﻿322 Dr. P. E. Shaw on the .Disruptive Voltage of Thin 



Table II. 



P.D. 

 in volts. 



















I. 

























i 



Paraffin 

 Oil. 



Trans- 

 former Oil. 



Oorl-liver 

 Oil. 



Pentane. 



Hex- 



ane. 



Hept- 

 ane. 



Octane. 



390 



1 ... 



1 ••• 



450 









345 



| breaks 



I breaks 



395 



400 



260 



320 



350 



300 



Y down 



( down. 



350 



250 





260 



310 



250 



| readily. 



J ... 



290 



220 



220 



275 





200 



I 



350 





170 



220 



235 



240 



175 



J 350 















150 



300 



210 



175 



145 



170 



140 



180 



125 



220 



170 



150 











100 



180 



110 



120 



90 



150 



100 



105 



75 



155 



80 













62 



125 















56 



95 















50 



75 



70 



50 



50 



80 



50 





25 



38 















As all experimenters on gaseous discharge have found, the 

 discharge-distance is an uncertain measurement. The same 

 order of error occurs for liquids. In the case of gases, dust 

 or want of polish increases the distance of discharge for a given 

 P.D. In liquids, solid or flocculent substances or minute air- 

 bubbles may be floating, or the presence of water, acid, alkali, 

 or salts will set up electrolysis and change the discharge- 

 distance. 



In the curves below (figs. 3, 4, 5) some liquids may be repre- 

 sented roughly by straight lines ; other liquids certainly require 

 irregular curves for proper representation. They may be 

 classed as follows : — 



Straight Curves. 

 Linseed Oil. 

 Eape „ 



Fusel „ 



Cod-liver „ 

 Pentane. 

 Octane. 



Armacell Varnish. 

 Ohmaline „ 

 Sterling- „ 



Irregular Curves. 

 Paraffin Oil. 



Transformer „ 

 Castor ,, 



Olive „ 



Turpentine. 

 Hexane. 

 Heptane. 



Though this classification is convenient, there is no sharp 

 demarcation between the two classes above. Only those 

 liquids are placed in the second chiss for which it woul be 

 absurd to suppose the departure from a straight line could be 

 due to experimental errors. 



