In a recent number of the Proceedings the writer made a futile attempt to solve some of the difficulties of the naval signal question, by proposing a system of day, night and fog signals conforming to the requirements of the American Morse code, adopted some two years ago as the service code; but any attempt to patch up the defects of that code, as applied to naval purposes, is a foredoomed failure, because it possesses just those undesirable qualities which every consideration of theory and practice condemns. The substitution of the English Morse code for the Myer was not regarded as successful, as it seems to have been dropped in turn for the American Morse. This last named was adopted by the Navy Department at the request of the Chief Signal Officer of the Army, in order that the two branches of the military service might communicate in emergencies. The principal reason which led the Army authorities to adopt the American code was, that it being the commercial code, in time of war skilled operators would be abundant from which to recruit the Signal Corps. That reason at least has no bearing on the question from a naval standpoint, for in time of war if there is anything more than another that is not wanted aboard ship it is a landsman with only one talent, and that for working a key. In accepting this allurement, however, the Army authorities left out of consideration the fact that in time of peace, especially now that enlisted men can purchase their discharge, the current is reversed, and the expert signalmen of the Army furnish an inexhaustible source of supply to the commercial demand for telegraph operators. In the Navy we had some such experience with seaman gunners as expert electricians, until the course at Newport was changed. Whether or not the Army authorities have made a mistake concerns us only in that if we are to be saddled with a code not adapted to our purposes, and ages behind the times from a naval standpoint, just because the Army has adopted it, then the sooner we can compromise with the Army authorities the better. As it is now we have a code which possesses the sole recommendation that in some remote emergency we can communicate with an Army station. It does not take much cleverness to see, however, that this very desirable thing could be accomplished by requiring only the signal officer and signal men of each ship to know the American Morse code for just such emergencies; but why require the whole Navy for years and years to use it, with all its glaring defects, when the solution is so simple?
The American Morse code is not adapted for naval purposes:
1st. On account of the space letters c, o, r, y and z, and in the character and. The space is bad in the wig-wag, for in the recovery from a front, representing the space, it is made so quickly that unless the signal sender is facing squarely (and often he cannot be, in communicating, say, with two adjacent ships or stations at once) it is difficult to distinguish this quick front from a dot or dash, as the case may be. In night flashing with winker lights it requires an extra light to make the space signal, or else the use of a very long dash, which is confusing and wastes time. In a display of red and white lights, like the Ardois system, a third color (green) must be introduced, as was attempted in a system proposed by the writer in a recent Proceedings, which plan as there proposed is acknowledged to be a failure. Anything which attempts to utilize the American Morse code is, however, bound to more or less of a failure. In fog signals the space must be indicated by a long dash, which may lead to some confusion, but at any rate wastes valuable time. If these reasons are not enough to condemn the American Morse code for naval purposes, then:
2d. The character (error is, in this code, seven dots; the numeral six is six dots; the letter p is five dots. Even if we get around the error by substituting some device, still, in some such system as the Ardois, by which all the elements of a character are exhibited in one display, we must have at least six lanterns in the permanent hoist with which to make the numeral six. When the Navy Department ordered the eight Ardois apparatus from abroad, a plan of the sixty-four sectors was furnished the makers by which to mark the disks of the signal boxes. It was found impossible to use the American Morse code, not only on account of the requirements of the error and the six, but of the impossibility of displaying the space letters. Accordingly a new alphabet and numeral code was invented. In other words, the American code has to have a supplementary code to do its work for it. Now the system of night signaling by means of a permanent hoist of lanterns similar to the Ardois has come to stay. The advantages offered of rapidity, rehability, and distinctness within the limits of distance imposed by squadron cruising, make it improbable that any device will very soon supersede it. All the principal navies of the world have adopted some such system, and the only question is as to the best apparatus, from a mechanical standpoint, for exhibiting the displays of lights needed for the different codes.
It is interesting to note, in this connection, that we are much behind other navies, in that we use five lanterns in a permanent hoist. Italy, Spain, Austria and France have adopted four lights. These, spaced in the distance formerly occupied by five or more, increase the range of visibility greatly, and with military masts and the short hoists offered by the auxiliary vessels of a fleet, such as torpedo-boats, etc., four lamps possess an enormous advantage over even five. Of course, using four lanterns implies the use either of a four-element code, or, as with the Italians, the consonants only, just as in the International code. Indeed, they have for day purposes the regular International flags, but the signal books are of course different. The principal night-signal apparatus used abroad is that on what is known as the Kaselovsky system, as improved by Lieutenant Sellner of the Austrian Navy. One of these four-lamp Sellner devices is now mounted on board the Chicago for trial as compared with the Ardois. It has, unfortunately, only twenty-seven sectors marked, and uses only the consonants. It is the type in service in the Italian Navy, and costs delivered in this country about $700. The Ardois with five lamps and sixty-four sectors is invoiced at about $1180. In point of workmanship it looks as if the Ardois were much superior, but there are excellent points in each that might well be embodied in an apparatus of home manufacture. The proposition is simply how to best and most economically mount four double (red and white) lanterns in a permanent hoist, and make certain displays of lights by means of keys. The Ardois lantern is far from satisfactory. American ingenuity can be trusted to solve the problem. An apparatus designed by Ensign F. J. Haeseler, U. S. Navy, possesses great merit, but the field is an open one. If the Navy Department will settle the vexed question of the numerous codes we would soon have smooth sailing. There is enough data in the office of Naval Intelligence and available in the Squadron of Evolution to warrant an official inquiry and settlement, by a board of officers or otherwise, of all the points at issue. The revision of the general signal book need in no way interfere with this, although it is certainly an open question as between the use of flags similar to the International and those in the present numeral day flag code. It is generally admitted that the present method calls for too many flags in a hoist. The principal argument against any change is the great expense of devising and printing new signal books. There is, however, a further reason for going slowly in the matter. There is a growing feeling that some device like collapsing shapes will eventually supersede flags for day signals, in which case a numeral code would seem to be necessary. It will, however, be observed that this question does not enter into the one under consideration, that of night signals, for a four-element code admits of both the International (consonants) and the numeral code.
The Germans use a three-element code and a hoist of three lights. The English Admiralty in June last ordered a three-lantern Sellner apparatus for experiment. Three lanterns restricts the usefulness of the system to numeral codes only, whereas four admits of both an alphabetical and numeral code, as will be shown.
Experience in foreign services has led to the universal condemnation of the green light for general signal purposes (that is, a green transmitted light as distinguished from a chemical one, as in the Very system). Indeed, it is even so well recognized that a white light is visible in thick weather farther than a red one, that in the latest Sellner apparatus four white lights only are used. What was the red light in a display is now a pulsating white one, readily distinguishable from the steady white light. The pulsations are given by taking the current from a specially arranged commutator, and it is claimed that a pulsating light is visible very much farther in thick weather than even a steady white one. Dynamos and apparatus are furnished cruising torpedo-boats in some foreign services, with the fixed hoist of four lanterns. In fact, the whole question seems to have passed the experimental stage in which we find ourselves. By a simple device, in the later types of the Sellner apparatus the signal made is automatically recorded on a slip of paper, so as to check errors and preserve a record.
We have now for night signaling-, both in the North and South Atlantic Squadrons, 1st, the Ardois alphabet and numeral code; 2d, the American Morse alphabet and numeral code, transmitted in one of three ways, viz., by three lights and a hand keyboard, by a torch wig-wag, or by fog whistle; 3d, two Very codes, one of three elements with brackets (the original), and a very successful and desirable experimental code of four elements. As the Coston signals have not as yet been officially suppressed, it might not be unfair to include this system and code in the list, as well possibly as the hoists of oil lamps in the general signal book. This is a somewhat chaotic condition of affairs, but the solution is absolute simplicity itself.
1st. Abolish the Coston, the oil lamps, and the original Very three element code.
The Coston lights are from time to time sent out into service to be further experimented with. The system is a back number and should be given its quietus officially. The Very signals, as a system, are the best long-distance nautical night signals in the world, but the original code introduced the bracket in order to get a three-element code with two colors. If, however, one ball fails, as is too often the case, it is necessary to repeat the whole signal. This is not only exasperating, but is a grave defect. With the four-element code, failures do not interfere, as two minutes delay are allowed between elements. The four-element code should unquestionably be adopted in place of the original, and a further change made of using white in place of green. There is no fault to be found with the chemical green light, but as the night signal colors are now red and white, uniformity calls for the change. There is another reason which will appear later on.
2d. Retire the American Morse code, excepting for expert signalmen to use in communicating with Army stations.
3d. Drop the present Ardois code as unnecessary.
4th. Adopt the four-element Very code, for all purposes, as the numeral code.
The Very system has also come to stay, and every watch officer, watch petty officer and signalman has to know it or should be required to know t. The four-element code is as follows:
1. RRRR 2. GGGG
3. RRRG 4. GGGR
5. RRGG 6. GGRR
7. RGGG 8. GRRR
9. RGGR 10. GRRG
To use this on the Ardois or other apparatus with four lanterns, the green would show white and the red show red, but it is urged that the Very signals be changed in that respect from green to white. To use the numeral code on the fog whistle or in the ordinary wig-wag, transcribe it so that for red read one, and for green or white read two, thus:
1. 1111 2. 2222
3. 1112 4. 2221
5. 1122 6. 2211
7. 1222 8. 2111
9. 122 10. 2112
In other words, where we now have in service four numeral codes, viz., the Morse, the Ardois, the Very three element, and the Very four element, substitute the last named for all—one for four.
5th. Adopt the Myer alphabet, dropping everything but the letters.
The service is a unit on a return to the Myer code. It was used through the war and for eighteen years subsequent. It is a four element alphabet, and requires only four lanterns in a permanent hoist. Its elements of one and two are more distinct to the ear on a fog whistle than a dot dash system; the eye catches them more distinctly with the night flashing or winker light, and an inexperienced operator can make one, two better on a key than he can a dot, dash. It only requires one winker light to transmit the Myer, whereas the American Morse takes three.
In other words, the five propositions here submitted mean simply, adopt what may be called the Myer- Very code for all purposes, and practically drop everything else of the kind.
It is amazing how much this simplifies matters. The Ardois apparatus now in service can be converted without expense to the new code by pasting the Myer-Very characters over the proper sectors, and by doing away with the lowest lantern of the five. On the Philadelphia, for instance, her new masts crowd five lanterns to the point of failure, but there is full room for four. Instead of a signal box with sixty-four sectors, we would in future need one with thirty or thirty-two, and instead of five lanterns only four. This means a great saving in cost. It is proposed, with the adoption of the Myer-Very code, that signals be read in the hoist from top downwards, instead of from bottom upwards, as in the Ardois at present. This last named can be changed in five minutes to read that way.
It will be noted that there are only thirty possible combinations of the numerals one and two limited to four elements as a maximum. On the other hand there are twenty-six letters and ten numerals to be provided for in the Myer-Very, or thirty-six in all. It happens that the Very duplicates in its numerals seven of the Myer consonants, z, f, j, g, v, m, and b; moreover, there is a still more fortunate circumstance in that the character 2212 occurs in neither. This is vital, as it comes in as the "interval" separating words and groups of numbers. The three numerals not duplicated in the Myer alphabet are 1, 3 and 8, and are available for general and code calls. The assignment is as follows:
Cornet (General call), 1111
Letters call, 1112
Numeral call, 2111
Interval, 2212
The cornet is the general call. When followed by an initial letter it calls a particular ship. The "letters" call indicates that the signals which follow are to be read as a spelled-out message. The "numeral" indicates that numerals follow. We have now in the signal book G.L.U. (geographical list use), T.D.U. (telegraphic dictionary use), S.B.U. (general signal book use), etc., and these are available for code signal calls to indicate in what book to look for the significance of the signal made. In the wig-wag, the Myer signal "numerals follow" will indicate that the signal is to be read as a numeral, or if no such signal is made, as alphabetical.
This whole scheme was, by order of the Commander-in-Chief of the Squadron of Evolution, recently submitted to a board of officers of which Captain Philip, U.S.N., was senior member, and was by that board unanimously approved. Experiments in the lines laid down are contemplated this winter. Four lanterns and thirty sectors will accomplish all that five lanterns and sixty-four sectors will, if the Myer-Very code is used. In deference, however, to the expressed fear that in restricting ourselves to thirty sectors we may be tying ourselves down for unforeseen contingencies, the writer would suggest that two or four sectors be added and a pulsating current (produced either by a commutator or an automatic make and break in these sectors) be introduced if necessary for certain unforeseen code calls or added characters. It is of very great importance that we come down to the four lanterns in a hoist instead of five.
The Myer-Very code would require the thirty sectors to be marked as follows, reading from top down:
(Note one is red, and two white. The corresponding Ardois sectors are given to read from top down.)
| Myer-Very | Ardois |
A | 22 | “Numeral” |
B or 0 | 2112 | “Gen” |
C | 121 | 2 |
D | 222 | “Error” |
E | 12 | “Final” |
F or 4 | 2221 | “Annul” |
G or 6 | 2211 | “Understand” |
H | 122 | H |
I | 1 | “Interval” |
J or 5 | 1122 | T |
K | 2121 | 3 |
L | 221 | “Cornet” |
M or 9 | 1221 | I |
N | 11 | O |
O | 21 | 1 |
P | 1212 | B |
Q | 1211 | C |
R | 211 | 9 |
S | 212 | 2 |
T | 2 | 0 |
U | 112 | P |
V or 7 | 1222 | L |
W | 1121 | Q |
X | 2122 | 4 |
Y | 111 | W |
Z or 2 | 2222 | Key No. 2 |
Cornet or 1 | 1111 | “tion” |
Letters or 3 | 1112 | X |
Numerals or 8 | 2111 | Blank 45 |
Interval | 2212 | — |
With regard to the Very night code there are several great improvements admissible. In the first place, the change of the green to white is very important for uniformity, and moreover there is no reason why, with the Myer-Very code, we should not be able to use the Very signals for the alphabet for long distances instead of limiting it to the numeral code only. This can be accomplished by using the green ball for the word interval and the red and white for the elements. The importance of all this is that it enables vessels not provided with electrical apparatus to use the Myer-Very code in communicating with vessels that have. With picket launches, wooden cruisers of the old type, and torpedo-boats this is a consideration. The Very pistol should be replaced in service by double-barreled breech-loading shot-guns with short barrels.
It is never out of place to urge that signalmen be made petty officers ranking with coxswains, and that the pay be increased to $30 per month; also that quartermasters be given $35, and chief quartermasters $50.
To summarize: The service at large favors a return to the Myer code, and we must eventually adopt the Very four-element code, so why not adopt the Myer-Very code and enable us to use four lanterns instead of five in a permanent hoist? All the principal foreign services are down to four, and the Myer-Very is a step in the right direction. It leaves the question of flags vs. shapes open for future settlement; it brings order out of chaos; it cheapens the present apparatus; it gives us an alphabet for distant night signaling; and, above all, it offers a substitute for the American Morse code, than which nothing worse can be devised.