TRANSIT CIRCLE piece and visible with the threads themselves. In order to render thread and scale visible by night, various contrivances are used, the most common of which is to introduce a flat oval ring with whitened surface into the central cube, and with its plane inclined at an angle of 45 with the axis, so that, receiving light thrown in through an orifice in the pivots, it will reflect sufficient into the field to show the threads as black lines upon a bright ground. Sometimes also the illumination is thrown upon the threads themselves, when they appear as bright lines upon a dark ground ; and in the great transit circle at Greenwich a very ingenious combina- tion of prisms enables the observer to produce either effect at pleasure. Upon each half of the axis, between the cube and the pivots, is a circle whose diameter is usually from one third to one half the length of the telescope. These circles with their several radii and cross bars are generally cast each in a single piece, to in- sure greater firmness and avoid unequal ten- sions. But the six-foot circles of the Green- wich instrument just mentioned, weighing about 300 Ibs. each, are made of two castings, the rim in one, and the whole system of radii and braces in another, the two being afterward firmly bolted together at 12 equidistant points. Upon a narrow band of silver inserted near the circumference of the circles are cut the gradu- ations required for the special office of each ; one, used only for pointing the telescope in any given direction, is divided so as to read with a vernier to single minutes, which is abundantly sufficient; the other circle, intended for the exact measurement of angles, is divided with the most scrupulous accuracy into arcs of two, three, or five minutes, as the case may be, and, once fixed upon the axis, should never during observations be 1 handled or subject- ed to unequal pressure or strain of any sort. Assuming now that these division marks are truly cut, we next look for the means of sub- dividing the small arcs into seconds and frac- tions of seconds, and find this accomplished by a system of " reading microscopes." These are microscopes of the ordinary compound con- struction, but each provided with a microme- ter screw carrying, as in the German instru- ments, a pair of close parallel threads between which the image of the division under con- sideration can be placed with great accuracy, or, as in Tronghton's form, two threads cross- ing each other at a very acute angle, which may be bisected by the division. The micro- scopes are so made that one revolution of the screw is equal to a minute, and the microme- ter head is divided into 60 equal parts, each of which therefore represents a second. There are usually four of these microscopes placed 90 degrees apart; but sometimes as many as six are used for greater certainty, both from the greater number of readings and from the probable reduction of the systematic errors of the primary division. The proper method of supporting these microscopes to insure their perfect stability has been a subject of much study. A favorite plan has been to place them on the periphery of another smaller circle which rests, accurately fitting, upon the axis itself, but is prevented from revolving with i by a small projecting bar caught below between two screws attached to the pier. Experience however, seems to have decided in favor o securing firmly and independently upon the pier itself, near tiie V-plate, a solid block o metal which serves as the centre of a strong square frame at whose corners the microscope are attached by adjusting screws. The micro- scopes are thus entirely disconnected from the circle ; and although every new adjustment of the axis will show itself in their record of the graduations, yet this produces no effect what- ever upon the mean of readings of opposite microscopes. In the Greenwich instrument, whose piers are broader than the circles them- selves, the microscopes are very long, arid are passed through the pier itself, converging from the rim of the circle until their eye pieces are collected within a very small space, where the observer reads them with convenience and ease. The graduated limb is bevelled to suit this arrangement, and from another point near the observer a small gas-burner radiates light through other openings in the pier in such a manner as to illuminate uniformly the field of each microscope ; a matter of very high prac- tical importance. To bring the instrument into its proper place in the meridian, it is necessary that the middle vertical thread of the fixed diaphragm be placed truly in the op- tical axis of the telescope, which is the central line of the cone of rays converging from the object glass. This may be effected by turning the telescope to a very distant fixed object, noting the exact position of this middle thread with reference to the images in the field, and then reversing the instrument, when the thread will probably occupy a different position, whereupon it must be brought by the adjust- ing screws of the diaphragm to a point mid- way between the first and second places, and the operation repeated until no change ap- pears upon reversal. Next, by means of a spirit level and the vertical adjusting screws of one of the V-plates, the axis of the instru- ment is rendered truly horizontal ; and finally, the approximate sidereal time being known, the telescope is directed to some star, also known, very near the pole of the heavens, and the axis moved by the horizontal adjusting screws of the other V-plate, until at the right moment the star and thread coincide exactly. The three errors thus corrected are denom- inated the errors in collimation, level, and azi- muth respectively. And now, by help of stel- lar observations under properly varied circum- stances, we are able not only to determine with great precision the small outstanding val- ues of these errors, which by no means remain constant for any length of time, but also to judge the clock that aided us, and finally the
