“In the system of levers made by Lord Rosse for his six-feet speculum, the primary, secondary, and tertiary systems were piled up one over the other, so that the distance from the support of the primary to the back of the speculum was about fifteen inches. This, as will be readily seen on consideration, introduced a new strain when the telescope was turned off the zenith, and had to be counterpoised by another very complicated system of levers. But in the Melbourne telescope, by the substitution of cast-steel for cast-iron, and by hanging the tertiary system from the secondary, and allowing it (_the tertiary_) to act in some places through the secondary, the whole system is reduced to three and a half inches in height, and the distance from the support of the primary lever to the back of the speculum is only one and three-quarter inch, by which means this cumbersome apparatus is entirely done away with. “The ultimate points of the tertiary system are gunmetal cups, which hold truly ground cast-iron balls with a little play, and when the speculum is laid on these it can be moved about a little by a person’s finger with such ease as to seem to be floating in some liquid.” It may perhaps be thought that it would be better to support these great specula on a flat surface, and it might be, if we could do so without extreme difficulty; but Lord Rosse has stated that if we attempt to support a large speculum on a surface extremely flat, a thread placed across that surface, or even a piece of dust, is quite enough to bend the mirror and render it absolutely useless. That will show the extreme importance of the support of the speculum. Let us then assume that we have the speculum and the tube perfectly adjusted. The next thing, in all constructions except the Herschelian, is to apply the second small reflector, concave in the case of the Gregorian, convex in the case of the Cassegrainian, and plane in the case of the Newtonian. This small mirror is generally supported by a thin strip of metal firmly fastened to the side of the tube, with power of movement parallel to the axis of the telescope, in the case of the Gregorian and Cassegrainian, for the purpose of focussing. In the Newtonian, the reflecting diagonal prism or plane mirror, inclined at an angle of 45° to the axis, is preferably supported in the manner suggested by Mr. Browning. See Figs. 77 and 78. In these B B B represent strips of strong chronometer spring steel, placed edgewise towards the speculum; by these the prism or small mirror D is suspended. The mirror thus mounted, does not produce such coarse rays on bright stars as when it is fixed to a single stout arm; it is also less liable to vibration, which is very injurious to distinct vision, or to flexure, which interferes with the accuracy of the adjustments. [Illustration: FIG. 77.—Support of diagonal plane mirror (Front view). ] [Illustration: