82 Professor Potter on the Aeromelric Balance, 



density, and temperature of a gas, by the consideration that 

 when it is applied to a mixture of gases, such as the atmo- 

 sphere, the value of k is different for every variation in the 

 proportions of the elements. 



The method of weighing a known volume of the air, which 

 has been admitted into a flask partially exhausted by the air- 

 pump and then weighed, is one of considerable labour, and 

 requiring the greatest degree of care in order to obtain correct 

 results. An instrument which would give the actual density 

 by a simple reading off is evidently so great a desideratum, 

 that the .instrument about to be described has probably in 

 some shape or other suggested itself to others, though no in- 

 strument of the kind is used at present amongst the recog- 

 nized meteorological instruments that I am aware of. 



The new r instrument is a modification of an old air-pump 

 experiment, which is used to show that the greater buoyancy 

 of more bulky over denser bodies of equal weights in the air 

 is a considerable and a measurable quantity. A metallic body 

 hanging at one end of a balance appears in the air to be 

 heavier than a closed hollow globe of glass hanging from the 

 other end ; but when placed under the receiver of an air- 

 pump, and the pump is worked, it is soon seen that as the air 

 is withdrawn the apparent excess of weight becomes less and 

 less, and eventually the hollow globe preponderates. In an 

 instrument for determining the variable buoyancy of bodies 

 in the lower stratum of the atmosphere, the points to be sought 

 for in the construction are, a sufficient sensibility and a means 

 of determining the results with readiness. A large closed 

 globe of glass might be weighed at different times in a fine 

 chemical balance ; but the process would be tedious, and from 

 perpetual changes in temperature, susceptible of less nicety 

 than that of a delicate bent lever balance, such as I have 

 adopted. 



There arise difficulties in the construction when extreme 

 sensibility is desired, which a knowledge of the mechanical 

 properties of balances, and a confidence in their theory, alone 

 can be expected to surmount. For instance, in the trials 

 which have been made with the instrument constructed by 

 Messrs. Watkins and Hill from my drawings, we have found 

 a position of stable equilibrium and another of unstable equi- 

 librium within an angle of 40 degrees. The theory showing 

 that no such positions of equilibrium can exist in a perfect 

 balance, the result indicated that the knife-edges of the balance, 

 or the agate planes in contact with them, though made with 

 all the care considered requisite in fine chemical balances, 

 were yet not sufficiently true for such a degree of sensibility ; 



