Figure 7. — Results of experiments in the determination 

 of the length of the seconds pendulum at Konigsberg by a 

 new method were reported by F. \V. Bessel in 1826 and 

 published in 1828. With this apparatus, he obtained two 

 sets of data from the same pendulum, by using two dif- 

 ferent points of suspension. The pendulum was about 10 

 feet long. The distance between the two points of suspen- 

 sion (a and 4) was 1 toise (about six feet). A micrometric 

 balance (c) below the bob was used to determine the in- 

 crease in length due to the weight of the bob. He pro- 

 jected the image of the clock pendulum (not shown) onto 

 the gravity pendulum by means of a lens, thus placing the 

 clock some distance away and eliminating the disturbing 

 effect of its motion. (Portion of plate 6, Mimoires publies 

 par la Socicte francaise de Physique, vol. 4.) 



experiments were undertaken to determine whether 

 or not the length of the seconds pendulum should be 

 adopted as the standard of length by the new govern- 

 ment of France. The bob consisted of a platinum 

 ball 1 6% Paris lines in diameter, and 9,911 grains 

 (slightly more than 17 ounces) in weight. The bob 

 was held to a brass cup covering about one-fifth of 

 its surface by the interposition of a small quantity of 

 grease. The cup with ball was hung by a fine iron 

 wire about 12 Paris feet long. The upper end of the 

 wire was attached to a cylinder which was part of a 

 wedge-shaped knife edge, on the upper surface of 

 which was a stem on which a small adjustable weight 

 was held by a screw thread. The knife edge rested on 

 a steel plate. The weight on the knife-edge apparatus 

 was adjusted so that the apparatus would vibrate 

 with the same period as the pendulum. Thus, the 

 mass of the suspending apparatus could be neglected 

 in the theory of motion of the pendulum about the 

 knife edge. 



In the earlier suspension from jaws there was un- 

 certainty as to the point about which the pendulum 

 oscillated. Borda and Cassini hung their pendulum 

 in front of a seconds clock and determined the time of 

 swing by the method of coincidences. The times on 

 the clock were observed when the clock gained or 

 lost one complete vibration (two swings) on the pen- 

 dulum. Suppose that the wire pendulum makes n 

 swings while the clock makes 2n-{-2. If the clock 

 beats seconds exactly, the time of one complete 

 vibration is 2 seconds, and the time of swing of the 

 2«+2 



wire pendulum is T= 



-=2(l + l/n). An error 



in the time caused by uncertainty in determining the 

 coincidence of clock and wire pendulum is reduced 



312 



HI LLETIN 240: CONTRIBUTIONS FROM THE MUSEUM OF HISTORY AND TECHNOLOGY 



