SUN AND THE EARTH'S MAGNETIC FIELD — FLEMING 203 



neers and magneticians. However, the accumulated reports indi- 

 cating connection between the storm and the power-line disturbances 

 proved convincing. Tlie evidence shows that the storm produced 

 earth-current gradients amounting to about 10 volts per mile. Thus 

 an increased practical importance of cosmical research in geomag- 

 netism has been shown, for the lengthy observations extending over 

 a century supply definite information on the probability of occurrence 

 of such storms and therefore on the extent to which it is advisable to 

 improve electrical installations to avert their effects. Because of the 

 theoretical and practical importance of obtaining complete records 

 of the infrequent great magnetic storms, a number of observatories 

 have added less sensitive equipment to the more sensitive types here- 

 tofore exclusively used. Thanks to these insensitive magnetographs, 

 we have complete photographic records of the very great disturbances 

 of April 1938, March 1940, and September 1941. 



The great storm of September 18-19, 1941, while not producing any 

 gi'eat power-line interference, was most remarkable in auroral dis- 

 plays over a great part of the United States. This storm is also of 

 interest in that H. W. Wells, of the Department of Terrestrial Mag- 

 netism of the Carnegie Institution of Washington, several days be- 

 fore its inception, had formally warned radio-transmission engineers 

 that disturbed geomagnetic and ionospheric conditions might be ex- 

 pected September 18. He was able to make this prediction through 

 close study of daily reports of areas, numbers, and locations of sun- 

 spots supplied by the United States Naval Observatory. Following 

 several days of slight disturbance, violent fluctuations in the direction 

 and intensity of the earth's magnetic field began about 11 ]). m., east- 

 ern standard time, September 17. Maximum activity was between 

 1 and 3 p. m. September 18. The usual difficulties with long-distance, 

 radio, telegraphic, and telephonic communications were experienced. 



The storm w^as accompanied by extensive auroral display — an oc- 

 currence which one living in our latitude has seldom the privilege of 

 seeing. A. G. McNish, of the Department of Terrestrial Magnetism 

 of the Carnegie Institution of Washington, who observed the display 

 carefully, gives the following account : 



But nature had scheduled her choice act for the evening hours. Those who 

 watched the evening Sun go down were aware of a strangely persistent glow in 

 the northwestern sky. As the sky darkened, distinct rays were visible in the 

 northern sky, brightening, fading, and continually changing. By 8 p. m. the en- 

 tire sky was filled with rays apparently converging to a point near the zenith 

 to produce a vivid coronal formation. Various forms of auroral activity, rays, 

 curtains, extensive arcs, and flickering rays resembling searchlight or air- 

 heacon beams continued until almost dawn September 19. Some auroral activity 

 was noted during the preceding and following nights, neither being comparable 

 with the display of September 18-19. 



