^ ' Captain Kater's account of the 



ations in length of the seconds* pendulum, doubts may be 

 inferred of the accuracy of the method which I employed in 

 the observations for ascertaining the length of the seconds* 

 pendulum in London, as well as in those which have been 

 made with the invariable pendulum. It is asserted, that 

 taking a mean between the disappearance and re-appearance 

 of the disk, is a more correct method of observation than 

 that which I have pursued ; and that the intervals between 

 the coincidences obtained by observing the disappearances 

 only of the disk, would be productive of error. 



With respect to the convertible pendulum, it will be seen 

 on referring to the Philosophical Transactions for 1818, that 

 the disk was made to subtend precisely the same angle as the 

 tail-piece of the pendulum ; so that at the moment of disap- 

 pearance of the last portion of the disk, its centre coincided 

 with the middle of the tail-piece, a circumstance which, in 

 my method of observing is indispensable, when the object is 

 to obtain the true number of vibrations made by the pendu- 

 lum in twenty-four hours. 



With the invariable pendulum, from causes unnecessary 

 here to detail, the circumstances were somewhat different, the 

 disk subtending a less angle than that of the tail-piece of the 

 pendulum ; in consequence of which, the interval between the 

 apparent coincidences was lessened, and the inferred number 

 of vibrations in twenty-four hours diminished about two- 

 tenths of a vibration ; but as the experiments with the inva- 

 riable pendulum are intended to be merely comparative, and 

 should therefore be made as nearly as possible in every 

 respect under similar circumstances, no part of the apparatus 

 being changed, nor any alteration made in the pendulum of 



