102 Mr. Baily on the Discordancies in the Methods 



MM.Borda, Bouvard, Mathieu, and himself, on the length of 

 the pendulum at Paris : which, by adding the known difference 

 in the number of vibrations between Paris and Greenwich, 

 will make the length of the simple pendulum, at this latter 

 place, equal to 39'1379 inches. So that we have here also an- 

 other discordance, amounting to more than '001 of an inch. 

 I am aware that there is supposed to be a slight constant dif- 

 ference between the observations at Greenwich and London : 

 but I believe that some recent experiments do not exactly ac- 

 cord with the theory on this point. This will, however, very 

 shortly be more fully determined. 



Were I to construct another convertible pendulum I would 

 make the distance between the two knife edges exactly 36 

 inches ; or as nearly so as the artist could effect it. I pre- 

 sume that Captain Kater selected the distance somewhere be- 

 tween 39*4! and 39"5 inches, in order that the proposed co- 

 incidences might be accommodated to the pendulum of the 

 clock : at least, this circumstance weighed materially with my- 

 self, when I ordered the knife edges to be placed as nearly as 

 possible 39*3 inches asunder. This, however, I find by ex- 

 perience to be a minor considei-ation ; and inconveniences of a 

 greater kind present themselves by adopting this plan. It is 

 much better to select the best and most convenient length for 

 the distance between the knife edges ; and to make the pen- 

 dulum of the clock subservient to the intervals required. Now, 

 ihe best and most convenient length must be the very measure 

 we are in search of; namely, the standard yard : for, when 

 any other distance is fixed on, we may be in doubt with what 

 part of the standard scale such distance ought to be compared. 

 Captain Kater took the mean of 20 several lengths of 39"4 

 inches each ; " commencing from zero of the scale, and ad- 

 vancing by single tenths through the space of two inches." 

 But, if he had made a comparison w^ith the same distance (39'4') 

 at the other C7ul of the scale, he would have found a difference 

 amounting to above '0003 of an inch, in that length. Now, 

 the distance, with which Sir George Shuckburgh himself com- 

 pared the several standard yards in his time, is marked on the 

 scale, and extends from 10 inches to 46 inches: and this I 

 conceive to be the proper and only unit of measure which 

 ought to be resorted to : unless we take the mean of the whole 

 scale, as hereafter alluded to. The fact is, that the standard 

 scale of Sir George Shuckburgh (although otherwise in excel- 

 lent preservation) is not exactly straight, but is very slightly 

 bent and twisted : and more so at one end (the zero end) than 

 at the other. The bar is nearly 68 inches long, 1 '^ inches wide, 

 and O** inches thick ; and formed oitwo distinct pieces of brass, 



fastened 



