EFFECT OF PRESSURE UPON ARC SPECTRA. 
155 
unsymmetrically reversed lines of Group III., it is not in general the most intense 
portion of the emitted light that is absorbed. 
This result might have been deduced from the displacement curves given in 
Diagrams VII. and VIII., which show that when the members of Group III. reverse, 
their reversals are at that pressure less displaced towards the red than are the 
emission lines. 
Since the values of the displacements of eS at 80 and 100 atmospheres correspond 
to those for the other lines of Group III., it is legitimate to assume that if its 
emission could be examined without the superposed absorption, its displacement 
curve would be identical with those of fl, f 2, &c. This being the case, it is clear 
that its absorption line is on the violet side of the most intense part of the emission 
line, and it again follows that for the lines of Group III. the most intense vibration 
emitted is not necessarily that which is most strongly absorbed by the vapour 
surrounding the arc. 
Only one line of Group II. shows reversal, and that is line 2, which reverses 
unsymmetrically at 10, 20, and at 25 atmospheres; its displacements at these 
pressures seem consistent with the absorption of its most intense emission, differing in 
this respect from the lines of Group III. 
Table VII. 
- 4 . 
Pressure in 
atmospheres. 
Line. 
Wave-length. 
Displacements in thousandths 
0 of an 
Angstrom unit. 
Reversed. 
Not reversed. 
15 
/1 
4236-09 
95 
355 
20 
/ 2 
4233-76 
108 
296 
30 
/1 
4236-09 
155 
344 
80 
e3 
4250-64 
293 
754 
From Table VII. the change in the frequency under pressure of the absorbing 
particles is less than one-half the change in the frequency of the primary radiation 
from the core of the arc, but as the measurements are taken from the region of 
anomalous displacements, and are from plates bearing different amounts of absorption, 
the readings are not strictly comparable ; the better method is to derive the values 
from Diagrams VII. and VIII., when the displacements of the reversal approximate 
closely to one-half that of the emission line. 
The suggestion that a possible cause of the anomalous displacements of the 
absorption lines is some interaction in the outer envelopes of the arc between the iron 
and the surrounding atmosphere of air is capable of experimental investigation by a 
study of the effect of different gases upon the spectrum, and this is on the programme 
