Enclosure-based methods (i.e., flux chambers) have been widely used in agricultural, ecological, geophysical, and engineering studies to estimate gas exchanges at the soil-atmosphere and the water-atmosphere interfaces. In this study, the flux chambers are analyzed using diffusion theory and mass balance principle. Mathematical models are developed to simulate the general behavior of both closed and dynamic chambers. Simulation for the closed chamber behavior shows that the flux from the enclosed soil matrix into the chamber decreases with time after chamber placement. This indicates that application of a simple linear model to calculate flux may underestimate the real flux, even though the concentration data obtained from the chamber headspace shows a relatively linear increase with respect to time. It is recommended that nonlinear models be considered whenever possible for calculating flux for closed chambers. Simulations of dynamic chambers show that (1) these chambers can reach a steady state rapidly after placement and (2) the proper measurement of flux depends on both chamber operational conditions and soil permeability to air. A dynamic chamber may underestimate the actual flux when operating on low permeable soils. On soils with high air permeability a dynamic chamber may give an underestimate of the actual flux when operating at low airflow rate but an overestimate when the airflow rate is high. Theoretically, both closed and dynamic chambers may produce accurate flux estimates if they operate under ideal conditions and appropriated models are used in flux calculations. In practice, however, a dynamic chamber should be more desirable. ¿ 1998 American Geophysical Union