prism, is a narrow slit, forming the sight through which the vision is directed when making an observation. On looking through this slit, and raising or lowering the prism in its socket, distinct vision of the divisions on the compass card immediately under the sight-vane is soon obtained, and these divisions, seen through the prism, all appear, as each is successively brought into coincidence with the thread of the sightvane by turning the instrument round, as continuations of the thread, which is seen directly through the part of the slit that projects beyond the prism.

The method of using the instrument is as follows:-The sight-vane s, and the prism P. being turned up upon their hinge-joints as represented in our figure, hold the instrument as nearly in a horizontal position as you can judge, or, if it be used with a tripod stand, set it as nearly as you can in a horizontal position by moving the legs of the stand, so that the card may play freely. Raise the prism in its socket till the divisions upon the card are seen distinctly through the prism, and, turning the instrument round, until the object to be observed is seen through the portion of the slit projecting beyond the prism in exact coincidence with the thread of the sight-vane, bring the card to rest by touching the spring n; and then the reading at the division upon the card, which appears in coincidence with the prolongation of the thread, gives the magnetic azimuth of the object observed, or the angle which a straight line, drawn from the eye to the object, makes with the magnetic meridian*. The magnetic azimuth of a second object being obtained in the same manner, the difference between these two azimuths is the angle subtended by the objects at the place of the eye, and, which is an important point, is independent of any error in the azimuths, arising from the slit in the prism not being diametrically opposite to the thread of the sight-vane.

For the purpose of taking the bearings of objects much

* The magnetic meridian now makes an angle of 24° with the true meridian, at London, the north point of the compass being 24 west of the true north point. This angle is called the variation of the compass, and is different at different places, and also at the same place at different times. Since this variation will affect equally, or nearly so, all azimuths observed within a limited extent and during a limited time, the angles subtended by any two of the objects observed, being the difference of their azimuths, will not be affected by the variation, and hence the map, or plan, may be constructed with all the objects in their proper relative positions; but the true meridian must be laid down, if required, by observations made for the purpose.

sup

above or below the level of the observer, a mirror, R, is plied with the instrument, which slides on and off the sightvane s, with sufficient friction to remain at any part of the vane that may be desired. It can be put on with its face either upwards or downwards, so as to reflect the images of objects considerably either above or below the horizontal plane to the eye of the observer; and, if the instrument be used for obtaining the magnetic azimuth of the sun, it must be supplied with dark glasses, D, to be interposed between the sun's image and the eye.

There is a stop in the side of the box, not shown in our figure, by touching which a little lever is raised and the card thrown off its center; as it always should be when not in use, or the constant playing of the needle would wear the fine agate point upon which it is balanced, and the sensibility of the instrument would be thereby impaired. The sight-vane and prism being turned down, a cover fits on to the box, which is about three inches in diameter, and one inch deep; and the whole, being packed in a leather case, may be carried in the pocket without inconvenience *.

THE BOX SEXTANT.

This instrument, which is equally portable with the prismatic compass, forming, when shut up, a box of about three inches in diameter, and an inch and a half deep, will measure the actual angle between any two objects to a single minute. It requires no support but the hand, is easily adjusted, and, when once adjusted, but seldom requires re-adjusting.

When the sextant is to be used, the lid, E, of the box is taken off and screwed on to the bottom, where it makes a convenient handle for holding the instrument. The telescope, T, being then drawn

B

E

T

* For much valuable information respecting the use of the prismatic com. pass, especially in military surveying and sketching, we can refer our readers to a Treatise on Military Surveying, &c., by Lieutenant-Colonel Basil Jackson, in which the subject is handled with great ability.

out, the instrument appears as represented in our figure. ▲ is an index arm, having at its extremity a vernier, of which 30 divisions coincide with 29 of the divisions upon the graduated limb, 17; and the divided spaces upon the limb denoting each 30 minutes, or half a degree, the angles observed are read off by means of the vernier to a single minute. The index is moved by turning the milled head, B, which acts upon a rack and pinion within the box. To the index arm is attached a mirror, called the index glass, which moves with the index arm, and is firmly fixed upon it by the maker, so as to have its plane accurately perpendicular to the plane in which the motion of the index arm takes place, and which is called the plane of the instrument. This plane is evidently the same as the plane of the face of the instrument, or of the graduated limb, l l. In the line of sight of the telescope is placed a second glass, called the horizon glass, having only half its surface silvered, and which must be so adjusted that its plane may be perpendicular to the plane of the instrument, and parallel to the plane of the index glass when the index is at zero. strument is provided with two dark glasses, which can be raised or lowered by means of the little levers seen at d, so as to be interposed, when necessary, between the mirrors and any object too bright to be otherwise conveniently observed, as the sun. The eye-end of the telescope is also furnished with a dark glass, to be used when necessary.

The in

The principle upon which the sextant is constructed has been proved at page 75; viz. that the total deviation of a ray of light, after reflection successively at the index glass and horizon glass, is double the inclination of the two glasses. Now the limb, 77, being divided into spaces, each of 15′ extent, and these spaces being figured as 30' each, the reading of the limb gives double the angle moved over by the index arm from the position in which the reading is zero, or double the angle of inclination of the two mirrors, if these mirrors be parallel when the reading is zero. If, then, the instrument be in perfect adjustment, and any object be viewed by it after reflection at both the mirrors, the reading of the instrument gives the total deviation of the rays of light, by which the vision is produced, or the angle between the bearing of the object from the center of the index mirror, and the bearing of the reflected image from the place of the eye, that is, between lines drawn respectively from the object to the center of the index glass, and from the reflected image in the horizon glass

to the eye. This angle is very nearly equal to the angle subtended by the object and its image at the place of the eye, differing from it only by the small angle subtended at the object by the place of the eye and the center of the index glass. This small angle is called the parallax of the instrument, and is scarcely perceptible at the distance of a quarter of a mile, while for distances greater than that it is so small that it may be considered to vanish. It also varies with the amount of deviation, and vanishes altogether whenever the center of the index glass is in a direct line between the object and the eye*.

To see if the instrument be in perfect adjustment, place the dark glass before the eye-end of the telescope, and looking at the sun, and moving the index backwards and forwards a little distance on either side of zero, the sun's reflected image will be seen to pass over the disc as seen directly through the horizon glass, and if in its passage the reflected image completely covers the direct image, so that but one perfect orb is seen, the horizon glass is perpendicular to the plane of the instrument; but, if not, the screw at a must be turned by the key, k, till such is the case. The key, k, fits the square heads of both the screws seen at a and b, and fits into a spare part of the face of the instrument, so as to be at hand when wanted This adjustment being perfected, bring the reflected image of the sun's lower limb in exact contact with the direct image of his upper limb, and note the reading of the vernier; then move the index back beyond the zero division of the limb, till the reflected image of the sun's upper limb is in exact contact with the direct image of his lower limb, and, if the zero of the vernier be now exactly as far behind the zero of the limb as it was at the former reading in front of it, so that the reading now on the part of the limb called the arc of excess, behind its zero divisiont, be the same as the former reading,

We have seen a method given for what is called correcting the parallax, when an observation is made at a short distance, by finding the deviation at this distance, when the angle between the object and its image is equal to zero; this deviation being given by the reading of the instrument, when the reflected image of the object observed exactly coincides with the object itself, seen through the unsilvered part of the horizon glass. This deviation, however, is not the parallax, even for a small angle between the object and its image, and, if the angle be not very small, the error introduced by the method will be greater than the parallax itself.

In reading an angle upon the arc of excess, the division to read on the limb is that next in front of the zero of the vernier, or between the zero of the vernier and the zero of the limb, and the divisions of the vernier itself are to be read from the end division, marked 30, and not, as usually, from

the instrument is in perfect adjustment; but, if not, half the difference of the two readings is the amount of the error, and is called the index error, being a constant error, for all angles observed by the instrument, of excess, if the first reading be the greatest, and of defect, if the second reading on the arc of excess be the greatest.

In the former case, then, the true angle will be found by subtracting the index error from, and in the latter by adding it to, the reading of the instrument at every observation.

This method of correcting for the index error is to be used with the larger instruments, hereafter to be described under the head of Astronomical Instruments; but in the box sextant this error should be removed by applying the key, k, to the screw at b, and turning it gently till both readings are alike, each being made equal to half the sum of the two readings first obtained. When this adjustment is perfected, if the zeros of the vernier and limb are made exactly to coincide, the reflected and direct image of the sun will exactly coincide, so as to form but one perfect orb, and the reflected and direct image of any line, sufficiently distant not to be affected by parallax, as the distant horizon, or the top or end of a wall more than half a mile off, will coincide so as to form one unbroken line.

To obtain the angle subtended by two objects situated nearly or quite in the same vertical plane, hold the instrument in the right hand, and bring down the reflected image of the upper object by turning the milled head B, till it exactly coincides with the direct image of the lower object, and the reading of the instrument will give the angle between the two objects.

To obtain the angle subtended by two objects nearly in the same horizontal plane, hold the sextant in the left hand, and bring the reflected image of the right-hand object into coin cidence with the direct image of the left-hand object.

It will be seldom that the surveyor need pay any attention to the small error arising from parallax; but, should great accuracy be desirable, and one of the objects be distant while the other is near, the parallax will be eliminated by observing the distant object by reflection, and the near one by

the zero division: thus, if the zero division of the vernier were a little further from the zero division of the limb, then the first division on the arc of excess; and if the twenty-seventh division on the vernier, or the third from the end division, marked 30, coincided with a division upon the limb, then the reading would be 33'.