EXPEDITION TO POINT BAKIJOW, ALASKA. 375 



assumed the arched form so common on other nights. Irregular-shaped masses of hazy light 

 appeared in various parts of the sky, principally in the N. and SK., which extended imperceptibly 

 upwards until they formed broad cloud-shaped masses in or near the zenith, and then after a time 

 faded away into the usual lumirous ha/e. Needles disturbed. Disappeared before daylight; 

 about 10 a. m. 



February 20, 188-', G a. m. to 7.30 , m. A faint patch appeared in the SK. and one in the N., 

 which soon extended towards each other and formed a faint arch, which rose slowly until it reached 

 the zenith, when it broke up into irregular-shaped masses, which arranged themselves round the 

 zenith in a form almost circular. A period of activity theii ensued, and numerous short rays shot 

 upward and converged directly overhead. While in this condition it was simply a corona with 

 the center wanting. A few tints of green, rose, and yellow were observed during this active 

 period, but they were of very brief duration. The display lasted about liftcen minutes, and then 

 gradually faded, and was no more visible after 7.30 a. m. 



February 27, 1882, 3 a. m. to 7 a. m. Impossible to determine the beginning or end of this 

 aurora, owing to the cloudiness of the sky. It was occasionally seen until 7 a. m. The magnets 

 were slightly disturbed. 



February 28, 1882, to . Too cloudy to permit observation. Auroral light was only 



seen ouce, near the zenith to the NE. Needles somewhat disturbed. 



March 1, 1882, 7 u. in. Seen through the clouds in the SE. at 7 a. in., but the rest of the night 

 the sky was clouded. 



March 2, 1882, fi.15 a. m. to . At 9.15 a. m. the clouds rolled off for a few minutes and left 



a rather bright auroral band visible, passing through the zenith in u NW. and SE. direction. 



March :.;. 1882, 3 a. m. to . Commenced faint and irregular, and at -1 a. in. there were two 



arches at right-angles to each other, the brightest running from N. toSE. Soon afterwards they 

 became broken up into segments, and soon faded into the usual luminous haze, and as the sky 

 soon" became obscured by clouds the termination of the display could not be ascertained. The 

 magnets, as usual, were agitated. 



March iaiid 5, 1882, to . On the 4th and 5tb, especially the former, thero was magnetic 



disturbance at times, but being cloudy uo aurora could be seen. 



March 0, 1882, 2 a. m. to 8 a. m. First observed about 2 a. m., when three somewhat sinuous 

 rays or bands extended from N. to SE. about 10 west of zenith. Between 2 and 3 a. m. there 

 was quite a brilliant interval when the sky in and near the zenith was covered with fleecy cloud- 

 shaped aurora. There was very little apparent motion, and after 3 a. m. there was a constant 

 decrease in brilliance, and after -I a. m. but little light was seen. The last was seen at_8 a. in., 

 when' a faint ray was visible in the W. The increasing cloudiness, however, prevented its termi- 

 nation from being observed. 



March 7. 1882, 1 a. in. to 9 a. m. Commenced as faint rays in theN. and SE., which soon formed 

 a narrow arch with a few streamers at its northern end. Occasional arches and scattered streamers 

 followed at intervals, but none were very brilliant and there was much less of the luminous haze 

 whic.ii has lii-en so common during last month. After 4 a. m. only an occasional ray appeared 

 until (i a. in., after which no more were visible until just as the dawn began to appear at 9 a. m. 

 (;>. !> a. m. local time), when a few rays appeared for a few moments just above the line of light 

 in the E. and parallel to the rays of light coming from below the horizon. 



March 8, 1.S82, 5.15 a. in. The night was cloudy and oidy one glimpse of auroral light was had 

 at 5.15 a. m. The magnets were considerably agitated. 



March 0. 1882, .'! a. m. to 8 a. m. Commenced about 3 a. m. while the sky was partly covered 

 by clouds. There were occasional displays of streamers, irregular curtains, and arches, accom- 

 panied by considerable motion. The streamers were long, pale and slender, and sometimes 

 approached the coronal form converging towards the zenith. The prevailing character, however, 

 was the diffused form distributed in patches all over the sky; the light threw the intervening 

 clouds into strong relief and seemed on many occasions similar to the diffused brightness of the 

 dawn. Tlr,' general motion was from N. to 8., but it was mostly very difiicuit to determine its 

 direction owing to the cloudiness and the extensive distribution of the 'ight. The sky became 

 entirely cloudy after 8 a.m. and no more of the display was observed. The magnets were very 



