With the growth of commerce and the vast extension of telegraph and railroad systems, the necessity of a reform in the usual methods of time-reckoning has made itself forcibly felt. This has expressed itself in various attempts to adopt systems of standard time which should embrace as wide an extent of longitude as possible, and at the same time vary but little from the local time of any place using one of these systems. Thus it was a comparatively easy matter for Great Britain in 1848 and Sweden in 1879 to adopt as legal standard times, the mean solar day beginning at midnight of the meridian of their respective government observatories. The solution of the problem in the United States and Canada has been more difficult, on account of their wide extent in longitude. Yet, at a railway time convention in 1883 a system was adopted, and shortly after carried into execution, which system embodies all the essential principles necessary for uniformity in the chronology and time-reckoning of the ordinary affairs of life. It will be remembered that this plan divided the two countries into districts by meridians one hour of time apart; that the division into districts was so made that the central meridian of each is an integral number of hours west of the Greenwich meridian; and that from the central meridian of each district the standard time of the whole district is to be reckoned.
In continental Europe, with railroad, telegraph and postal routes extending through different countries separated by purely artificial boundaries, the inconvenience has been even greater than in the United States; but national pride and jealousy have proved serious obstacles to systematic reform. Chief among these has been the necessity, in any thorough reform, of the selection of an initial meridian which would be universally recognized.
Out of the discussion which these attempts have involved, has developed the idea of a universal day. Such a day would begin, for the whole world, at a given instant of time at the initial meridian; but by this scheme the various epochs of the day would begin one hour later for each hour of longitude to the west, and there has been no serious attempt to adapt it to the practical affairs of everyday life. The railway time system of the United States is only a simple modification of this idea. With standards of time carried out for the whole world on this system, the process of finding the universal date and time would be a very simple one, and the variation of local date and time reduced to a simple rule.
The necessity of a universal prime meridian, met in all the proposed plans for securing uniformity in time-reckoning, has always been urged by seafaring men for a different purpose. It is not necessary to more than refer to the utility and convenience to navigators of the adoption of a universal zero point from which to reckon longitude. The need of this has been a prominent factor in the success of the movement towards uniformity in the methods of time reckoning.
The various commercial and scientific societies of Europe, in which these questions have been discussed, have from time to time suggested the United States as the most favorable nation, for obvious reasons, for taking the initiatory steps in calling an international conference for selecting this prime meridian. The subject, in this country, had been so generally discussed and the utility of the proposed reforms so generally recognized, that Congress, in 1882, passed an act authorizing the President to call an international conference for the purpose of carrying the project into execution. After the arrangement of the necessary diplomatic preliminaries, the Conference met in Washington, in October, 1884. All its acts are of special interest to nautical men, one of them proposing no less radical a reform than the abolition of the astronomical day.
The second resolution reads as follows: "That the Conference proposes to the governments here represented the adoption of the meridian passing through the center of the transit instrument at the Greenwich Observatory, as the initial meridian for longitude." The vote of San Domingo was the only negative vote; the delegates from France and Brazil abstained from voting.
The third resolution was, "That from this meridian, longitude shall be counted in two directions up to 180 degrees, east longitude being plus and west longitude minus." This reversal of the uniform custom of navigators in regard to the sign for the application of longitude arose from its application in reducing universal time to local time; it is of little importance to navigators, as the precepts for its application to local or Greenwich time are too simple to cause any confusion as to the meaning of the resolution.
The third and fourth resolutions were in regard to the universal day, defining it as beginning at the moment of mean midnight of the initial meridian, coinciding with the beginning of the civil day and date of that meridian; the hours to be counted successively from 0 to 24.
In the sixth resolution, "The Conference expresses the hope that as soon as may be practicable the astronomical and nautical day will be arranged everywhere to begin at mean midnight."
This resolution, though anticipated, has met with strong opposition from the majority of eminent astronomers, particularly Europe. It is not difficult to see why they should oppose any innovation of the long-established and natural method of counting astronomical time. No one has so forcibly presented the objections from an astronomical point of view as Professor Newcomb; I therefore quote at length his letter to Commodore Franklin, protesting against the adoption of the new system, January 1, 1885.
"The first of these recommendations proposes a change in the method of counting astronomical time which has come down to us from antiquity. The practice of taking noon as the moment from which to count the hours originated with Ptolemy. This practice is not, as some distinguished members of the Conference seem to have supposed, based solely upon the inconvenience to the astronomer of changing his day at midnight, but was adopted because it was the most natural method of measuring solar time. At any one place solar time is measured by the motion of the sun and is expressed by the sun's hour angle. By uniform custom hour angles are reckoned from the meridian of the place, and thus by a natural process the solar day is counted from the moment the sun passes over the meridian of the place, or over the standard meridian. For the same reason the sidereal day is counted from the moment the vernal equinox passes over the meridian of the place, and thus the two times correspond to the relation between the sun and the equinox.
"It would appear that the Conference adopted the recommendation under the impression that the change would involve nothing more than the current method of reckoning time among astronomers, and could therefore be adopted without serious inconvenience…
"A change in the system of reckoning astronomical time is not merely a change of habit, such as a new method of counting time in civil life would be, but a change in the whole literature and teaching of the subject. The existing system permeates all the ephemerides and observations which fill the library of the astronomer. All his text-books, his teachings, his tables, his formulae, and his habits of calculation are based on this system. To change this system will involve a change in many of the precepts and methods laid down in his text-books.
"But this would only be the beginning of the confusion. Astronomical observations and ephemerides are made and printed not only for the present time, but for future generations and centuries. If the system is changed as proposed, the astronomers of future generations, who refer to these publications must bear this change in mind in: order not to misinterpret the data before them. The case will be yet worse if the change is not made by all the ephemerides and astronomers at the same time epoch. It will then be necessary for the astronomers of the twentieth century, using observations and ephemerides of the present, to know, remember, and have constantly in mind a certain date different in each case at which the change was adopted…
"It is difficult to present to others than astronomers who have used the published observations, the confusion, embarrassments and mistakes that will arise to their successors from the change. The case can be illustrated by saying that it is of the same kind—though in less degree than—the confusion that would arise to readers and historians in the future, if we should reverse or alter the meaning of a number of words in our language, with a result that the reader would not know what the words meant unless he noticed at what date the book was printed…
"The change will affect the navigator as well as the astronomer. Whether the navigator should commence his day at noon or midnight, it is certain that he must determine his latitude from the sun at noon. The present system of counting the day from noon enables him to do this in a simple manner, since he changes his own noon, into the astronomical period by the simple addition or subtraction of his longitude. To introduce any change whatever in the habits of computation of uneducated men is a slow and difficult matter, and is the more difficult when a complex system is to be substituted for a simple one. I am decidedly of the opinion that any attempt to change the form "of printing astronomical ephemerides for the use of our navigators would meet with objections so strong that they could not be practically overcome."
The objections from an authority of such eminence in astronomical matters seem to be shared by the majority of astronomers in Europe; but it should be noted that these objections are raised only in so far as they affect astronomers and their work. At the last meeting of the Astronomische Gesellschaft (at Geneva, in August, 1885) the discussion of the 6th resolution of the Conference was limited by the emphatic declaration of the President to its consideration from a purely astronomical standpoint; thus limited, although no formal resolution of the society was adopted, a large majority expressed themselves as opposed to the change. Those speaking in favor of the change were, however, among the most eminent astronomers of the society; they recognized the difficulties it would entail to astronomers, but were willing to make the sacrifice for the sake of gaining uniformity in methods of reckoning time; to these may be added Professor J. C. Adams, of Cambridge, Professor Christie, Astronomer Royal of Great Britain, as well as Professor Oppolzer, of Germany. The last-named proposes to give practical effect to his views by adopting the new reckoning in an extensive list of solar and lunar eclipses which he is now preparing for publication. Professor Adams pointed out in the proceedings of the Conference that there were noted exceptions even to the universality of the old method; such, for instance, as Delambre's Tables of the Sun, Burg's, Burkhart's and Damoiseau's Tables of the Moon; Bouvard's Tables of Jupiter, Saturn and Uranus; in all of which mean midnight is used as the epoch of the tables. Also La Place in his Mecanique Celeste uses Paris mean midnight as the origin of the astronomical day.
American astronomers have been as unanimous in favor of the change as those of Europe against it. Shortly after the conference, Commodore Franklin, Superintendent of the Naval Observatory, sent out a circular letter soliciting expressions of opinion on the subject. So far as I have been able to learn, with the exception of Professor Newcomb's letter above quoted, favorable answers were returned. Most of them, however, and this may be said of all who favor the change, advise waiting until a certain date can be fixed upon by international agreement. Nautical almanacs are published, or in course of publication, up to 1890; and the change could not well be carried into effect before that date. Professor Tietjen, who directs the publication of the Berliner Jahrbuch, has said that in his opinion such a change would not find place there before 1900. Thus it will be seen that astronomers are about evenly divided as to numbers, if not as to ability.
It may be mentioned here that the change was formally adopted at the Greenwich Observatory, January 1, 1885, thus initiating the confusion liable to arise from the selection of different dates for the inauguration of the change, the dangers of which are so forcibly pointed out in Professor Newcomb's letter. At the same time the adoption of this action by so prominent an observatory will tend to compel its ultimate adoption by all.
An examination of the opinions of astronomers will lead, I think, to the conclusion that those astronomers who are chiefly engaged in combining and discussing the vast mass of observations from various sources and widely different times, are generally opposed to any change; while those who perform the practical work of making the observations are as unanimous in its favor. To the latter the change would be a simple one, and give rise to little or no inconvenience. To the former the inconveniences introduced and the liability to error would be felt for a long time. It would seem, though, that they to whom the change might prove burdensome, are fitted by their skill, education and training to avoid the mistakes to which there would be liability.
However, it is not the purpose of this article to discuss the opinions of astronomers, or the effect such change would have on purely astronomical work, but rather to call the attention of naval officers and nautical men generally to the effect it would have on the various problems of the navigator; whether it will introduce, as Professor Newcomb says, a complicated system for a simple one, and consequently cause liability to errors, or whether it will be a gain in simplicity by avoiding the use of the two dates aboard ship, one for the log and one for the navigator. These are questions which the education and experience of naval officers ought to fit them to discuss. If the discussion leads to a general expression of opinion by intelligent navigators that the change will introduce more simple and direct methods, then the opinions of the uneducated navigator, whose opposition Professor Newcomb predicts, ought not to stand in the way of its adoption. It is to be expected that they would object to any change in their habits of computation, no matter for what ultimate gain; if, on the contrary, the change introduces methods which, although simple to educated men, would perplex and confuse those of little mathematical education, these objections ought to be respected.
At the first glance it would seem to be self-evident that the use of two dates to represent the same instant of time, and the necessity of reducing the civil date of the ship's log to the astronomical date of the Nautical Almanac, is not a simple system, and would be attended by liability to error. I think it will be found that navigators of considerable experience have at times made mistakes in taking data from the almanac, by confusing the two dates. Yet, on the other hand, in nearly all the methods of finding a ship's position at sea, or the chronometer error by observations on shore, the hour angle of a celestial object from the upper meridian is either directly the result of the computation, or is used in the computation for finding some other required quantity. Under the proposed system the hour angles obtained would have to be reduced to the lower meridian, or from the lower to the upper meridian to be used in the computation. Whether this reduction would be attended by as much inconvenience and liability to error as the simple process of changing the date, ought to be discussed. It should be considered what changes in the precepts and rules contained in all works on nautical astronomy, and in the tables employed, will be necessary to make perfectly plain to the thumb-rule class the transition to the new system.
A general idea of this can be obtained by an examination in detail of the various problems of navigation. By far the most frequently used are those involving observations of the sun: 1st, for latitude by meridian altitude; 2d, for longitude by time sight; 3d, for latitude by off-meridian sights; and 4th, for latitude by circum-meridian altitudes. Observations of the moon, planets and stars, for latitude or longitude, though comparatively infrequent, are generally used only on those times when observation of the sun has been impossible and it is important to find an approximate position of the ship. On such an occasion, a serious error due to the changed methods might be disastrous. It will first be noticed as a factor in every problem that the Greenwich date and time would be indicated simply by the application of the longitude to local date and time; that the latter is the civil date and time, except that the P. M. hour must be increased by twelve hours. Under the present method, the civil date and time are first reduced to local astronomical date and time, by adding twelve hours to the A. M. time, and decreasing the number indicating the civil date by a unit, for forenoon time; for P. M. time the date and time are shown immediately by the civil date and time. The change substitutes the simple rule that there is but one date, and that is the civil or local date.
The tabulated data for the first problem, finding the latitude by a meridian altitude of the sun, are given on page I of the Nautical Almanac, for Greenwich apparent noon; for obvious reasons, this could be advantageously left as it is, the navigator bearing in mind only the fact of the coincidence of civil and astronomical time. The use of page I is confined almost to this one problem, and the data when used for other purposes only required approximately. In all other cases, and perhaps in this, the data would be tabulated, in accordance with the new system, for Greenwich mean midnight, or for times reckoned from that instant.
In the second problem, the data would be taken from page II of the Nautical Almanac and corrected for the G. M. time of observation; the hour angle would be found from the usual tables; the only change would be in the addition of twelve hours to the hour angle resulting from an afternoon sight, instead of that from the forenoon sight.
In the third problem, the hour angle of the sun would be required for the computation; this would be found for an A. M. sight, by subtracting the L. A. time from twelve hours; for a P.M. sight, by subtracting twelve hours from the L. A. time. Here again a simple change in an old precept is made.
The hour angle required in the fourth problem, to reduce the altitude of the sun to meridian altitude, would be found as before, by comparing the Greenwich time of observation as shown by the chronometer with the Greenwich mean time of apparent noon. The only difference being that noon occurs at 12 hours instead of at 0 hours, as before.
So far as these problems are concerned the changes are very simple; in those involving observations of the moon, planets or stars, the effects of the changes are not so obvious, involving, as they do, changes of hour angle into sidereal time, and the reverse. In time sights of these bodies, the data required from the Nautical Almanac would be found and corrected for the Greenwich mean time of the observation, as before. The resulting hour angle simply designated (in hours, minutes and seconds) as east or west of the meridian, would be reduced to the lower meridian by subtracting it from or adding it to 12 hours, according as the observation was east or west of the meridian. This hour angle would be converted into local sidereal time and this again into local mean time, by the usual formulae.
In all other problems in which the hour angle of these bodies is required for the computation, it would be found, as before, by subtracting the right ascension of the body from the local sidereal time; this hour angle, though, is referred to the lower meridian; it would be reduced to upper meridian by subtracting it from 12 hours for an observation east of the meridian, by subtracting 12 hours from the hour angle for an observation west of the meridian.
The following precepts would be found sufficient to make plain the use of the Nautical Almanac and the necessary tables with the origin of the astronomical day at midnight:
I. The astronomical day begins at mean midnight, and coincides with the civil day and date; the hours are counted successively, beginning at midnight from 0 to 24.
II. All data of the Almanac are given for the astronomical date and time as defined in I.
III. The local astronomical time is given directly by the civil time in the forenoon; in the afternoon, by the addition of 12 hours to the civil time.
IV. All hour angles resulting from observations of the moon, planets or stars, expressed simply as east or west hour angle, are to be reduced to the lower meridian by subtracting the east hour angle from 12 hours, or by adding 12 hours to the west hour angle.
V. Hour angles thus reduced are to be reduced to local sidereal time by the addition to the hour angle of the right ascension of the body for the Greenwich mean time of the observation.
VI. Local sidereal time is reduced to local mean time, as follows: Subtract from the local sidereal time the right ascension of the mean sun for Greenwich mean midnight of the given date plus the correction for the Greenwich mean time; or subtract from the local sidereal time the right ascension of the mean sun for Greenwich mean midnight corrected for longitude; the resulting sidereal interval is corrected as usual by Table II of the American Ephemeris, or any other table for converting sidereal into mean time interval.
VII. To reduce local mean time to sidereal time, add to the local mean time the right ascension of the mean sun for Greenwich mean midnight of the given date, corrected for the Greenwich mean time.
VIII. To find the hour angle of a body, for use in computation:
- For the sun: for a forenoon sight, subtract the local apparent time of observation from 1 2 hours; for an afternoon sight, subtract the local apparent time from 12 hours.
- For the moon, planet or star: find the local sidereal time by VII; from the local sidereal time subtract the right ascension of the body observed for the Greenwich mean time of the observation; the result is the hour angle from the lower meridian. For an observation east of the meridian, subtract this result from 1 2 hours: for an observation west of the meridian, subtract 12 hours from the result.
It will be readily seen that they affect the whole literature of nautical astronomy. The changes, taken by themselves, are not difficult to understand; yet any one using the existing text-books on the subject would be obliged to make in them the changes necessary to adapt them to the different origin of reckoning time; a difficult thing for one to whom the subject is a new one. Or, studying them as they now are, he would be obliged to change important precepts which had been learned with difficulty. To rightly estimate how difficult this would be, one must look at it not from the standpoint of the skilled navigator, but as one to whom the subject of nautical astronomy is full of perplexities.
It is easy for one who has a thorough knowledge of the theory of the subject to make the necessary precepts for himself, or to readily see the bearing of new ones given for the use of tables under the new system; but would the majority of navigators come under this head?
It may seem trivial and unnecessary to examine in detail all the problems of nautical astronomy to see what effect general changes would produce; yet it serves to show that the meridian of the place is the natural origin to which are referred the various quantities used in or derived from the computations; by this the problems are simple and direct. It is for this reason that the nautical or astronomical day is the most convenient; that it is made to begin at noon. To change its beginning to midnight is only an apparent gain in uniformity. It is not designed for the purpose of chronology, and is not so used; and there is no reason why it should coincide with the day used for that purpose. The civil day is used aboard ship in all cases when it is most suitable; the nautical day only by the navigator in those problems of navigation where it simplifies astronomical calculation. As such, its retention would not conflict with the purposes of those who aim to secure uniformity for chronological and commercial purposes.