The purpose of this investigation was to determine if turbulent mixing created by the ejector causes change in the size distribution of particles passing through it. The results using dry solid sodium chloride (NaCl) particles show no discernible difference in the geometric mean diameter and the geometric standard deviation of particles passing through the ejector. Similar results were found for the dioctyl-phthalate (DOP) particles. In other words, there was no apparent shift in the center and spread (measured by the geometric standard deviation) of the particle size distributions. The ratio of the total number concentrations before and after the ejector indicates a dilution ratio of approximately 20 applies equally for both DOP and NaCl particles. The result suggests the dilution capability of this ejector was not affected by the particle composition. When the variability of the particle counts (in coefficient of variation, CV) in individual size ranges was examined, we found that the bin-to-bin variability was much greater for DOP than for NaCl particles. Although the first and second moments of a distribution (i.e., the geometric mean diameter and geometric standard deviation) were not altered significantly for the DOP particles by the ejector, the skewness measure indicated otherwise. The ejector may modify the particle size distribution of deformable particles such as DOP. The results suggest that it may be difficult to precisely characterize the size distribution of particles using the ejector technique if the particle is not solid. This result has an implication for the use of the ejector as a dilution tool, and one needs to know that the particles to be diluted with such a device are reasonably dry and/or non-deformable under the turbulent condition inside the ejector. (C) 2002 Elsevier Science Ltd. All rights reserved.