66 VHILOSOPHICAL TRANSACTIONS. [aNNO 1/50. 



have only 2 poles; but about the hole In the middle the curves formed of steel-dust 

 were a little confused. These always came exactly to the same point, after vi- 

 brating a long time ; and if drawn ever so little on one side, would return to it 

 again without any sensible difference. We may therefore conclude, that a re- 

 gular parallelopiped is the best shape for a needle, as well as the simplest ; with 

 the holes for the caps as small as can well be contrived ; or if it can be made to 

 answer the purpose without any hole at all, it will be still more perfect. Yet the 

 common shape has one advantage which this has not : for being made broad at 

 the ends, and slender in the middle, its weight is removed as far as possible from 

 the centre : on which account, if it once points true, the friction at the centre 

 cannot so easily put it in motion ; and its vibrations, when in motion, will be 

 slower ; so that their limits may be more nicely observed, and the middle point 

 between them is that where it would stand, if at rest. 



Being unwilling to part with these advantages, Mr. K. contrived a light circle 

 of brass, of the same diameter with the card, which will supply a weight acting 

 at the greatest distance from the centre of motion, and also serve to support the 

 card ; which may now be made of thin paper, without any thing to stiffen it. 

 So that the extraordinary weight of the brass ring is compensated in a great mea- 

 sure by the lightness of the card. This ring is of service in another respect ; for 

 being fixed below the card, and the needle above it, the centre of gravity is 

 placed low enough to admit of the cap being put under the needle ; by which 

 the hole in the needle becomes unnecessary ; and the latter being placed above 

 the card, renders it easier to be touched with a pair of bars. 



Having thus completed the needle and card to his satisfaction, what chiefly 

 remained, was to contrive such a cap and point as will have the least friction 

 and be most likely to continue in a state of perfection. The caps in use are 

 either of brass, a mixed metal, like that of a reflecting telescope, crystal, or 

 agate. The first 2 will only admit of brass points, and the latter are rather too 

 expensive for common use. He therefore thought of trying glass caps : he had 

 3 of them made by a' glass-blower, 2 of which he got polished : they were all set 

 in brass, so as to screw into the same needle, which had also one of agate fitted 

 to it. He compared them with that of agate, by trying with each of them how 

 many vibrations the same card and needle would make, when drawn aside Q0°, 

 on the same point ; which was a very small sewing needle. 



The number of vibrations with the agate cap, on the first trial, were 39, then 

 37, then 39 again ; with one of the glass caps it made 23, and then 20. This 

 difference from the agate cap was so great, that he concluded the point must 

 be damaged, and therefore chose a finer ; on which the same glass cap made 4 1 

 vibrations; then 43; and another glass cap made 47, and the next time 43. But 

 the agate cap with this point made 51, 57, and 58 vibrations. The unpolished 



