342 Royal Society. 



tolerablj^ well. This niaj' be seen on inspection of the Plates, 

 where the curves of Sumatra, Java and the Cocos, which are ex- 

 treme southerly stations, agree exceedingly well with those at Sin- 

 gapore. 



The next observations referred to are those of the horizontal 

 force at different seasons, As the observations at the different 

 stations were made at different times of the year, it appeared ne- 

 cessary, as in the case of the declination, to ascertain what changes 

 might take place in the different seasons at a station where an un- 

 interrupted series had been observed for a considerable time. Ac- 

 cordingly, Singapore, where observations were made hourly during 

 three years, was chosen as the point to which all the stations of the 

 Survey were referred. Unlike the declination, which changes its time 

 of extreme western position, not only in every season, but in nearly 

 every month of the year, the maximum horizontal force at Singa- 

 pore occurs with great regularity either at 22 or 2'J> hours, whilst 

 the minimum at 9 or 10 hours is but very faintly defined, lu 

 other respects also the curves preserve a perfect parallelism, if we 

 except only the greater range during the spring. 



In the curves representing the changes of the horizontal force 

 at the different stations of the Survey a general similarity is per- 

 ceptible ; a very decidefl maximum at noon, with a minimum very 

 faintly marked. The irregularity, where observable, appears to be 

 principally attributable to the short time during which the observa- 

 tions were taken. At Moulmein, the large oscillation appears to be 

 principally due to the very great change of temperature. The coeffi- 

 cient of temperature of the magnet was determined at Moulmein, 

 and yielding the same results at other stations, is considered satis- 

 factory as regards accuracy. The range of the thermometer at 

 Moulmein was upwards of iMy° \ and in the middle of the day, in 

 the tent, the heat was almost insujjportable, being upwards of 105°. 

 The oscillation of the horizontal force is much greater at Moulmein 

 than at any other station. The minimum oscillation is at Singapore, 

 and increases both northward and southward. During the Survey the 

 portable bifilar was compared with the large Observatoiy bifilar, and 

 the results are given. The angle of torsion employed was always 

 the same, equal to 60°; and the threads by which the hollow cylin- 

 drical magnet was suspended were the same during nearly the whole 

 of the Survey, having been in use for upwards of three years. 



The barometer was observed at every station. At the principal 

 stations the standard was in use, and at the olliers the portable, from 

 the greater trouble and risk attending the landing the large baro- 

 meter. The curve of the barometer, corrected to 32°, is exceedingly 

 regular, the time of maximum and minimum being identical for 

 every station on the Survey, viz. 21 or 22 hours for the maximum, 

 and 3 or 4' hours for the minimum; the oscillation at every station 

 amounting to about one-tenth of an inch. The curve of the ten- 

 sion of vapour, as determined by the dry and wet bulb, is irre- 

 gular, which is communicated to the l)arometric curve wiien ap- 

 plied to it; but this irregularity may be owing to the very short 



