1904- Xo. 2. THE DIURNAL VARIATION OF TERRESTRIAL MAGNETISM. 21 



of the place. Xorth of Pawlowsk, the curve has a strongly- 

 marked maximum at noon; in Pawlowsk this remains quite constant 

 for the space of a couple of hours; but when we then come to 

 WilhelmshaA'en, there is a depression in the noon-maximum of the 

 curve. This depression increases farther south, until at last we 

 have a prominent minimum at noon bet\veen two maxima. 

 2. The effect of the sun's position in relation to the magnetic north 

 pole is also very clearly apparent. The short vertical stroke with 

 the dot above, giving the noon of the magnetic pole, marks 

 another, secondary maximum in each separate curve, though this 

 is least conspicuous as regards the most southerh' stations. 

 If we now pass on to the stations in the northern polar section, we find 

 great irregularities, which are without doubt due to the proximity of the 

 magnetic pole; for the curve for Los Angeles, which lies as much as 36° 

 south of the magnetic pole, is just as regular as the curve for Tiflis and 

 Zi-Ka-Wei, while the curves for Point Barrow, Jan Mayen, and Godthaab 

 correspond in having their greatest maximum very near the moment at 

 which the sun reaches the meridian, not of the place, but of the mag- 

 netic pole. 



This correspondence is still more clearly apparent in Fig. 3, where 

 the portions of the respective curves for the three above-named stations, 

 which represent the interval between 6 a. m. and 6 p. m., magnetic-polar 

 time, are placed side by side. 



The curves for Kingua Fjord and for Fort 

 Rae, which are the two stations lying nearest 

 to the magnetic pole, seem to be difficult to 

 make out. This is possibly connected with 

 the fact that the feeble directing force of the 

 horizontal needle makes the readings somewhat 

 uncertain, even upon calm days. It should 

 also be remarked that to many of the readings 

 in the series of observations from these two 

 stations during the period of 48 hours here 

 dealt with, is appended a sign indicating that 

 the needle has been in motion. 



Lastly we have the two stations in the 



southern hemisphere, belonging to the southern »• 3- 



anti-polar section. The curves for these exhibit the same regularity as the 

 curves for the stations in the northern anti-polar section. The efiect of 

 the sun's position in relation to the magnetic south pole can only just be 



