REPORT OF THE SECRETARY. 19 



regions, are valuable additions to our knowledge of the physical geog- 

 raphy of the earth. 



I regret to be obliged to state that since the publication of the paper 

 on the winds at Van Kensselaer Harbor, some doubt has arisen as to 

 the proper interpretation of the original record. It is stated by Dr. 

 Kane that the observations of wind were uncorrected for magnetic 

 variation. In consequence of this statement a correction was applied 

 by Mr. A. Schott to reduce them to the true meridian. Mr. Sonntag, 

 one of the principal observers, after his return from Mexico, asserted 

 that the observations of the wind were recorded in reference to the 

 true meridian, and therefore required no correction. The same state- 

 ment was subsequently made independently by Dr. Hayes. An ap- 

 pendix has therefore been prepared for the series giving the correc- 

 tions to be applied to the tables, in order that the results may be in 

 conformity with either assumption. The weight of testimony would 

 appear to be in favor of the supposition that the records of the wind 

 at Van Eensselaer Harbor were recorded with reference to the true 

 north ; but the question cannot be fully settled until other observations 

 from the same place are obtained. 



The next paper in the 12th volume of Contributions is on the 

 fluctuations of the level of the surface of the North American lakes. 

 It has long been known that the great interior fresh water seas, of 

 North America are subject to variations of level. From the observa- 

 tions given in this paper and others previously published, the fluctua- 

 tions are of three kinds: First. A general rise and fall, extending 

 through a period of many years, which may be called the secular 

 variation of level. It evidently depends on peculiar changes in the 

 meteorology of the country drained, and although it may probably 

 have a regular period of return, this has not yet been determined. 

 Second. An annual rise and fall, the period of which is completed in 

 about twelve months, which, is caused by the changes of the seasons, 

 can be predicted with considerable certainty, and is properly called 

 the annual variation. Third. An irregular movement, producing 

 frequently a sudden elevation, from a few inches to several feet. This 

 is of two kinds, one evidently due to the wind, and the other result- 

 ing from rapid undulations in calm water. Both classes may be 

 styled transient fluctuations. To these a fourth may be added, ac- 

 cording to a late publication by Colonel Graham, United States Army, 

 which is a true lunar tide. The author of this paper professes to 

 have condensed from all sources within his reach information respect- 

 ing the fluctuations of the water since the settlement of the country. 



