EFFECT OF PRESSURE UPON ARC SPECTRA. 
129 
The rubber pipes for supplying the jackets with water may be seen, one in front at 
the bottom of the cylinder, and the other behind it at the top. Above the cylinder a 
ventilating chimney serves to carry off any noxious vapours which might otherwise 
escape into the room, and a block and tackle arrangement facilitates the removal of 
the cylinder covers. 
6. Precautions necessary in talcing the Photographs. 
The object of the experiments was the direct comparison of the spectrum of the 
iron arc when subjected to high pressures with the spectrum emitted under ordinary 
conditions. To facilitate the examination of the relative positions of the lines in the 
two cases it was desirable to photograph the two spectra in as close proximity as 
possible, and for this purpose the usual form of comparison-shutter was adopted. 
This caused the central strip of the plate to be exposed to the arc under pressure, 
and above and below this strip the comparison spectrum to be photographed; or 
vice versa. The shutter is described in Part I., p. 117. The dispersion of the second- 
order spectrum (1 millim. = 1'3 A.U.) was employed in this research. 
The following precautions were taken to insure the freedom of the photographs 
from any fictitious displacement of the lines :— 
1. To prevent jarring of the camera when the shutter was operated, the latter, 
which was originally fixed to the camera box, was detached from it and separately 
supported, as already described. 
2. To make sure that the conditions were the same for the two different positions 
of the shutter, some photographs were taken with the pressure spectrum within the 
central strip and others with the comparison spectrum in the centre. No systematic 
difference was found between the two. 
3. To insure the absence of an apparent shift due to the illumination of the grating 
being different for the two exposures, the arrangement of mirrors and lenses already 
described was adopted. Since the mirror N and the lens L, fig. 10, were fixed, the 
light always passed through the lens L in the same direction when the image of the 
arc was in focus on the slit. This is the necessary condition for constant distribution 
of illumination over the grating. That no conditions might be different for the two 
exposures, the comparison spectrum was taken from the same metallic poles within 
the cylinder. 
4. As a fictitious displacement of the lines might also be caused by a change in 
temperature of the grating (see Part I., p. 122), a careful watch was kept upon a 
thermometer placed near it. For some weeks it was found impossible to take photo¬ 
graphs on account of the fluctuations of the temperature, of which complete records 
were kept, and, although one or two reliable photographs at low pressures were taken 
at night towards the end of 1905, it was not until the Spring of 1906 that the 
temperature of the room settled down sufficiently for long exposures to be made, a 
VOL. ccviii.—a, s 
