RESULTS OF THE MAGNETIC SURVEY OF FUNAFUTI ATOLL. 35 



Base Station. 



At Funafuti, a quarter of a mile south of the Mission House, a-base station was 

 selected. Here observations witli the absolute instruments were made at the 

 commencement and towards the close of the series. The differential declination 

 magnetometer was also set up, and hourly observations made. The base observations 

 with the Fox apparatus were also made here, including the formation of a table of 

 equivalent weights. 



Diurnal Variation of the Declination. 



Althougli some difficulties arose with the observation of the differential magneto- 

 meter from the instability of the stands upon which the instruments were mounted, 

 and from suspension threads breaking, so much care was taken in constantly 

 verifying the zero, that the following results may be accepted as a very close 

 approximation to the truth : — 



Extreme range : June, 3' 25"; July, 2' 23''. 



The needle attained its extreme deflection — 



June : East extreme, fJ.SO ^.m. ; west extreme, 1.0 p.m. ; 2nd east extreme, 7.0 p.m. 

 July : East extreme, 6.30 a.m. ; west extreme, noon ; 2nd ieast extreme, 7.0 p.m. 



The general characteristics of the diurnal variation therefore approach those 

 observed at Kew for example during the summer months, and are as might be 

 expected in south latitude during the three months when the winter solstice occurs. 



Normal Values. 



One great object of the magnetic survey was to ascertain if any local magnetic 

 disturbance existed, and, if observed, to point out to the geologists the areas of 

 greatest disturbance as those where rocks might be nearest the surface and be 

 most eligible for borings. The question then arose as to how the best normal 

 values could be determined in such an isolated spot. For the declination a close 

 approximation was obtained by swinging the ship in deep water both to the north- 

 ward and southward of the island, the effects of the iron in the ship and possible 

 disturbance of the land beino- thus eliminated. In the case of the inclination and 

 force the vertical disturbing force caused by the iron in the ship was a bar to similar 

 results, swinging the ship being obviously of no avail. 



The mean of the absolute observations at the land stations was therefore taken 

 as normal, the declination having the valuable check of the swinging before 

 mentioned. 



F 2 



