734 



lilRKKI.AN'I). THE NORWEGIAN Al'RORA POLARIS EXPEDITION, I QO2 1903. 



where c/ 1 and q> are quantities which depend on the resistance of the cable and the soil and on the way 

 in \vhich the cables are connected with the ground. These quantities, </\ and </> may easily be verv 

 large numbers. 



During the stay at Kaafjord and Bossekop the resistance of the soil between the earth plates was 

 rvpcutcdlv measured by using a Wheatstone bridge arrangement with alternating current and telephone 



The results of measurements are here given in tabular lorni. 



TABLK CIX. 



Date 



Resistance 



N~S E-W 



Date 



Rusistanrr 

 N-S K-\V 



Nov. 15 . 



,, 26 . 



Dec. 4 . 



6 . 



I OOO 



700 



700 



I 1 50 



850 



400 

 1 400 

 I 500 

 I 600 



500 



1 500 



1 200 



Jan. 





; '5 

 17 



3 r 

 Feb. . i 



700 



I OOO 



75 



.(OO 

 'J OOO ,, 27 



i 500 Mnri'li 7 



i 7 -'3 

 500 April i 



i 500 



i 250 

 Hoo 



'5 



600 

 600 

 150 

 650 



400 



-i 30 

 400 



1 1OO 



800 



'5 

 500 

 600 

 400 

 600 

 35 

 13 

 400 



We notice that the resistance of the soil undergoes great variations, but always in such a way that 

 the resistance is about equal in both circuits. 



As the earth-connections for the two cables were made as equal as possible, we should probably 

 at any moment be able to put 



Thus I think when we take J > as a relative measure of the earth-current, we ought to get approxi- 

 mately the right direction of the current. Values of J c found at different times with different conditions 

 of the soil however need not be exactly comparable. 



CONSTANTS FOR THE EXPERIMENTAL ARRANGEMENTS. 



145. The three galvanometers used we shall call si, 1> and />'. 



The sensitiveness of the galvanometers was measured by disconnecting them from the cables and 

 exposing the instruments to a current of known strength. The deflections were indicated by marks 

 made by the spot of light on the photogram. The scale-values for the instruments were observed at 

 intervals and were found to keep constant within the limits of experimental error. 



The scale value and inner resistance of the galvanometers are given in the following table: 



TABLK CX. 



