Chemistry and Physics. 145 



(2.) The flame behaves like an electrolyte to the metal electrodes 

 which arc immersed in it (Matteucci). 



(3.) The explanation of the electricity <>f flames is found in the 

 thermoelectric difference of the electrodes (Buff). 



Julius Elster and Hans < T eitel believe thai previous observers 

 have overlooked the influence which the heated layer of air out- 

 side the flame exerts. On account of the great resistance of this 

 air, observers who have used a galvanometer could not detect a 

 difference of potential. They, however, used a Thomson's elec- 

 trometer and were enabled 'to study the phenomena, believing 

 that they were the first to use a Thomson's electrometer for this 

 purpose/ The same method, however, was used by J. Trowbridge 

 (this Journal, 1873). 



Their conclusions are as follows : 



(J.) The length polarization of the flame (Hankel) is only an 

 apparent phenomenon and is due to the unequal immersion of the 

 wires which serve as electrodes. 



(2.) The flame appears in section to be strongly polarized— and 

 the electrode in the surrounding layer of air appears to be 

 always positive toward the electrode in the flame. 



(3.) The electromotive force is independent of the size of the 

 flame. 



(4.) The change of polarity of the flame is due to change of 

 position of the electrodes. 



(5.) The electromotive force of the flame is dependent upon the 

 nature of the metal from which the exploring electrodes is made, 

 and also upon the nature of the burning u\as. Aluminum and 

 magtie>'umi call forth strong electrical elfects. If the electrode 

 in the layer of air is covered with a salt the electrical effect is 



('<>.) With the use of water electrodes and exclusion of metals, 

 the electrical phenomenon of flames is also seen. The electrode 

 in the outer layer of heated air is always positive to that in the 

 flame itself 



(7.) Flames can be combined in series just as batteries are 

 joined for intensity.— Ann. tier Physik vnd Chemie, No. 6, .pp. 

 193-222, 1882. J. T. 



9. Kerr's Phenomenon— H. Broxgersma has repeated Kerr's 



Mag., No. 4 (50), p. 337, lS7-->. and has corroborated Kerr's 

 ults which certain observers have failed to repeat. Bron- 



ntal imnortance in connection with the elect rodynamic theory 

 light.— Ann. der Physik und Chemie, No. 6, pp. 222-233, 



