The apparatus adopted by Fresnel for the _fixed_ light may be regarded as perfect; but his _revolving_ light has been now superseded by the holophotal apparatus of Mr. Thomas Stevenson. The inclined mirrors and lenses employed in Fresnel’s apparatus are done away with in Mr. Stevenson’s, shown in the diagram, in which, by the single agency of lenses, _L_, and totally reflecting prisms, _p_, all the rays are rendered parallel. In this form the _whole_ glass frame, consisting of lenses and reflecting prisms, revolves round the central lamp. As Fresnel’s lighthouse prisms only gathered the light vertically, they could not produce the sheaf of rays required for the revolving light unless when combined with others which gathered the rays horizontally. The first lighthouse in which single prisms were made to revolve was the Horsburgh light, near Singapore, the apparatus of which was designed by Mr. Stevenson in 1850. In this form of revolving light apparatus the prisms are generated about a horizontal instead of a vertical axis, as in fixed light. The forms of the beams of light issuing forth from Fresnel and Stevenson’s apparatus are shown opposite to the diagram of each, and marked _x, y_. [Illustration: STEVENSON’S FIXED LIGHT VARIED BY FLASHES.] In France there is frequently employed what Fresnel called a “_fixed light varied by flashes_.” This effect is produced, as already explained, by causing panels of glass, curved horizontally but not vertically, to revolve outside of Fresnel’s fixed apparatus, as shown in the diagram. The ordinary fixed apparatus only acts in the vertical plane, while the straight panels only act in the horizontal plane. So that when the fixed apparatus is alone visible the rays are only gathered from the vertical plane, and the light is comparatively weak; but when the panels come opposite the eye, the rays are gathered from both planes into one powerful beam, as in a revolving light. Here, as in the former case, two agents are employed, causing great loss of light and great unnecessary expense, where one, if of the proper form, is sufficient. Mr. Stevenson’s modification of the holophotal arrangement for this purpose is shown in the diagram. It consists of alternate panels of the fixed light and holophotal apparatus; and thus, by single agency, a weak fixed light and a stronger revolving light are shown time-about to the mariner, which is the required characteristic. [Illustration: FRESNEL’S FIXED LIGHT APPARATUS.] [Illustration: FRENCH FORM OF APPARATUS FOR A FIXED LIGHT VARIED BY FLASHES.] If, for the sake of further diversity, the lights are to be coloured, we content ourselves in the case of a fixed light, as already stated, with enclosing the flame in a green or red tube. In the French eclipsing apparatus, polished sheets of coloured glass are placed, on one side or the other, against the lenses intended to emit the flashes of colour.