166 Mr. Ivory on the Ellipticity of the Earth 



The pendulum which beats seconds at Paris, at the tempe- 

 rature of 1 5° Cent., being represented by unit, M. Duperrey 

 has determined the relative lengths, at the same temperature 

 and at the level of the sea, at the five following stations : 



Toulon 



Isle of Ascension 

 Isle of France. . 

 Port Jackson . . 

 Falkland Islands 



Latitude. 



43° 7' 20" N. 

 7 55 48 S. 

 20 9 23 

 33 51 40 

 51 31 44 



Relative Length. 



0-99950585 

 0-99729881 

 0-99789022 

 0-99871430 

 1-00025995 



Adopting 39-12929 for the length of the pendulum at Paris 

 in English inches, at the temperature 62° Fahr., the fol- 

 lowing table shows the lengths of the pendulums at the sta- 

 tions of M. Duperrey, in English measure, at the standard 

 temperature, and likewise the errors according to my formula 

 published in this Journal for October 1826. 



The pendulums in this table at Ascension and the Isle of 

 France coincide very nearly with the lengths previously de- 

 termined by Captain Sabine and M. de Freycinet. From 

 some cause or other, they are both anomalous, and they are 

 treated as such by M. de Freycinet and M. Duperrey, being 

 left out in computing the ellipticity which they adopt. But 

 the other three pendulums fall within the limits laid down in 

 this Journal for October 1826, and increase the number of 

 instances that come under my formula to 31, out of 40 the 

 total number of experiments that have been made. My for- 

 mula was originally constructed from 26 experiments; and 

 all the pendulums since determined by M. de Freycinet, Mr. 

 Foster, and M. Duperrey, agree with it within the prescribed 

 limits, except in the three instances of the islands of Mowi, 

 Guam, and the Isle of France, which are greatly irregular 

 and irreconcileable with the other experiments. It is there- 

 fore very probable that the mean ellipticity of the earth will 

 ultimately be found to approach very near ^^ as deduced 

 from my formula. 



' At 



