, in this experiment.The Michelson Interferometers you will use in this laboratory are precise and expensive research-grade instruments. Despite their heavy construction, however, they must be handled with great care, since the dimensional tolerances that have to be maintained are on the order of a wavelength of light, or a few times 10-5 cm.
Be particularly careful not to touch the mirrors or beam splitter. The mirrors are front-surface coated and so they are particularly susceptible to damage.
A. APPARATUS
Your instructor will acquaint you with the various components on the base of the instrument. Notice that each of the two mirrors is adjustable. One has two thumbscrews on the rear of its mount; these adjust the mirror angle so that the necessary condition of near perfect parallelism between its image in the beamsplitter and the second mirror can be achieved. These adjustments will have been set previously for you; don't touch them until you do the slight adjustment mentioned below.
The second mirror is adjustable in distance from the beam splitter. You will notice a micrometer driving a lever arm that, in turn, pushes on the mirror from a point very near its fulcrum. The micrometer reading that gives l2 - ll = 0 is marked on the base of the instrument.
The light source you will use for the first part of the experiment is mounted on a bracket that extends from the main base. It contains two sources, each controllable from the switch box. One is an ordinary white incandescent bulb that emits at all visible wavelengths. The second is a mercury vapor bulb that emits light at a few discrete wavelengths; we say that these are "lines" in the violet, blue, green, and weakly in the yellow.
Included also in your equipment are two filters that transmit light over
wavelength bands of different width. The green plastic filter transmits a much
wider band than the blue filter; you will determine their approximate
bandwidths,
, in this experiment.
B. PROCEDURE
Turn on the mercury lamp, and slip the green filter over its window. The
function of the filter here-is to allow passage of the green spectral line
while blocking the others. The spectral width 
of what comes through, however, is not determined by the filter, but rather by
the mercury atoms themselves, which produce an extremely narrow bandwidth.
Assuming that the interferometer is in proper adjustment, you should now be able to see in the viewing port a system of vertically oriented stripes, or "fringes".
The reason that you see several fringes rather than a uniform illumination over the field is that the image of mirror Ml ` which we designate Ml in the figure, is not exactly parallel to M2 Thus, the difference in path lengths l2 - ll varies from one point in the field to the other.
Carefully make a slight adjustment in the nearest screw on Ml .
Note that the number of fringes in the field can be set to any value you
please. Now, with about ten fringes or so in the field, make a slight
adjustment of l2 - ll with the micrometer. For each
complete fringe that passes a given point in the field, l2 -
l1 has changed by 
. Why does the
fringe pattern appear to move across the field?
Attempt to place an upper limit on 
for the mercury green line by seeing how large l2 - ll
can be made with the fringes still visible.
Next, turn on the white light, and remove the green filter. You will see no fringes at all until you adjust l2 - ll to the neighborhood of zero, and then make a very careful search. Go slowly; the few fringes are easy to miss.
What can you conclude about the coherence length of white light from the fringes you see? Is the coherence length you infer consistent with what you already know about the bandwidth?
Place the green filter over the white light, and re-estimate coherence. Notice that visually, the present green light and the mercury line are the same, but that their coherences are very different.
Now, substitute the blue filter. Infer its bandwidth from the number of fringes you see.