A two-component aerosol system is investigated using the MSPEC code, which models the dynamic behavior of particle composition as a function of particle size. The predicted aerosol concentration behavior is shown to be sensitive to several parameters and model choices, in contrast to the situation for singlecomponent aerosol systems, where these parameters and models appear to play a distinctly uncritical role. In addition, the predicted aerosol concentration behavior is shown to significantly diverge from that predicted by MSPEC using a “single-component” model mode that assumes uniform particle composition across the size distribution. This latter mode is common to codes presently used for nuclear accident source term evaluations. These findings point to the need for an expanded experimental data base, both to validate multiple-component aerosol behavior codes and to supply the necessary data to drive them.