1880.] Physical Constants of Liquid Hydrochloric Acid. 117 



When the plate becomes a thin sheet we may put a=~Rb, and make b 

 infinitely small ; we shall then have 



d? n dP n dP, 



R — -2™ _ = 2 



7TW 



R 



2 Trio 



If we express P and P as functions of p, 0, « + - — 0, 



. p =-Kt)* 



— 5 1 being found on the original system of variables. There seems 

 <?0 / 



to be a mistake in Maxwell's integration here, but it does not affect the 

 representation of the result by means of the spiral trail of images which 

 he has given. 



Spherical Conductor. 



In the case of a sphere or spherical shell, the vector potentials and 

 currents will have the same general forms as before, and the solution is 

 contained in the equations 



yjr =ar sin 9% $=- d *= -! V 2 P, 



Y d0' cZ0 4tt 



<r X =<o(P+P ), ^V 2 P="f(P + Po), 



47T Cl(p 



From the general formulas expressing completely the effect of the 

 shell on external points, we pass to one whose thickness (c) is infinitely 

 small: putting then Re=<x, 3P = Pr^, Q^P^, p = alogr, we obtain 

 finally 



dp d<fi d$ 



which may be integrated as before. It indicates a spiral trail of images 

 differing somewhat from the former ; that due to a single pole will form 

 a curve lying on a cone, and would be, if the cone were developed, a 

 series of portions of an equiangular spiral. 



IV. " On the Physical Constants of Liquid Hydrochloric Acid." 

 By Gerrard Ansdell, F.C.S., Chemical Assistant at the 

 Royal Institution. Communicated by Professor Dewar, 

 F.R.S. Received January 22, 1880. 



In a paper read before the Royal Society, on June 18th, 1879, I 

 described the results of some experiments made with liquid acetylene, 

 obtained by compressing the gas in one of M. Cailletet's ingenious 



