Complete physiological understanding of HFV requires knowledge of four general classes of information: 1) the distribution of airflow within the lung over a wide range of frequencies and VT (sect. IVA), 2) an understanding of the basic mechanisms whereby the local airflows lead to gas transport (sect. IVB), 3) a computational or theoretical model in which transport mechanisms are cast in such a form that they can be used to predict overall gas transport rates (sect. IVC), and 4) an experimental data base (sect. VI) that can be compared to model predictions. When compared with available experimental data, it becomes clear that none of the proposed models adequately describes all the experimental findings. Although the model of Kamm et al. is the only one capable of simulating the transition from small to large VT (as compared to dead-space volume), it fails to predict the gas transport observed experimentally with VT less than equipment dead space. The Fredberg model is not capable of predicting the observed tendency for VT to be a more important determinant of gas exchange than is frequency. The remaining models predict a greater influence of VT than frequency on gas transport (consistent with experimental observations) but in their current form cannot simulate the additional gas exchange associated with VT in excess of the dead-space volume nor the decreased efficacy of HFV above certain critical frequencies observed in both animals and humans. Thus all of these models are probably inadequate in detail. One important aspect of these various models is that some are based on transport experiments done in appropriately scaled physical models, whereas others are entirely theoretical. The experimental models are probably most useful in the prediction of pulmonary gas transport rates, whereas the physical models are of greater value in identifying the specific transport mechanism(s) responsible for gas exchange. However, both classes require a knowledge of the factors governing the distribution of airflow under the circumstances of study as well as requiring detail about lung anatomy and airway physical properties. Only when such factors are fully understood and incorporated into a general description of gas exchange by HFV will it be possible to predict or explain all experimental or clinical findings.