\ ' y^ y.y\^^ |'w-<yw' 5, »^ If $ -^ » » 'w* %<^ 



ON THE 



EEDUCTION OF THE BAEOMETEK 

 TO SEA LEVEL. 



f\:^.Oi \ a 



BY CHARLES CARPMABL, M.A., F.R.A.S., 



(LATE FELLOW OP ST. JOHN'S COLL., CAMB.) 



Deputy Superintendent of the Meteorological Service of Canada. 



The application of an approximately correct reduction to baro- 

 metric readings, taken at various levels, in order to reduce them to 

 •what they would have been at one specified level, is absolutely 

 necessary for their intercomparison. In the following paper several 

 formulae which have been employed for this purpose are examined ; 

 and tables are appended by means of which, with very little calcu- 

 lation, a sufficiently correct redixction may be obtained, and which 

 are, moreover, peculiarly adapted to the computation of tables of 

 reduction for individual stations. 



Guyot's Tables* D, XYI. and XIX'., are commonly employed, on 

 this continent, for the purpose of effecting the reduction. These give 

 the height, in English feet, of a column of air corresponding to a tenth 

 of an inch in the barometer at various temperatures, the barometric 

 pressure at the base of the column being from 22 inches to 30-4 

 inches. 



A formula is given for use with Table XVI., which may be 

 written 



where R represents the required reduction in inches, Z the differ- 

 ence of height between the two stations, or the height above the sea 

 (expressed in feet), iVthe number in the table, /3 the observed reading 

 of the barometer reduced to 32° Fahr., and b the pressure on which 

 the tabular number iV" is based,t that is, 30 inches. 



* Meteorological and Physical Tai)les. Third edition. Washington, 1S59. By Arnold Guyot, 

 P.D., } "^ .D., Professor of Geology and Physical Geography, College of New Jersey. 



t Gui jt defines what is here represented by &, as " the normal height of barometer atthesea- 

 Ie"vel," and in an example which he gives, he employs 30 in. It is, however, only because the 

 table is based on a barometric reading of 30 in., that this value of 6 is to be employed. 



{Proceed. Can, Inet. Vol I. Part /,) 



r. 



