184 Lord Rayleigh on the Equilibrium of Liquid 



These conclusions are confirmed by numerous experiments; 

 and no experiment has been found to contradict them. 



The decision in favour of this theory was supplied by the 

 fact that glowing and cold wires separated only by heated air, 

 with the exclusion of combustion-gases ; showed an electrical 

 difference. Here also the latter is dependent on the nature, 

 and the quality of the surface, of the electrodes employed, and 

 on their state of incandescence. A too strong heating of the 

 wires, and, therewith, also of the separating air stratum, proved 

 unfavourable to the development of free electrical tension — a 

 circumstance prooably due to the augmentation of the con- 

 ducting-power of that separating stratum. In accordance with 

 this, wires introduced into the flame, so long as they are both 

 immersed in the combustion-gases (which are relatively good 

 conductors), never give the maximum of potential-difference ; 

 rather this enters only when one of the wires comes into con- 

 tact with only the outer air stratum of the flame (which is 

 endowed with a very high resistance). 



The occurrence of a thermoelectric counterforce within the 

 galvanic flame-arc is also naturally explained by the above 

 theory. 



The questions proposed at the commencement are therefore 

 to be answered thus : — Hankel's theory is not in accordance 

 with experiment; and the two kinds of excitation assumed 

 by Buff and Matteucci must be regarded as simultaneously 

 causing the apparent electricity of flame. 



Wolfenbiittel, February 1882. 



XX. On the Equilibrium of Liquid Conducting Masses charged 

 with Electricity. By Lokd Rayleigh, F.R.S.* 



IN consequence of electrical repulsion, a charged spherical 

 mass of liquid, unacted upon Iry other forces, is in a con- 

 dition of unstable equilibrium. If a be the radius of the 

 sphere, Q the charge of electricity, the original potential is 

 given by 



If, however, the mass be slightly deformed, so that the polar 

 equation of its surface, expressed by Laplace's series, becomes 



r = a(l+F 1 + F 2 +...+F"+...), 

 * Communicated by tbe Autbor. 



