Given the neoclassical assumptions of optimizing economic agents, perfect information, perfect competition, and productive efficiency, timber supply is a dynamic process. Different discrete-time dynamic timber supply models and their solution methods are compared and their common elements derived. A continuous-time model is derived, but not solved. The discrete-time timber supply model is then incorporated into a dynamic multi-sector model and a dynamic general equilibrium model. In the multi-sector model, all household's utility functions are aggregated into a single community utility function which is maximized subject to the technology of the economy. The technology for the forest sector is the same as in the discrete-time dynamic timber supply models. Wood is treated as an intermediate input into the production of consumer goods. The technology of the consumer goods sectors is based on the technology used in computable general equilibrium models. The optimal steady state problem for this model is discussed, and the solution for an example problem is presented. Disaggregating the utility function is necessary for modeling true general equilibrium. This greatly complicates the problem of Ending numerical solutions, but enriches the model considerably. The formulation of the general equilibrium model as an optimization problem is described, but proved rather difficult to solve. The optimal steady state problem can be solved using an algorithm developed by Scarf (1967) for finding fixed points of continuous functions. The fixed-point approach provides a reliable solution method and appears to have more potential for modeling departures from perfect competition than the optimization approach. The equivalence of the two approaches is discussed.

Ph. D.