On Gases under the Influence of the Electric Discharge, 407 



they place carbonic acid differently. This may be due to the 

 difference in the nature of the discharge employed, which may 

 have decomposed the substance in the one case and not in the 

 other. 



Masson's* values were obtained by means of the method of 

 Faraday. It is interesting to note that the value he gives for 

 oil of turpentine is, as it ought to be, equal to the square of 

 the specific resistance which I have given. 



XL VII. On the Thermic and Optical Behaviour of Gases under 

 the Influence of the Electric Discharge. By E . Wiedemaitn. 

 [Concluded from p. 380.] 



5. Connexion between Spectral Phenomena, Elevation of Tem- 

 perature, and Quantity of Electricity. 

 I HA YE attempted to solve the problem of the connexion 

 between the quantity of electricity passing through a gas of 

 given pressure, the quantity of energy produced, and the cor- 

 responding changes in the spectral phenomena only in one 

 particular case, which, however, seemed to me to be of especial 

 interest. I have determined what amount of energy is necessary 

 to change the band spectrum of hydrogen into the line spectrum. 



The observations were made in the following way: — A 

 calorimeter was passed over the horizontal capillary tube of a 

 discharge-tube, and the spectrum was observed, as near as 

 possible to the place where it fitted on the tube, by means of a 

 spectroscope with a horizontal slit. The band spectrum was 

 then changed into the line spectrum by interposing air-sparks. 

 If the spark-length is gradually increased, beginning with 

 none, the band spectrum is at first very bright ; but soon the 

 three hydrogen-lines make their appearance beside it, then 

 the ground becomes fainter and the hydrogen-lines brighter, 

 until at last the band spectrum disappears almost suddenly. 

 The change is so sudden that it may often be recognized by 

 the eye without the spectroscope. This order of phenomena 

 has already been observed by Wiillner and others. When the 

 spark-length necessary had been determined, the rise in tem- 

 perature of the calorimeter t and the number of discharges 

 were determined. From this the rise of temperature for one 

 discharge at the given pressure was calculated, and then, 

 by dividing by the pressure, what the rise of temperature 

 would have been if the pressure had been only 1 millim. 



If a denote the rise in temperature of the calorimeter in a 

 minute caused by a current of deflection 100, W the water- 

 equivalent, z the number of discharges in a second with the 



* Ann. de Chim. et de Phys. 3rd ser. t. zxx. p. 25. 



2G2 



