1886.] 



A Tlieory of Voltaic A ction. 



309 



the slightest puff of breath over them caused the formation of a film, 

 which, though quite invisible, easily formed enough current to send 

 the index quite off the scale. 



32. I now made some experiments to ascertain whether the film cell 

 could be polarised by the current from a Leclanche element connected 

 to the plates. 



The following examples show that this is so, and that the polarisa- 

 tion effect is so great as at first even to overpower the ordinary 

 electromotive force of the film cell. 



Column I gives the mode of connexion of the poles of the cell to 

 the plates ; P the carbon pole. 



II. — The deflection observed on the galvanoscope. 



III. — The zero point of scale of reflecting galvanometer. 



IV. — Its initial swing on having the terminals of film cell changed 

 to it. 



V. — The point at which the index rested after its first swing. 

 (1.) Hygro metric state of air=0*8 — 



I 



II 



III 



IV 



1 





P to Cu 



25 



17 





30 



Falling. 





71 



20 





16 



Eising, 







Eeadjusted plates. 





P to Cu.. .'. .. 



12 



30 





45 



Falling. 





68 



32 





22 



Eising beyond 32. 







Eeadjusted plates. 







16 



40 



60 



47 



Falling. 



P to Zn 



74 



44 



31 



40 



Eising. 



(2.) Hygrometric state of air=0'826 — ? 



I. 



II. 



III. 



IV - 



Y. 1 



! 



P to Cu 



52 

 74 



12 

 21 



60 

 14 



30 



Falling. 



Eising till after some 

 minutes the deflec- 

 tion passed the zero, 

 showing the ordinary 

 current from film cell. 







[32a. Several of the foregoing experiments go to show that a cessa- 

 tion of chemical action at the surface of the plates implies a subse- 



