256 



and noted as before ; the metal coating being removed, the same jar 

 had an equal extent of water applied to its opposite surface coating. 

 The method of effecting this is described. The author states that it 

 was so perfect as to shield the experiment from all interference of 

 vapour from the water surface, so that the jar completely retained 

 the charge without any dissipation, and in no sense differed in this 

 respect from a metal-coated jar. 



The results of this experiment are not a little remarkable. The 

 exploding distance of the 1 00 measures, whether with the metal or 

 with the water coating, did not materially differ, except in apparent 

 force, being for the metal *22, for the water '2. The exploding 

 spark from the water coating, instead of the sharp ringing sound 

 attendant on the exploding spark from the metal coating, is weak and 

 subdued, and is often like the sound of fired damp gunpowder. The 

 intensity or attractive force is also in each case alike, or very nearly ; 

 being for the metal coating 144, for the water 142. The residuary 

 charges differed considerably, being for the metal coating about 2*25 

 measures, or about ^th part of the total charge; for the water 

 coating 14 '5 measures, or about the ^-th of the total charge. The 

 residuary charge with a water coating is more than six times as great as 

 with a metal coating. The thermo-electric effect with the metal 

 coating was 10, with the water coating nothing; 200 measures, or 

 double the charge, had no effect on the thermo-electrometer. 



In this experiment it does not appear requisite that both the 

 coatings should be water ; one coating may be metal, as in the first 

 forms of the electrical jar. The author could not, at least, discover 

 any material difference in the results, and concludes that if the first 

 forms of the electrical jar with an internal coating of water had been 

 continued, we should have had but small experience of the effects of 

 artificial electrical discharge on metallic wires. 



Imperfect conducting substances employed as coatings to the elec- 

 trical jar have very similar but very exaggerated effects. With coat- 

 ings of paper we have a striking example of retention of charge. A 

 jar exposing 5*5 feet of coated glass, first coated with metal and sub- 

 sequently with paper, gave the following results under a charge of 

 100 measures. 



Exploding distances, as in the former case, nearly the same, being 

 23 and "25 ; attractive forces or intensity also nearly the same, 



