r THE 



LONDON, EDINBURGH and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[THIRD SERIES.] 



AUGUST 1850. 



VIII. On the Aerometric Balance, an instrument for measuring 

 the Density of the Air in which it is situated. By Professor 

 Pott eh, A.M., late Fellow of Queen's College, Cambridge*, 



THAT the density of the air is an important element in 

 every discussion of atmospheric phenomena, will at once 

 be admitted ; but it will be often maintained also, that we have 

 the means of determining it from observations of the wet and 

 dry bulb thermometers and the barometer. A little reflection 

 will, however, soon convince us that this is not absolutely true, 

 since it supposes a regular constitution of the atmosphere for 

 all localities and all states of the weather, which certainly does 

 not exist. Thus, in a room where many persons have been 

 for some time assembled, the carbonic acid gas coming from 

 their lungs in respiration will increase the density of the air 

 in the room; but this will not affect the thermometer, or in a 

 sensible degree the barometer. 



There are many natural localities on the earth where ex- 

 halations affect the density of the air, but of which the pre- 

 sence escapes detection by the before-named instruments ; and 

 at all places we shall find, on consideration, that the density 

 of the air depends on circumstances not necessarily shown to 

 exist by them. If, for instance, the barometric pressure and 

 the temperature were given, we might still have an upper cur- 

 rent of wind which had a particular hygrometric state and a 

 particular proportion of its constituent gases, whilst the lower 

 current of wind was of another composition. The varying den- 

 sity of the lower current would clearly not be indicated by 

 the usual meteorological instruments. 



These cases would be comprehended in the discussion of 

 the formula p = kp{\ + a0°) for the relation of the pressure, 



* Communicated by the Author. 

 Phil. Mag. S. 3. Vol, 37. No, 248, August 1850. G 



