To enable the observer to keep the eye end of the bubble at f till the required contact is observed, a light mahogany rod, about feet in length, attached to the sextant, and parallel to the telescope, is pressed against some fixed object on deck, which enables hint gently to elevate or depress the telescope till the bubble is brought into the required position, and kept there as long as may be necessary. For this purpose, an iron staunchion, about six feet in length, should be made to screw, when required, into different parts of the deck near midship, with a sliding projection, about two or tin ce inches in length, which may be fixed by a finger-screw at any required height, so as to afford a convenient prop, against which the sextant rod may be pressed by the observer when taking ob servations. To show the cross hairs, and the position of the bubble, when taking night observations, a small lamp. made for this purpose, is applied to the right side of the eye tube by means of a brass rod fixed to the lamp, which slides in a square socket, attached to the cylinder on the right of the holder of the teles cope. The quantity of light thrown upon the bubble and cross hairs, is easily increased or diminished, by moving the lamp rod a little forward or backward in the socket. The same lamp, when detached, enables the observer to read off his observations.
The screw o ni, acting• through the near end C of the level frame, gives it its vertical adjustment; and the two screws at p, acting horizontally against each other through the farther end D, give it its lateral adjustment. The accuracy of the vertical adjustment may be examined by comparing meridian altitudes of a celestial object, taken by means of the level, with those taken at, or nearly at, the same time, by means of an artificial horizon. At sea. the accuracy of this adjustment may be examined by moving the index backwards off the limb, as many minutes as are equal to the dip of the horizon, and then observing whether the reflected horizon of the sea is brought up to the horizontal hair in the centre of the field, when the eye end of the bubble is at f in the level tube; if not, its distance from it is equal to the error of the vertical adjustment, which may either be corrected by the screw, or allowed for, like an index error of the sextant.
To examine the latter adjustment, screw the object tube FI3 firmly into the sextant holder of the telescope by means of the screw at y, or fix it steadily, by other 'means, in a horizontal position, which is easily deter mined by the vertical adjustment of the level. Move the united eye and middle tube a few degrees round in it to the right and left; and observe whether the bubble, formerly in the middle, now moves to either end of the level tube. if it does not, the later"al ad justment is already made. is it does, correct the ob served motion by means of the adjusting screws at p. If this adjustment is not made, a slight deviation of the plane of the sextant from the vertical plane, which the observer cannot detect, when shut out from the horizon of the sea, may cause a considerable error in the observed altitude. To prevent this, let a plummet he suspended behind the plane of the sextant, which will readily detect any deviation of the instrument from the vertical plane.
If Ell, the middle tube of the telescope, be moved forward or backward in the object tube FB, so as to place the object glass a little too near, or too far from the cross hairs in the centre of the field, the image of the observed object may thus be brought nearer to or farther from the eye than the intersection of these cross hairs, without causing any apparent indistinct ness of the image. In this case, when the eye is slightly elevated or depressed, it will cause the con tact of the image with the horizontal hair to appear either too close or too open, and may thereby cause an error of one or more minutes in the observation, according to the distance of the image on either side of the cross hairs.
To avoid this source of error, care must be taken to mark on the middle tube Ell a line E 2 to which the middle tube should be moved, so that the image of a celestial object may be formed exactly at the cross hairs; for then, any elevation or depression of the eye will cause no sensible change of the apparent contact of the limb of the image with the horizontal hair. The proper distance of the object glass is a constant quantity for all celestial objects, but it varies with the distance of terrestrial objects. As considerable care and application are necessary, in order to acquire cor rectness and facility in the practice of this method of observation, it will be proper, when practicable, that the observer should accustom himself to take obser vations by this method on shore, before he proceeds to sea." In using the preceding very ingenious instrument, the eye sees, by direct vision. the contact of the sun with the horizontal wire, and by oblique vision the contact of the bubble with one or both of the marks on the level. This double and simultaneous observa tion is difficult to make; but independent of this diffi culty, there is a property of in virtue of which an object seen obliquely disappears, as if it had been completely annihilated.
On this and other grounds, says Dr. Brewster, Mr. Adam's eye tube has always appeared to me suscepti ble of improvement. The first idea of this kind which occurred to me is shown in Fig. 20. where the field of view, ABCD, is contracted, and consists of a perfora tion in the reflecting mirror. The parts are then ad justed, so that when the wire AB touches the sun, the bubble AMBN is concentric with the field of view ACBD. This approximation of the bubble to the observed limb of the sun is an obvious advantage; hut as it is liable to the objection formerly stated against oblique vision, I thought of the method shown in Fig. 21. In this method I dispense entirely with a metal lic reflector, and I form the image of the bubble by a plate of parallel glass PD, lying between the eye and the field CD, and inclined 45° to the axis of the tube. By this means the bubble EF may be brought in con tact with the wire AB; and the parts are adjusted, so that the axis of the telescope is horizontal when the wire AB is in contact with the lower end of the bub ble and the upper limb of the sun. The only objec tion to this construction is, that the glass plate 1'1) reflects little light; but this may be completely reme died, by placing the darkening glass anterior to the field of view, or by throwing an additional light upon the bubble of the level.
For farther information on sextants and artificial horizons, see Hooke's reflecting quadrant, in his ani madversions on Ilevelius, 4to, Loud. 1675; Newton's paper on a Reflecting Instrument Like Hadley's, Phil. Trans. 1742, p. 153; Hadley's Sextant, in Phil. Trans. 1731, p. 147, and 1732, p. 52; Ewing's Im provement of Hadley's Quadrant, american Transac tions, vol. i. p. 126; Maskelyne on nudity's Quad rant in Phil. Trans. 1772, p. 99; Magellan Descrip tion des ()duns rt SCrtan3 4to; Atwood's Theory of sextants in Phil. Trans. 1781, p. 375; Lud lam on Hadley's Quadrant, 8vo; Ward on Correcting the ,S'e•tant for the Back Observation, Phil. Trans. 1733, vol. xxxviii. p. 167; Leigh in Phil. Trans. 1733, vol. xl. p. 417; Short on Scrson's Top, Phil. Trans. 1751, p. 352; Little's artificial Horizon, in Irish Transactions, vol. viii. p. 77; Gould's Patent artificial Horizon, Repertory of arts, ii. vol. i. p. 98; Mr. Adam's Sextant in Edinburgh Journal ofSeience, No. vii. p. 95; Dr. Brewster's Improvement on it, Id. No. xii. p. 250; Ducont's artificial Horizon, Id. No. x. p. 341.