1839] Terrestrial Magnetic Force in Southern India. 223 



netized to saturation by a powerful magnet, and employed during two 

 months previously to my departure, for the purpose of discovering if 

 their intensities remained constant. Eiich needle was fitted into a brass 

 stirrup (see fig. 2), for tbe purpose of being attached to a few filaments 

 of untwisted silk, by which it was suspended when in use ; and when 

 unemployed, was fitted with an armature of soft iron. When the ob- 

 servation was to be made, the needle was suspended in a square wooden 

 box, of four inches on the side, and li inches high; the top of this box 

 was fitted with a plate of glass, in the centre of which the end of a hol- 

 low glass cylinder of eight inches in length (a piece of broken baro- 

 meter tube) was cemented; to the upper end of this cylinder, one end 

 of the silk filaments supporting the needle was secured, so as, passing 

 through the cylinder, to leave the needle suspended by the other end at 

 half an inch above the bottom of the box, when its oscillations could be 

 seen through the glass plate at top, or through two glass windows, 

 placed opposite to each other in the sides. At the bottom of the box 

 inside, aline 'was drawn in the direction of the two windows; and at 

 angles of 20 degrees on either side of this, were drawn two other lines; 

 the centre line being made to correspond with the direction taken up by 

 the needle when at rest (the magnetic meridian), and the other lines, 

 exhibiting azimuths of 20® to the east and west of this, shewing the 

 arcs at which the observations were always to commence. On one side 

 of this box a delicate thermometer was fixed, whose indications could 

 likewise be seen through the glass plate at top. The pocket chrono- 

 meter by Arnold was one of ordinary goodness; in addition to which 

 Mr. Caldecott had provided himself with a pocket chronometer by 

 Barraud, and an artificial horizon and sextant, for determining the lati- 

 tudes of places which could not be identified on the map. We will now 

 mention one or two 



Particulars with regard to the modes of ohservingi 



Hitherto, it had been my custom to watch the decreasing vibrations of 

 the dipping needle, and register its indication when stationary. Such a 

 mode of observation, however, often led to unsatisfactory results ; for, 

 on tapping the apparatus gently, without un-centering the needle, it fre- 

 quently exhibited an altered position, to the amount sometimes of forty 

 or fifty minutes ; and an approximate result only could be obtained, by 

 taking the mean of the readings after repeatedly tapping and re-centering 

 the needle. This circumstance led to an alteration in the mode of ob- 

 serving, which consisted of first centering the needle on the agate planes, 



