TROWBRIDGE. — RONTGEN RAYS. 2G1 



force, therefore, employed to charge the jars, the less resistance would 

 occur at the brushes where the change from multiple to series occurred, 

 and therefore the less the loss of energy due to Joule's heat. I there- 

 fore investigated the length of spark obtained from the Plante machine 

 by continuing the curves plotted by De La Rue and Miiller,* and found 

 that, starting with 10,500 volts, and increasing the Ley den jars pro- 

 gressively, the length of sparks plotted as ordinates and the rise in 

 voltage (iV^O gave a straight line which was an extension of those 

 obtained by them. Furthermore, as will be shown later, differences 

 between points and planes, and points and paraboliform surfaces dis- 

 appeared with high voltages. In order to ascertain if the discharges 

 through Crookes tubes when the Rontgen rays were apparently pro- 

 duced most strongly were oscillatory, I first placed a Geissler tube in 

 tlie circuit with the Crookes tube, and carefully observed the appearances 

 at the two electrodes of the Geissler tube. The electrodes were quite 

 alike in appearance, and indicated an oscillatory discharge. I then re- 

 placed the Geissler tube by a small spark gap, and photographed it in 

 the rapidly revolving mirror. 



The photograph showed ten clearly defined oscillations, with a period of 

 about one ten-millionth of a second, with the Crookes tube and the cir- 

 cuit 1 employed. Furthermore, applying the method of estimating re- 

 sistances by the method of damping, I found that the resistance of 

 the rarefied medium was less than five ohms. The energy therefore at 

 the moment of the emission of the Rontgen rays was not far from 

 3,000,000 horse power acting for one millionth of a second. The 

 Crookes tube which I employed was of the focus tube pattern (King's 

 College, London). I also employed a Crookes tube with an aluminium 

 mirror of about two centimeters' focus. The resistance of this tube to 

 the discharge was approximately the same as that in which the mirror 

 had a focal length of about five centimeters. There seemed to be no 

 advantage in shortening the distance between the anode and the kathode 

 in a Crookes tube. Struck by the fact that the distance between the 

 electrodes did not appear to make any appreciable difference in the re- 

 sistance offered by the Crookes tube to oscillating currents, I replaced 

 the tube by a spark gap in air of six inches in length, and photographed 

 the spark in another gap iu air in the same circuit. This latter gap was 

 6 mm. in length. The photographs showed on the average the same 

 number of oscillations, both when the additional spark gap was six inches 



* Proc. Roy. Soc, Vol. XXXVI. p. 151, 1883-84. 



