler collection efficiencies for aerosol particles as large as 100 Ám at wind speeds of 0.5, 1.0, and 4 m/s. Samplers were positioned on a manikin rotating slowly within a wind tunnel. The samplers were all purportedly conductive, the 37-mm plastic cassette samplers having been painted with a conductive coating. The results of this experiment indicated negligible intersampler variability within each sampler type. However, each sampler type exhibited bias relative to the sampling convention. Within the EC study, the most reliable data pertain to the lower ambient wind speeds 0.5 or 1.0 m/s, which therefore become the focus of this column. A recent study 15 of wind speed characteristic of indoor workplaces indicates a mean wind speed of about 0.2 m/s, including normal worker, motion, and a 95 percentile range of up to 0.8 m/s. Thus the range considered here evidently extends beyond indoor workplace conditions to a degree. The samplers discussed in this column were ranked in terms of the bias relative to the convention as follows: Experimentally determined sampling efficiencies versus aerodynamic diameter were modeled mathematically. Correction factors were computed by minimizing the absolute value of bias of mass concentration estimates relative to the ideal inhalability convention over aerosol size distributions characterized by a mass median diameter MMD of <50 Ám and a geometric standard deviation GSD of 1.5. This range of size distributions is considered representative of components of aerosol size distributions found in the workplace16 see also Reference 8 and the CEN Prestandard on Workplace Atmospheres: Assessment of the Performance of Instruments for Measurement of Airborne Particles. The correction factors were optimized so as to accommodate wind speeds of 0.5 and 1.0 m/s simultaneously. Samplers were considered acceptable regarding sampling efficiency if, for any size distribution in the test range, bias is 50 percent at 95 percent confidence in the sampler characterization. This column then provides a list of samplers together with conversion factors for sampling according to the inhalable aerosol sampling convention. The suggestions are based on limited data existing at present. As information on these or other samplers and physiological inhalability itself develops, the list, together with its calibration factors, may change.