The traditional dichotomy between cycle theory and growth theory is no longer considered justifiable except in pedagogically simplified models. The expression ‘cyclical growth’ is therefore more perspi­ cuous than such artificial phrases as ‘trends and cycles’ and ‘business cycles and economic growth’. For that former expression at once reveals a new epistemological propensity to treat business cycles as an integral part of secular growth or to regard cycle theory as a special branch of general dynamic economics. Despite this methodological innovation there still remain a number

of substantive problems to be thrashed out, especially the conceptual problems of linear versus non-linear assumptions about cycle-growth makers and exogenous versus endogenous approaches to oscillatory mechanisms. This chapter1 is addressed to the proposition that the theory of cyclical growth must give emphasis to the non-linear and endogenous factors affecting economic change if it is to serve the scientific purposes of prediction and control most effectively. Specifically we shall discuss (a) the general nature of a comparativedynamic system capable of generating cyclical growth, (b) the oscil­ latory mechanism of a self-limiting type, and (c) the diagrammatic analysis to implement the abstract formulations of the preceding sections.