332 Electro-Magnetic Experiments. 



provide themselves often, on long voyages, with a set of magnetic 

 bars, by means of which the strength of their compass needles, if im- 

 paired or lost, may be restored. But it is possible, nay it is proba- 

 ble, that the same stroke of lightning which destroys the magnetism 

 of the compass needles may also spoil that of the steel bars. A cop- 

 per trough of modern dimensions, a zinc plate, a little sulphuric and 

 nitric acid, or, if that was considered too dangerous, some sal- 

 ammoniac, and withall a soft iron wire, bent in the shape of a horse- 

 shoe, and wound round with copper wire, would constantly insure 

 the certitude of restoring magnetism on ship board. 



After making these experiments, I was anxious to know if the 

 power of the magnet was susceptible of still farther increase by aug- 

 menting the strength of the galvanic apparatus. A second trough, 

 the acting superficies of which was about six square feet, was added 

 to the first. The zinc plate of one trough was connected with the 

 zinc of the other, as also the copper of the one with the copper ol 

 the other. Thus the superficies brought to action was about seven- 

 teen square feet. But the horse-shoe magnet had acquired, it would 

 appear, its maximum of strength, for its magnetic force was not ma- 

 terially increased by this augmentation of galvanic power. 



I afterwards tried, with the same view, the powerful galvanic appa- 

 ratus of Colonel Offerhaus, which I formerly described in Dr. Brews- 

 ter's Journal.* But no increase of galvanic power was capable ol 

 increasing the strength of the magnet beyond a certain limit. 



Another iron wire, bent in the shape of a horse-shoe, and similar 

 in every respect to that with which the preceding experiments were 

 made, was surrounded with a spiral coil of brass, twisted from lelt to 

 right, (dextrorsum.) The effect was exactly the same as with cop- 

 per wire, except that the spiral being wound to the right hand side, 

 the north pole was, as might have been anticipated, on the side con- 

 nected with the zinc. 



A brass horse-shoe, wound round either with an iron, a brass, or 

 copper wire, did not show the least effect. I did not, indeed, expect 

 the least result from this experiment, which was made at the instance 

 of a friend. 



It has been shown by the former experiments, that it is ol litti 

 consequence whether a brass or a copper spiral be used. An iron 



Edinburgh Philosophical Journal, t 6, p. 52. 



