﻿the Rays producing Aurora Boreal u. 21$ 



given out by some known radioactive substances. We notice 

 that down to a height of about 70 kin. the value of (It) 

 diminishes rapidly with increase of velocity, so that «-rays 

 possessing a velocity about twice that of KaO would get down 

 to about 75 kin. 



Measurements of the height of aurorse have been made by 

 a great many observers, and quite recently Stormer * has 

 obtained measurements by means of a photographic method. 

 Various observers give values which differ considerably ; 

 but it can be considered as quite certain that aurorae are most 

 frequently found at a height interval, say, between 70 and 

 300 km. Stormer's observations gave a maximum of frequency 

 at about 115 km. above the ground. Comparing calculated 

 and observed values, we find that the range of ordinary a-rays 

 in matter is of the right order of magnitude to explain the 

 height of the main hulk of aurora? horeahs. 



The Diurnal Distribution of Aurora? and its Bearing 

 on the Radiation Theory. 



5. It has usually been held an indisputable fact that 

 aurorse occur more frequently on the evening than on the 

 morning side of the night. If there is such an asymmetry 

 of distribution with respect to the plane through the sun 

 and the magnetic axis, we might have a point of attack for 

 the determination of the sign of the electric charge carried 

 by the solar rays. Birkelaud finds in the law of distribution 

 stated an argument in favour of a negative radiation. 



When the study of the structure of the aurora? led the 

 writer to suppose a radiation of the a-ray type, it became a 

 question of importance to make further investigations into 

 the distribution of aurorse, and, provided it followed the rule 

 stated, to see whether it would involve the necessity for a 

 negative charge. 



in the following pages I have attempted to give the 

 diurnal distribution of certain auroral forms, and also to 

 compare results with theory so far as such coordination is 

 possible at the present stage of theoret'cal investigation. 



The determination of the diurnal distribution of aurora 

 borealis meets with a great many difficulties. First of all, 

 daylight will prevent aurora? from being observed during the 

 day ; secondly, the great variety of forms and intensities 

 makes it very difficult to get a quantitative measure of 

 aurorse which is free from any arbitrariness. Faults may 

 also enter into the results from the sky being not equally 

 well watched at different parts of the night. 

 * C. Stunner, C. R. 1911. 



Q2 



