322 Obituary Notices of Fellows deceased. 



became luminous on plunging it into a charged jar, and also on with- 

 drawing it. 



In the same note-book from which quotations have been made 

 above there are recorded detailed observations on the spectra shown 

 in thermometer tubes, made at various dates up to February 25, 1873. 

 These latter formed the subject of a short paper in the " Philosophical 

 Magazine" for April, 1873 (4th series, vol. 45, pp. 241-245), entitled 

 ' On Spectral Lines of Low Temperature,' which passes from observa- 

 tion to theory. It would appear that in writing this paper Lord 

 Salisbury had regarded the phenomena described as more dependent 

 on conduction than, in the light of modern experience of electric 

 oscillations, we could now admit ; while the changes in the spectra 

 are discussed on the supposition that they are controlled by the 

 temperature of the residual gas. The absence of the hydrogen 

 spectrum he believed to be due to the low temperature of the discharge.* 

 He set about determining this temperature. Experiments with a 

 vacuum tube without electrodes, and having the bulb of a small air 

 thermometer sealed into it, gave some indication of increased tem- 

 perature when the discharge passed through the tube. 



In consequence of a brush discharge being observed at the top of 

 the thermometer, it was apparently thought that the illumination 

 of the thermometer was accompanied by conduction through 

 the glass. An experiment was made subsequently in which an 

 exhausted tube, five feet long, and without electrodes, but with 

 the ends covered with tin foil, was placed with one end in contact 

 with the prime conductor of an electrical machine. On working the 

 machine the tube was illuminated until the charge became constant, 

 and on discharging the conductor a flash was again seen in the tube. 

 With a Holtz machine, having the electrodes near together, so that 

 frequent discharges occurred, the tube was brilliantly luminous when 

 one end was placed in contact with one of the electrodes, the other 

 end of the tube being insulated. If a brass ball connected with earth 

 was placed near the tin foil covering of the distant end of the tube, 

 sparks passed at each discharge of the Holtz machine. This 

 tube, made in July, 1874, is an early example of the electro- 

 statical variety of electrodeless vacuum tubes. From letters by 

 Mr. J. T. Bottomley in "Nature," vol. 23, 1881, pp. 218 and 243, it 

 appears that Sir W. Thomson and he had independently observed the, 

 same phenomena. 



* H. Keyser, " Handbuch der Spectroscopie," vol. i., 204, mentions these 

 experiments of Lord Salisbury as the first in point of time in which the conclusion- 

 was dra\nn that a gas at low temperature can emit a bright spectrum. 



