Intelligence and Miscellaneous Articles. 155 



zation were to be expected. The force fell, during the first half of 

 this time; from 4-83 to 0*40, and then gradually rose to 1*37, to 

 sink again toward the close of the period mentioned ; but the alte- 

 rations took place slowly and without any quick leaps. 



It now appeared that if for only five minutes two Daniell ele- 

 ments were closed between the one outer plate A and the middle 

 plate B, which carried hydrogen to the plate B, a change occurred in 

 the behaviour of the other side of B toward the second outer plate C. 



For example, in a series of experiments, immediately before 

 closing the current through A and B the pressure-difference between 

 B and C had been — # 6 ; after the current had been passing for 

 five minutes, while plate C remained electrically insulated, the pres- 

 sure-difference was + 5-2 ; it then rose in the next three hours, the 

 plates being insulated, to 17-1; and 18 hours later it was again — 3. 



In other experiments the current between A and B was closed 

 during 12 or 18 hours. The difference of potential between B and 

 and was then greatest immediately after the interruption of that 

 current, and gradually diminished during the hours that followed. 



"When the current between A and B received the opposite direc- 

 tion, so that it urged oxygen toward B, the opposite result occurred 

 on the other side of B, with the same quickness, and partly with 

 same, partly with greater intensity than in the previous hydro- 

 gen polarization. Whether this difference in the quantity of 

 the effect is or is not conditioned by the antecedent hydrogen 

 polarization cannot be decided from the experiments hitherto made. 

 As sometimes the state of the plates B and C has been equalized 

 by joining them to the circuit, a sensible quantity of hydrogen may 

 then have been also carried over to C, and thence may arise the more 

 powerful action of B, polarized with oxygen. In general, quanti- 

 tative determinations of the action were frustrated by the circum- 

 stance that the gas once forced into the plate only very slowly 

 becomes equally diffused in its interior, and can only be removed 

 again very slowly. Moreover the action is quite evident when only 

 a single Daniell element between A and B is employed ; and, on 

 the other hand, the action of two Bunsen elements was but a 

 trifle greater than that of two Daniells. The cause of this may be, 

 that when electrolytic gas-production commences, the escaping 

 portions of the gases withdraw themselves from the force which 

 presses them into the platinum ; and hence, probably, a further re- 

 inforcement of the electromotive force augments the decomposition 

 of the water, but the pressure of the gases into the platinum is 

 increased very little or not at all — as indeed it is known that the 

 polarization of the plates, when once it has arrived at gas-develop- 

 ment, can be raised but little higher. 



When plates B and C were put for a short time in conducting 

 connexion after B had been charged with gas from A, immediately 

 after interruption of the conduction the difference of potential be- 

 tween the two was =0, but then rose again on the same side as it 

 had been before the conducting connexion, in precisely the same 

 way as this usually takes place when both plates have been polarized 

 by an electric current led directly through them. 



