210 Prof. E. Edlund on the 



at the result that the resistance opposed by a gas to the pro- 

 pagation of electricity cannot be proportional to the intensity 

 of the current, as is the case with solid and liquid conductors. 

 It is quite as impossible to admit, with Hittorf, that the resist- 

 ance of gases is in inverse proportion to the intensity of the 

 current ; for in that case the resistance in a column of gas 

 through which an infinitesimal current passes would be infi- 

 nite. Now, at the discharge of a condenser, or the closing of 

 a galvanic pile, the current is at first excessively slight. If 

 Hittorf 's> hypothesis were true, the resistance of the gas would 

 at first be excessively great, and the current could not begin 

 to circulate. Consequently the resistance of a gas cannot in 

 any case be inversely proportional to the intensity of the cur- 

 rent. It is in the nature of things that the effective resistance 

 of a column of gas should be proportional to the length of the 

 latter. Therefore, if I denotes the length of the column, and 

 r the resistance in the unit ef length, the resistance will be 

 proportional to rl ; and from what has just been said, r is 

 neither directly nor inversely proportional to the intensity of 

 the current or to the tension of the electrodes. Quite the con- 

 trary, experiments show that the tension necessary for the dis- 

 charge is proportional to I, whence it follows that r is inde- 

 pendent of the tension. 



If we admit, in accordance with what has been said above, 

 that the resistance of gases is independent of the intensity of 

 the current, all the differences above stated, between gases, on 

 the one hand, and both solid and liquid bodies, on the other, 

 can be accounted for by the unitarian theory. 



If the resistance or counterpressure opposed to the propaga- 

 tion of the current by a column of the gas of the length 1 and 

 section 1 be called k, the total counterpressure in a similar 

 column having the section a will be equal to ka, not to ki as is 

 the case with both solid and liquid conductors. On multiply- 

 ing this expression by the velocity h of the electricity, the 

 product will be proportional to the mechanical work which is 

 accomplished in the column during the unit of time. Now 

 i = 8ah, an expression in which, as has been said above, 8 is a 

 constant. We obtain, then, for the work done in a column of 



length 1 and section a the expression -~-, to which the quantity 



of heat developed in the same column must be proportional. 

 The quantity of heat developed in a column of gas will there- 

 fore be proportional to the intensity of the current, but inde- 

 pendent of the section of the column. 



The resistance being determined by the counterpressure 

 which the conductor opposes, on the unit of section, to the 



