An artificial horizon for sea use has been wanted for the last hundred years; in fact, ever since the sextant came into general use. It is a matter of common occurrence for mist or fog to obscure or shroud the horizon and thus render an observation impossible though the sun or the stars be in plain view. Many efforts have been made to provide a substitute for the natural horizon. A few of these are based on the prin: ciple of the gyroscope, or the pendulum; but the vast majority use a liquid to fix the horizontal. Such devices have failed for shipboard use, because they are seriously disturbed by the slightest movement of the vessel. The fact is often overlooked that as the sextant is a double-reflecting instrument each degree of the circle is represented by two degrees of arc and, therefore, every motion of the head of the sextant doubles the motion of the reflected image both in speed and extent. Hence, the requirement that contacts of heavenly bodies be made on the exact center of a bubble or other indicator of the horizontal is a feat that only the exceptionally steady few may accomplish; but even they cannot maintain such contacts long enough to be perfectly sure of their measurements. In the accompanying engraving we illustrate an attachment for sextants which is adapted to overcome the usual difficulties. An artificial horizon is provided which consists of two bars that exactly span the periphery of the sun's image in the horizon glass, and are maintained in horizontal position by means of a pendulum forming part of the attachment. This is shown in detail in the line cut. At A is a pivoted balance beam which carries the horizon bars B, the latter extending through a slot in the casing. The pendulum comprises a pair of plates, G, pivoted to a block at their upper ends, while their lower ends carry a heavy weight, D. A plate, E, is secured to the weight, B, and is formed with an opening through which extends a rod, F, projecting from the beam, A. The pendulum maintains its vertical position when the sextant is held in the hand, and by means of the rod, F, the horizon bars are kept at a proper horizontal level. The faults justly attributed to a pendulum when hung to the bulkhead or other fixture of a ship, do not prevail when it is suspended on an instrument in the grasp of a human pendulum who intuitively beats time to the movements of the vessel beneath him and maintains an upright position. The attachment adds about ten ounces to the weight of the sextant; but this is by no means an objection, as it tends to steady the hand of the observer. However, the principal feature of this attachment is the double-barred horizon or "spanner" horizon, as it has been called. If contacts are not properly made, that is, if the contact be too "hard" on either leg of the spanner, the error will be seen by the absence of contact on the other leg, and also by the orb being unequally divided by a wire which is stretched centrally between the bars. In fogs when the sun's limbs are not defined, a fair approximation may be obtained by centering the dim circumference. When quick-flying scud or slow-moving clouds obscure first one limb of the sun and then the other, its rise or fall is readily perceived on the respective legs of the spanner. At night the central wire may be used for observations. The spanner is protected against the wind by a casing of aluminium and a shutter G serves to protect the horizon glass from rain or spray. The attachment is the invention of Capt. Henry A. Gadsen, of 6 Binden Road, Raves-court Park W., London, England.