Interference Fringes by a Highly Dispersive Medium. 327 
the centre, and still more out-of-step to the left of the centre. 
If we had but the two systems indicated by the solid and 
dotted lines, it is obvious that the systems would come into- 
step again to the left of the centre, a condition which would 
occur if D; consisted of two infinitely narrow lines very 
close together. In the actual case the presence of waves of 
length intermediate between those of B and C make such a 
recurrence of visibility to the left of the centre impossible, 
and we have distinct fringes to one side only of the original 
centre of the system. On increasing the density of the 
sodium vapour, the point D of maximum visibility moves 
further along to the right, and to keep the fringes in the 
field it is necessary to turn the screw of the instrument in 
such a direction as to cause the system to move in the same 
direction as the shift due to the sodium vapour. 
Now the sodium vapour accelerates the helium light, since 
its refractive index is less than unity for light of shorter 
wave-length than that of D,, consequently the reduced path 
is less. To shift the fringes in the same direction as that 
resulting from the shortening of the path through the sodium 
vapour, we must lengthen the other or air-path, which is pre- 
cisely what was found to be the case as I have already 
said. 
If the Dz line lay on the other side of the D lines, the 
shift would be in the opposite direction, 7. e. to the right, and 
we might at first sight expect the point of maximum visibility 
to shift to the left of the centre. We must, however, re- 
member that in this case the change of velocity is greatest 
for the shortest waves on the side C of the line; consequently 
the system indicated by the solid line will suffer the greatest 
displacement, and we shall have coincidence at D, to the 
right of the centre, exactly as before. To test this point 
experimentally, the interferometer was illuminated with light 
from the monochromatic illuminator, a narrow band on the 
green side of the D lines being utilized. The formation of 
sodium vapour in one of the paths gave rise to the same 
changes as were produced in the case of the helium light, it 
being necessary to increase the air-path to prevent the fringes 
from disappearing. If the fringes were made very narrow, 
so as to occupy only a small portion of the field, the wander- 
ing of the system to one side could be easily watched, as the 
sodium vapour was formed. It must be understood that 
only.a very small actual displacement occurs, the wandering 
of the system being merely a change in position of the region 
over which fringes can be seen. On repeating the experi- 
ment with a band cf approximately monochromatic light on 
