its lower altitude and its lack of forest and its smaller area, has a less prolific yield than the Genesee at Portage, thus again making the error on the safe side.

From December, 1892 to August, 1893, in the absence of gagings, he deduced the run-off from the rainfall, and subsequent to the loss of the dam from November, 1897 to November, 1898, inclusive, the flow from Mount Morris was deduced from the record at Rochester, which from embracing much drainage having lower altitude and smaller rainfall than the portion above Portage, again tends to bring the estimate of the discharge of the river at Portage on the safe side.

In the nine years of probable monthly flow, thus deduced, it is of great interest to note that the year of smallest run-off, 1895, and the year of largest run-off, 1894, fall within the period covered by the daily gagings at Mount Morris, wherefore these records of daily fluctuations during this period may prudently be used as a safe guide in computing the storage requirement and the power yield for the proposed site. We will later examine the rainfall records from 1825 to 1907 for possible warning against any period

more severe.

STORAGE REQUIREMENTS AT PORTAGE.

The best method of computing the storage requirements for a given drainage area from records of stream flow is by the mass curve method. (See plate 18.) *

Plotting this mass curve from the nine years run-off data just described, it immediately is plain that the most severe period in providing uniform delivery of water for power, or in other words, the series of years of smallest yield, is from July 1, 1894 to June 1, 1898.

Starting with reservoirs full after the heavy rainfalls of March, April and May, 1894, a uniform discharge at the rate of 800 cubic feet per second from the entire drainage area of 950 square miles, could have been steadily maintained throughout every day of this extreme drought. This is equivalent to a total run-off of 11.4 inches per year in depth from the tributary watershed. The divergence of the mass curve from the draft line shows that maximum reservoir depletion at the end of the drought would have been equivalent to 5.0 inches depth on entire drainage area of 950 square miles, which in other words is 11,700,000 cubic feet.

The proposed reservoir above Portage according to table on p. 708-9, N. Y. State Engineer's report of 1896, would contain this quantity with water at elevation of 1,190 feet above sea level and is therefore of ample size to store some of the surplus of the year of heavy rainfall and to carry the surplus over for the year of drought and with dam built of the height that I propose would have 20 feet in depth of reservoir at its top where content per foot of depth is greatest, always available for flood control.

In the year 1895, according to the concurrent testimony of all the rainfall records of the region around the watershed of the Genesee above Portage (viz. at Humphrey, Arcade, Angelica, South Canisteo and Alfred) the rain

*NOTE. The 800 second feet draft line on plate 18 is too large, but due to the weight of other considerations it is thought best to let it stand.