Chronic rat inhalation studies have shown that a number of different particle types can induce significant adverse effects, including impaired lung clearance, chronic pulmonary inflammation, pulmonary fibrosis, and lung tumors. These effects occurred when highly insoluble particles of low solubility and low cytotoxicity were inhaled in long-term studies. Inhaled concentrations ranged from a few milligrams per cubic meter up to 250 mg/m(3). This wide range of inhaled concentrations may indicate that the particulate compounds have differed largely in their toxicity. This view appears to be supported by the fact that cytotoxic crystalline SiO2 shows very similar effects after much lower inhaled concentrations. However, although administered doses are customarily expressed in units of mass, this may not be the appropriate dose-metric for a correlation with observed effects. For example, effects on alveolar macrophage (AM) mediated clearance of particles could best be correlated with the volumetric lung burden of different particle types, suggesting that the particle volume phagocytized by AM is an appropriate dose parameter for this endpoint. On the other hand, the inflammatory response induced by a number of different particle types could best De correlated with tile surface area of the particles retained in the alveolar space. In addition, total surface area of retained particles was the best dose parameter for a correlation when the endpoint was lung tumors. In all of these studies crystalline SiO, did not fit into the overall exposure-response or dose-response relationship, clearly demonstrating that SiO2 is a very different (more cytotoxic) particle type. Particle size and surface area can play important roles in tile response to inhaled particles, which is especially relevant for ultrafine particles, Inhalation studies with rats exposed to aggregated ultrafine TiO2 and carbon black showed that both compounds induced lung tumors in rats al considerably lower gravimetric lung burdens than larger sized TiO2. However, the different ultrafine particle types did also show differences in the strength of response that cannot be explained by differences in surface area only. Analyses of inhalation studies with ultrafine particles show that the movement of particles from alveolar spaces into interstitial sites appears to reflect the ability of inhaled ultrafine particle aggregates (TiO2; carbon black) to break down into smaller units, or even singlet particles. Further data are needed to evaluate the importance of interstitial cell-particle interactions for the long-term effects. The lung tumor response in rats after chronic high-dose particle inhalation has been suggested to be a at-specific response that may not be relevant to humans. However, lacking an understanding about mechanistic events, the rat model should not be dismissed prematurely. What should be questioned instead is the relevance of using excessively high exposure concentrations of particles in a rat study. Exposure-response and dose-response relationships for different endpoints indicate the existence of a threshold below which no adverse effects may occur. Such a threshold could be explained by overwhelming specific defense mechanisms in the respiratory tract, such as particle loading of macrophages (prolongation oi particle clearance), or limitations of pulmonary antioxidant capacities (inflammatory response). It appears, however, that duration of exposure plays a significant role that can result in a shift of exposure-dose-response relationships and a shift of a threshold when these relationships are compared at the end of a subchronic study versus the end of a chronic study. This shift will cause difficulties for defining a threshold as well as a maximum tolerated dose from results of a subchronic particle inhalation study.