The study of oligopoly has progressed along two lines. The first deals with the behaviour of a firm viewing existing rivals as responsive to its actions. The critical assumption in this analysis is the firm's conjecture about rivals' reactions to its price or quantity decisions. Cournot [4] provided the classic supposition that rivals will maintain their current level of output in response to a change in the given firm's output level. This implies price matching by rivals. Questions addressed include those of existence and stability of equilibrium in an oligopolistic market, its possible convergence to the competitive solution as the number of firms increases indefinitely, and whether the approach is monotone (quasi-competitiveness). A recent synthesis is provided by Ruffin [11]. The second strand of inquiry focuses on the firm's behaviour regarding potential rivals, especially actions designed to preserve positive profits. The strategy of pricing to retard or preclude entry, limit pricing , has received the most intense study. A summary of earlier theoretical developments is provided by Bhagwati [2]; more recent contributions include Gaskins [6], Pyatt [10], Baron [1], and Kamien and Schwartz [8, 9]. The first of these two lines of inquiry, focused on interaction among existing rivals, may be viewed as a short-run theory of oligopoly. The second approach, with its attention to consequences of potential rivalry, forms the complementary long-run theory. This paper constitutes an attempt to bridge these parallel developments, an objective shared with Fisher [5]. In the next section we present a model of a firm thatviews its existingrivals in accordance with the classic Cournot assumption and views potential rivals in the manner posited in our paper [8]. The timing of rival entry is regarded as a random variable whose probability distribution is dependent on current industry price. After obtaining necessary conditions for maximization of the firm's long-run expected profits, we extend the first four theorems of Ruffin's paper to the case in which potential rivalry is recognized. The classic Cournot oligopoly model is compared with the present one and our results are summarized. 2. THE MODEL