﻿Atomic Hydrogen and the B aimer Seines Spectrum. 539 



current of 15 or 20 amperes in the transformer. The 

 actual intensity of the secondary spectrum (at its minimum) 

 was about 1/50 of the intensity which it had at the ends of 

 the tube. 



As we shall see later on, the probable explanation of 

 the peculiarities thus far outlined is as follows : — The 

 secondary spectrum is emitted by the hydrogen molecules ; 

 the Balmer lines by the atoms. 



With a heavy current in the tube, the dissociation into 

 atoms is nearly complete and permanent, no appreciable 

 recombination occurring during the very brief current- 

 pauses which occur when the transformer potentials are 

 near the zero value, the duration of which can be observed 

 by viewing the discbarge in a revolving or " wabbled " 

 mirror. This time is of the order of 1/1500 sec, and, 

 as we shall see presently, about 1/5 of a second is required 

 for the recombination of the atomic hydrogen. 



At the ends of the tube, molecular hydrogen is con- 

 tinuously supplied by the bulbs, the metallic electrodes 

 acting as catalyzers causing the instantaneous recombination 

 of the dissociated hydrogen. The probability of the truth 

 of this explanation will appear when we come to the subject 

 of the action of metallic wires on the discharge. 



With a feeble current, however, atomic hydrogen is not 

 formed rapidly enough to permit of high concentrations, 

 and the secondary spectrum predominates in consequence. 

 With the heavy current, about 1/50 sec. is required for 

 the dissociation of all of the molecular hydrogen in the 

 central portion of the tube. This we know from the study 

 of the duration of what I called the " secondary spectrum 

 flash/' 



It was found that with the tube operating under such 

 conditions a direct-vision prism showed only the Balmer lines 

 in the central portion, the intervening regions being quite 

 black (the "black stage"") ; if the current was interrupted 

 for a moment and then turned on again, the secondary 

 spectrum appeared as a brilliant flash, which lasted 

 from one to three or four half-cycles of the current, 

 according to the pressure of the gas in the tube. At 

 high pressures the duration of the flash was longer, as 

 was to be expected, more time being required for the 

 complete breakdown of the gas into atoms. 



The further extension of the Balmer spectrum depends 

 upon ascertaining the cause of , and abolishing, the secondary 

 spectrum and the faint continuous background, and giving 

 a sufficiently long exposure. The most promising line of 



