that when it ought not to be visible in the absence of vapours,
because the line AE encounters the rotundity of the Earth, it will be
perceived in the line AF by refraction. But this angle EAF is scarcely
ever more than half a degree because the attenuation of the vapours
alters the waves of light but little. Furthermore these refractions
are not altogether constant in all weathers, particularly at small
elevations of 2 or 3 degrees; which results from the different
quantity of aqueous vapours rising above the Earth.

And this same thing is the cause why at certain times a distant object
will be hidden behind another less distant one, and yet may at another
time be able to be seen, although the spot from which it is viewed is
always the same. But the reason for this effect will be still more
evident from what we are going to remark touching the curvature of
rays. It appears from the things explained above that the progression
or propagation of a small part of a wave of light is properly what one
calls a ray. Now these rays, instead of being straight as they are in
homogeneous media, ought to be curved in an atmosphere of unequal
penetrability. For they necessarily follow from the object to the eye
the line which intersects at right angles all the progressions of the
waves, as in the first figure the line AEB does, as will be shown
hereafter; and it is this line which determines what interposed bodies
would or would not hinder us from seeing the object. For although the
point of the steeple A appears raised to D, it would yet not appear to
the eye B if the tower H was between the two, because it crosses the
curve AEB. But the tower E, which is beneath this curve, does not
hinder the point A from being seen. Now according as the air near the
Earth exceeds in density that which is higher, the curvature of the
ray AEB becomes greater: so that at certain times it passes above the
summit E, which allows the point A to be perceived by the eye at B;