ON THE SPECTRUM OF HYDROGEN. 
387 
(12) Summary. 
(1) A discussion is given of the conditions under which the Balmer series and the 
secondary spectrum of Hydrogen are produced. 
(2) The wave-lengths of about 1,200 lines in the secondary spectrum have been 
measured. 
(3) It has been found possible to classify the lines into different physically related 
groups under different conditions depending on the pressure of the gas in the discharge 
tube, the electrical conditions of excitation, and the presence of Helium. 
(4) These methods of classifying the lines have been compared with the results 
obtained by other investigators relating to the Stark and Zeeman effects, and with the 
regularities observed by Fulcher. 
(5) A comparison has been made of the wave-lengths of lines of different classes in 
the secondary spectrum with the Fraunhofer and Sunspot spectrum. No evidence of 
the presence of the secondary Hydrogen in the Sun has been obtained. 
(6) A new method of measuring the widths of spectrum lines has been developed. 
(7) It has been shown that the secondary spectrum is due to the Hydrogen molecule. 
(8) Experiments have been made which appear to show that when electrical dis¬ 
charges are passed through vacuum tubes, a partial separation of the gases takes place, 
and this appears to afford a satisfactory explanation of a number of phenomena which 
have been observed. 
(9) A number of observations relating to the appearance of impurity lines in vacuum 
tubes are discussed. 
We wish to express our thanks to the Department of Scientific and Industrial 
Research for a grant which has been made to one of us (S. B.) during the course of this 
investigation. 
3 h 2 
