PNEUMATICS. 173 



3S2. The ordinary use of the barometer on land as a weather 



value of the indicator is extremely limited and uncertain. It has been 



barometer as a already Stated that the weio-ht of 100 cubic inches of air is 



weathtr indi- , ^ ../^ m , , • , • , . . 



cator? about 60 grams, lo obtam this result, it is necessary that tlio 



experiment should be performed at the level of the sea, and it 

 is also requisite that the temperature of the air should be about 60° Fahren- 

 heit's thermometer, and that the height of the column of mercury in the ba- 

 rometer tube should be 30 inches. As these conditions vary, the weight, or 

 pressure of the atmosphere, and consequently the height of the mercury in 

 the barometer tube must also vary. Especially will the lieight of the mer- 

 curial column vary with every change in the position of the instrument as 

 regards its elevation above the level of the sea. A barometer at the base of 

 a lofty tower will be higher at the same moment than one at the top of the 

 tower, and consequently two such barometers would indicate different com- 

 ing changes in the weather, though absolutely situated in the same place. No 

 correct judgment, thertfore, can be formed relative to the density of the at- 

 mosphere as affecting the state of the weather, without reference to the situ- 

 ation of the instrument at the time of making the obser\'ation. Consequently, 

 no attention ought to be paid to the words "fair, rain, changeable," etc., fre- 

 quently engraved on the plate of a barometer, as they will be found no cer- 

 tain indication of the correspondence between the heights marked, and the 

 state of the weather. 



_ . . The barometer, however, may be generally relied on for 



To what extent „ . , . . ,. . -. , /. , , 



may the ba- fiimishing an indication of the state oi the weather to this ex- 



roiTieter here- tent: — that a fixll of the mercury in the tube shows the ap- 

 licd on for fore- ' •' ^ 



tcUinj:; changes proach of foul weather, or a storm; while a rise indicates 

 in the weather ? ^,^^ approach of fair weather. 



At sea, the indications of the barometer respecting the weather, are gener- 

 ally considered, from various circumstances, more reliable than on land: the 

 great hurricanes which frequent the tropics, are almost always indicated, some 

 time before the storm occurs, by a rapid fall of the mercury. 



„ , 383. If a barometer be taken to a point elevated above the 



How mar the . , i -n /• ,i i 



barometer be surface of the earth, the mercury in the tube will fall ; because 



mMng'^ ^°\he ^ ^® ascend above the level of the sea, the pressure of the 



height of atmosphere becomes less and less. In this way the barometer 



mouutams? maybe used to determine the heights of mountains, and tables 



have been prepared shov\-ing the degrees of elevation corresponding to the 



amount of depression in the column of mercury. 



vTh^t is the 384. The absolute height to which the at- 

 ofThrltmlfs-' ninsphere extends above the surface of the 

 phere? earth is not certainly kno\vn. There are good 



reasons, however, for believing that its height does not 

 exceed fifty miles. 

 This envelope of air is about as thick, in proportion to the whole prlnhe, as 



