There's a small error in the diagrams shoring the deadweight loss from the average cost pricing solutions. The area below P*, above the MC curve, and in between Q_AC and Q* should also be included in the deadweight loss triangle under average cost pricing.