At'lTITEM^iDRAS OBSERVATORY. 



305 



or miscomputatioii'pbut rather- to a. latent- discordance betAve'eri 'the hy^a- 

 thesis (the, increase as the square of the sines of the latitudes y and- the 

 true cury^^Qf. tl^,e.meridip^^;at^pifdes. ...^An^d- such ■^, discordancfi' probably 

 ^does exists ; gjnc^^jcji^t hjjPpthesis.is;gj:qiuid:ed^qn/t|-iQ-s^^^ hon^qgeneity 



of the earth. .,• - ;•- 



SS. HowEV■^Efl9jth@'^^r^t'numbe^xDf cbmbinatfonSji ^v^^^ have u^ed^ 

 with latitudes both highferrand- lo^Wer^Hiato *Mgdfas^pt'd-Y)!''Sk in a great me^- 

 'Sur.^,thii3;dQfc(9:^^.for^it^appearsi; that th'^e^iregiilts w 



in :Oi?e.. case, -exceed also.mast''in- the-ethep; ^ in fe' i^^t^nS^^ 4vheV^ 

 Manilla is the place referred to, vvhere *the''^f^eatest^'^''<^"a^ito^rl^'3'bJbsert'- 

 ,^ble; fp^r it will J3,e, found, that the mean .result of, ^1^^ V^c>l€;S?t fp.r•^ the 

 equator coincides exadlly with Mr. Boy<Jui;R.f .^<^jli:j1' determination..^, . ,,;• 



39' Hence I attribute Mr.- Oentil's objs^'fVktibrii'"t!iat his experi- 

 " mehts at fc/n'/Za do noV tally with those, by -Mr.' Bouguer," to his not 

 having combined them wdtharsdfRcient number, iof other results, in different 

 higher and lower latitudes, -, — -^" >- -- - „^. ,_ ,,. 



■-•'Lev. 



40 



nifcriT'oqxD, vd l:ff 3; 



. By taltih^ th^t mean of the respective combinattonsi ■ we have the 



foTlowihg lengths j at tKe pdte and'equator. 



Jii'X 01 





Names of Places. 



TT 



Porto Be 

 .Paftcticher.r'y, . ... 

 Madras, ., .'. .' .', 

 Manilla, 



At the Equaloi 



38.98667 

 35,98012. , J 

 3(^.98711 ^' ' 

 38.99507 



At the Pole. 



ElUpticifij. 



39.21517 



39.207^22 

 39.20013 







^^iiiil''- TH*i''mean i^eSt^l't -of' these- operations',^ -'as fwell a^-thbSfe^ iiiade at 

 -Mii?r^]'s^^ari?kly,-giVe^MiJr a' diflference of '•t=?^' of hii^ih^i^'l^'' fiiiiri-^what 

 'M-i^^'BcJuftu^^^R has nriiidfe'^itftb ^,"at the equit6?r-l^y aSttiJiFe'i'p^VfmentS. 

 Xet us novY consider, what ratio between the".e^ftli,'s diameters ought, to re- 



E 4 



