8 THE METEOROLOGY OF JAMAICA 



in order to show the temperature of the instrument for the 

 following reason. 



The density of mercury varies with its temperature, so 

 that a column of 30 inches of mercury at a temperature of 

 50 weighs more than a column of 30 inches of mercury at 

 80. In order, therefore, to compare the readings of baro- 

 meters at different temperatures, it is necessary to reduce 

 all these readings to what they would have been, supposing 

 that the mercury had always the same temperature. 



It has been universally agreed to adopt 32, the tempera- 

 ture of freezing water, as the temperature of reference for 

 mercurial barometers. 



Table I. gives the decimal parts of an inch to be subtracted 

 from the reading of the barometer for every degree between 

 60 and 90 ; it takes into account the expansion of the brass 

 frame as well as the expansion of the mercury. It will 

 be seen that the reduction varies with the height of the 

 barometer; thus when the attached thermometer is 75 we 

 must subtract 0"108 inch from the reading of the barometer 

 when it is about 26 inches, but we must subtract 0'125 inch 

 when the reading is about 30 inches. 



Now, although Torricelli showed that the atmosphere 

 exerted a pressure which could be measured by a barometer, 

 it was left to Pascal to show that this pressure was due to 

 the weight of the air. The atmosphere rests upon the 

 surface of the sea and land in much the same way that 

 the ocean rests upon its bed. The pressure at great depths 

 in the ocean is enormous ; the atmospheric pressure at the 

 surface of the sea is about 15 pounds on every square inch ; 

 and this pressure continually diminishes as the elevation 

 above the level of the sea increases. 



In the year 1648, at the suggestion of Pascal, Perier 

 ascended the Puy de Dome, a mountain near the centre 

 of France ; and he found that the barometer fell almost 

 four inches as he ascended from the foot of the mountain 

 to its summit. The pressure at the summit was relieved 

 of the weight of the air below ; and it only remained to 

 ascertain the weight of a given quantity of air in order to 

 compute differences of elevation by means of barometric 

 observations. 



