ON PENDULUMS. 81 



Mr. Troughton's apparatus, alluded to in his paper is differ- Troughton's 

 ent from thofe, and when he dial! favour me with the account p 

 of it, and the experiments lie is making, the world will fee that 

 it anfwers its purpofe with fuch facility and precifion as do ho- 

 nour to his great [kill in thefe fubjefts. 



Circular Pendulum of Huygens.] The difficulties attending Remarks on 



the conftru&ion of this oendulum feem to be principally in the Hu y£ ens s CIT ' 



1 . cular pendulum. 



firing or chain. Perhaps fine metallic wire might delerve the 



preference. It would not probably be difficult to bring the 

 curve A B fufficiently near the true figure to anfwer its general 

 purpofe, through the changes of force in the firft mover, and 

 of refiftance in the air. Practical men have not thought the 

 cycloidal cheeks applied by Huygens, and fince him by others, 

 to the common pendulum, of any utility; but have rather di- 

 rected their attention to fmall vibrations, or an equalizing of 

 (he firft mover by periodically detaching the train. Huygens 

 liimfelf alfo propofed this expedient. It is generally admitted 

 to be an advantage that the regulating inftrument, whether pen- 

 dulum or balance, mould perform the greateft part of its mo- 

 tion unconnected with the train. The circular pendulum is 

 confiantly fo connected. 



As the ingenious author has not detailed his reafons for Whether it be 

 thinking that pendulums of this kind will not be affedted by e , s ', a %° 

 change of temperature, I would (imply remark that it does not temperature. 

 appear to me why an expanfion of the firing mould not caufe 

 the pendulum of Huygens to revolve more flowly, or that an 

 expanfion of the curves in the other figures would not dimi- 

 nifh their curvature and produce the fame retardation. 



The tube may be enclofed in a circular vejfel.] As the refiff- On the method 

 ance of the air will vary no lefs than a tenth or a fifteenth part pendulum" fcc/ 

 according to the ftation of the barometer, we might expedt 

 time pieces to be confiderably affecled by its irregularity. In 

 aftronomical clocks, with heavy pendulums and fhort vibra- 

 tions, this quantity is extremely minute, as is proved by their 

 very correct performance. But there are fa£ls attending the 

 performance of our bed portable chronometers, which fhew 

 that it is not in tbem an inconfiderable obje£t. Several years 

 ago I entertained a notion that the refiftance of the air might 

 be rendered equable, or in fact done away, by boxing up the 

 balance, as is here propofed by our author. But from his 

 own excellent paper on the blaft ventilator inferted in the 



Vql, X,— February, 1805. G fourth 



