3.20 M. F. Riess on the Passage of Electric Currents 



calculated from four thermometric observations; in the fourth line 

 the heating effect ^vith a ])ositive knob is given, that with a positive 

 surface and equal pressure of air being supposed equal to 100. 



Heating effects in the discharge circuit when the latter is 

 interrupted by rarefied air. 



The increase of heat by reversing the discharge in the glass 

 cylinder, and thus making the knob the positive instead of the 

 negative electrode, decreases quickly with increasing pressure of 

 air, and is no longer perceptible at a pressure of 2^ inches. An 

 interesting corroboration of the theorem, that in these experi- 

 ments two essentially different kinds of electric discharge are 

 involved, is presented by the circumstance, that, in the line A, 

 the greatest heat occurs when the air is most dense (3i inches 

 pressure), and, in the line B, when the air is least dense (IJ^th 

 of an inch pressure). 



In fact, when the greater part of the partial discharges were 

 glimmering, the admission of air must have prevented the glim- 

 mering, and thereby rendered these discharges effective in the 

 generation of heat. When, on the contrary, the majority of the 

 partial discharges were discontinuous, the increased quantity of 

 air had the effect of increasing the interval between the several 

 partial discharges, and, consequently, of diminishing the heating 

 effect. Insomuch as, in reference to this heating effect, these 

 opposite actions simultaneously took place in every experiment, 

 it cannot appear surprising that sometimes the one, sometimes 

 the other action was in excess. Thus, in order to explain the 

 observation in the line A with ygths of an inch pressure, we 

 must assume that by changing the pressure of air from -^^ to 

 ■j^2, the effect due to the prevention of the glimmering is in 

 excess, whilst at -^ths of an inch pressure the retardation of the 

 discontinuous discharges is perceptible in the observation. From 

 3^ inches pressure upwards only one kind of discharge, the dis- 

 continuous, appears to exist, whatever may be the condition of 

 the electrodes ; and, in fact, at 2~ inches pressure it was no 

 longer possible to detect any difference in the luminous phseno- 

 mena in the cylinder when the knob was positive and when it 

 was negative. In both cases a bright, broad expanse of light 

 was alone visible in the space between the electrodes. 



