122 



ABSOLUTE OBSERVATIONS AND ADJUSTMENTS. 



The observations indicated tliat tlie zero value of the declinometer scale began to change 

 slowly in April, and continued to move in the same direction until July. This was probably 

 due to a movement of rotation in the wooden pillar caused by absorption of moisture in the 

 spring. I noticed a movement in the same direction with the transit instrument, which, 

 when directed to a fixed mark on one day, was often found on the following day to be pointing 

 two or more minutes to the eastward of it. 



Inclination. 



Tor observations of inclination a dip circle by Barrow, London, with 3^-inch (9 cms.) 

 needles was used. 



Table 5 gives the results of these observations. At every observation both ends of the 

 needle were read in each of the usual four positions ; the poles were then reversed and the 

 readings repeated. 



When the inclination was observed at Kew with this instrument, before leaving England, 

 an almost identical value was afforded by needles 1 and 2, and no difference in the 

 results obtained from them was observed at Fort Rae. The instrument was so much injured 

 on its journey back that it was not possible to make any observations with it after its return 

 to Kew. 



An inspection of the observations showed that the value of the inclination varied at 

 different times of the day, and they were accordingly grouped by hours with the following 

 results. 



Table 6. 



Hourly Means of Inclination. 



Hour. 



!) 



10 .i.m. 



I o to II „ 



11 » 



II to 12 ,, 



12 „ 



12 to I p.m. 

 I 



2 

 2 

 3 



4 



4 



4 to 5 



S 



5 to 6 



6 



6 to 7 



7 



7 to 8 



I to 



3 to 



The last column of the above table gives the value of the inclination as calculated from the 

 absolute horizontal and vertical forces, X and Y being the mean values of the whole of the 

 year's observations at those hours. 



It corroborates the fact of the great diurnal variation of the inclination, and the mean 

 value for the month or year will probably be more accurately found from the mean values 

 of the vertical and horizontal forces than from the observations of inclination, which are too 

 few in number and are not fairly distributed over the 24 hours. 



