for any theory is to understand some segment of reality. If it is beyond man's capability to change this reality, his environment, then at least with his knowledge of it he can better adapt to it. In other cases where man does have some power over the environment, the understanding of it points the way for directed change or planning. If diffusion concerns the spread of a phenomenon through a population over space and through time, then planners would presumably find diffusion valuable to at least two kinds of planning problems: (1) dissemination of knowledge and encouraging use of existing or planned facilities or services and, more generally, (2) inculcating in a population some desired change in attitudefor example with respect to willingness to integrate, or maintenance of property. In this paper we will be concerned with the usefulness of diffusion theory, in its present state, for resolving such problems, and with how diffusion theory can be generalized or improved to make it more clearly useful. The paper is divided into three parts. We first deal with the significant aspects of the application of diffusion theory to past phenomena. Next, we treat the application of diffusion theory to the planning process, and finally, we discuss some potentially valuable generalizations of diffusion theory and their consequences for the usefulness of the theory, particularly to planning. DIFFUSION IN THE DESCRIPTION OF PAST EVENTS