734 THE ELECTEIC BELL. 



quently of the hammer. If the end of the screw s were in direct 

 contact with the keeper, / 2 being altogether removed, the current 

 would be always interrupted at the very instant when the keeper, in 

 consequence of the attraction of the electro- magnet, has moved through 

 the smallest possible distance ; the keeper would then go back again 

 immediately, because the current has ceased. It would again make 

 contact, and again be attracted through quite an insensible distance, 

 and so on ; in other words, the ' amplitude ' of the oscillations of the 

 keeper would be so small that there would only be a buzzing sound, 

 caused by the vibrations of the keeper and the attached pieces of 

 metal. But since the end of the screw * presses gently against the 

 spring / 2 , the tension of the spring is only gradually relaxed when 

 the keeper is drawn towards the electro-magnet ; the consequence 

 is that the contact between / 2 and s, and hence the duration of the 

 current, is maintained for some little time, and that when the 

 contact really ceases the keeper has already moved through a sen- 

 sible space and acquired an appreciable velocity. The amplitude of 

 the oscillations of keeper and hammer is thus rendered sufficiently 

 great for the hammer to strike against the bell ; when the current 

 stops, the keeper with its appendages is forced back into the former 

 position by the tension of the spring f lt the end of s presses once 

 more against / 2 , the circuit is again complete, the effect is repeated, 

 and the ringing is continued. 



Fig. 370 shows a key of easy construction. Two strips of sheet 

 brass, one of them (in the figure the one on the right) made 

 elastic by hammering and slightly bent, are screwed to a small boai'd 



FIG. 370 (an. proj. ; \ real size). 



by means of two wood-screws. The ends of the connecting wires 

 are bent into loops, and these are clamped beneath the heads of the 

 two outer screws. By pressing the bent strip down, contact is 

 established with the straight one, and the circuit is closed ; on 

 releasing the strip the circuit is again opened, and the bell stops 

 ringing. 



A key of this kind may frequently with advantage be interposed 

 in the circuit in many experiments on galvanic electricity and 

 electro-magnetism ; as, for example, in those on the heating effect 

 of the current, and in using the various kinds of apparatus repre- 

 sented in figs. 343 to 345, 351 to 354, 361 and 362. It is best in 



