One point worthy of notice in this connection is the comparatively frequent 

 reference to arches in the sketches and log book. The same point is referred to by 

 Muwson* in his discussion of the auroral observations at Shackleton's Winter Quarters 

 in 1908. Our observations show that arches were observed at an altitude of 70, 

 or over, on 59 occasions in this period, while Mawson gives details of 118 occurrences 

 at altitudes over 60 in the paper referred to. 



Another point which appears to be of interest is the relative number of 

 aurorae of all types which were observed at high altitudes, as it is not unreasonable 

 to suppose, if we agree that auroral manifestations close to the zenith are, on the 

 whole, closer to the observing station than those at low altitudes, that any physical 

 factors closely connected with aurorae will be of enhanced importance at these times. 

 During the whole of the period in question, auroras were seen at altitudes of 70 and 

 over on only 85 occasions. 



It is equally interesting to note, from a perusal of the aurora log, the very small 

 number of occasions when colour and movement (other than slow movement) is 

 mentioned, a complete contrast in regard to " colour " being furnished by the Cape 

 Adare observations. In all, at Cape Evans in 1911, colour (red and green) was 

 recorded on only 14 occasions, and movement, other than slow movement, on only 

 24 occasions. 



These figures are put on record, as it seems not improbable that they may assume 

 importance in connection with any comparison of the relative suitability of stations 

 for auroral observations. 



3. SHORT PERIOD VARIATIONS AND PERIODIC CHARACTER OF THE AURORA. 



Even a cursory study of the aurora log shows that the aurora undergoes 

 considerable variations both in general form, in details of form and in intensity, 

 during the course of a few minutes. The variations are, however, more pronounced 

 in certain cases than in others. Thus, in the case of isolated streamers, short period 

 variations are most pronounced, both as regards changes in position and in intensity, 

 variations in form and position being more evident the higher the altitude of the 

 aurora, and the variations in intensity being apparently greater, the greater the 

 absolute brilliancy. The most stable forms of aurora are apparently the glows and 

 arches so frequently observed, while the least stable in form are those of the curtain 

 and streamer type. It is reasonable to assume that the variations of " glow " aurorae 

 will generally not be very evident, on account of the low altitude and consequent great 

 distance of these auroral forms from the observing station ; while the constancy in 

 form of the arches is explicable on the assumption that these are formed of connected 

 rays or streamers viewed nearly end-on, the apices of the streamers being directed 

 away from the observing station. 



'Auroral Observations at tin' ('iipr Royds Station," 'Trans. Roy. Soc. of South Australia,' 

 Vol. XL.. 1916. 



