ON PENDULUMS, 79 



Reference to the Figures. 



Fig. ]. An exact copy of the figure of Huygens's circular 

 pendulum, from the Horologium. B G A, the edge of the 

 paraboloidal lamina : F E, the parabola produced by its means : 

 F, the ball. 



Fig. 2. The firft conftruction propofed in place of Fig. 1. 

 A C, the parabolical glafs tube : A, the mercury: C, the air- 

 vent: B B, the encloflng circular vefiel reprefented in fection. 



Fig. 3. The third construction propofed infiead of Fig. 1. 

 AC, the metallic rod bent into a parabolical figure: B, the 

 fpherical weight : V, the roller on which it moves along A C : 

 the parts in this figure are reprefented double, to (hew the belt 

 method of adopting this plan as a regulator for wind-mills and 

 other engines. 



Fig. 4-. The propofed conftruction for an ofcillatory pendu- 

 lum, which, it is imagined, will not be afiecled by change of 

 temperature. A, the ofcillaling cylinder: B, the cycloidal 

 fup port. 



Fig. 5. A fection of the (rough mentioned in the fecond pro- 

 pofed conftruction, with thefphere in it. 



Annotations. W. N. 

 Pyrometers, p. 74.] The common pyrometers of the (hops are The beft pyro- 

 indeed liable to the objections of our author; b ut ingenious men ^ve e faIIacious 

 have long ago obviated them in their experiments. Deluc, inrefuks. 

 the Philofophical Transitions for 1777, gives a method of af- De,uc ' s mcthod « 

 certaining the relative expanfions of two different metals by 

 heat. He fufpended one of the bars to an arm proceeding 

 horizontally from an upright deal plank, and he fupported the 

 other bar by refting its lower end upon a fmall cock or Mage 

 proceeding from the lower end of the former. A microfcope 

 was attached to the plank in fuch a manner that, while it was 

 conftanlly fupported by an horizontal arm, it could be moved 

 fo as to keep the laft mentioned or (landing bar in the focus of 

 radiation for diftinct vifion. Heat was applied to the bars, 

 and the microfcope was directed to a point on the (landing bar. 

 When by repeated drifting the microfcope a point was found 

 which was neither raifed nor deprefled by the changes of tem- 

 perature, the refpective lengths of the bars were inverfely as 

 their expanfive powers. There was no mechanifm, and the 

 plank would remain at the fame temperature during the expe- 

 riments, 



