LYMAN. — THE SPECTRUM OF HYDROGEN. 131 
is used between discharge tube aud spectroscope both parts of the apparatus are of 
course exhausted together and both are filled with hydrogen together. If a window 
separates the two, the tube must be exhausted by the mercury pump and filled from 
the separate supply of hydrogen. In either case the most laborious part of the adjust- 
ment lies still ahead, for the spectra from both slits must be in focus at the .same time 
and the position of the plate-holder can only be 
determined by trial. It is therefore necessary to 
take a series of spectrographs, removing the face 
plate after every trial in order to change the ad- 
justment of the spectroscope, and replacing the plate 
on each occasion air-tight in order to exhaust and 
fill with hydrogen. As can be easily understood 
from the figure, that the conditions of adjustment 
should be fulfilled both slits must lie on the circle 
whose diameter is the grating's radius of curvature 
and the plate must form a part of the arc of this 
circle. By construction, the curve to which the plate 
is bent passes through the slits. There are then two degrees of freedom of adjustment 
the draw tube can be run in and out and the disc can be turned about the axis A A 
these two motions will suffice to bring the slits and plate into their correct theoreti 
cal positions. 
Tedious as is the method of trial above described it has seemed better to adopt i 
Fig. 2. 
• 
complicate the apparatus by the introduction of de 
.- 
focus from outside the receiver. Such devices might permit the focus to be changed 
without admitting the air, but the author is not at present prepared to face the problem 
of moving joints which must be maintained air-tight. Once the spectroscope i> in ad- 
justment the face plate, if it is of the improved form, can remain permanently in place. 
As regards pumping the apparatus, and as to the extent to which it is necessary 
to wash with hvdroeen with a direct connected discharge tube, the following example 
_ The receiver and drying tubes were first exhausted to .7 mm. of 
The tubes were then shut off and filled with hydrogen ; after the gas had 
may be of 
mercury. 
stood over the drying material for two or three minutes it was admitted to the re- 
ceiver. A second filling of hydrogen was let into the drying tubes and in turn run 
into the receiver. The tubes are of such a capacity that two fillings raise the pressure 
by about 15 cm. of mercury. The pump was then applied and the pressure reduced 
to .45 mm. j hydrogen was admitted, the pump again applied until the pressure again 
