99 



INTRODUCTION 



BY 



L. C. BERNACCHI. 



A RECORD of the auroras visible during the two winters 1902 and 1903 was kept. The observations were 

 generally made by the officer who was on meteorological duty for the night a duty in which all the 

 members of the " Discovery's " wardroom participated. 



Ordinarily, the observations consisted of noting the time, position of the aurora, its altitude and 

 amplitude, intensity, form, movement and duration. These observations were entered in a special 

 journal kept for the purpose, and a rough " chart " of the surrounding hills was supplied each night for 

 drawing in the position with regard to the magnetic meridian (see Plate 7). Whenever the display was 

 fairly extensive, the physicist was called and special observations were taken, such as measurements of its 

 intensity, width of bands, altitudes, and times of special movements. 



On the whole the displays, although very frequent, were extremely poor, and were generally in the 

 following forms : 



1. Faint lights, with no defined forms. 



2. Luminous patches, which frequently presented the appearance of clouds. 



3. Incomplete arcs, or segments of arcs, of which the brilliancy was not uniform nor the border 



regular. From these arcs rays would frequently shoot up intermittently. 



4. Rays, or vertical shafts, separated from each other at a greater or less distance, frequently 



described as streamers. 



5. In one or two exceptional cases irregular bands, formed of rays or vertical shafts, pressed close 



together and forming " draped aurorae." 



The faint lights and luminous patches were of the most varied dimensions, sometimes very small, and at 

 other times occupying almost the whole of the eastern (geographical) sky ; their brilliancy was rarely much 

 more intense than that of stars of the 4th magnitude, or even the Milky Way. They formed, as it were, 

 a white veil over the sky, through which stars of small magnitude were plainly visible. A clearly defined 

 arc, formed of a homogeneous luminous mass touching the horizon at both extremities, was rarely seen. 



From the middle of the moon's first quarter to the middle of its last quarter the aurorse were generally 

 quite invisible. Spectroscopic observations of the aurorse were not successful, due, apparently, to the weak 

 intensity of the light. On some occasions the characteristic yellow line near D was seen by means of a 

 direct-vision spectroscope, but, although plates were exposed on many nights in the prismatic camera, the 

 times of exposure varying from a few minutes to twenty-four hours or more, not the slightest trace of the 

 spectrum could be discovered on developing the plates. The spectrum plates (Cadet) appeared to be in 

 fairly good condition, good photographs of the spectrum of krypton gas and of the sun and atmosphere 

 having been obtained on them. 



.The observations of atmospheric electricity taken during the displays reveal no special effect referable to 

 the aurora. 



An examination of the journal shows that the largest number of aurorse occur during the mid-winter 

 months, June and July, and that there is some indication of May being relatively a quiet aurora month in 

 both years, but this may be purely an accident. The small number observed in March and September is, 

 of course, due to the large amount of daylight. A daily variation of the aurora at Winter Harbour is 

 probably shown, the maximum occurring at about 2h. a.m., which is also near the time of the mean 

 maximum altitude of the display. The time of maximum intensity appears to depend upon the latitude, 



o 2 



