293.  The galvanometer was one or the other of those formerly described (87. 205.), but the glass jar covering it and supporting the needle was coated inside and outside with tinfoil, and the upper part (left uncoated, that the motions of the needle might be examined,) was covered with a frame of wire-work, having numerous sharp points projecting from it.  When this frame and the two coatings were connected with the discharging train (292.), an insulated point or ball, connected with the machine when most active, might be brought within an inch of any part of the galvanometer, yet without affecting the needle within by ordinary electrical attraction or repulsion.

294.  In connexion with these precautions, it may be necessary to state that the needle of the galvanometer is very liable to have its magnetic power deranged, diminished, or even inverted by the passage of a shock through the instrument.  If the needle be at all oblique, in the wrong direction, to the coils of the galvanometer when the shock passes, effects of this kind are sure to happen.

295.  It was to the retarding power of bad conductors, with the intention of diminishing its intensity without altering its quantity, that I first looked with the hope of being able to make common electricity assume more of the characters and power of voltaic electricity, than it is usually supposed to have.

296, The coating and armour of the galvanometer were first connected with the discharging train (292.); the end B (87.) of the galvanometer wire was connected with the outside coating of the battery, and then both these with the discharging train; the end A of the galvanometer wire was connected with a discharging rod by a wet thread four feet long; and finally, when the battery (291.) had been positively charged by about forty turns of the machine, it was discharged by the rod and the thread through the galvanometer.  The needle immediately moved.

297.  During the time that the needle completed its vibration in the first direction and returned, the machine was worked, and the battery recharged; and when the needle in vibrating resumed its first direction, the discharge was again made through the galvanometer.  By repeating this action a few times, the vibrations soon extended to above 40 deg. on each side of the line of rest.

298.  This effect could be obtained at pleasure.  Nor was it varied, apparently, either in direction or degree, by using a short thick string, or even four short thick strings in place of the long fine thread.  With a more delicate galvanometer, an excellent swing of the needle could be obtained by one discharge of the battery.

299.  On reversing the galvanometer communications so as to pass the discharge through from B to A, the needle was equally well deflected, but in the opposite direction.

300.  The deflections were in the same direction as if a voltaic current had been passed through the galvanometer, i.e. the positively charged surface of the electric battery coincided with the positive end of the voltaic apparatus (268.) and the negative surface of the former with the negative end of the latter.