﻿402 Intelligence and Miscellaneous Articles, 



gravity of the air, and calculating therefrom the expansion at the 

 time. The former method is in universal application in the mer- 

 cury barometer ; the latter one has not hitherto been used in barome- 

 tric measurements. 



The apparatus which I have prepared and employed, and by which 

 the atmospheric pressure can at all times be determined, consists of 

 a scale-beam, to the ends of which are screwed two bodies nearly 

 equal in weight but greatly differing in volume, — a hollow sphere 

 and a solid cylinder*. On one end of the beam is a mirror which 

 is approximately at right angles to the axis of the beam. At some 

 distance from the apparatus is a telescope with a vertical scale, 

 the image of which in the mirror is observed by means of the tele- 

 scope. It is clear that when there is any change in the expansion 

 of the air in the vicinity of the apparatus, the beam will indicate va- 

 rying angles with the horizon, which angles may easily be read off 

 in the mirror by means of the telescope. 



The variations of the scale-beam in consequence of alterations in 

 the pressure will not amount to much, if the dimensions of the [appa- 

 ratus are moderate ; but the use of Poggendorff and Gauss's method 

 of reading affords such a degree of accuracy that, as a brief calcula- 

 tion shows, under assumptions which are easily realized, the changes 

 in the position of the beam can be measured with far greater cer- 

 tainty and accuracy than the height of the mercurial column in the 

 ordinary barometer, provided the whole construction is light, and 

 that its centre of gravity is at a short distance from the knife-edge 

 of the beam. 



We will assume that the beam in dry air at the temperature 0° and 

 under a pressure of 760 millims. is in equilibrium, and that we have : — 



Q, the true weight (reduced to vacuum) of the hollow sphere ; 



Qj, that of the solid cylinder ; 



V and v , the corresponding volumes at 0° ; 



<p and (j) v the corresponding coefficients of expansion ; 



a, the coefficient of expansion of air ; 



B, the expansion of air ; t, the temperature ; 



T, the expansion of the aqueous vapour in the air ; 



G, the weight of the entire construction; 



h, the distance of its specific gravity from the axis of rotation ; 



u, the length of the arm of the lever. 

 The tangent of the angle of deflection of the beam is then 



l0g(tf = , 



G^ 

 a 



* Thus far the apparatus I have employed is in principle identical with 

 Guericke's dasymeter or manometer. That arrangement is, in fact, a ba- 

 lance in which equilibrium is established by means of weights. Gerstner 

 (senior), who has constructed the most accurate manometer, used it in de- 

 termining the specific gravity of air under varying conditions of pressure 

 and temperature. This he effected by adjusting a sliding weight on a 

 graduated scale-beam. The apparatus fell into disuse in consequence of 

 the subsequent more accurate measurements. 



