76 Great Lakes Levels (Continued from Page 59) lost from the lakes by evaporation. A hot, dry summer with high and persistent winds will result in an abnormal loss, difficult however to calculate. It is probable that this cause has contributed to some extent to the low levels in recent years. 3. A third cause of general changes in the elevations of the lakes is the uneven rate at which the rainfall reaches these bodies of water. Gen- erally speaking the precipitation over the Lake region is heavier in sum- mer than in winter; this is par- ticularly the case in the western dis- tricts. Even if the rainfall for an entire year is about normal there are always months, or series of months, in which it will be above or below normal. During the long winter of this north- ern region much of the precipitation is held back on the land in the form of snow and ice, while ice and ice jams diminish still further the flow of incoming streams and outgoing waterways. From all these causes it results that during the course of each year the lake levels are subject to a consistent seasonal rise and fall which usually reaches its lowest stages during the winter months and its highest during the summer months. The extent of this annual fluctuation varies greatly from year to year: the minimum ever recorded ranges from 0.53 foot on Lake Superior to 0.87 foot on Lake Erie, while the maximum runs from 1.94 feet on Lake Huron to 3.65 feet on Lake On- tario. Wind has Temporary Effect 4. The fourth cause of change in lake levels, and the one which pro- duces the greatest temporary and local changes, is the action of the winds. . When these blow over any lake long enough and with sufficient velocity to drive the surface water forward in greater volume than is carried by the lower return current, there is a raising of the elevation of the water on the lee shore and a lowering of it on the weather shore. This effect is most pronounced on Lake Erie, because it is the shallowest of the Great Lakes and affords the poorest opportunity for the impelled upper waters to return through reverse cur- rents beneath the depth disturbed by a storm. As a result, strong westerly winds pile up the water in Buffalo harbor at the eastern end of the lake, while easterly winds drive it out and lessen the depth and hence the flow of the Niagara river. The effect seldom lasts more than twelve hours MARINE REVIEW but it has raised the water in the ‘harbor as much as 9.7 feet above the adopted datum plane, and low- ered it as much as 3.5 below that plane. It has quite often happened that the highest and lowest stages of water recorded in any month have occured on the same day, or only a few days apart. On the other hand, the winds pro- duce exactly the reverse effect at the western end of Lake Erie, where the harbors of Sandusky and Toledo are particularly affected. In the latter the observed wind-produced fluctua- tions, in combination with the pre- vailing water stage, range between ex- tremes of 9 feet above to 6.5 feet below datum. ‘The wind fluctuations on the other lakes are less pronounced, though they are materially augmented in bays and at the lake extremities where the impelled water is concen- trated in a restricted space by the converging shores, and sometimes by a gradually sloping bottom. On Lake Superior the wind-produced changes may amount to as much as. 2 feet in violent storms but seldom ex- ceed 1 foot above or below normal level. Green Bay harbor on Lake Michigan and Saginaw harbor on Lake Huron are instances of bays where wind effects are strongly felt. 5. A. fifth cause, resulting in tem- porary changes of lake levels, is the variation in barometric pressure. These variations cause at times oscillations (known as seiches) in the water surface ranging from a few inches to several feet and return to normal within a few hours. These seiches are as a rule neither of large amount nor of sudden occurrence on Lake Erie. 6. The sixth cause includes the artificial conditions—those produced by man and not as in the preceding cases by nature—which affect lake levels. The best known of these is the diversion through the Chicago sanitary drainage canal of an amount of water from Lake Michigan, which has aver- aged 8660 cubic feet a second for the last five years. Accord to the latest computations this is estimated to have caused a lowering of the level of Lake Michigan of between six and seven inches, and is more than double the amount of diversion that was authorized in a temporary permit granted by the secretary of war in 1901. After many years of litigation the government in 1925 secured a final decision from the Supreme Court compelling the sanitary district of Chicago to comply with,the regula- tions of the war department, and November, 1926 that department at once took active steps which will result in reducing the amount of diversion as rapidly as due regard for the health of the citizens of Chicago will allow. As previously stated, Lake Michigan in August, 1926, was. 1.96 feet below the average August stage of the last ten years. Of this 23%4-inch fall, the Chicago diversion at its recent rate of flow (which is about 8250 second feet) may be held responsible for about 6 inches—an amount of great importance to lake commerce when the levels are as low as during re- cent years. Another important artificial cause is the sluice gates and dikes erected jointly by the United States and Can- ada in the St. Marys river at the head of the falls by which the vol- ume of the outflow from Lake Su- perior is regulated. In order to pre- vent this lake from falling so low as to interfere with navigation, some of the water that would ordinarily have flowed into the lower lakes has been held back at times, thus adversely affecting the levels of these lakes. Compensate for Deepened Channels Deepening of the connecting water- ways between ‘the lakes might have had an important influence on their levels but it has been almost, if not entirely, compensated for by other works. It is not believed that the cutting down of the forests on their shores has diminished the amount of water reaching the lakes. It may be noted here that tides, the chief cause of water fluctuations in our coast harbors, have no_per- ceptible effect on any of the Great Lakes. . As for the future, since no man can foretell what vagaries will accur in the precipitation over the lake region in the next few years, no accurate prediction: is possible. How- ever, as in the past periods of low water have always been followed by periods of high, it is only a question of time before the lakes begin to rise from their present low stage. It is quite possible that this sea- son may see the turn of the tide. It is at least an encouraging sign that in August all the lakes, except. Superior (which is artifically reg- ulated) are slightly higher than they were a year ago, the first month in which this has accurred for several years. As already stated, the level of Lake Superior is now regulated by works built at its outlet. The feasibility of controlling the levels of Lake Michi- gan, Huron, Erie, and Ontario by