e interposition of a large dark satellite. Recent
observations of these eclipses, through glass of different colors, show
variations in the time of obscuration. Apparently, some of the rays
reach the earth sooner than others, although all leave the star at the
same time. As the entire time may amount to several centuries, an
excessively small difference in velocity would be recognizable. A more
delicate test would be to measure the intensity of different portions of
the spectrum at a time when the light is changing most rapidly. The
effect should be opposite according as the light is increasing or
diminishing. It should also show itself in the measures of all
spectroscopic binaries.

A third method of great promise depends on a remarkable investigation
carried on in the physical laboratory of the Case School of Applied
Science. According to the undulatory theory of light, all space is
filled with a medium called ether, like air, but as much more tenuous
than air as air is more tenuous than the densest metals. As the earth is
moving through space at the rate of several miles a second, we should
expect to feel a breeze as we rush through the ether, like that of the
air when in an automobile we are moving with but one thousandth part of
this velocity. The problem is one of the greatest delicacy, but a former
officer of the Case School, one of the most eminent of living
physicists, devised a method of solving it. The extraordinary result was
reached that no breeze was perceptible. This result appeared to be so
improbable that it has been tested again and again, but every time, the
more delicate the instrument employed, the more certainly is the law
established. If we could determine our motion with reference to the
ether, we should have a fixed line of reference to which all other
motions could be referred. This would give us a line of ever-increasing
length from which to measure stellar distances.

Still another method depends on the motion of the sun in space. There is
some evidence that this motion is not straight, but along a curved line.
We see the stars, not as they are now, but as they were when the light
left them. In the case of the distant stars this may have occurred
centuries ago. Accordingly, if we measure the motion of the sun from
them, and from near stars, a comparison with its actual motion will
give us a clue to their distances. Unfortunately, all the stars appear
to have large motions whose law we do not