284 Mr. G. A. Hemsalech on the Origin of the 



carbon tube. Under these conditions, and with a rod tem- 

 perature not far from 3000° C, they obtained a steady 

 ionization current of about 3J amperes. It seems to me 

 that this experiment proves conclusively that at the higher 

 furnace temperatures an electric current must pass through 

 the ionized vapours contained between the extremities of the 

 resistance-tube. 



I have repeated the second experiment of Drs. Harker 

 and Kaye with a slight modification, so as to approach more 

 nearly the actual working conditions prevailing in the present 

 investigation. The furnace was of the type already described 

 in the preceding paper (§ 2), the carbon tube having an 

 internal diameter of 14 mm. and an effective length of 

 4 inches between the graphite blocks. A carbon rod 4 mm. 

 in diameter was mounted in such a way that it could be 

 moved along the axis of the furnace whilst remaining in a 

 concentric position with regard to the tube. Thus the radial 

 distance between the carbon rod and the inner furnace-wall 

 was always as nearly as possible 5 mm. The carbon rod was 

 connected to one end of the furnace-tube with an ammeter in 

 the circuit, as shown in fig. 1; and it was provided with a 



Fig. 1. 



"i Furnace Tube 



Carbon Rod 



Method of measuring' Ionization Current. 



division, so that the amount of its penetration into the tube 

 could be ascertained. The experiments were made at a 

 temperature of 2200° C; at this temperature the carbon rod 

 does not bend but remains perfectly straight throughout the 

 time necessary for the various manipulations and readings 

 involved. The difference of potential at the ends of the 

 furnace-tube, at this temperature, was about 9 '5 volts, or 

 equal to a voltage drop of 1 volt per centimetre approximately. 

 The following table gives the values of the ionization currents 

 obtained under these conditions for various positions of 

 the rod. 



