720 



SCIENCE 



[N. S. Vol. XLIV. No. 1142 



auroral focus thus determined is 16° 54' and 

 the azimuth of that point, 22° 42' E. The 

 accuracy of the observation, however, can not 

 be considered better than about one degree. 



It is of ■ course desirable to compare these 

 values with the magnetic elements of the place 

 of observation. 



The Canadian government is now engaged 

 in making a magnetic survey, and has made 

 observations at nearly five hundred stations in 

 Canada, but none recently on Prince Edward 

 Island. Fortunately, the department of ter- 

 restrial magnetism of the Carnegie Institu- 

 tion of Washington has been able to provide 

 values for the magnetic declination and in- 

 clination at Charlottetown, P. E. I., which 

 were determined in 1908. They are as follows : 



Declination 23° 46'.4 W. for the epoch 1908.8 

 Inclination 74° 59'.3 N. for the epoch 1908.8 



Dr. L. A. Bauer has kindly given me the 

 average rates of annual change of both the 

 declination and inclination during the period 

 from 1908 to 1916, as well as the direction of 

 the isogonics and isoclinals for 1916.8, which 

 makes it possible to give the declination at the 

 place of observation as 24° 36" W., and the 

 inclination 75° 04' N". A comparison of the 

 focus point of the aurora and the above values 

 is as follows : 



While the accuracy of the determination of 

 the auroral focus is only one degree, it is 

 about as close as other determinations. The 

 observation may be unique, owing to the fact 

 that the focus was formed by streamers in the 

 southern as well as in the northern sky, that 

 the point in the heavens was determined from 

 the focus itself, and not from a corona, and at 

 a station as far south as latitude 46°. 



Elaborate observations have been made 

 during several Arctic expeditions of the azi- 

 muth of the summits of aurora arcs, but 

 there seems to be no definite coincidence be- 

 tween the azimuth measured and the magnetic 



meridian, the angular differences being often 

 many degrees, sometimes as great as 20° to 

 40°. ~No explanation has been given for this 

 anomaly. The corona center has been meas- 

 ured at a number of stations at high latitudes, 

 and as a rule has been found to agree with the 

 magnetic zenith to within about one degree. 



At Cape Thorsden (78° 28' N\ Lat.) the 

 mean of a considerable number of observations 

 made the angle between the auroral focus and 

 the lines of the earth's magnetism 1° 7', the 

 magnetic inclination being 80° 35', while the 

 coronal center had an altitude of 79° 55'. 

 Somewhat smaller differences have been re- 

 ported at other far northern stations. 



The height of this aurora may be taken at 

 about sixty-five miles above the surface of the 

 earth, if the results of Carl Stormer's auroral 

 expedition are accepted, as recorded in 

 Nature, 6 and reproduced in Fig. 3. Approxi- 

 mately 2,400 of these determinations have 

 been transferred from the chart in Pro- 

 fessor Stormer's report and used to make the 

 curve in Fig. 4. It is seen that the maximum 

 height for the aurora according to this set 

 of observations is between 55 and 80 miles. 



The position of the auroral focus thus shows 

 the direction of the field of terrestrial magnet- 

 ism at about sixty-five miles above the sur- 

 face of the earth. 



The lines of force of the earth's magnetism 

 as determined by the auroral focus should 

 curve downward and pierce the earth's crust 

 at about the place of observation, 14 miles 

 north of Charlottetown. It is therefore proper 

 to compare the direction of the magnetic field 

 shown by the auroral streamers with the mag- 

 netic declination and inclination at that place. 

 The observed values of both declination and 

 inclination at 65 miles altitude are less than 

 the values at the surface of the earth (each 

 by about two degrees). This is exactly what 

 should be expected since above the surface of 

 the earth the lines of force curve towards the 

 south pole as in the case of any magnet. 

 Assuming the aurora to be at this height, the 

 point on the surface of the earth directly be- 

 neath the apparent focus was about 14 miles 



e Natwre, Vol. 97, No. 2418, March 2, 1916. 



