84- 



near the knife-edge, at the opposite end, and may be fixed at plea- 

 sure ; the third weight is a small slider, intended to move near the 

 centre of the bar, upon which are engraved divisions of one twentieth 

 of an inch, seen through an opening in the slider. The support of 

 the pendulum consisted of agate planes bedded in bell-metal. 



In proceeding to the details of the experiments, the author acknow- 

 ledges his obligations to Henry Browne, Esq. F.R.S., who permitted 

 him to use his house in Portland-place, and his excellent clocks, for 

 the purposes of the investigation. The greatest daily variation of the 

 clock used as a standard of comparison did not exceed three tenths 

 of a second between the months of February and July. 



By the method of coincidences which Captain Kater employed, the 

 number of vibrations made by the pendulum in twenty-four hours 

 might be obtained in the space of eight minutes to within half a 

 second of the truth ; and the usual correction was applied for the 

 extent of the arc of vibration. 



The pendulum being suspended with the great weight above, the 

 number of vibrations in twenty-four hours was determined ; and if 

 it differed when the pendulum was inverted, it was equalized by 

 moving the second weight, and finally adjusted by the slider, every 

 allowance being made for the temperature, and the height of the 

 barometer being noted. Thus the number of vibrations in twenty- 

 four hours, of a pendulum equal in length to the distance between 

 the knife-edges at a given temperature and barometrical height, was 

 ascertained. 



The next sections of Captain Rater's communication refer to the 

 apparatus and methods employed for the measurement of the distance 

 between the knife-edges ; for the comparison of the British standard 

 measures of the highest authority ; and to the expansion of the pen- 

 dulum, which was found to be -00000996 of its length for each degree 

 of Fahrenheit's thermometer. 



After describing the methods of deducing the length of the pen- 

 dulum vibrating seconds, and the corrections for the buoyancy of the 

 atmosphere, the author makes it appear, that the distance of the 

 knife-edges, at the temperature of 62 Fahr., by the mean of three 

 several sets of measurements, the greatest difference between any 

 two of which did not amount to T-r.W-o-th of an inch, was, upon Sir 

 George Shuckburgh's scale, 39'44085 inches. From a table inserted 

 in this paper of twelve sets of experiments, each set consisting of 

 four, from which, and from the preceding measurements, the length 

 of the seconds' pendulum in vacua is calculated, it appears that seven 

 of these sets are within T ^Vmrth of an inch of the mean result ; two 

 a little exceeding -ra-.-o-o-o-th of an inch ; and of the remaining three, 

 the greatest difference is less than -ro-.-S-o-Tj-th of an inch ; so that the 

 mean result must, it is presumed, be very near the truth. 



To the length thus found, the, author next applies a correction for 

 the height of the place of observation above the level of the sea. The 

 advantages of his different methods are then explained ; and the con- 

 clusion of the whole is, that the length of the pendulum vibrating 



