330 Mr Connell on the Action of 



a coil of about thirty circuits of insulated copper wire ; * and its 

 indications were quite sufficient for my purpose, as my principal 

 object was to detect such electric currents as might be the cause 

 of chemical decomposition, and not more trifling streams, if such 

 exist. When the current was powerful, as from 216 pairs of four- 

 inch plates, the effect was considerably more marked, although 

 still limited. When, however, minute quantities of certain solu- 

 ble foreign bodies, particularly of potash, were added, the con- 

 ducting power was much augmented, as was shewn by the in- 

 creased action on the galvanometer, and by the great increase in 

 the decomposing action. I have already shewn that 1^ dram of 

 absolute alcohol yielded with difficulty .2 or .3 of a cubic inch of 

 hydrogen in an hour's action of the power of 216 pairs ; whilst 

 the very same alcohol, with only g-}^ of potash dissolved, and 

 with a power of seventy-two pairs, yielded one cubic inch of gas 

 in one-fourth of that time. 



It is remarkable how minute a quantity of potash is capable 

 of causing symptoms of decomposition to be seen in alcohol, un- 

 der circumstances where otherwise it cannot be observed. Thus, 

 if absolute alcohol be acted on in a watch-glass by a power of 

 fifty pairs of two-inch plates, by simply approaching the plati- 

 num foil poles to one another horizontally, no action whatever is 

 observed when the alcohol holds nothing in solution ; but the 

 presence of even -g oV o" P ar ^ °f potash renders a minute stream of 

 bubbles visible from the negative pole. 



I formerly mentioned, that, when a solution of iodine in al- 

 cohol is acted on by a moderate power, no symptoms of decom- 

 position are observed. The effect on the galvanometer is, how- 

 ever, increased by the presence of the iodine. 



* I compared the indications of this galvanometer with those of two astatic sew- 

 ing needles, each two inches long, one of which was placed in the centre of a coil 

 of twenty circuits, and found the indications of the long needle fully the more de- 

 licate. 



7 



