EXPEDITION TO POINT BARROW, ALASKA. 383 



tin 1 clouds broke away sufficiently to allow it to be visible. It was very pale (01 o 1 ) ;m<l in the form 

 of :i sheaf of narrow, quiescent, white, ha/.y bands, .stretching across the sky in the NE. from a 

 point near Arcturus in the N. through Ursa Major, Auriga, and Perseus. At 2.17 the clouds were 

 merely open enough to show traces of aurora in the form of pale streaks in I'rsn Major, in the N. 

 and overhead. At 3 a. in. the sky was quite clear, and only pale, quiet, white bands, radiating 

 from a point in Bootes near the NW. horizon and converging lo a point in Taurus near the 

 SE. horizon, covered nearly the whole sky. The light was much dimmed by the full moon. 

 At 3.17 and nt 4 the aurora was essentially unchanged, though some bands faded and others were 

 formed, and the whole was much obscured by haze at 4. At 4.17 there was rapidly developed near 

 the southern horizon, dimly visible through the haze, a twisted horizontal band with a rapid motion 

 and indications of color which must have had a brightness of 3. The stars near it were invisible. 

 After this the aurora faded, none being observed at 5 a. m. 



October 1 , 1SS2, 1 a. m. to . At 1 a. in. there was a break in the clouds in the N. and a por- 

 tion of a vertical band of aurora (brightness 1 to 2) was visible. No stars were to be seen. The 

 openiug soon closed, and the weather continued rainy most of the night. A large and long-con- 

 tinued magnetic disturbance indicated a considerable aurora. 



October 8, 1882, to 8.30 a. m. The sky was alternately clear and overcast during the 



early part of the night, but no aurora was noticed until 4.45 a. in., when bright bands, white and 

 motionless, crossed the zenith from N. to S. The sky soon clouded and no aurora was observed at 

 the 5 a. rn. observation. At G a. m. a quiet yellowish baud had passed up from Orion on the SE. 

 horizon nearly to Cassiopeia, then declining towards Andromeda (brightness about 1). At 7 a.m. 

 a small yellowish patch in Taurus (brightness about 1) was all'the aurora visible. At 8 a. m. an- 

 other pale yellowish green arch stretched from the N. to the E. point of the horizon, reaching an 

 altitude of about 20. At 8.30, though the sky was perfectly bright and clear, no aurora was to bo 

 seen. Neither the galvanometers nor magnetic instruments indicated any disturbance. 



October 10, 1882, 2 a. m. to G a. m. Light snow fell during most of the night, but the sky cleared 

 at intervals. At 2 a. m. traces of aurora were visible through the hazy clouds, in the form of a 

 quiet band running across from N. to SE. near the zenith. At 4 a. m. the sky cleared off, showing 

 a bright patch in Aries near the S. horizon with some horizontal bauds in SW., brightness 2 to 

 3. At 4.17 bands, brightness 2, beginning in Lyra and passing through Cygnns. There was no 

 rapid motion or vibration, only a slow drifting and breaking of the bauds, which quickly faded and 

 new ones developing, especially one somewhat twisted and undulating from Lyra, through Cygnus 

 and Andromeda, to Aries on the S. horizon. At 5 a. m. the arch was in the SW., running from W. 

 to S., yellowish in color and vibrating rapidly, also twisting up and down vertically. At a. m. 

 a bright band passed from XE. to SE. with a rapid lengthwise vibratory motion, several times 

 parting in the middle. The clouds then became too thick for the aurora to be seen. A consider- 

 able magnetic disturbance commenced at G a m., continuing twelve hours. 



October 11, 1882, 3 u. m. to 4.30 a. m. Light suow was falling up to 3 a. m., when the- hazy 

 nimbus cloud broke rapidly away, disclosing a white hazy band (brightness to 1), quiet, stretch- 

 ing across from Hercules in the N., through (Jyguus, Lyra, Cassiopeia, and Andromeda, ending in 

 clouds near Aries in SE. This was a little brighter at the SE. end at 3.17, when the horizon again 

 thickened up. At 4 a. m. the sky was again clear, but the aurora had faded to a pale band on the 

 edge of the clouds in the SW., and in 10 minutes there was only a vague luminosity in the E. and 

 SW. During the rest of the night the sky was cloudy. 



October 12, 1SS2, 12 midnight, October 11, to 9 a. m. The sky cleared suddenly, disclosing an 

 arch in the XE., with its crown in Andromeda, and its extremities buried in the clouds. Its bright- 

 ness was 2, and it continued to rise and spread till, at 1 a. m., a broad, bright sinuous band ran from 

 the X. to SE. horizon, occupying mostly the constellations Corona Berenices, l"isa Major, < \imelo- 

 pardalis, Perseus, and Aries. Until 1.17 the only change was a slow spreading and undulation, 

 moving from the zenith eastward and slowly back again. The N. and SE. ends remained quiet, 

 the SE. end the brightest, while the center changed into one, two, and three bands of vertical 

 streamers and back again to wavy bands. The brightness of the band was from I! to i, and the 

 edges were tinged with rose and green. There was a magnetic disturbance, wilh increased decli- 

 nation and decreased horizontal force. At 2 a. m. there was merely a quiet arch, with streamers 



