from Long Vacuum Tubes. 745 



condition, the central portion shows a bright flash of the 

 secondary spectrum, on closing the switch : the duration of 

 this flash is of the order of -^ sec. : on opening the switch 

 the gas recovers its initial condition in about 4- sec, and 

 shows the " flash"" again on closing the switch. 



The secondary spectrum is attributed to molecular hydro- 

 gen by recent investigators (Morton, Stark, and others). 

 The first rush of the current appears to break down the 

 molecular-' hydrogen with the emission of the flash, leaving 

 only atomic hydrogen in the centre of the tube (which 

 gives the Balmer lines). On stopping the current, molecular 

 hydrogen appears to be re-formed in about 4 sec. This 

 break-down, however, appears to require the presence of a 

 trace of water-vapour. A very weak current, the secondary 

 spectrum is fairly strong at the centre of the tube, with the 

 Balmer lines weak or absent. Increase of current strength 

 weakens the secondary spectrum which passes through a 

 minimum of intensity as the current becomes stronger : the 

 Balmer lines increase steadily in intensity as the current is 

 augmented. 



This can perhaps be explained as due to the fact that 

 molecular hydrogen is re-formed as rapidly as it is broken 

 down into atomic hydrogen, in the case of a feeble exciting 

 current. As the current strength increases the proportion of 

 atomic hydrogen becomes greater, the secondary spectrum 

 weakens, and the Balmer lines become stronger. This, 

 however, does not go on indefinitely, for presently, with 

 further increase of current, the secondary spectrum begins 

 to brighten and both spectra increase in intensity at about 

 the same rate. About the only comment that can be made 

 in regard to this circumstance is that apparently the 

 percentage of molecular hydrogen cannot be reduced below a 

 certain minimum value. 



With dry hydrogen and a tube thoroughly freed from 

 water-vapour, the Balmer lines disappear entirely from the 

 entire tube (with the exception of a very faint trace of H a ). 

 In this stage a strong group of lines appears in the ultra- 

 violet between the 4th and 5th Balmer lines. The origin of 

 these lines has not been determined. 



With the tube in this, ihe white stage, operation with 

 condenser discharges gives a red discharge which shows the 

 Balmer lines strongly developed. This indicates that 

 momentary currents of several thousand amperes are able to 

 break up the molecular hydrogen even in the absence o\ 

 water-vapour. 



I have been aided in this investigation by a grain from the 

 Buinford fund of the American Academy of Arts and Sciences. 



Phil. Mag. S. 6. Vol. 42. No. 251. Nov, 1921. 3 D 



