nciple, though a simple one,
escaped him, and it was first discovered by Willebrord Snell, about
the year 1621.

Less with the view of dwelling upon the phenomenon itself than of
introducing it in a form which will render subsequently intelligible
to you the play of theoretic thought in Newton's mind, the fact of
refraction may be here demonstrated. I will not do this by drawing the
course of the beam with chalk on a black board, but by causing it to
mark its own white track before you. A shallow circular vessel (RIG,
fig. 4), half filled with water, rendered slightly turbid by the
admixture of a little milk, or the precipitation of a little mastic,
is placed with its glass front vertical. By means of a small plane
reflector (M), and through a slit (I) in the hoop surrounding the
vessel, a beam of light is admitted in any required direction. It
impinges upon the water (at O), enters it, and tracks itself through
the liquid in a sharp bright band (O G). Meanwhile the beam passes
unseen through the air above the water, for the air is not competent
to scatter the light. A puff of smoke into this space at once reveals
the track of the incident-beam. If the incidence be vertical, the beam
is unrefracted. If oblique, its refraction at the common surface of
air and water (at O) is rendered clearly visible. It is also seen that
_reflection_ (along O R) accompanies refraction, the beam dividing
itself at the point of incidence into a refracted and a reflected
portion.[4]

[Illustration: Fig. 4.]

The law by which Snell connected together all the measurements
executed up to his time, is this: Let A B C D (fig. 5) represent the
outline of our circular vessel, A C being the water-line. When the
beam is incident along B E, which is perpendicular to A C, there is no
refraction. When it is incident along _m_ E, there is refraction: it
is bent at E and strikes the circle at _n_. When it is incident along
_m'_ E there is also refraction at E, the beam striking the point
_n'_. From the ends of the two incident beams, let the perpendiculars
_m_ _o_, _m'_ _o'_ be drawn upon B D, and from the ends of the
refracted beams let the perpendiculars _p_ _n_, _p'_ _n'_ be also
drawn. Measure the lengths of _o m_ and of _p_ _n_, and divide the one
by the other. You obtain a certain quotient. In like manner divide
_m'_ _o'_ by the corresponding perpendicular _p'_ _n'_; you obtain
precisely the same quotient. Snell, in fact, found this quotient to