r 



THE BAROMETER. 



the tube B A, which is above the level of F. The bore of the tube being in 

 this case equal in every part of its length, it is clear that, through whatever 

 space the surface E falls, the surface F will rise, and vice versa. Hence it is 

 obvious that the variation in the height of the barometric column will alwavs 

 fat double the change in the height of either surface E or F ; for if the surface 

 F fall, the surface E must rise through the same space. They are thus rece- 

 ding from each other at the same rate, and therefore their mutual distance will 

 be increased by the space through which each moves, or by double the space 

 through which one of them moves. 



In the same manner, if F rise, E must fall, the two points mutually approach- 

 ing each other at the same rate ; so that the distance between them will be dimin- 

 ished by the space through which each moves, or by double the space throngh 

 which one of them moves. The change, therefore, in the height of the 

 barometric column will always be double the change in the position of the 

 level F. 



Upon the surface at F floats a small ball of iron, suspended by a string, 

 which is carried over a pulley or small wheel at P, and counterpoised by the 

 weight at W, less in amount than the weight of the iron ball. When the sur- 

 face F rises, the iron ball being buoyant, will be raised with it, and the coun- 

 terpoise W will fall ; arid when the surface F falls, the weight of the iron ball 

 being greater than the weight of the counterpoise W, will cause it to descend 

 with the descending surface, and to draw the counterpoise W up. It is evi- 

 dent that, through whatever space the iron ball thus moves in ascending or 

 descending, an equal length of the string will pass over the wheel P. Now 

 this string rests in a groove of the wheel in such a manner that by its friction 

 it causes the wheel to revolve, and consequently the revolution of this wheel indi- 

 cates the length of string which passes over its groove, which length is equal to 

 the change in the level of the surface F. Upon the centre of this wheel P an 

 index H is placed, which, like the hand of a watch, plays upon a graduated cir- 

 cular plate. Let us suppose that the circumference of the wheel P is two 

 inches : then one complete revolution of the wheel will correspond to a change 

 of two inches in the level F, and therefore to a change of four inches in the 

 barometric column. But in one revolution of the wheel P, the hand or index 

 H moves completely round the circle ; hence the circumference of this circle 

 corresponds to a change of four inches in the barometric column. Now, the 

 circular plate may easily be made so that its circumference shall measure forty 

 inches ; consequently ten inches of this circumference will correspond to one 

 inch of the column, and one inch of the circumference will correspond to the 

 tenth of an inch of the column. In this way variations in the height of the 

 column amountiag to the tenth of an inch are indicated by a motion of the hand 

 H over one inch of the circumference of the plate. By further subdivision, a 

 still greater accuracy may be obtained. 



In this form of the barometer it is evident that the preponderance of the iron 

 ball assists the atmospheric pressure in sustaining the column. This cause of 

 error, however, may be diminished almost indefinitely by making the prepon- 

 derance of the ball over the counterpoise W barely sufficient to overcome the 

 friction of the wheel P. 



Again, when the atmosphere is diminished in weight, and when the surface 

 F has a tendency to rise, it is compelled to raise the ball ; and there is this 

 obvious limit to the indications of the instrument, namely, that a change so 

 slight that the difference ef pressure will not exceed the force necessary to 

 elevate the ball, will fail to be indicated. 



For scientific purposes, the vertical barometer is preferable to every other 

 form of that instrument. In the oblique barometer the termination of the mer- 



