the Action of Galvanic Batteries. 35 



teries employing porous cups show irregularities more or less 

 marked. It will be noticed that the variations may be divided 

 into two classes, one a general undulation, the other a series of 

 rapid fluctuations which make up the undulations. There are, 

 likewise, two separate actions which cause these irregularities : 

 first a diminution in the current strength, caused by the pores 

 of the partition becoming filled with the base, and thus pre- 

 venting action until it has been dissolved and fresh acid can 

 attack the zinc ; second, a diminution of the acid at the posi- 

 tive pole, and a consequent decrease in current. This action 

 is due to electrical endosmose. The undulations are due to the 

 first of these causes ; the fluctuations to endosmose. It will be 

 observed that the fluctuations in all cases begin as soon as the 

 current is made, and before it is possible that the partition can 

 have become so impregnated as to cause an interruption to the 

 current. This can be well seen in figures 4 and 5. This shows 

 that the fluctuations and undulations must be due to different 

 causes. In figure 4 the undulation is very marked. This 

 figure is the photograph of the action of a battery using a very 

 dense porous cup. Figure 3 shows the action with a cup, the 

 same size as that used in figure 4, made of ordinary unglazed 

 paper. We find here no action whatever. We should suppose 

 from the above experiments that the more dense the partition 

 the more liable would it be to become clogged, whereas with a 

 very porous cell there should be very little resistance offered. 

 To prove this, a cup of very porous earthenware was compared 

 with one much denser ; figures 7 and 8 show the results. In 

 the case of the porous cup the action is without undulations, 

 while with the other the undulations are quite marked. 



To investigate the action of endosmose, a very irregular open 

 circuit battery was employed, consisting of a solution of bichro- 

 mate of potassium and sulphuric acid, into which were plunged 

 a piece of carbon and a porous cup containing zinc surrounded 

 by mercury. In the action of a galvanic battery there are 

 three separate actions which take place : 



1st. A decomposition of the electro-positive ion at the posi- 

 tive electrode. 



2d. A decomposition of the electro- negative ion at the nega- 

 tive electrode. 



3d. The electric current carries whatever comes in its way 

 from the positive to the negative electrode. 



It is this last which is called endosmose, and which we wish 

 to investigate. 



It was discovered by Breda and Logeman that in a continu- 

 ous liquid this third action disappeared, and only when a porous 

 partition was introduced was this phenomenon observed. This 

 agrees with the above stated theory ; for no such fluctuations 



