But even if the reservoirs did not fail with disastrous results, the margin was a narrow one and the lesson is equally plain. It has long been an engineering principle that an earth embankment must not be overtopped. Twenty-four years ago, the Johnstown disaster, due to insufficient spillway capacity, impressed this upon the whole world. And it is an interesting parallel, that this was caused by an old reservoir originally built by the state for canal purposes, and later abandoned and used for pleasure purposes. Yet in Ohio there were four earth embankment reservoirs, one of which had no spillway and a far from sufficient discharge pipe; two of which filled up so that the banks were overflowed; and one which did not overflow, but which filled up sufficiently so that waves were driven over the embankments. Nor was the rainfall one beyond the range of probability. The March storm probably broke all records for combined intensity, duration and extent. But for small drainage areas such as these (52 to 114 square miles) the rainfall was not unprecedented. At least two storms have occurred in Ohio during the past forty years in which the rainfall in forty-eight hours was greater than that recorded in any forty-eight hours of the late storm, at any station, excepting Piqua, which is below the reservoirs in question. And in at least one storm in the same period, the rainfall for twenty-four hours was within .06 inch of the highest twenty-four hour rainfall of last month. The faults in these reservoirs, then, were not due to a lack of knowledge as to what to expect, but only to failure to apply knowledge already gained. In this case, of course, state ownership put an extra responsibility on Ohio to see that its property was not a menace to its citizens. But, in any case, the state is the only institution which can see that such structures are provided with the necessary facilities to make them safe. Johnstown ought to have taught the necessity of examining reservoirs and dams, and of enforcing suitable standards of design and construction. Yet if we examine the statute books of Ohio we find no legislative provision of this kind whatever. Nor is there any provision for the study, mapping and gauging of the water resources. This is a necessary preliminary to a full understanding of the possible menace from uncontrolled waste. With only two or three exceptions, conditions are precisely the same throughout the country. Even in Pennsylvania, which has probably suffered more grievously from dam failures than any other state, there is as yet no public knowledge of the design and condition of all dams, and no authority in any official or body to correct a dangerous condition. Must we wait for another Johnstown or an Austin to change these things? Or will we learn from what might well have occurred in Ohio, and make a repetition of such disasters impossible? The lesson is plain. Will we profit by it? BOOKS