Sidney Winter's [1971] Satisficing, Selection, and the Innovating Remnant was directed toward the establishment of an important and seminal theorem. The import of the theorem was that, givan a number of unexceptional sufficient conditions, a Markov process will, with certainty, result within a finite time in a textbook industry equilibrium. All firms in the industry will achieve zero profits, and there will be no potential entrants who could achieve positive profits by entering the industry. The conditions that were claimed to be sufficient were the following. 1. All firms in the industry have access to a single technology. 2. The industry faces a vector-valued price function mapping inputs and outputs into prices. 3. Some input-output vectors will yield net profits at scales sufficiently close to zero, and there are scales sufficiently large that no input-output vector can yield net profits. 4. A standard competitive equilibrium exists. 5. Each firm in the industry has a production technique and a capacity at each period. The capacity determines the scale of the realized input-output vector. 6. The change in the number of plants held by each firm at time t + 1 depends probabilistically on the actual profitability (for incumbents) or notional profitability (for potential entrants) of its production technique at time t. In other words, the capacity changes of firms are modeled as a first-order Markov process. 7. The weak axiom of revealed preference holds with respect to the price function. The Winter theorem rests on the proposition that until equilibrium is reached, there is a positive probability that the difference between current output and equilibrium output will diminish and implicitly that that diminution will not be reversed. In fact, I can find no proof of the italicized statement in any of Winter's specifications of the theorem. Indeed, the italicized statement is a necessary condition for Winter's theorem to be correct, but the statement does not hold in the conditions specified by Winter.