250 Mr. Ivory's Disquisition concerning the Length of the 



Clermont. If the lengths in the table turned out to be cor- 

 rect, we should then have one instance at least, well authenti- 

 cated by actual observation, of that local attraction which is 

 often made to play so great a part in the pendulum experi- 

 ments. On the other hand if any errors should be detected 

 in the present lengths, and the new experiments should agree 

 better with the others that have been so successfully executed 

 from Unst to Formentera, it would be ascertained that a 

 great portion of the earth's surface coincides with an elliptical 

 spheroid so nearly that it must ever elude the utmost efforts 

 of human ingenuity to discover any difference between them. 

 It is desirable too that all the experiments in the table were 

 re-examined with respect to the necessary corrections, and 

 more especially that they were all reduced to the level of the 

 sea by one uniform rule. But the table, as it now stands, 

 seems to prove that the earth is more regularly an elliptical 

 spheroid than has generally been supposed. 



It remains to notice the experiments excluded from the 

 table, because the errors pass beyond the prescribed limits. 

 They are six in number, and are with good reason placed by 

 themselves on account of their irregularity, whether that be 

 caused by inaccuracy of observation, or by local attraction. 







Observed 



Com- 

 puted 

 Pendu- 

 lum. 



Excess 





Stations. 



Latitude. 



Pen- 



ofCal- 



Observers. 







dulum. 



culation. 





Galapagos 



o , // 



32 19 N. 



39-01717 



39-01172 



- -00545 



Hall *. 



St. Thomas 



24 41 



39-02074 



3901171 



-•00903 



Sabine. 



Ascension. . 



7 55 48 S. 



3902410 



39-01566 



- -00844 



Sabine. 



Sierra Leone 



8 29 58 N. 



39-01997 



39-01624 



--00373 



Sabine. 



Jamaica .... 



17 56 7 



3903510 



39-03144 



-•00366 



Sabine. 



Drontheim 



63 25 54 



39-17456 



39-17838 



+•00382 



Sabine. 



The five experiments that stand first in the table, are in- 

 consistent with other experiments whether made near the 

 equator, or at a distance from it, which prove that the equa- 

 torial pendulum must be about 39*01 . Captain Sabine, at 

 p. 359 of his work, assigns 39*01 as an approximation at least 

 to the mean length of the pendulum at the equator ; and, at 

 p. 341, he seems to infer that the ellipticity wrll be the same, 

 whether the equatorial pendulum be 39*01 or 39*01568. No- 

 thing, however, is more certain than that the first of these 

 lengths is totally inconsistent with Captain Sabine's ellipticity 



* Phil. Trans. 1824. The circumstances in which this experiment was 

 made, do not lead us to expect great precision. 



and 



