period come singly, such as exceptionally heavy rainfall on ground previously saturated or rendered impervious by frost. 4. The gagings of rate of flow have apparently been somewhat crude and inexact, but accepting the quoted estimates as correct, it appears that a rate of discharge at Rochester greater than 40,000 cubic feet per second causes great damage, with danger of greater damage than has heretofore occurred, as for example, if a jam of trees, ice or debris should occur in the archways of the bridges at Rochester. 5. Floods giving rate of flow of 20,000 cubic feet per second at Rochester may submerge large areas of the flats between Mt. Morris and Rochester. 6. A flood rate which measures 40,000 cubic feet per second as it passes Mt. Morris may become so modified by spreading out over these flats, by its temporary storage and delay thereon, that notwithstanding the drainage area at Rochester is 2.3 times as great, the rate of discharge in cubic feet per second at Rochester may be only about half as great as at Mt. Morris. 7. The project for flood relief on the Canaseraga flats by straightening the channel has not yet been studied by means of recording gages on the Genesee and its main tributaries, or with precision to show just what bearing quicker drainage for local relief this will have upon the rate and height of flood discharge at Rochester and on the flats immediately above, and in absence of precise data, I venture no opinion on this question. It is believed by those who promoted this project that this cutting of a new channel and straightening the Canaseraga will hasten the transit of the Canaseraga water and get it out of the way largely before the flood flow from above Portageville arrives. According to the Portage or the Mt. Morris storage reservoir projects presented in the State Engineer's Reports of 1896 and 1895, respectively, it was proposed to discharge the water from the storage reservoir at such a rate as would produce a substantially uniform flow of water for power at Rochester throughout the year. This reservoir, about 100 feet in total depth, would ordinarily be filled to overflowing in April and in years of ordinary rainfall would have become drawn down only a few feet by the date of July 6th, on which date the great flood of 1902 occurred, and if the proposed Portage reservoir be operated, according to the schedule plainly proposed on page 713 of Mr. Rafter's report of 1896, this storage reservoir thus conserved for power supply would have been so nearly full on the occasion at least of some of the great floods of March 2, 1902; April 4, 1896; March 20, 1894; June 1, 1889; March 16, 1875; March 17, 1865; June 20, 1813, that it could have served only partially to retrain and lessen the flood at Rochester or in the valley below Mt. Morris. Higher Dam and Deeper Reservoir Now Proposed. I have therefore as a first study proposed to build the dam, say, ten feet higher and to reserve ten feet or twenty feet at the top of the reservoir, where its area is greatest, solely for flood relief, with a strict rule that the reservoir should after a flood be emptied as rapidly as the river channel could carry off its water without overflowing the banks until the reservoir is drawn down to this power level, ten or twenty feet below the spillway crest, so that the emptied portion of the reservoir should stand ready to restrain whatever rainfall of dangerous proportion might come next. The precise depth to be thus reserved for flood control, whether ten feet or twenty feet or more, has not been worked out, for I consider many more data and observations all along the flooded region are first desirable. It would be well worth the cost to install and maintain a series of at least six self-recording river stage registers — one at Rochester, one at Mt. Morris, one at Fillmore, one near the upper end of the Canaseragı Valley, and two at intermediate points by which the relative time of passing of the flood wave could be observed and permit of a more complete study of the progress of the flood. In the absence of such observations, estimates are uncertain about the precise effect on flood heights of the quickened flow on the Canaseraga or the retarded flow above Portage, save that it is sure beyond question that all detention of flood water behind the Portage dam would aid materially in flood relief below Mt. Morris and also in the Canaseraga Valley. It appears from the report of the Flood Commission of 1902 that the solution of the problem of flood control at Rochester was by no means fully worked out or even attempted by its members. They chiefly collected some facts as to what had happened. This is a large and many-sided problem and requires more surveying and more study of local conditions than I have been able to give. The widespread destruction and danger likely to follow a repetition of such floods as have already occurred, or the far greater flood which plainly is likely to come at some future time, warrant the expenditure for fully studying this problem. PLAN OF PORTAGE WORKS PROPOSED. The main features are as shown by the plans on plates 22, 23 and 24. First. A dam three-fourths of a mile upstream from the upper Genesee Falls, entirely outside the limits of the land purchased by Mr. Letchworth, and forming a part of the Genesee Park. This is made of an arched form resting on solid ledge abutments, for strength as well as for better architectural effect. An exceptionally long spill way is provided at the westerly end, so that the greatest flood may be discharged with a small increase in height of reservoir. The height of dam is proposed at an elevation ten feet greater than that proposed by Mr. Rafter, in order that there may be a greater reserve available for flood control, in case a flood should occur while a large volume was being held in storage for power purposes. Second. Sluiceways are provided in this dam, such that any desired quantity may be withdrawn for scenic purposes over the Genesee Falls. These sluiceways would also serve for rapidly lowering the water in the reservoir to a depth of about ten feet below the spillway level, immediately after the peak of a freshet on the main river had passed, so as to prepare storage space in the top of the reservoir as soon as possible against chance of a second flood. The dam is to be of concrete, and except for the addition of the arched form, is proposed to be of substantially the same type as those recently designed for New York city's water supply at Ashokan and Kensico. The elevation of these sluiceways and the elevation of the intake to the proposed rower tunnel, are to be set so high that the reservoir cannot be |