The authors thank Jeanne Galbo and Cyndi Saunders for their expert word processing support i n the preparation of this manuscript. The authors would also like to thank the anonymous reviewers for their thoughtful and insightful comments on the submitted manuscript.
EMPIRICAL PARTICLE DOSIMETRY IN THE RAT 541Zhang and Yu (1 993) presented empirical equations for nasal fractional deposition in humans and small laboratory animals (mouse, hamster, rat, and guinea pig) using the published data of Raabe and colleagues (1 977, 1988). They did not, however, consider fractional deposition in the tracheobronchial (TB) or alveolar (A) regions of the respiratory tract.All of the empirical models of regional fractional deposition in humans and the ET model of Zhang and Yu (1993) fit independent equations to deposition efficiency (number or mass depositedhumber or mass entering that region) for each region. This approach treats the respiratory tract as a series of linked filters through which the aerosol passes (Rudolf et al., 1983;Heyder & Rudolf, 1977). Asgharian et al. (1995) recently published an empirical model of fractional A deposition in the human, monkey and rat. Their model differs from the other models referenced already in that their equations for the A region are calculated directly from deposition fractions rather than from deposition efficiencies. In contrast to regional deposition efficiencies that depend only on ventilatory parameters, airway geometry, and particle characteristics, regional deposition fractions depend on fractional deposition in preceding regions as well. This limits the usefulness of such a model but does provide a way to use published fractional deposition data in empirical models even when, as i s often the case, the regional efficiencies cannot be derived. In the model of Asgharian et al. (1 995), fourth-order polynomials were essentially eye-fit through published data on A deposition following exposure to monodisperse or near monodisperse particle size distributions. In a technique similar to that used by Kobrich et al. (1994) for humans, Asgharian and co-workers (1 995) then generated equations to estimate A deposition of polydisperse spherical particles by first integrating the monodisperse deposition equation with a lognormal distribution describing the polydisperse particle size distribution. Fourth-order polynomials were then fit to the resulting predicted polydisperse deposition fractions to provide an empirical model for polydisperse particle fractional deposition in the A region. The model of Asgharian and colleagues (1995) does not describe fractional deposition in either the ET or TB regions, although they discuss that the same procedures could be applied to these regions.In this article, we present an empirical model for the rat to calculate predicted fractional deposition of particles in the aerodynamic size range. This model consists of three independent deposition efficiency equations (ET, TB, and A regions). Multiplying predicted deposition efficiencies (or their complemen...