14 Prof. E. Edlund on the Electrical 



seems nevertheless to show positively that there, in the same 

 way as in the voltaic arc, an obstacle must exist to the propa- 

 gation of the current. If we designate this obstacle by r, the 

 total obstruction to the propagation of the current through 

 the rarefied gas from one electrode to the other must conse- 

 quently be =r + r l l; and it is necessary that the electric ten- 

 sion of the electrodes be able to overcome this sum in order 

 that a discharge may be possible. 



According to the observations which have been cited in the 

 preceding pages, the tension necessary for the discharge dimi- 

 nishes as the gas is rarefied, until a certain limit is reached, 

 after which the tension must be augmented for it to be pos- 

 sible to effect the discharge. The degree of rarefaction at 

 which this turning-point is met with is dependent on the dis- 

 tance between the electrodes, the width of the tube, the amount 

 of surface of the negative electrode, and several other circum- 

 stances. Now, as, from the preceding, r x diminishes constantly 

 with the increase of rarefaction, the fact in question, demon- 

 strated experimentally by several physicists, can only be ex- 

 plained by the first term, r, increasing with the rarefaction. 

 I consequently assume that the term r increases when the 

 density of the gas is diminished. In this manner the sum 

 r + r-J, can reach its minimum at a certain density; and when 

 this takes place, the minimum of electric tension is sufficient 

 to produce the discharge. If a Ruhmkorff induction-appa- 

 ratus or an ordinary battery endowed with great electromotive 

 force be employed as the source of electricity, the intensity of 

 the current is found to increase as the rarefaction is continued, 

 until the above-mentioned turning-point is reached, after 

 which the intensity begins again to diminish. If the rarefac- 

 tion be carried far enough, r will have so increased that no 

 tension will be sufficient to cause electricity to pass. Now 

 this by no means comes, as it has been assumed to come, from 

 excessively rarefied gas being an insulator, but from the resist- 

 ance r having become so great. The fact that the degree of 

 rarefaction at which the point in question is met with depends, 

 as Gaugain has observed, on the distance between the elec- 

 trodes is explained by the above-mentioned expression of the 

 resistance, into which / enters as a factor. 



If the gas has a pressure corresponding with the ordinary 

 barometric pressure, r is very small in proportion to r 1} and 

 can be neglected in comparison with the latter term. It is 

 then seen that the tension necessary for the discharge must be 

 proportional to the distance; that is to say, we arrive at the 

 well-known old law of the distance of the discharge. If the 

 pressure be diminished, r increases, whilo 1\ decreases ; and 



