TRANSACTIONS OF THE SECTIONS. 



Section A.— MATHEMATICAL AND PHYSICAL SCIENCE. 



President of the Section— The Rev. Professor Salmon, D.D., D.C.L., LL.D., 



■ F.R.S., M.R.I.A. 



THURSDAY, AUGUST 15, 1878. 



The following Papers were read : — 



1. Report of the Committee on Underground Temperature. 

 See Reports, p. 178. 



2. Report of the Committee on the Mechanical Equivalent of Heat. 

 See Reports, p. 102. 



An Account of some Experiments on Specific Inductive Capacity* 

 By J. E. H. Gordon, Assistant Secretary of the British Association. 



4. On the Effect of Variation of Pressure on the Length of Disruptive 

 Discharge in Air. By J. E. H. Gordon, Assistant Secretary of tlw 

 British Association.-^ 



The author has worked with a 17-inch induction coil, and with sparks varying 

 from 6 to 10 inches at atmospheric pressure, and up to over 36 inches at low 

 pressure. 



Two " aurora tubes " were used, and the method of working was to allow the 

 discharge to take its choice of two paths, one short at high pressure, and the other 

 longer at lower pressure, and thus to compare the resistance to discharge at the 

 different pressures. In the tables which accompany the paper, experiments at 55 

 different pressures, ranging from 2'42 inches to 29 - 7 inches, are given. 



The general results deduced are — 



(1) From a pressure of about 11 inches to that of the atmosphere, Harris's^ 

 law — namely, that the spark length varies inversely as the pressure — is approxi- 

 mately true. 



(2) No law can be said to be more than approximately true, as when the electric 

 density has almost reached tHe discharging limit, any slight accidental circumstance 

 will cause the discharge to take place. 



(3) Below about II inches a much greater electromotive force per unit length 



* See ' Proc. Roy. Soc. 1878-79.' The experiments described in this paper were 

 shown at the Royal Dublin Society's Soiree the same evening. 



t 'Phil. Mag.' Sept. 1878. t <phil - Trans.' 1834. 



1878. P v 



