TABLE XVII.—Showing the daily discharge, readings of water gauge, and direction of wind in the Niagara River, 1868. TABLE XVII.-Showing the daily discharge, readings of water gauge, &c.-Continued. Youngstown. feet. Wind, miles per hour. per sec'nd. second. TABLE XVIII.-Showing the daily discharge, reading of water gauge, and direction of wind in the St. Lawrence River, 1868. It will be noticed that the force and direction of the wind have a marked effect upon the amount of discharge, and that, except in the Niagara, there is not much change from any other cause, the decrease due to the slight fall in the general level of the river being marked by the wind effect. The effect of the wind on the discharge, form of curve, &c., has been left for another year, when more observations will be available. The difference between the mean of the daily discharge from these 10 2. 63 Calm. 1. 011 272, 633 1.306 11 2. 71 D., 3.7 0.987 266, 647 1.276 12 2. 56 Up, 4.9 0.936 251, 847 1.208 14 2. 56 Up, 1.3 0.985 263, 153 1. 272 21 Means.. 2.97 tables and that obtained directly from the observations is very small, will be seen by a comparison of the tables. as The change in the velocity and discharge at Niagara for the season is, however, very marked, and there are daily changes, which do not appear to be due to the wind at the place of observation. It will be noticed that there are two tables of velocities at this point, one for the first month, and the other for the remainder of the season. These were thus divided, as it was found that during the first period there was but little variation, while during the latter the velocity and discharge rapidly decreased. In the table of daily discharge are two columns, giving the wind and gauge reading at Buffalo. It will be noticed that the wind at Buffalo appears to have more effect upon the discharge than that recorded at the place of observation, and that the decrease in the discharge follows the fall of the Buffalo guage. The explanation of this seems to be, that between these two points are the falls of the Niagara, where the whole body of water pours over in a thin sheet, as over a weir, and that a small difference in the depth there would make a large difference in the level below. In fact, Captain Robinson, an old resident of Niagara village, states as the result of his observation that one inch difference of depth on the crest of the fall will make a difference of a foot in the level below. This seems to be corroborated by the fact that in the winter of 1866–7, when, on account of the water being blown back from the mouth of the river at Buffalo by a strong northeast wind, there was a fall in the level at the ferry dock below the falls of twenty-feet, the difference of depth on the crest of the falls could have been only a small proportion of this, for, except at the apex of the Horseshoe, the depth of water on the crest cannot exceed three or four feet; and though at that time the difference in the amount of water passing the falls was quite perceptible, the crest was not uncovered in any part. It is very unfortunate that accurate measurements were not taken at that time, as such a large difference rarely occurs. Between the falls and Lewiston, the points to which steamers can ascend, the river is quite narrow, and therefore very rapid, and the great change of level at the foot of the falls is very much decreased; and at Youngstown, five miles below, where the observations were taken, almost the only effect is seen in the change of velocity, and therefore of discharge, as the river almost immediately widens into the lake, and very little difference of level is perceptible. Thus we must look to the change of level on the crest of the falls for the difference in the daily discharge. It will be seen, that when the wind is southwest at Buffalo, thus heaping up the water at the mouth of the river, the discharge is proportionally increased, and vice versa. Again, the fall of six inches in the Buffalo gauge makes a proportionate decrease on the crest of the falls, thus notably decreasing the discharge. Besides this decrease, arising from the fall of water at Buffalo, there is, also, during cold weather, a formation of ice above the falls, which has a tendency to check the flow of water. During the breaking up in the spring, when large quantities of ice are forced into the river from the lake, the water in Lake Erie is rapidly rising, while in the lower part of the river, as shown by the gauge at Fort Niagara, it often remains at a low stage till the latter part of April or May, when it suddenly rises a foot or more in a few days. Therefore we see that though the summer discharge is greater than that at Ogdens burg, it probably is decreased enough in winter to make the mean for the year sufficiently smaller to allow for the inflow from the water-shed of Lake Ontario. This also accounts for the slight difference noticed in the last report between the outflow of the Niagara and St. Clair Rivers; for the observations at the former place were taken in August and September, when the discharge had only begun to decrease, and, as is seen in the more extended observations the past year, had become about equal to that of St. Clair. A few observations in winter would be very interesting if it were We cannot, therefore, at present directly compare the outflow of the possible to take them, but the running ice would make it very difficult. Niagara as observed with that of the other rivers. In the following table are given the outflow of the several rivers as found by floats in 1867, corrected for difference of level of water and by meter in 1868. TABLE XIX.-Showing the discharge per second in the different rivers for 1867 and 1868. No observations were taken the past year in the Sault River, as it was thought that the float observations could be corrected, so as to compare its outflow with that of the other rivers; but, as before mentioned, the comparisons between the meter and floats are not sufficiently extended to deduce the law of variation. For Niagara the outflow for the first and latter part of the season are given, the latter only comparable with that of 1867, being taken during the same months. The outflow of 1867, for the St. Clair and St. Lawrence, will be seen to be about ten per cent. too large; this again proves the correctness of the assumption that the floats give too great a velocity. In Generals Humphreys and Abbot's report on the outflow of the Mississippi, they give the following formula for the computation of the discharge when the mid-depth velocities are known. The form of the vertical curve being considered a parabola, the coefficient b was found equal to 0.1856, but in the following tables its value will be found for each depth and is always larger than that amount. |