light, the rings then ceasing altogether. It needs but a moment's
reflection to understand that the colours of thin plates, produced by
white light, are never unmixed or monochromatic.

[Illustration: Fig. 14]

Newton compared the tints obtained in this way with the tints of his
soap-bubble, and he calculated the corresponding thickness. How he did
this may be thus made plain to you: Suppose the water of the ocean to
be absolutely smooth; it would then accurately represent the earth's
curved surface. Let a perfectly horizontal plane touch the surface at
any point. Knowing the earth's diameter, any engineer or mathematician
in this room could tell you how far the sea's surface will lie below
this plane, at the distance of a yard, ten yards, a hundred yards, or
a thousand yards from the point of contact of the plane and the sea.
It is common, indeed, in levelling operations, to allow for the
curvature of the earth. Newton's calculation was precisely similar.
His plane glass was a tangent to his curved one. From its refractive
index and focal distance he determined the diameter of the sphere of
which his curved glass formed a segment, he measured the distances of
his rings from the place of contact, and he calculated the depth
between the tangent plane and the curved surface, exactly as the
engineer would calculate the distance between his tangent plane and
the surface of the sea. The wonder is, that, where such infinitesimal
distances are involved, Newton, with the means at his disposal, could
have worked with such marvellous exactitude.

To account for these rings was the greatest optical difficulty that
Newton, ever encountered. He quite appreciated the difficulty. Over
his eagle eye there was no film--no vagueness in his conceptions. At