The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

134 THOMPSON'S COAST PILOT. During each of the first three winters a portion of the daily observations was lost, owing to the tide-gauge being frozen fast in its box, but they constituted only a small number in propor- tion to that embraced in the series. During the subsequent winters, artificial means were resorted to, to prevent this freezing. _ " These observations were instituted chiefly for the purpose of ascertaining with accuracy the amount of the annual, and also of the secular variation in the elevation of the lake sur- face, with a view to regulating the heights of break-waters and piers to be erected for the protection of vessels, and for improving the lake harbors." After a series of close observations, from 1854 to 1858, Lieut.- 'Colonel Graham observes : | "The difference of elevation of the lake surface, between the periods of lunar low and lunar high-water at the mean spring tides, is here shown to be two hundred and fifty-four thous- andths (254) of a foot, and the time of high water at the full and change of the moon, is shown to be thirty (30) minutes after the time of the moon's meridian transit. "We therefore, in accordance with custom in like cases, indi- cate as the establishment for the port of Chicago, : h. m. t Foot, 0 30, " Although this knowledge may be of but small practical advantage to navigators, yet it may serve as a memorandum of a physical phenomenon whose existence has generally hereto- tore been either denied or doubted. " We think it probable that, if the effect of unfavorable winds and all other extraneous forces which produce irregular oscilla-: tions in the elevation of the lake surface could be fully elimin- ated, a semi-diurnal lunar spring tide would be shown of as much as one-third of a foot for the periods of highest tides. 'The time of low water and the relative times of duration of the flood and ebb tides are given only approximately. The extreme rise of the tide being so little, the precise time of the change from ebb to flood, and hence the duration of the flow of each, can only be accurately determined by numerous obser-