This consideration brings us at once to the intricate question of reasonable rates, which involves the matter of reasonable design and construction of the property. In most cases the working capacity of the plant must be much greater than the average annual demand for the service performed, as so-called "peak loads" of relatively short duration must be provided for. The magnitude of these peak loads, however, varies with the subsequent development of the territory, which is necessarily conjectural; hence it follows that a comparatively large amount of capital is often invested in an enterprise for the purpose of taking care of such anticipated temporary demands, and on this investment a "fair return" should be granted. This condition is particularly noticeable in municipal water-works plants, where provision must be made for supplying water for fire service to an extent which may be several times greater than the normal hourly rate of consumption. In the case of railroads, such demands can usually be met by adding to the rolling stock at moderate expense, while in a water-works the outlay is relatively greater because the entire plant must be adapted in the outset to the anticipated maximum delivery in the course of a comparatively long period of time. The problem of rate-making has been excluded by the author from his present paper on valuation; but, inasmuch as he is so well qualified for the task, and also because the non-physical value of the property depends mainly on the rates obtained for the service rendered, it is hoped that he will deal with this feature in a subsequent paper, thereby bringing out a discussion on the obscure subject of "fair return." It is noticeable that these phrases occur frequently in judicial opinions, but the fundamental principles on which a definite conclusion should be reached are seldom set forth clearly. RICHARD T. DANA, M. AM. SOC. C. E. (by letter).—The solution of this problem includes practically all the factors in the general subject of economics, in which engineering occupies a large but by no means preponderant part. Mr. Riggs has done some very valuable and pioneer work in contributing this paper at this time; and the writer, in calling attention to what appears to be a radical error in it, does not wish to be taken as attempting to detract in any way from its great value as a whole. It is most important, in the inception of such an investigation, on the part of the members of this Society, to remove from the subject the stumbling blocks as they appear. The author makes the following statement: