SELF-REGISTERING INSTRUMENTS. In the determination of scientific data, it is of the utmost importance that the instruments of observation be as free from error as possible, or that the sources and amount of error bo accurately known. It is further necessary that the means of observation be irreproachable, and, if possible, con tinuous. Where an instrument requires to have its indications recorded night and day, a large staff of observers is required, and they are liable to error, even when carefully trained to their work. [EouarioN, PERSONAL.] The magnetic instruments at the Greenwich Observatory were formerly observed every two hours ; but the results were of course liable to error, and occasionally the magnetic variations were too rapid and transient to be recorded. Hence attempts—and many of them successful ones—have been made from time to time to make instruments of observation record their own results. We have seen under ANEMOMETER the contrivances for registering the direction and pressure of the wind, together with the amount of rain ; and under BAROMETER some of the mechanical arrangements are stated for making that instrument self-registering. The maximum and mini mum thermometers in ordinary use will be described under TOERMO METER. Our object in this placo is briefly to notice the vast improvement that has taken place of late years in consequence of the application of photography to the purposes of self-registration. A description of the apparatus employed for the self-registration of the changes in position of the declination magnet, of the horizontal force magnet, and of the vertical force magnet, will show the nature of the arraugements. A prepared photographic paper is wrapped round a cylinder, the axis of which is placed parallel to the direction of movement to be registered, and the cylinder is turned round at a uniform rate by clockwork. The light is supplied by a gas-lamp furnished with a copper chimney, In which is a small alit capable of adjustment by a screw. It is on the breadth of this slit that the breadth of the register-line depends. This light falls upon a concave speculum, which rests in a stirrup connected with the magnet to be observed, so as to partake of all its angular movements. The pencil of light is reflected from the mirror to a plauo convex Iona, placed near to, and parallel with, the axis of the cylinder, which lens condenses the line of light to a definite spot of light on the paper. The source of light being fixed, it is evident that the move ments of the spot of light will correspond with those of the magnet, to the right and left in a horizontal plane in the case of the declination magnet and horizontal force magnet, and up and down in a vertical plane in registering the movements of the vertical force magnet. Hence, I as the cylinder is constantly being moved round by clockwork, there is traced upon the paper a curve of which the abscissa, measured in the direction of a line round the cylinder, is proportional to the time, while the ordinate, measured in the direction of the axis of the cylinder, is proportional to the movement of the magnet. A base-liue, from which
to measure the ordinates, is traced upon the paper by the action of a spot of light proceeding from another gas-lamp placed near the cylinder and passing through a slit fixed to the carrier of the cylinder.
In the photographic registration of the barometer, the instrument is arranged in the siphon form, and a float, resting on the surface of the mercury in the shorter limb, hangs in a notch on the short arm of a delicately poised lever. At the end of the long arm of this lever is an opaque screen containing a small aperture, which a pencil of light passes. Now it is evident that, as this screen moves up and down with the oscillations of the mercury, a line will be traced on the pho tographic paper by the pencil of light -transmitted by the screen. There is also an arrangement for tracing a base-line on the photographic paper. The cylinder carrying the sensitive paper is mounted on a turn-table, which is carried round by the hour-band of a clock placed concentrically beneath it. The paper is covered by a second cylinder to prevent it from becoming dry, during the twenty-four hours that the apparatus remains in action. The cylinders are also covered with a blackened zinc case to prevent all light from falling on the paper, except that from the pencils which describe the register and the base line.
In the photographic registration of the thermometer and psychro meter, the bulbs of the instruments are freely exposed beneath a table on which is a revolving cylinder covered with sensitive paper, while the sterns pass up through the table, and are placed between the cylin der and a source of light, which, by means of a cylindrical lens, falls on the stem of the thermometer in a narrow vertical line, and, passing through that part of the bore which is above the mercury, blackens the sensitive paper. The boundary between the affected and the unaffected party of the paper indicates the position of the mercury in the stem of the instrument. Fine wires are placed across the slit through which the light passes, and coarser wires are attached to every tenth degree, as well as to the points 32°, 54°, 76% and 98° Fakir. The shadows of these wires protect the portion of the paper on which they fall from the action of the light, so that the darkened surface of the paper is traversed by is series of parallel pale lines, and the relative positions of the broad and narrow lines show the tempe rature indicated by the register. The whole of the apparatus is protected by an outer wind- and water-tight zinc case.