as a Measure of Electricity. ^SH 



one of iron, the other of steel, there was produced a deeidetl 

 deviation : since, then, the only difference in the metals arises 

 from an alloy of from a ^^^ to a j^^y part of the whole, it ap- 

 pears that this is sufficient to alter their electric relations. — 

 The powerful affinity of potassium for oxygen, made it highly 

 probable, that, in the galvanic circuit, it would become strongly 

 negative with all the metals. On my first trial with disks of 

 potassium and zinc, the potassium took fire before I could ob- 

 serve the effect; this difficulty I afterwards obviated by alloy- 

 ing it with mercury ; on making the contact the needle deviated 

 through nearly a right angle : The same effect was produced 

 with copper ; it was needless to try it with the other metals, 

 for being negative with respect to zinc, and zinc being nega- 

 tive with respect to all the other metals, there can be no doubt 

 that in the galvanic circuit, potassium is the most strongly 

 negative metal with which we are acquainted. — It is per- 

 haps scarcely necessary to remark, that, if any proof of the 

 metallic nature of potassimn were wanting, this experiment 

 would have afforded it. 



In using the magnetic needle as a measure of galvanic 

 effects, we may either observe the deviation at a standard di- 

 stance of the connecting wire from the needle, or assume a 

 standard angle and measure the distance. — The latter method 

 seems to have the advantage, as enabling us to use a smaller 

 and therefore a more delicate needle, with this additional con- 

 venience, that the scale is increased in proportion as the length 

 of the needle is diminished. — I therefore constructed an in- 

 strument, having a connecting wire fixed upon a moveable 

 slide divided into inches and tenths, to which a vernier might 

 be added if necessary. My first object was to ascertain the 

 divisions on the scale, corresponding to variations in the angle 

 of deviation ; — for this purjiose, the moveable wire was placed 

 at different distances from the needle, increasing in arithmeti- 

 cal progression, and the corresponding deviations were ob- 

 served. As the effects decrease very rapidly during the gal- 

 vanic actions, the experiments were made as quickly as possi- 

 ble, proceeding from a distance of ^ an inch to 10^ inches, 

 and again returning to the first distance. On taking the mean 

 of several trials, made in this manner, I found that the tan- 

 gent of the deviation varies inversely as the distance of the 

 connecting wire from the magnetic needle. 



It is well known that in a galvanic arrangement, intensity 

 is given by the number, quantity by the magnitude of the 

 plates; hut I am not aware that any notice has been taken of 

 the eflL'Cts ju'oduced by varying their distances from each other. 

 — On placing a moveable plate of zinc opposite a fixed cop- 

 per- 



