138 Astronomical Society. 



mites ; but if an interval of about 10 minutes should be preferred, 

 the distance should be about 39 4 inches -. and so in proportion. 



In order to adjust a pendulum of this kind, it must be placed on the 

 agate planes, and the number of vibrations determined in the usual 

 manner^ beginning, for the sake of regularity, with the knife-edge A) 

 then inverting the pendulum, and determining the number of vibrations 

 with the knife-edge B. In these preliminary experiments it is not ne- 

 cessary to extend the observations beyond one coincidence ; neither is 

 it requisite to apply any other correction than for the arc of vibration, 

 and for the temperature of the room, since all the other sources of 

 error will be common to the two positions of the pendulum, and 

 therefore may be rejected in these first trials. If it should be found 

 (as will in fact be the case) that the knife-edge B makes a less num- 

 ber of vibrations in a day than the knife-edge A, we must file away 

 the bar at the end B, until the two knife-edges are perfectly synchro- 

 nous. The amount to be taken away can be ascertained by experi- 

 ment only i and as we approximate towards the truth we must be 

 more cautious in using the file, and more accurate in making the ob- 

 servations. In this last step of the process it is difficult in all cases 

 to determine the exact quantity that has been filed away j and we 

 may sometimes overdo it, so as to cause an inequality in the vibra- 

 tions of an opposite kind, and thus render it necessary to add a small 

 quantity to the end B. In order to meet this difficulty, Mr. Baily 

 caused a small hole to be drilled at the end B, into which a screw 

 was fitted j and by means of a small piece of sheet lead inserted un- 

 derneath, the adjustment could be carried to any required degree of 

 accuracy. 



The principal advantages attending the placing of two knife-edges 

 on one and the same bar, Mr. Baily states to be as follows : viz. 1st. 

 That we are thereby possessed of two separate and independent pen- 

 dulums ; the results of which may be used separately or conjointly at 

 pleasure : and each of which is a more complete check on the other 

 than when formed of separate pieces of metal, that may probably be 

 of different specific gravities, and of different expansive qualities. 2nd. 

 That the knife-edges, being once rendered synchronous, will always 

 remain so, into whatever part of the world the pendulum may be 

 carried -, and thus enable us> if it should be required, to determine 

 the length of the simple pendulum at any point of the globe at which 

 it may be swung, by merely measuring the distances between the 

 knife edges. 3rd. That we are thus furnished with the means of 

 ascertaining whether the pendulum has sustained any accidental in- 

 jury j since such a fact would be immediately discoverable from the 

 inequality in the number of vibrations between the two knife-edges. 

 And, even in case of such an unforeseen misfortune occurring, the 

 ratio between the two would from that moment remain the same in 

 all parts of the world, and answer the same useful purpose of compa- 

 rison during the remainder of the voyage. Whereas, in the pendu- 

 lum as usually constructed, the effect of such an injury (if not sus- 

 pected) might be attributed to the errors of observation : and indeed 

 the fact itself could not be ascertained until the return of the ob- 

 server to some place where the pendulum had been previously swung j 



leaving, 



