BAROMETRICAL MEASUREMENT OF HEIGHTS. 



421 



The elevation of a place in the interior of a continent where regular meteorological 

 observations are made, may be ascertained by taking the yearly means of the ba- 

 rometer reduced to the freezing point, and of the temperature of the air, as data for 

 the upper station, and the yearly means of the I'educed barometer and of the free 

 thermometer at the level of the sea, as the data for the lower station. The Hypso- 

 metric Tables then will give the difference of level. As observation, however, has 

 shown that the mean height of the barometer at the level of the sea is not the same 

 m ail latitudes, it is necessary to take for such a comparison the mean height of the 

 barometer which belongs to the latitude of the station the elevation of which is to be 

 computed, or that which is nearest to it. 



Table XII., published by Schouw, in Poggendorf's Annaleri, and in the Comptes 

 Rendus de V Academic des Sciences^ Tom. HI. p. 573, gives in Paris lines the mean 

 height of the barometer in various latitudes. The reduction into millimetres is 

 from Martins's French translation of Kaemtz's Meteorology^ p. 278 ; the correspond- 

 ing values in English inches, and the new stations, Savannah, Ga., Philadelphia, Pa., 

 and Cambridge, Mass., have been added. The mean heights last mentioned have 

 been derived from three years of observations at Savannah, by Dr. John F. Posey, 

 from June, 1853, to June, 1856, published in the American Almanac; from four 

 years of hourly observations at Girard College, Philadelphia, by Prof. A. D. Bache ; 

 and from ten years of observations at Cambridge Observatory. They have been 

 reduced to a common absolute standard and to mean tide-water at the respective 

 places. 



These mean barometric heights, corrected for the variation of gravity in latitude, 

 according to the proposition of Poggendorf, by the formula 5 = Z» 45 ( 1 — 0.0025935 

 cos 2 (/)), where h is the height of the barometer in latitude 0, and ^45 the corre- 

 sponding height at the forty-fifth degree of latitude, are found in another column. 

 For computing the elevations, the uncorrected heights are to be used. 



The mean barometric pressure, as shown by Table XIII. from Kaemtz's Precis de 

 Meteorologie, French translation, p. 281, is not the same in all seasons, and the 

 monthly means differ by a quantity which also varies with the latitude. If, therefore, 

 the height of an inland station is to be ascertained frotn the barometrical means of 

 one or more months only, the computation must be made with the mean pressure in 

 the corresponding months at the level of the sea ; or if this is not known, the yearly 

 means taken from Table XII. must be corrected for the difference between the monthly 

 means of the given month, or months', and the annual mean in the same latitude, as 

 derived from the comparison of the numbers in Table XIII. 



Example. 



Suppose an inland station, in latitude 40° N. ; the mean barometric pressure for 

 July is 26.30 inches, and its elevation is to be computed from it. 



Table XII. gives for latitude 40°, at Philadelphia, reduced to the level of the sea, 

 30.053 inches. Table XIII. gives as the mean for July, at the same place, 759.80 

 millimetres, and for the year, 760.25 millimetres (both not reduced to the level of 

 the sea), difference — 0.45 millimetres = — 0.017 English inches, which is to be 

 subtracted from the annual mean, 30.053, to reduce it to the mean of July ; or 



D 81 



