030 SCIENTIFIC RECORD FOR 1885. 



eter, })rovided with two platinum wire electrodes, the ends being 3°"" 

 apart, and with two bauds of tin foil, was connected to the mercury 

 pump while the terminals of an induction coil were connected alter- 

 nately with the electrodes and with the tin-foil bands. When the press- 

 ure in the tube was above SG"""" no discharge could be observed be- 

 tween the tin-foil coatings, though that between the platinum wires be- 

 came very brilliant. At about 1™°^ a luminosity between the armatures 

 was observable, which increased to 0-004™'" when it was intense, the 

 spark between the electrodes being feeble. At 0-00030'"'" the electrode 

 spark appeared only occasionally, while the armature discharge was 

 constant and very bright. The two discharges give exactly oi)posite 

 results, that between the electrodes diminishing with the exhaustion 

 and that between the armatures increasing with it. The author can 

 explain this and other similar experiments only by the supposition 

 that the resistance of the vacuum itself diminishes as the rarefaction 

 increases and that there is developed simultaneously at the electrode a 

 condition which hinders the passage of the electricity into the air from 

 the metal. {Phil. Mag., February, 1885, V, xix, 125 ; J. Phys., June, 

 1885, II, IV, 273.) " I 



Goldstein has made the following remarkable experiment on the prop- 

 agation of electricity through a vacuum. A Geissler tube has for its 

 negative electrode either a platinum loop or a carbon filament from a 

 Swan lamp. By means of a battery these loops are raised to incan- 

 descence, and then the discharge of an induction coil is sent into the 

 tube. A shunt circuit connected with the two spherical terminals of 

 a spark interrupter permits the graduation of the length of spark sent 

 through the tube. Measured in this way the resistance of the Geissler 

 tube is at least one hpndred times less when the electrodes are heated 

 to incandescence than when the cathodes are cold. The incandescence 

 of the positive electrode is without influence. {Ber. AJ:. Wien, 1884, 

 58; J. Phys., April, 1885, II, iv, 182.) 



This result appears entirely analogous with certain phenomena ob- 

 served by Edison in the spring of 1884 in his incandescent lamps. Insert- 

 ing a platinum electrode in the lamp, betweeti the- sides of the carbon 

 loop, he noticed that when the lamp was brought up to incandescence 

 a galvanometer connected on the one skle to this platinum plate and on 

 the other to the positive conductor, showed a deflection increasing with 

 the degree of incandescence. The carbon filament was, in this case, an 

 incandescent electrode as above, and the electromotive force of the 

 machine, about 110 volts, was under these conditions sufficient to cause 

 a discharge through the vacuum. {Science, October, 1884, iv, 374; Na- 

 ture, April, 1885, xxxi, 545.) 



Lehmann concludes that the apparent difference of properties of 

 positive and negative electrification in vacuum tubes is due entirely to 

 secondary actions dependent upon the state of charge which the air 

 takes in consequence of its friction against the electiodes. This elec- 



