The Cournot-Bertrand model of oligopoly was introduced in the 1970s. It involved some firms being price setters while other firms were output setters. However there remained a question about the identity of the entity which changed the prices of the output setting firms. An alternative formulation of this problem is set out where in an industry there is a price leader and a price follower. But the price follower is a Stackelberg leader. The price follower tries to estimate the reaction function of the price leader through an iterative learning process and incorporate the former into its profit maximisation exercise. Thereby the price follower tries to factor in the consequences of changes in its output on the price it receives. The equilibrium, local stability and analogy of the model with respect to the Cournot-Bertrand model of oligopoly are examined. Such a set-up is compatible with constant returns to scale of the production process of the price follower.