120 CORRELATION OF PHYSICAL FORCES. 



somewhat similar manner, over fused tin, with but partial 

 success ; he also made many other attempts to obtain a perfect 

 vacuum ; his main object being to ascertain what would be 

 the effect of electricity across empty space : he admits that he 

 could not succeed in procuring a vacuum, but found electricity 

 much less readily conducted or transmitted by the best vacuum 

 he could procure than by the ordinary Boylean vacuum. 



Morgan found no conduction by a good Torricellian 

 vacuum ; and, although Davy does not seem to place much 

 reliance on Morgan's experiments, there was one point in which 

 they were less liable to error than those of Davy. Morgan, 

 whose experiments seem to have been carefully conducted, 

 operated with hermetically-sealed glass tubes and by induced 

 electricity, while Davy sealed a platinum wire into the ex- 

 tremity of the tube in which he sought to produce a vacuum. 

 I have found in very numerous experiments which I made to 

 exclude air from water, that platinum wires, most carefully 

 sealed into glass, allow liquids to pass between them and the 

 glass ; and this gives some reason to believe that gases may 

 equally pass through ; indeed, I have observed such effect in 

 the gas battery when it has been in action for a long period. 

 Davy supposed that the particles of bodies may be detached, 

 and so produce electrical effects in a vacuum ; and such effects 

 would more readily take place in his experiments, where a 

 wire projected into the exhausted space, than in Morgan's, 

 where the induced electricity was diffused over the surface of 

 the glass. 



M. Masson found that the barometric vacuum does not 

 conduct a current of electricity, or even a discharge, unless 

 the tension is considerable and sufficient to detach particles 

 from the electrodes ; and by adopting a plan of Dr. Andrews, 

 viz. absorbing carbonic acid by potash, Mr. Gassiot has suc- 

 ceeded in forming vacua across which the powerful discharge 

 from the Rhumkorf coil will not pass. 



The limited but somewhat varying height at which the 

 Aurora Borealis appears (according to such parallactic mea- 

 surements as can be made), is not improbably due to the trans- 

 mission of electricity, between the polar and equatorial 

 regions of the atmosphere, being only effected at a certain 



