148 



ANNUAL REPORT SMITHSONIAN INSTITUTION, 1936 



as far south as the State of New York the curtain may sometimes be 

 seen south of the zenith. 



The altitudes of these displays have been skillfully measured in 

 Norway by St0rmer, with a number of observers connected by tele- 

 phone, who took photographs at the same instant from different 

 places at a measured number of miles apart (fig. 1). The photo- 

 graphs (pi. 1) show the Northern Lights in each case with a back- 

 ground of the stars of the Great Bear, but owing to parallax the 

 lights are seen in different positions relative to the stars on the 



FlODRB 1. 



-Four stations in Norway selected by Dr. Carl Stormer and connected by tele- 

 phone. From Oslo to Kongsberg is 63 kilometers. 



different photographs. A simple calculation determines the altitude 

 of the aurora. About 60 miles is the most common result, that is, 

 60 miles from the surface of the earth, not from the observer. 

 Sometimes the tops of the streamers may be 250 miles above the 

 earth, and I believe that the lowest determination is an altitude of 

 40 miles. The record height for the top of a streamer is 1,000 kilo- 

 meters, more than 600 miles. Similar measurements were made in 

 Canada by Sir John McLennan and others, and the results there 

 were in excellent agreement with the earlier determinations in 

 Norway. 



It is a strange fact, proved by St0rmer, that those auroras which 

 have the greatest altitudes, ranging (base to top) from 350 miles to 

 630 miles, occur in a sunlit portion of the atmosphere far above the 



