M 



Bowditch on the Oblateness of the Earth, 



considerably from the preceding. 



Onthe other hand by combia- 



ing the measured degrees of the tneridian, in a similar manner, he 



finds the oblateness to be 



t 



'^12 



*; wbich agrees much better with 



1 



^05 



1 



the quantity ^^r^ obtained by the pendulums, directly contra- 



335 



ry to what it ought to be by the theory, which teaches us that the 



J V_ 



method by pendulums ought to be more accurate than by the 

 measures of the decrees of the meridian. This difference will be 

 in part rectified if we use the corrected measure of the degree of 

 Lapland as found by Svanberg, but even with this correction the 



method of pendulums will differ the most from — -r which we have 



supposed to be the correctest value. This difficulty will, however, 

 be wholly obviated if we correct the calculation for the two numer- 

 ical mistakes mentioned above, by which means the oblateness 



1 1 



will be increased from-^rr^o toTrr--- and then the results of the 



methods by the lunar equations, the pendulums, and the dei^rees 



of the meridian will be respectively 



1 



1 



-? 



30d'31.-7 



and 



1 



i2i 



•.which are in 



the natural order imlicated by (lie theory. Having given this 



sketch of the principles of the computation, I shall now proceed to 



r 



point out the mistakes alluded to, and in order that we may fully 

 understand the subject, I shall give a brief outline of the method 



V. 



Thisisoiven in Book HI, § 4l5of his ''Jfecanique Celeste^** using tljc de- 

 grees measured at the Equator, Cape of Good Hope, Pennsjlvania, Italj by Bos- 

 covich and Le Maire, in France by Delanibre and Mechain, in Austria by Lies- 

 ganig, and in Lapland by Maupertuis, If we use the corrected measure of tiie de- 

 gree of Lapland as found by Svanberg and given in Rees' Cyclopedia namely 

 57196 toises for the latitude C6^ 20' 10''. The oblateness will be decreased from 

 ^^ to nearly ^^, I took its length of the whole arch measured equalto L5°, as 

 given by Dr. Rees, as I had not Svaaberg's work to refer to. 



