Object-glass, or
that part of it or of the Eye which is not cover'd, may be consider'd as
a Wedge with crooked Sides, and every Wedge of Glass or other pellucid
Substance has the effect of a Prism in refracting the Light which passes
through it.[L]

How the Colours in the ninth and tenth Experiments of the first Part
arise from the different Reflexibility of Light, is evident by what was
there said. But it is observable in the ninth Experiment, that whilst
the Sun's direct Light is yellow, the Excess of the blue-making Rays in
the reflected beam of Light MN, suffices only to bring that yellow to a
pale white inclining to blue, and not to tinge it with a manifestly blue
Colour. To obtain therefore a better blue, I used instead of the yellow
Light of the Sun the white Light of the Clouds, by varying a little the
Experiment, as follows.

[Illustration: FIG. 13.]

_Exper._ 16 Let HFG [in _Fig._ 13.] represent a Prism in the open Air,
and S the Eye of the Spectator, viewing the Clouds by their Light coming
into the Prism at the Plane Side FIGK, and reflected in it by its Base
HEIG, and thence going out through its Plane Side HEFK to the Eye. And
when the Prism and Eye are conveniently placed, so that the Angles of
Incidence and Reflexion at the Base may be about 40 Degrees, the
Spectator will see a Bow MN of a blue Colour, running from one End of
the Base to the other, with the Concave Side towards him, and the Part
of the Base IMNG beyond this Bow will be brighter than the other Part
EMNH on the other Side of it. This blue Colour MN being made by nothing
else than by Reflexion of a specular Superficies, seems so odd a
Phaenomenon, and so difficult to be explained by the vulgar Hypothesis of
Philosophers, that I could not but think it deserved to be taken Notice
of. Now for understanding the Reason of it, suppose the Plane ABC to cut
the Plane Sides and Base of the Prism perpendicularly. From the Eye to
the Line BC, wherein that Plane cuts the Base, draw the Lines S_p_ and
S_t_, in the Angles S_pc_ 50 degr. 1/9, and S_tc_ 49 degr. 1/28, and the
Point _p_ will be the Limit beyond which none of the most refrangible
Rays can pass through the Base of the Prism, and be refracted, whose
Incidence is such that they may be reflected to the Eye; and the Point
_t_ will be the like Limit for the least refrangible Rays, that is,
beyond which none of them can pass through the Base, whose Incidence is
such that by Reflexion they may co