DAILY VARIATION OF POTENTIAL GRADIENT. 



319 



Fig. 93. Daily variation of potential gradient. 



The same conclusion is reached if we analyse the two curves into Fourier's saries. 

 The equations to the two curves are 



Karasjok P=L39 + 39 sin (9 + 177'') + 23 sin (28 + 1-58 ) 



Cape Evans P= 86 + 12 sin (e+ .33')+ .5 sin (29 + 228") 

 in which in each case 9 is measured from local midnight and equals 1-5" for each hour. 



Now the diSerence in phase between the two first terms is 177°-33''=144°, which is 

 equivalent to 9-6 hours and the difference in phase between the two second terms is 

 1.58°-228''=290"', which is equivalent to 9-7 hours. Hence if we change the time at Cape 

 Evans by 9-6 hours, the phase of both terms becomes practically identical at both stations. 



Now is it only coincidence that this value of the chaiage in phase is almost the same 

 as the diiierence in magnetic declination at the two places ? The declination at Karasjok is 

 approximately 0" and at Cape Evans 153" E., i.e., a difference of 153", which is very near 

 the phase corresponding to 9-6 and 9'7 hours (144" or 145"). 



This result is very suggestive, but we are not justified in discussing it in detail, for many 

 more observations at places having different values of declination are necessary before we 

 shall be able to say with certainty that there is any relationship between the direction of 

 the earth's magnetic field and the daily variation of the potential gradient. 



Atmospheric radio-activity. 



Forty-six measurements of the radio-activity were made during the winter months, May 

 to August, and twenty-eight during the summer months, December to February. The method 

 used was the same as that employed on the Terra Nova on her voyage from England to 

 New Zealand.* The object of the work was to compare the atmospheric radio-activity in 



Simpson and Wright. Proceedings of Royal Society, A, Volume 85, page 175, 1911. 



