1871.] 



through Rarefied Media, fyc. 



237 



The tube principally used in these experiments is shown full size in pho- 

 tograph No. 1 (PI. II.) ; it contains two aluminium wire rings, the one y^inch 

 in diameter, the other T 9 <y inch, and separated T ? - inch : the tube was 1 \ inch 

 in diameter, 3| inches in length ; it was one of Geissler's manufacture, was 

 very well exhausted, and professed to contain hydrogen. A U-shaped glass 

 tube containing glycerine and water was placed in circuit. Two aluminium 

 wires inserted in this tube gave a ready means of reducing or augmenting 

 the resistance at pleasure. Glycerine affords an easy means of producing 

 very great resistances. 



The battery used in this experiment was a Daniell's battery, each cell 

 having a resistance of from 50 to 100 ohms. The resistance of the glycerine- 

 and-water tube was between 2 and 3 megohms ; this latter resistance was 

 made large, in order that the resistance of the tube and battery might be 

 neglected without entailing error. 



The following laws were found to govern the passage of the current : — 

 1st, each tube requires a certain potential to leap across ; 2nd, a passage 

 for the current having been once established, a lower potential is sufficient 

 to continue the current ; 3rd, if the minimum potential, which will maintain 

 a current through the tube, be P, and the power be varied to P+l, P + 2, 

 &c. to P + ^» the current will vary in strength, as 1, 2, &c. n. 



Tables I. & II. (p. 242) illustrate this ; there is a little irregularity in 

 the figures due to the irregularity of the battery, although it was recharged 

 for the occasion. 



It thus appears that a certain amount of power is necessary to spring 

 across the vacuum ; after that it behaves as an ordinary conductor, ex- 

 cluding that portion of the battery whose potential is P, and which is used 

 to balance the opposition of the tube. In these experiments P was 304 

 cells. The tube in question could not be persuaded to allow a current of 

 less than 323 cells to pass ; but when once the current had established a 

 channel, on lowering the potential by short circuiting portions of the 

 battery, so as not to break the circuit, the current would flow when the 

 battery was reduced to 308 cells. By, however, passing a current from 

 600 cells, in the manner shown in PL III. fig. 1, through the second tube, 

 filled with pure glycerine, and offering several thousand megohms resistance, 

 an extremely feeble current, too weak to affect the galvanometer, kept a 

 channel open by its passage ; with this arrangement the figures in Table II. 

 were obtained, which are more regular at the commencement, and a power 

 of P 1 would pass across the tube. 



The positive pole alone was observed to be luminous when the current 

 was very minute, and the negative only was luminous when the current 

 was strong. The following experiments were tried, and the results, which 

 have been photographed, accompany this. 



A current was passed through the U tube and the vacuum ; the U tube 

 contained pure glycerine, and had a very large resistance, which was gra- 



