LYMAN. — THE SPECTRUM OF HYDROGEN. ].°>U 
result is as above described. The annoyance of disconnecting the discharge tube from 
* 
its pump, removing the face plate from the receiver, detaching the dischar: tube from 
the face plate and cleaning the window, followed by the same set of operations in the 
inverse order, must be experienced to be thoroughly appreciated. When it is remem- 
bered that with a disruptive discharge this process must be gone through after about 
four exposures the difficulty of this part of the research will be understood. 
In practice the discharge tube was filled with hydrogen and a spectrograph taken 
without capacity, next a spark gap was introduced in series with the tube, and capacity 
was put in parallel until the gas showed the four line spectrum clearly. The appear- 
ance of the tube was constantly watched with a direct vision spectroscope. 
A similar set of experiments were tried for air. In both cases the material of the 
electrodes was altered in various experiments. It is important to observe that the 
nature of the electrode does not seem to affect the nature of the phenomena. 
The effect of the introduction of capacity with hydrogen is to introduce five sets of 
new lines. These lines lie between X 1900 and X 1400 ; under favorable circumst anc<^ 
they are strong and characteristic. The appearance of the principal spectrum remains 
unaltered, except for a very slight weakening. 
The effect of capacity on the spectrum of air is very different. The band spec- 
trum is weakened to such an extent as to be almost wholly destroyed and five sets of 
new lines are introduced. These new or secondary lines are identical with those which 
appear in hydrogen. Though some of these lines are always present both in hydrogen 
and in air, with the disruptive discharge, they vary very much in intensity from ex- 
periment to experiment. This variation with the condition of the research, add. d to 
fact that the secondary lines appear both in hydrogen and 
1 
nake 
certain that they owe their origin to some impurity common to both gases. The 
nature of this impurity can only be decided after the spectra of the other principal 
gases have been examined. At present it seems safe to state that (1) there i no second- 
ary spectrum of hydrogen in the region below X2000; (2) that the introduction of 
capacity almost totally destroys even the primary spectrum of air ; (3) that new an 
characteristic lines do 
both air and hydrogen, and that 
lines are probably due to some impurity. In weighing the evidence here presented it 
must be remembered that these results have been checked by experiments performed 
under very varying conditions. The pressure and purity of the gases, the shape and 
character of the discharge tube, the material of the electrodes, and the time of ex- 
posure are all factors which have undergone investigation. 
v 
