24 Dr. E. N. da C. Andrade on the Electrical 



the conductivity of the vapour when the electrodes do not 

 stand in the vapour itself, which should throw some further 

 light on the matter. 



§ 6. Apparatus for the experiments at High Pressure. 



The colourless flame of air and gas burnt from a flat 

 quartz flame-piece between two thick brass electrodes stand- 

 ing outside the flame : the bead of salt could be brought at 

 will into the flame by a small magnetic arrangement. The 

 disposition of the apparatus was much as described else- 

 where *, the same burner, electrodes, and influence machine 

 being used. The flame, however, burnt in a closed vessel, 

 in which the pressure could be varied from atmospheric to 

 six atmospheres ; the vessel was made in cylindrical form of 

 iron lined with copper, and was 40 cm. in diameter and 

 60 cm. high ; the top and bottom were removable, being- 

 bolted on with screws and nuts ; the joints were made air- 

 tight with thick rubber rings. In the bottom of the a 

 were an inlet for gas and two for air, one being to supply 

 the burner with air, and the other to keep the air in the 

 vessel constantly renewed. Insulating plugs in the bottom 

 of the vessel bore the wire to the insulated electrode, the 

 wire for supplying the current to small magnet, and the two 

 wires connected to the parallel " probes r of fine platinum 

 wire used lor measuring the fall of potential in the flame ; in 

 the last case the insulations were of amber. All the plugs 

 were protected from dirt and moisture by brass caps. The 

 flame was observed through a small plate-glass window. 

 The cover of the vessel was provided with a pressure gauge 

 and a cock to regulate the escape of air : this cock communi- 

 cated with a copper chimney which came directly over the 

 flame, and collected the foul air. The vessel was surrounded 

 for two thirds of its height by a cooling jacket of thin tin- 

 plate, through which a constant circulation of cold water was 

 maintained. The coal gas and air used were compressed in 

 a neighbouring room by means of a small compression pump, 

 and stored in large iron vessels at 10 atmospheres pressure, 

 from which they were brought to the experimenting room 

 through lead pipes. 



The flame is lighted and the cover bolted on. Compressed 

 air is then lei into the vessel : as the pressure rises the supply 



of air and gas t<> the flame imiM be carefully increased, in 

 order to keep the flame burning at a constant size. With a 

 little practice the flame may be raised to a pressure el' five 



E, V d:i 0. Andrade, he. eU. p. - 



