52 Prof. L. Vegard : Results of Nortlilight Investigations 



Thus we find that the height distribution curves for the 

 three most frequent forms show two distinct maxima, one at 

 100 km. and one at 106 km., and these maxima cannot be 

 explained from errors. 



The existence of these maxima is of great importance. 

 First of all they show that the three most frequent forms 

 Diffuse Arcs, Drapery-shaped Arcs, and Draperies, which 

 give exactly the same maxima, must be produced by the 

 same kind of electric rays. The fact that the same maxima 

 do not come out in the case of Rays and Pulsating Aurorse 

 does not at all mean that these two forms are produced by 

 another type of rays. In the case of the faint and diffuse 

 pulsating forms the maxima are likelv to be wiped out as an 

 effect of errors. 



In the case of the ray-form the distribution curve is 

 drawn higher up into the atmosphere from causes which 

 will be discussed later on. 



Secondly, the existence of two maxima indicates that a 

 great part of these cosmic electric rays which produce the 

 aurorce is made up of two groups of rays, each of which has 

 its own quite definite penetrating power. 



In the case of long bands which stretch across the sky, it 

 would be of interest to know whether the height of the 

 lower boundary keeps constant all along the band. We have 

 examined a number of such cases which should give the most 

 favourable conditions for an exact determination. 



The lower boundary was divided into two halves and the 

 mean height was taken of the points belonging to each part. 



As the mean of 14 long bands we found 106'0 km. of the 

 western part, 107-2 km. of the eastern part. This result 

 shows that inside the limit of possible errors the height of 

 the band- form is on an average equal, in the western and 

 eastern part, and there should be no change of the properties 

 of the electric rays as we pass along the band. 



§ 3. The Direction of the Ray -Streamers. 



It is well known that the direction of the auroral streamers 

 very nearly coincides with that of the magnetic lines of 

 force. The direction is given by the altitude and azimuth 

 of the radiation point which is shown by the coronal-form. 

 The earliest observations, which had to be undertaken by 

 direct observations on the sky, were less reliable in indi- 

 vidual cases, on account of the great variability of the 

 aurorse. A series of photographs of coronal-forms were for 

 the first time taken at the Haldde Observatory, and from 



