1839] 



Terrestrial Magnetic Force in Southern India. 



269 



These results appear, on the \Yhole, as accordant as results from inten- 

 sity observations generally come ; for, independent of the known cor- 

 rections and unavoidablo errors which occur to observations of this na- 

 ture, there are without doubt others of whose disposition we are igno- 

 rant ; a fact, that \\\\\ be rendered sufficiently obvious by consulting the 

 foregoing observations, in which the greatest error of observation from 

 known causes, I estimate can never amount to 4-tenths of a second of 

 time ; under these circumstances it will be turning the above results to 

 the best account, if we now deduce the intensity of the magnetic equator ; 

 for, thus employed, we may fairly hope to get rid of errors of 

 every kind. To arrive at the most accurate result, we must here 

 have recourse to theory ; and although the theory that the tan- 

 gent of the dip is equal to twice the tangent of the magnetic lati- 

 tude {tan. ^ =1 2 tan. \) is derived from principles which appear, all 

 circumstances considered, to be not strictly true ; still, for the small 

 space over which we have occasion to employ it, and that only in a dif- 

 ferential sense, its accuracy is fully equal, and indeed superior, to our 

 wants. If to the above formula we add, that, k = A cos. \, where A 

 represents the total magnetic force; we arrive, after a little reduction, 

 at the expression 



from which we will now compute the 



Total vitensUy of the Magnetic force at the Magnetic Equator, horizon 

 taJ force at Madras = I. 



Place. 



Ongole 



Ramapatam .. 



Alloor 



Nell ore 



Woojilly 



Sooloorpet. . . 



Poodaway 



Madras 



Sadras 



AUumparva . 

 Pondicherry. 

 Porto Novo. . 



Sheally 



Tranquebar .. 



Observed values 

 of A. 



0,9920 

 ,9879 

 ,9845 

 ,9922 

 ,99:^3 

 ,9856 

 ,9933 



1,0019 

 ,0082 



0,9938 

 ,9980 

 ,9966 

 ,9939 

 ,9939 



Mean value. 



0,9921 



Difference. 



- ,0001 



- ,0042 



- ,0076 

 -t- ,0001 

 4 ,0012 



- ,0065 

 + ,0012 

 -t- ,0098 

 -\r ,0161 



,0017 

 + ,0059 

 -Y- ,0045 

 -f .0018 

 -i- ,0018 



