5 2 



BIRKELAND. THE NORWEGIAN AURORA POLARIS EXPEDITION, 19021903. 



TABLE IV. 



Scale-values for Dyrafjord. 

 H = 0.120 0) r d = i734mm. io d H = d = 3-47- 



Note. December I, 1902, the sensibility of the variometer for V. I. made a little greater. 



December 1 5, 1902, compensation for the variometer for V.I. altered. 



February 23, 1903, the curve longer from the base-line for V. I. ; otherwise unaltered. 



January 27, 1903, fixed new mirror for V. I. ; otherwise unaltered. 



It appears from the above table that the sensibility for H has remained nearly constant all the time. 

 No decided variation in the temperature is noticeable, nor yet any decided variation with time. 

 It is therefore most natural to let Eh be constant all the time. The mean of the scale-value is 



;.= 5-5 1 - 



E, also remains fairly constant all the time. For we obtain the following: 

 Nov. 25, 1902, to Dec. i, 1902, 5.50 

 Dec. i, 1902, to Dec. 15, 1902, e v = 4.73 

 Dec. 15, 1902, to Jan. 27, 1903, e, = 4.92 

 Jan. 27, 1903, to Apr. 15, 1903, = 4.61 + 0.0094, < 



For this last period from the 27th January to the i5th April, we have a fairly regular increase 

 of with time. The formula set up is calculated by the method of least squares. 

 / here stands for the number of days reckoned from the 27th January. 



(') It must be remarked that this value is somewhat uncertain; for owing to the illness of Saeland, whose knee became 

 stiff while at Dyrafjord, no complete absolute determination was made. A deflection experiment was made, and this, combined 

 with a knowledge of the magnetic moment of the magnet employed, gave the value here given, which moreover is in accordance 

 with the terrestrial-magnetic charts. 



