The parts of the telescope which are moved by the driving-clock weigh about 100 tons, and it was necessary to provide means of reducing the great friction on the bearings of the polar axis. To accomplish this, large hollow steel cylinders, floating in mercury held in cast-iron tanks, were provided at the upper and lower ends of the polar axis. Almost the entire weight of the instrument is thus floated in mercury, and in this way the friction is so greatly reduced that the driving-clock moves the instrument with perfect ease and smoothness. The 100-inch mirror rests at the bottom of the telescope tube on a special support system, so designed as to prevent any bending of the glass under its own weight. Electric motors, forty in number, are provided to move the telescope rapidly or slowly in right ascension (east or west) and in declination (north or south), for focussing the mirrors, and for many other purposes. They are also used for rotating the dome, 100 feet in diameter, under which the telescope is mounted, and for opening the shutter, 20 feet wide, through which the observations are made. A telescope of this kind can be used in several different ways. The 100-inch mirror has a focal length of about 42 feet, and in one of the arrangements of the instrument, the photographic plate is mounted at the centre of the telescope tube near its upper end, where it receives directly the image formed by the large mirror. In another arrangement, a silvered glass mirror, with plane surface, is supported near the upper end of the tube at an angle of 45 deg., so as to form the image at the side of the tube, where the photographic plate can be placed. In this case, the observer stands on a platform, which is moved up and down by electric motors in front of the opening in the dome through which the observations are made. [Illustration: Fig. 14. Spiral nebula in Andromeda, seen edge on (Ritchey). Photographed with the 60-inch telescope.] Other arrangements of the telescope, for which auxiliary convex mirrors carried near the upper end of the tube are required, permit the image to be photographed at the side of the tube near its lower end, either with or without a spectrograph; or with a very powerful spectrograph mounted within a constant-temperature chamber south of the telescope pier. In this last case, the light of a star is so reflected by auxiliary mirrors that it passes down through a hole in the south end of the polar axis and brings the star to a focus on the slit of the fixed spectrograph. ATMOSPHERIC LIMITATIONS