With the close of the last century there came about a most active period of telegraph cable laying in the Atlantic Ocean under the flags of the four great commercial and maritime countries. In 1898 the French laid the then longest stretch of cable from Brest to Cape Cod; among the British expenses of the Boer War was a cable from the Cape to England, touching foreign soil only at the Cape de Verde Islands; in 1900 the first German trans-Atlantic cable was laid from Emsden to Coney Island via the Azores, and at the same time the Commercial Cable Company of New York put down their connecting cable from Nova Scotia to the Azores.
This last cable, made and laid by Siemens Brothers & Co., of London, was without doubt the most perfect telegraph cable ever manufactured and its laying stands as the most scientific, skillful and successful exhibition of that branch of the seaman's art.
It was the exceeding good fortune of the writer to have been the guest of Mr. William Siemens, the head of the Siemens Company on board their cable-laying steamer Faraday during the laying of the last section of the Commercial Company's Canso-Fayal cable, in the summer of two, his invitation having been extended through and most generously approved by Clarence W. Mackay, now president, and by Mr. George G. Ward, manager of the Commercial Cable Company.
Of an equally high order with the making and laying of this cable was the navigation of the Faraday while paying it out, and this too will stand at the head of scientific and skillful navigation of the ocean during the 19th century.
Among the Faraday's navigation appliances was the "Distance-Wire" which gave the exact distance run over the ground: a deep-sea sounding wire (No. 20, BWG) was paid out continuously under a carefully regulated strain and in passing over a measuring wheel recorded the run with an accuracy comparable to that of measuring distances in land surveying. Furthermore the "lee-way" could be quite accurately checked by observing the trend of this "ground-log" wire over the stern.
Another feature was the daily noon telegraphic time signal from the Canadian Government observatory at Quebec, which gave the chronometer comparison on the syphon-recorder tape.
On deck no less than nine experienced navigators took sights from early dawn until dark and half-hourly azimuths on the Sir William Thompson compass with its accurate shadow-pin. In sextant work these observers habitually used the long astronomical telescopes in their sextants; these and all other operations were favored by clear weather and a smooth sea in which the 10,000 ton Faraday with her deep bilge-keels and the restraining pull of the cable over the stern rode as steady as the proverbial church. All sights were worked out with 7-place logarithms and a mean of the various positions was accepted after throwing out any which differed a mile from that mean.
But in spite of all these, there came a bright day with a seeming perfect horizon when every "line of position" was bad and although the nine sextants gave the same meridian altitude to within a half minute, yet the noon latitude was nearly three miles north of the line on which the ship was supposed to be. With unfaltering judgment, Mr. Brittle, the cable engineer in charge, held his course and the next day's sights put the ship back on the line which had been laid off two days before.
All these scientific appliances and all the care and skill of these trained navigators had been bowled over by a change in the atmosphere whereby the sea horizon was raised above its normal place, and the actual "Dip of the Horizon" was less than that taken from the "Dip Table."
Having no means of measuring the actual Dip, the sights were all rejected and the location of the cable on the chart was plotted by "dead reckoning" from the "Distance-Wire" records, as if the sun and stars had not been seen that day.
With this experience fresh in mind, the writer conceived the idea of the Navigator's Prism by which the Dip of the Horizon can be measured as readily and twice as accurately as the index-error is taken on the horizon, of using this measured Dip to correct the observed altitude and of thus eliminating a most treacherous element of uncertainty in observations at sea.
The Navigator's Prism is plain glass prism of square section with the end faces bevelled off at angles of 450 with the long axis and at right angles with each other, as shown to scale in Fig. 1.
A ray of light in the plane of the prism, striking as indicated in the figure, is turned through 18o degrees and returns in a Path parallel to its first direction.
Held vertically, with the lower bevelled face at the height of the eye, Fig. 2, the observer sees in the prism the back horizon (inverted) and at the same time sees in front of him (on either side of the prism) the front horizon: the two horizons are separated by the angle of twice the Dip.
To measure this angle with the sextant, the Prism is secured to the sextant so that the lower face is in front of the index- glass, Fig. 3, and the index-bar is moved until the direct (front) horizon and the reflected (back) horizon are seen in line: the corrected reading of the sextant is twice the angle of the Dip.
The details of the attachment are shown in Fig. 4. The Prism is protected, except the faces A and B, by a cover of aluminum to which is attached the clamp for securing it to the sextant. The clamp may fit over the upper leg of the sextant (which covers the axial pin of the index-bar) or, if preferred, it may fit directly on to the wooden handle so that no additional weight is brought on any part of the frame of the sextant.
It might appear at first glance that any error in the construction of the Prism would be eliminated by taking two observations with the Prisms reversed end for end, but such is not the case since the error would appear in the same direction both times. It becomes necessary then to determine the error (if any) in the construction of the Prism. This error, measured by the maker of the Prism, is engraved on it and is bound to remain constant.
To measure the Dip: measure the index correction of the sextant; clamp the Prism to the sextant, and, holding the sextant vertical (as in taking a sight), turn the tangent screw until the reflected and direct horizons are in line; read the sextant and apply the index correction of the sextant and the Prism correction to the reading and divide this corrected angle by two and the result is the angle of the Dip.
To take a sight it is not necessary to take off the Prism for it may be turned back out of the way, as in Fig. 5, leaving the index-glass unobstructed.
The observed altitude is corrected first for index error and then the measured Dip is subtracted and the remaining angle is the observed altitude above the true horizon.
There is no adjustment of the Prism to the sextant; this can be shown when the two horizons are seen to be in line, by slacking up the clamps and moving the Prism when, so long as the reflected horizon remains in the field of view, its coincidence with the front horizon is intact. If the Prism were turned about its vertical (long) axis through a considerable angle the horizons would separate but in practice the reflected horizon disappears from the field of view before this separation becomes large enough to be noticed.
DIP TABLES.
The Dip of the Horizon set forth in Table XIV, Bowditch's Practical Navigator, is tabulated from a formula deduced by Chauvenet ("Spherical and Practical Astronomy," Volume I, page 177), and does not differ materially from the Dip Tables in other recognized Epitomes.
The formula is:
Dip = D — .0784 D
D = the Dip of the Horizon, neglecting the atmospheric refraction.
X = the height of the eye of the observer above the water.
a = the radius of the earth.
The determination of D is mathematical and exact, although in practice there may be small errors in the assumed height of the eye due to variations of the draft and trim of the ship and of the position of the observer.
But the refraction coefficient 0.0784 D is deduced from a series of assumptions and ancient experimental data, and at best is supposed to hold only for a mean refractive state of the atmosphere, and many cases of abnormal Dip of the Horizon have been recorded.
The following table sets forth some of the observations of the Dip of the Horizon as measured with the Navigator's Prism by the officers of the U. S. S. Alert.
The Dip is shown to have varied more than 10 minutes of arc in an interval of 5 days in practically the same place, and this too with only small thermometric and barometric changes.
These observations prove conclusively that the Dip of the Horizon from any given height is variable and that it should be measured at the time of taking a sight: the Navigator's Prism attached to the sextant provides a simple, accurate and handy instrument for this purpose.
USE OF THE NAVIGATOR'S PRISM IN PILOT WATERS.
The late Lieutenant-Commander F. R. Brainard, U. S. Navy, in a small volume entitled "Sextant and other Reflecting Mathematical Instruments," Van Nostrand Co., 1881, defined an Inter-Range as a line joining two distant objects on which line the observer is located between the two objects; this in contradistinction to an ordinary Range when the observer is on the extension of the line joining two objects. (See page 75.)
Later on (page 112) he points out that an instrument by which the Dip could be measured would be useful in navigation and surveying by enabling the observer to make use of Inter-Ranges.
This the Navigator's Prism does: held horizontally with one bevelled face in front of the eye, Fig. 6, the observer sees in the Prism the landscape behind him, and at the same time sees ahead of him (above and below the Prism), the front landscape. Each object of the rear landscape appears in the same vertical line with some object in the front landscape and the observer is on the straight line joining each pair of these objects which become an Inter-Range.
When an exact Inter-Range has been established the coincidence of the front and rear objects is not altered by turning the Prism on its three axes, nor by moving it bodily in any direction, so long as the reflected object remains in the field of view it remains in coincidence with the front object. If revolved about the line of sight, the reflected field is seen to revolve about the rear object as a center and the reflected horizon appears inclined to the front horizon.
Fig. 7 shows a method of attaching the Prism to a binocular telescope to assist the vision. The bevelled end of the Prism partly covers one object glass and the relative brightness of the direct and reflected landscape is regulated by moving the Prism in the direction of its length, thereby altering the ratio of the covered to the uncovered areas of the object glass. The horizontality of the Prism is secured by turning the clamp about the barrel of the telescope until the reflected horizon is seen parallel to the front horizon.
While observing with this one eye is closed, but the Prism need not be removed for the ordinary use of the binocular because the image produced by the Prism fades away when the direct image is reinforced in brightness by the second barrel of the binocular.
Observations with these are facilitated by pointing the glasses upward with a slight inclination, so that the front horizon appears in the lower part of the field of view with the back horizon showing above its skyline; then by slowly lowering the telescope to the horizontal, the coincidence of the various points of the two pictures is readily observed.
USE OF THE INTER-RANGE IN PILOTING AND SURVEYING.
The exact determination of a ship's being on a line between two points, and the approximate determination of how far she is off that line, have heretofore been so difficult that little attention has been paid to this method of navigation. The use of the Range (by which was always meant two objects lying in the same direction from the observer) has long been well understood, but every two visible objects between which vessels pass, are available as an Inter-Range. It is evident that the occurrence of useful Inter-Ranges is much more frequent in pilot waters than that of useful ordinary Ranges. Especially is this true in chart work where the borders of the shore are presented with accuracy and detail, and afford many pairs of recognizable marks between which ships pass, whereas, on the distant background (where the rear-marks of the ordinary ranges must be sought) it is often difficult or impossible to locate on the chart a rear-mark.
Lieutenant Armistead Rust, U. S. Navy, in an article in the PROCEEDINGS OF THE U. S. NAVAL INSTITUTE, No. 102, June, 1902, on a "Direction Indicator" has shown many of the practical ways of using Inter-Ranges, all of which can be done by the Navigator's Prism which, with its absolute accuracy, its freedom from adjustment and its handiness, is offered as an instrument for the development of this new method of navigation in pilot waters.