258 Prof. De la Rive on the Propagation of Electricity 



the pressure and nature of the gas. Thus, even when there is 

 no sensible dark space near the negative electrode, the thermo- 

 meter placed at that part rises less than the one near the positive 

 electrode, showing that still the gas is there most expanded and 

 has the highest conductivity. The difference of temperature is 

 thus a more sensitive criterion, than even difference of lumino- 

 sity, of the greater or less electrical resistance of different parts 

 of the gaseous column. The absolute temperature is in general 

 lower in hydrogen, at all degrees of rarefaction, than it is in 

 nitrogen or atmospheric air, which offer more resistance to the 

 passage of electricity. Moreover the difference between the two 

 thermometers was never as great in hydrogen as it was in nitro- 

 gen or atmospheric air. Thus in hydrogen the greatest differ- 

 ence was 4°*5 R., attained when the pressure was between 9 and 

 12 millims., the thermometer having risen in two minutes from 

 21° to 28°-5 R. at the negative electrode, and from 21° to 31° R. 

 at the positive. In nitrogen the maximum difference of 5° was 

 reached under a pressure of 5 millims. (the negative thermometer 

 rising from 20° to 24° R., and the positive from 20° to 29° R.). 

 In air the maximum difference was 6° under a pressure of 6 mil- 

 lims. (the negative thermometer rising from 18° to 26° R., and 

 the positive thermometer from 18° to 32° R.). Under 20 mil- 

 lims. pressure the difference was only 2°*5 R. in hydrogen (rise 

 of thermometers from 21° to 26°, and from 21° to 28°-5 R.) ; in 

 nitrogen it was only half a degree (rise of thermometers from 20° 

 to 25°, and from 20° to 25°- 5 R.) ; and in air there was no dif- 

 ference (both thermometers rising from 19° to 28°). When 

 there is no longer any difference between the indications of the 

 two thermometers, or only a very small difference, the appear- 

 ance of the luminous jet may be observed to be perfectly uniform 

 throughout its whole extent. 



The results of some experiments are given in the following 

 Table: — 



Atmospheric Air. 



Duration of each experiment, two minutes. 



Pressure. 



Positive 

 thermometer. 



Negative 

 thermometer. 



Difference. 



millims. 



2 



4 



6 



8 

 10 

 15 

 20 



R. 



16 to 25 

 18 „ 31 

 18 „ 32 

 18 „ 31 



18 „ 31 

 18£„ 31 



19 „ 28 



R. 



16 to 21 

 18 „ 25§ 

 18 „ 26 

 18 „ 27£ 



18 „ 28 

 18 J„ 29 



19 „ 28 



R. 



4 



6 



3^ 



3 



2 







