This reply covers four topics: (a) a brief comparison of the Westgren and Schrader (WS) and the Ladd and Martin (LM) formulations of blending processes in multiproduct plants, (b) a brief comparison of objective functions in the two models, (c) derivation of shadow prices of characteristics from the WS model, and (d) reevaluation of evidence on sign-optimality of the present corn-grading system. The statement of blending processes in the WS formulation of the second blending problem is clearly superior to the formulation in the LM paper. However, the LM model maximizes net revenue. WS maximize gross revenue. For reasons presented later, I believe that maximization of net revenue is more appropriate. The rest of this reply deals with a WS model in which the objective function has been changed to net revenue and which is referred to as the "revised WS model." The shadow prices for corn yielded by the revised WS model can be used to determine if a