2 Captain Kater*s account of the 



plates, on which the pendulum had rested during the expe- 

 riments, had suffered penetration by the knife edges. The 

 length of the seconds pendulum however, deduced from 

 these first experiments, did not differ more than two ten- 

 thousandths of an inch from the final determination. 



It may also be seen, that in repairing the knife edges after 

 the first set of experiments had been concluded, one of them 

 was broken ; and when replaced by another, the distance 

 between the knife edges was increased about one hundredth 

 of an inch ; so that two results, differing by so small a quan- 

 tity from each other, may be considered as having been 

 obtained by two different instruments. 



The Philosophical Transactions for 1819 contain an account 

 of experiments for determining the variation in length of the 

 seconds pendulum at the principal stations of the Trigonome- 

 trical Survey of Great Britain. For this purpose I constructed 

 an invariable pendulum, the number of vibrations made by 

 which being observed in London, and also at the principal 

 stations of the Trigonometrical Survey, the length of the 

 seconds pendulum, at such stations, could readily be deduced 

 from the previously known length of the pendulum vibrating 

 seconds in London. 



In this manner, the length of the pendulum vibrating 

 seconds at Unst was found to be S9,i7i46 inches, and at 

 Leith Fort 39,15554 inches. 



The " Reciieil d' Observations Geodesiques /' &c. which may 

 be considered as a fourth volume of the " Base du Systeme 

 metrique," affords me an opportunity of comparing these 

 last results with those of M. Biot, whose very valuable 

 labours for determining the length of the pendulum vibrating 



