Fitz Gerald and Joly — Measurement of Small Pressures. 131 



There are several uses to which these methods of measuring 

 small pressures may be put. One of the most important is the 

 determination of the densities of gases by balancing columns of 

 them against one another. This method of measuring the density 

 is quite independent of condensation of the gas upon surfaces which 

 introduce such difficulties into balloon and displacement methods. 

 Mr. Joly has suggested it as a method of measuring high tempera- 

 tures by measuring the density of a known column of air heated to 

 the high temperature. For these purposes, the indiarubber film, 

 we believe, can be got to work satisfactorily. It need only be used 

 as an indicator, to show when the pressures on its two faces are 

 equal. The change in pressure can be measured by a known small 

 compression of the air at one of its faces. Mr. Joly has made some 

 experiments on the possibility of using a horizontal column of dusty 

 air as an indicator to show whether two columns of gas balance one 

 another, but his results were not very encouraging. This method, 

 or one depending on the blowing about of fine fibres, such as silk, 

 cobwebs, &c, is, however, we believe, capable of being used to 

 determine when the pressures in two vessels are exactly equal. 



One of the most desired measuring instruments at present is 

 one by which it may be possible to measure very small pressures 

 accurately when the total pressure is very small. The M'Cleod 

 gauge is almost the only apparatus at present in use for this pur- 

 pose. It has obvious objections. It is doubtful what exactly its 

 indications mean. To what extent is there condensation of gas on 

 the surfaces ? To what extent does the mercury vapour condense ? 

 The presence of other condensible gases is fatal to its indications. 



Some apparatus such as we have sketched out might be used 

 instead of the M'Cleod gauge ; but, of course, none of them can 

 ever tell us when we have attained to a perfect vacuum. We 

 must be content to compare pressures with that in some standard 

 vacuum : it is evidently impossible by any methods such as we 

 have been proposing ever to measure absolute pressure ; we can 

 only by these means measure differences of pressure, and so we can 

 only compare one vacuum with some standard vacuum, such as a 

 potash vacuum, which we provisionally take as a standard. 



Note.— During the discussion on this Paper, Mr. Vereker suggested mica as 

 suitable substance of which to construct films. 



