520 Dr. J. H, Vincent : Electrical Experiments 



the tube with the mercury not in electrical contact. After 

 some further trouble with the disengagement of gas, the tube 

 again lit up and kept alight for fonr minutes when the tube 

 fniied. The frequency was 38'5 ; it took a mean current of 

 3 amperes. A similar tube of soda glass ran successfully for 

 8 minutes, and others (jacketted with w T ider glass tubes) for 

 17 and 20 minutes respectively before the tube failed. 



A quartz tube (10 cm., 4*3 mm., '46 mm.) ran very 

 satisfactorily as a low voltage mercur)^ lamp. When in 

 series with the above coil and a moving coil voltmeter 

 shunted with one ohm, the mean current from three large 

 accumulators in series was *5 ampere. A soft-iron volt- 

 meter was connected to the lamp terminals ; this read 6 volts, 

 the reading of the same instrument on the cells alone was 

 6* 2 volts. The light occupied about 2 mm. length of the 

 capillary bore in the middle part of the tube, and was of a 

 white colour with a very faint bluish tinge. When the lamp 

 was first lit, it had a frequency of 106 which gradually rose 

 until wdien about 117 the lamp went out. This phenomenon 

 could be repeated as often as desired. By removing the 

 shunted voltmeter from the circuit, the resistance external 

 to the lamp was practically reduced to that of the inductance 

 coil, i.e. *1 ohm. This alteration had the effect of rendering 

 the light permanent. The lamp continued to give out a 

 bright pleasant light until the current was switched off. 

 Under the new conditions the soft-iron voltmeter across the 

 lamp read 6*5 volts. 



A telephone receiver connected in series with 12,000 ohms 

 across the lamp gave out a somewhat raucous deep note 

 corresponding to the frequency of the lamp ; the octave of 

 this note was also present. Inserting small resistances in 

 the lamp circuit raised the frequency. By gently heating 

 the middle of the tube with the flame of a Bunsen burner, 

 the frequency was lowered. Stronger heating caused the 

 lamp to emit the green glow. The green light was given 

 out by the capillary for a length of about 1*5 cm. on each 

 side of the bright star of light in the centre of the tube, but 

 separated from it by about 2 cm. 



Similar experiments were performed with another quartz 

 tube (6 15 cm., 5*5 mm., *4 mm.). This tube gave a bright 

 intermittent arc about 1*5 mm. long at its centre when run 

 from five large accumulators in series. In addition to the 

 inductance coil, the resistance of which was # 1 ohm, 2*09 ohms 

 extra resistance was needed to get the best effect. Thus the 

 resistance in the circuit in addition io that of the lamp was 

 about 2'2 ohms. The mean current was *6 ampere, and a 



