NEW PENDULUM. 227 



Reckoning from the top the firft aaion is downwards, and Defcrlptlon of 

 confifts of the fpring, a fliort wire 0,2 diameter, and a Jong pg^^j^i^j^ by 

 wire 0,1 diameter; thefe all of fteel firmly connc6led, reach the drawing* 

 down within an inch of the centre of the bob, and occupy the 

 middle line of the whole apparatus. To the lower end of the 

 middle branch, is fatlened the lower end of the interior brafs 

 tube, 0,6 in diameter, which terminates a little ftiort of the top 

 of the exterior tube, and produces the firft dilatation upwards. 

 From the top of the interior tube depend two wires 0,1 dia- 

 meter, whofe fituation is in a line at right-angles to the fwing 

 of the pendulum, and reach fomewhal lower than the attached 

 tube itfelf, which they pafs through without touching, and ef- 

 fect the fecond expanfion downwards. The fecond action up- 

 wards is gained by (he exterior tube, whofe internal diameter 

 juft allows the interior tube to pafs freely through it: its bot- 

 tom is conneded with the lower ends of the laft deicribed wires* 

 To complete tiie correction, a fecond pair of wires of the fame 

 diameter as the former, and occupying a pofition at right angles 

 to them, ad downwards, reaching a little below the exterior 

 tube, having alfo palTed through the interior one without 

 touching either. The lower ends of thefe wires are fattened 

 to a fiiort cylindrical piece of brafs, of the fame diameter as the 

 exterior tube, to which the bob is fufpended by its centre. 



Fig. 3. is a full fize fedion of the rod, in which the three 

 concentric circles are defigned to reprefent the two tubes, and 

 the rectangular pofition of the two pair of wires round the raid- • 

 die one, are fiiewn by the five fmall circles. By copying this 

 arrangement, from the elegant conftru6lion of your own half 

 feconds pendulum (Philf. Journal for Auguft, 1799} I avoided 

 much trouble, which muft have occurred to me, unlefs indeed, 

 I had been impelled on the fame idea, by the difficulty of con- 

 triving the five wires to adl all in a row, with fufficient free- 

 dom and in fo fmall a fpace. Fig. 4. explains the part which 

 clofes the upper end of the interior tube: the two fmall cir- 

 cles are the two wires which depend from it, and the three 

 large circles ftievv the holes in it, through which the middle 

 and other pair of wires pafs. 



Fig. 5. is defigned to explain the part which flops up the 

 bottom of the interior tube, the fmall circle in the centre is 

 where the middle wire is fattened to it, the others the holes for 

 -|iie other four wires to pafs through. Fig, 6. is the part which. 



Q2 . clofes 



