A ray of light can do the same; everybody will admit that in each case,
if there is no gravitation, light will certainly extend itself in a
rectilinear way. If we limit the light to a flicker of the slightest
duration, so that only a little bit, C, of a ray of light arises,
or if we fix our attention upon a single vibration of light, C, while
we on the other hand give to the projectile, B, a speed equal to that
of light, then we can conclude that B and C in their continued motion
can always remain next to each other. Now if we watch all this, not
from the movable compartment, but from a place on the earth, then we
shall note the usual falling movement of object A, which shows us that
we have to deal with a sphere of gravitation. The projectile B will,
in a bent path, vary more and more from a horizontal straight line,
and the light will do the same, because if we observe the movements
from another standpoint this can have no effect upon the remaining
next to each other of B and C.



DEFLECTION OF LIGHT

The bending of a ray of light thus described is much too light on the
surface of the earth to be observed. But the attraction of gravitation
exercised by the sun on its surface is, because of its great mass, more
than twenty-seven times stronger, and a ray of light that goes close by
the superficies of the sun must surely be noticeably bent. The rays of
a star that are seen at a short distance from the edge of the sun will,
going along the sun, deviate so much from the original direction that
they strike the eye of an observer as if they came in a straight line
from a point somewhat further removed than the real position of the