Mirror

During polishing the curve is checked at frequent intervals, first to see that it has been ground to the radius desired, and then to see if it is approaching the paraboloidal shape, which it has ultimately to reach; for this final change from the true spherq is usually all effected in the polishing. The method devised by Foucault is always used for these tests. A pin point of light is placed to one side of the centre of curvature of the mirror. If the mirror is a perfect sphere the reflected light will all pass through a point on the opposite side of the centre, and an eye at this point, or even a little further from the mirror but in a line with this point, will see the whole mirror filled with light. Also if a straight edge be moved up from one side to cut the beam at this point, the whole mirror will darken uniformly.

If the mirror is not a perfect sphere, no point can be found at which the mirror can be made to darken uniformly; but if it is a surface of revolution points can be found at which given zones will darken all over at the same time.

To obtain a flat surface, it is necessary to work three mirrors, call them A, B, C. These are then tested in pairs by the "Newton's ring" method. If A and B touch all over, and B and C and also C and A, they must all three be flat.

Elliptic Mirrors

(generated by the rotation of an ellipse about its major axis) have been successfully applied to cinemato graph projection, and to "flood-arcs" for theatre lighting. The mir rors are made of glass, and are silvered at the back, and then cop per-plated to protect the silver. The carbons between which the arc is formed are supported in the axis of the mirror, which is pierced in its centre to allow the positive carbon to pass through; the crater of the negative carbon faces the mirror, and is adjusted at its focus. An enlarged image of the crater would then be formed at the farther focus of the mirror; but before reaching this focus, the light passes through a concave lens, which further magnifies the image of the crater, so that it is now large enough to cover the whole of the picture to be projected, which is 0.7 by 0.9 inches. The picture thus becomes virtually a self-luminous

one, and the light from it enters the projecting lens as a slightly divergent beam. This method of dealing with the light differs fundamentally from that in the ordinary projecting lantern.

Searchlight Mirrors.

Another very important application of the concave mirror is to the naval searchlight. Here it is re quired to project a nearly parallel beam, so that an intense light may be concentrated upon any desired distant object. The mir ror must therefore have a paraboloidal figure, as this is the surface that has the property that a ray from the focus, striking the sur face at any point, will be reflected in a direction parallel to the axis of the parabola. The parabola in this case must subtend a large angle at the focus, if it is to reflect the bulk of the light from the arc, and it will therefore differ greatly from the shape of a sphere. Thus the methods used to produce the parabolic mirror of a telescope (which only subtends a few degrees at its focus) can not be used in this case. Such surfaces are difficult to grind.

To evade these difficulties, a French Colonel Mangin in 5774 described a very ingenious mirror made in the shape of a concavo convex lens with spherical surfaces, silvered on the convex side. By a proper choice of the radii of the two surfaces, the mirror behaves as a parabolic one, and gives a nearly parallel beam. These mirrors were used for some twenty years, but they were expensive, very liable to be cracked by the heat, and of course would be shattered by a single bullet. Mirrors made of sheet metal, pressed or spun to the required curvature, and coated on the inside with silver or gold, or other reflecting metal, and polished, do not reflect so much light.

A method of grinding a glass mirror to have a surface with any desired conic section was patented by Schuckert in 1888. He uses a very small tool, which can therefore fit the surface with sufficient approximation, and is constrained to travel in a conic-section. Other ways have since been devised for obtaining a glass mirror of the required shape. (R. S. CO