The sextant is figured and described in almost every book of navigation, and is so commonly to be met with that we shall be very brief. A is a plane glass, called the index glass, silvered behind, and perpendicular to the face of the instrument. It is fixed on a centre perpendicular to the instrument, aud moves with the index bar A rt, the end of which, a, slides over the graduated arc. c is a plane glass, the lower half of which, next the instrument, is silvered, and the upper half left clear. It is called the horizon-glass, and should be parallel to the index-glass when the index points to 0° at the beginning of the arc. n is a telescope for viewing the objects observed. This should be of good quality and with an inverting eye-piece. In the common quadrants there is merely a plate with a small hole for directing the sight. Suppose a ray of light to proceed from the eye, it will proceed in the direction of the telescope ; and if it falls on the upper or unsilvered part of the horizon-glass, it will pass forward in a continued straight line until it falls upon some exterior object. But if the ray falls upon the silvered part of the horizon-glass it will be reflected to the index-glass (the horizon-glass is so placed as to make equal angles with lines from the eye and index-glass), and again reflected from the index-glass, outwards (that is, from the observer), until it meets some external object. Now instead of supposing the rays to pass from the eye, suppose them to come from external objects to the eye ; then there will be two images presented at the same time, one formed by the rays which pass through the unsilvered part of the horizon-glass, and another formed by the rays which have been pre viously reflected by the two glasses ; and it is easily shown from the elementary principles of optics, that when two objects are thus appa — — rently seen in coincidence, the angle which they subtend at the point of meeting is exactly double the angle which is contained between the planes of the index and horizon glasses, if these be supposed to be prolonged until they meet. Hence, if the index be at zero when the gizmos are parallel, and if the arc be divided into half-degrees which are numbered and subdivided as whole degrees, the are actually read off, after any such observation as is here described, will measure twice the inclination between the glasses (for this inclination is evidently the same as the angle between the parallel position of the index-glass and its new position), and, in consequence of the optical principle above mentioned, this will be the true measure of the angle between the objects when they are so distant that the place of the eye and the intersection of the rap may be considered to be the same. The sextant, therefore, furnishes the means of measuring the angle between any two well-defined objects, in whatever direction they may be placed (so that the angle does not exceed 140'), and without requiring more steadiness, than is necessary for seeing the objects distinctly. E and F are sets of dark glasses of varying intensity, which may be turned before either the index- or horizon-glass when required; a, a tangent screw for giving slew motion to the index-bar ; the reading micro scope and reflector. The handle below is visible in the drawing, and there is a screw behind the collar of the telescope for giving an up-and down motion to the telescope, which alters its distance from the plane of the instrument without affecting its parallelism thereto.
The adjustments of the sextant may be divided into those which more properly concern the artist and those which aro to be attended to by the It should be distinctly understood that no other instrument requires so much care and skill in its construction and use as the sextant. There are scarcely half a dozen makers in England, and probably not double that number in all Europe, who are fit to turn out a sextant; and though skilful sextant observers are somewhat more plen tiful, they are still exceedingly rare. The fabric of the sextant should be strong and light. A blow which would not seriously damage a thee dolct or circle, where three or more readings correct pretty nearly the errors which arise from such injury, would entirely ruin a sextant ; but while giving strength, the maker must be on his guard to avoid weight.. Troughton and Simma's double-frame sextants of 8 inches radina are a very good model, those of 10 inches are too large ; single-frame sextants may be had of 7, 6, or 5 inches radius. All these alum have or may have the same telescope, glasses, and shades. Theses which we have hitherto seen seem to admit of some improve. ment in their scheme. Mr. Dollond casts his sextanta in one piece of hard metal, which is an improvement, and very beautiful instruments they are. The limb of the instrument should be perfectly plane, and the axis on which the index-glass turns exactly perpendicular to it. If a purchaser is sufficiently confident in his own judgment to rely on that, rather than on the reputation of the maker, he should release the clamp, observe whether the index-bar moves wally and without shako along the whole arc, and also whether the vernier seems to press equally in every position. The telescope should be of the best quality,
with a large aperture and of as short a focal length as these conditions permit. It should show the edge of the tun and moon quite sharp and distinct with the highest powers employed. The index-glass is generally the most faulty member of a common sextant; the faces are not parallel, and consequently when rays fall very obliquely upon it (in which case those reflected from the front surface are nearly as copious as those from the silvered back), there are two faint and sepa rated images seen instead of one bright one. The artist tries his index glasses by the following test : The sun is viewed with a high power after very oblique reflection from the index-glass, and those glasses only are retained in which the image is single and sharp. The rejected glasses find their way to the second-rate makers. The horizon-glass, which is smaller and on which the rays fall at a larger angle, does not, if imperfect, affect the images so much, but it too should be examined. The ahades or dark glasses, which are either interposed between the glasses or placed before the horizon-glass, are next to be examined. The two images of the sun are brought into proper contact, after applying the highest power to the telescope and putting the proper dark glans on the eye end. One of the lightest shades is then inter posed between the Index and horizon glasses, and if the contact is slightly disturbed, this shows the form of the shade to bo prismatic. The shade is turned in its cell until the effect is perpendicular to the plane of the sextant. The corresponding shade before the horizon glass is now interposed arid tried as the former one, and turned round exactly as before, until its effect is also perpendicular to the plane of the instrument. Leaving this latter shade untouched, and substituting the next dark shade in place of that first employed, the process is again repeated, with this caution, that the error, if any, caused by the pris matic form of the last shade, is to be made exactly contrary to that of the first shade, as well as perpendicular to the plane of the sextant. In this way, changing when necessary the dark glass at the eye end of the telescope, all the dark shades are carefully examined and set right, and it need not be said that those which are very prismatic or which affect the sharpness of the images, are to be rejected. The index-glass is now to be set at right angles to the plane of tho instrument. This is done by looking very obliquely in the index-glass, when the edge of the limb or the divided arc seen by reflection will form an uninter rupted continuation to that seen directly, if the glais is perpendicular; otherwise the are will appear broken where the direct and reflected images meet ; by touching the screws seen in the figure, this error may be corrected. Repeating the observation at diffirent parts of the limb, will show whether the axis has been set at right angles to the limb. The horizon-glass is to lie made parallel to the index-glass by bringing the two images of the sun to coincide, when, if there is any lateral overlapping, it may be remedied by turning a screw (shown in the figure) which is in the mounting of the horizon-glass. The index error, as it is called, is of no importance whatever, though many observers are afraid that when large it vitiates the observation. This is an idle prejudice, but one which the maker is compelled to respect. hiaviug ascertained the diameter of the sun by measuring it forwards and back wards, he sets the index to the proper angle and then files away a little of one of the pins against which the index.glaes is pressed, until he brings the images nearly into contact Contacts should now be formed, first by turning the tangent-screw so as to separate the images, and then by bringing them together. If the readings do not agree, the centre moves too stiffly or the index-bar is too weak. Finally, the telescope is to be act parallel with the plane of the instrument. There are two parallel wires In the focus of the telescope, which are to be placed parallel to the instrument, and then as large an angle as possible is to be taken between two distinct objecta, making the contact exactly in the middle of the wires. The contact must now be examined when the objects are first at one side and then at the other of the field of view. If they separate equally when at equal distances from the centre, the position of the telescope la correct ; if not, there are adjusting screws In the collar which carries the telescope by which this -error may be corrected. For some of these Latter adjustments mechanical means may be and sometimes are applied. The instrument should however comply with the foregoing testa, which any one can try who understands the use of the instrument.