Each tap of the observer's finger completed for an instant an electric circuit, and recorded a mark on the 'chronograph.' This is a large metal cylinder covered with paper, and turned by a carefully-regulated clock once in every two minutes. Once in every two seconds a similar mark was made by a current sent by means of the standard sidereal clock of the Observatory. The paper cover of the chronograph after an hour's work shows a spiral trace of little dots encircling it some thirty times. These dots are at regular intervals, about an inch apart, and are the marks made by the clock. Interspersed between them are certain other dots, in sets of ten; and these are the signals sent from the telescope by the transit observer. If, then, one of the clock dots and one of the observer's dots come exactly side by side, we know that the star was on one of the wires at a given precise second. If the observer's dot comes between two clock dots, it is easy, by measuring its distance from them with a divided scale, to tell the instant the star was on the wire to the tenth of a second, or even to a smaller fraction. Whilst, since the transit was taken over ten wires, and the distance of each wire from the centre of the field of view is known, we have practically ten separate observations, and the average of these will give a much better determination of the time of transit than a single one would.

But let the watcher be ever so little too slow in setting his telescope, or ever so little late in placing himself at his eye-piece, and the star will have passed the wire, and as it smoothly, resistlessly moves on its inexorable way, will tell the tardy watcher in a language there is no mistaking, 'Lost moments can never be recalled.' The opportunity let slip, not until twenty-four hours have gone by will another chance come of observing that same star.

It is the stars that are chiefly used in this determination, partly because the stars are so many, whilst there is but one sun. If, therefore, clouds cover the sun at the important moment of transit, the astronomer may well exclaim, so far as this observation is concerned, 'I have lost a day!' The chance will not be offered him again until the following noon. But if one star is lost by cloud, there are many others, and the chance is by no means utterly gone. Beside, the sun enables us to tell the time only at noon; the stars enable us to find it at various times throughout the entire night; indeed, throughout both day and night, since the brighter stars can be observed in a large telescope even during the day.

There are two great standard clocks at the Observatory: the mean solar clock and the sidereal clock. The latter registers twenty-four hours in the precise time that the earth rotates on its axis. A 'day' in our ordinary use of the term is somewhat longer than this; it is the average time from one noon to the next, and as the earth whilst turning round on its axis is also travelling round the sun, it has to rather more than complete a rotation in order to bring the sun again on to the same meridian. A solar day is therefore some four minutes longer than an actual rotation of the earth, i.e. a sidereal day, as it is called, since such rotation brings a star back again to the same meridian.

The sidereal clock can therefore be readily checked by the observation of star transits, for the time when the star ought to be on the meridian is known. If, therefore, the comparison of the transit taps on the chronograph with the taps of the sidereal clock show that the clock was not indicating this time at the instant of the transit, we know the clock must be so much fast or slow. Similarly, the difference which should be shown between the sidereal and solar clocks at any moment is known; and hence when the error of the sidereal clock is known, that of the solar can be readily found.

It is often quite sufficient to know how much a clock is wrong without actually setting its hands right; but it is not possible to treat the Greenwich clock so, for it controls a number of other clocks continually, and sends hourly signals out over the whole country, by which the clocks and watches all over the kingdom are set right.

In the lower computing room, below the south window, we find the Time-Desk, the head-quarters of the Time Department. This is a very convenient place for the department, since one of the chronometer rooms, formerly Bradley's transit room, opens out of the lower computing room; the transit instrument is just beyond; it is close to the main gate of the Observatory, and so convenient for chronometer makers or naval officers bringing chronometers or coming for them, whilst just across the courtyard is the chronograph room, with the Battery Basement, in which the batteries for the electric currents are kept, and the Mean Solar Clock lobby, with the winch for the winding of the time-ball at the head of the stairs above it. These rooms do not exhaust the territory of the department, since it owns two other chronometer rooms on the ground floor and first floor respectively of the S.-E. tower.

At the time-desk means are provided for setting the clock right very easily and exactly. Just above the desk are a range of little dials and bright brass knobs, that almost suggest the stops of a great organ.

Two of these little dials are clock faces, electrically connected with the solar and sidereal standard clocks, so that, though these clocks are themselves a good way off, in entirely different parts of the Observatory, the time superintendent, seated here at the time-desk, can see at once what they are indicating. Between the two is a dial labelled 'Commutator.' From this dial a little handle usually hangs vertically downwards, but it can be turned either to the right or to the left, and when thus switched hard over, an electric current is sent through to the mean solar clock. If now we leave the computing room and cross the courtyard to the extreme north-west corner, we find the Mean Solar Clock in a little lobby, carefully guarded by double doors and double windows against rapid changes of temperature. Opening the door of the clock case, we see that the pendulum carries on its side a long steel bar, and that this bar as the pendulum swings passes just over the upper end of an electro-magnet. When the current is switched on at the commutator, this electro-magnet attracts or repels the steel bar according to the direction of the current, and the action of the clock is accordingly quickened or retarded. To put the commutator in action for one minute will alter the clock by the tenth of a second. As the error of the clock is determined twice a day, shortly before ten o'clock in the morning, and shortly before one o'clock in the afternoon, its error is always small, usually only one or two tenths. These two times are chosen because, though time-signals are sent over the metropolitan area every hour from the Greenwich clock through the medium of the Post Office, at ten and at one o'clock signals are also sent to all the great provincial centres. Further, at one o'clock the time balls at Greenwich and at Deal are dropped, so that the captains of ships in the docks, on the river, or in the Downs may check their chronometers.

The Time-Ball is dropped directly by the mean solar clock itself. It is raised by means of a windlass turned by hand-power to the top of its mast just before one o'clock. Connected with it is a piston working in a stout cylinder. When the ball has reached the top of the mast, the piston is lightly supported by a pair of catches. These catches are pulled back by the hourly signal current, and the piston at once falls sharply, bringing the ball with it. But after a fall of a few feet, the air compressed by the piston acts as a cushion and checks the fall, the ball then gently and slowly finishing its descent. The instant of the beginning of the fall is, of course, the true moment to be noted.

The other dials on the time-desk are for various purposes connected with the signals. One little needle in a continual state of agitation shows that the electric current connecting the various sympathetic clocks of the Observatory is in full action. Another receives a return signal from various places after the despatch of the time-signal from Greenwich, and shows that the signal has been properly received at the distant station, whilst all the many electric wires within the Observatory or radiating from it are made to pass through the great key-board, where they can be at once tested, disconnected, or joined up, as may be required.

The distribution of Greenwich time over the island in this way is thus a simple matter. The far more important one of the distribution of Greenwich time to ships at sea is more difficult. The difficulty lay in the construction of a clock or watch, the rate of which would not be altered by the uneasy motion of a ship, or by the changes of temperature which are inevitable on a voyage. Two hundred years ago it was not deemed possible to construct a watch of anything like sufficient accuracy. They would not even keep going whilst they were being wound, and would lose or gain as much as a minute in the day for a fall or rise of 10° in temperature. This was owing to the extreme sensitiveness of the balance spring—which takes the place in a watch of a pendulum in a clock—to the effects of temperature. The British Government, therefore, in 1714 offered a prize of the amount of £20,000 for a means of finding the longitude at sea within half a degree, or, in other words, for a watch that would keep Greenwich time correct to two minutes in a voyage across the Atlantic. In 1735, James Harrison, the son of a Yorkshire carpenter, succeeded in solving the problem. His method was to attach a sort of automatic regulator to the spring which should push the regulator over in one direction as the temperature rose, and bring it back as it fell. This he effected by fastening together two strips of brass and steel. The brass expanded with heat more rapidly than the steel, and hence with a rise of temperature the strip bent over on the steel side. This was the first germ of the idea of making watches 'compensated for temperature;' watches, that is, which maintain practically the same rate whether they are in heat or cold, an idea now brought to great perfection in the modern chronometer.

A visitor who should make the attempt to compare a single chronometer with a standard clock would probably feel very disheartened when, after many minutes of comparison, he had got out its error to the nearest second, were he told that it was his duty to compare the entire army here collected, some five hundred or more, and to do it not to the second, but to the nearest tenth of a second. Practice and system make, however, the impossible easy, and one assistant will quietly walk round the room calling out the error of each chronometer as he passes it, as fast as a second assistant seated at the table can enter it at his dictation in the chronometer ledgers. The seconds beat of a clock sympathetic with the solar standard, rings out loud and clear above the insect-like chatter of the ticking of the hundreds of chronometers, and wherever the assistant stands, he has but to lift his eyes to see straight before him, if not a complete clock-face, at least a seconds dial moving in exact accordance with the solar standard.

The test to which chronometers are subjected is not merely one of rate, but one of rate under carefully altered conditions. Thus they may be tried with the XII pointing in succession to the four points of the compass, or, in the case of chronometer watches, they may be laid flat down on the table or hung from the ring or pendant, or with the ring right or left, as it would be likely to be when carried in the waistcoat pocket. But the chief test is the performance of a chronometer when subjected to considerable heat for a long period. This is a matter of great consequence, since a chronometer travelling from England to India, Australia, or the Cape, would necessarily be subjected to very different conditions of temperature from those to which it would be exposed in England. They are therefore kept for eight weeks in a closed stove at a temperature of about 85° or 90°. At one time a cold test was also applied, and Sir George Airy, the late Astronomer Royal, in one of his popular lectures, drew a humorous comparison between the unhappy chronometers thus doomed to trial, now in heat and now in frost, and the lost spirits whom Dante describes as alternately plunged in flame and ice. The cold test has, however, been done away with. It is perfectly easy on the modern ship to keep the chronometer comfortably warm even on an Arctic expedition. The elaborate cold testing applied to Sir George Nares' chronometers before he started on his polar journey was found to have been practically quite superfluous; the chronometers were, if anything, kept rather too warm. The exposure of the chronometer in the cooling box, moreover, was found to be attended with a risk of rusting its springs.

Once the determination of the longitude at sea became possible, it was clearly the next duty to fix with precision the position of the principal places, cities, ports, capes, islands, the world over. Of all the work done in this department none has ever been done better, in proportion to the means at command, than that accomplished by Captain Cook in his celebrated three voyages. As has already been pointed out, it is the extent and thoroughness of the hydrographic surveys of the British Admiralty which have largely contributed to the honour done to England by the international selection of the English meridian, and of English standard time, as in principle those for the whole civilized world. The generosity and public spirit therefore which led the second Astronomer Royal to help forward and support his rival, has almost directly led to this great distinction accruing to the Observatory of which he was the head.

Three different methods have successively been used in the determination of longitudes of distant places. In each case the problem required was to ascertain the time at the standard place, say Greenwich, at the same time that it was being determined in the ordinary way at the given station. One method of ascertaining Greenwich time when at a distance from it was, as stated in Chapter I., to use the moon, as it were, as the hand of a vast clock, of which the sky was the face and the stars the dial figures. This is the method of 'lunar distances,' the distances of the moon from a certain number of bright stars being given in the Nautical Almanac for every three hours of Greenwich time.

As chronometers were brought to a greater point of perfection, it was found easier and better in many cases to use 'chronometer runs,' that is, to carry backwards and forwards between the two stations a number of good chronometers, and by constant comparison and re-comparison to get over the errors which might attach to any one of them.

But of late years another method has proved available. Distant nations are now woven together across thousands of miles of ocean by the submarine telegraph. The American reads in his morning paper a summary of the debates of the previous night in the House of Commons at Westminster. The Londoner watches with interest the scores of the English cricket team in Australia. It is now therefore possible for an astronomer in England to record, should he so desire, the time of the transit of a star across the wires of his instrument, not only on his own chronograph, but upon that of another observatory, it may be 2000 miles away. Or, much more conveniently, each observer may independently determine the error of his own clock, and then bring his clock into the current, so that it may send a signal to the chronograph of the other station.

In one way or another this work of the determination of geographical longitudes has been an important part of the extra-routine work at Greenwich, part of the work which has built up and sustained its claim to define 'longitude nought'; and many distinguished astronomers, especially from the leading observatories of the Continent, have come here from time to time to obtain more accurately the longitude of their own cities. The traces of their visits may be seen here and there about the Observatory grounds in flat stones which lie level with the surface, and bear a name and date like the gravestones in some old country churchyard. These are not, as one might suppose, to mark the burial-places of deceased astronomers, but record the sites where, on their visits for longitude purposes, different foreign astronomers have set up their transit instruments. Now, however, a permanent pier has been erected in the courtyard, and a neat house—the Transit Pavilion—built over it, so that in all probability no fresh additions will be made to these sepulchral-looking little monuments.

It might be asked, What reason is there for a foreign observer to come over to England for such a purpose? Would it not be sufficient for the clock signals to be exchanged? But a curious little fact has come out with the increase of accuracy of transit observation, and that is, that each observer has his own particular habit or method of observation. A hundred years ago, Maskelyne, the fifth Astronomer Royal, was greatly disturbed to find that his assistant, David Kinnebrook, constantly and regularly observed a star-transit a little later than he did himself. The offender was scolded, warned, exhorted, and finally, when all proved useless to bring his observations into exact agreement with the Astronomer Royal's, dismissed as an incompetent observer. As a matter of fact, poor Kinnebrook has a right to be regarded as one of the martyrs of science, and Maskelyne, by this most natural but mistaken judgment, missed the chance of making an important discovery, which was not made until some thirty years later. Astronomers now would be more cautious of concluding that observations were bad simply because they differed from what had been expected. They have learnt by experience that these unexpected differences are the most likely hunting-ground in which to look for new discoveries.

In a modern transit observation with the use of the chronograph it will be seen at once that before the observer can register a star-transit on the chronograph, he has to perceive with his eye that the star has reached the wire, he has to mentally recognize the fact, and consciously or unconsciously to exert the effort of will necessary to bring his finger down on the button. A very slight knowledge of character will show that this will require different periods of time for different people. It will be but a fraction of a second in any case, but there will be a distinct difference, a constant difference, between the eager, quick, impulsive man who habitually anticipates, as it were, the instant when he sees star and wire together, and the phlegmatic, slow-and-sure man who carefully waits till he is quite sure that the contact has taken place, and then deliberately and firmly records it. These differences are so truly personal to the observer that it is quite possible to correct for them, and after a given observer's habit has become known, to reduce his transit times to those of some standard observer. It must, of course, be remembered that this 'personal equation' is an exceedingly minute quantity, and in most cases is rather a question of hundredths of seconds than of tenths.

It will be seen from the foregoing description how little of what may be termed the picturesque or sensational side of astronomy enters into the routine of the Time Department, the most important of all the departments of the Observatory. The daily observation of sun and of many stars—selected from a carefully chosen list of some hundreds, and known as 'clock stars'—the determination of the error of the standard clock to the hundredth of a second if possible, and its correction twice a day, the sending out of time signals to the General Post Office and other places, whence they are distributed all over the country; the care, winding, and rating of hundreds of chronometers and chronometer watches, and from time to time the determination of the longitude of foreign or colonial cities, make up a heavy, ceaseless routine in which there is little opportunity for the realization of an astronomer's life as it is apt to be popularly conceived.

Yet there is interest enough in the work. There is the charm which always attaches to work of precision, the delight of using delicate and exact instruments, and of obtaining results of steadily increasing perfection. It may be akin to the sporting passion for record-breaking, but surely it is a noble form of it which has led the assistants, in recent years, to steadily increase the number of observations in a normal night's work up to the very limit, taking care the while that their accuracy has in no degree suffered. In longitude work also 'the better is the enemy of the good,' and there is the ambition that each fresh determination shall be markedly more precise than all that have preceded it. The constant care of chronometers soon reveals a kind of individuality in them which forms a fresh source of interest, whilst if a man has but a spark of imagination, how easily he will wrap them round with a halo of romance!

Glance through the ledgers, and you will see how some of them have heard the guns at the siege of Alexandria, others have been carried far into the frozen north, others have wandered with Livingstone or Cameron in the trackless forests of equatorial Africa.

More striking still are those pages across which the closing line has been drawn; never again will the time-keeper there scheduled return to the kindly inquisition of Flamsteed Hill. This sailed away in the Wasp, and was swallowed up in the eastern typhoon; that went down in the sudden squall that smote the Eurydice off the Isle of Wight; these foundered with the Captain. The last fatal journey of Sir John Franklin to find the North-West Passage leaves its record here; the chronometers of the Erebus and Terror will never again appear on the Greenwich muster roll. Land exploration claims its victims too. Sturt's ill-fated expedition across Australia, and Livingstone's last wandering, are represented.

Sometimes an amusing entry interrupts the silent pathos of these closed pages. 'Lost by Mr. Smith on the coast of Africa,' reads at first sight like a rather thin attempt of some one to shift the responsibility of his own carelessness on to the broad shoulders of Mr. Nobody. In reality it probably gives a hint of the necessary, dangerous, and exciting work of slave-dhow chasing which gives employment to our ships on the African coast. 'Mr. Smith' was no doubt a petty officer who was told off to carry the chronometer for a boat's crew sent to search for a slave-dhow up some equatorial estuary. Probably the dhow was found, and the Arabs who manned it gave so stout a resistance that 'Mr. Smith' and his men had other things to do than take care of chronometers before they could overcome them. We may take it that the real story outlined here was one of courage and hard fighting, not of carelessness and shirking.

Stories of higher valour and nobler courage yet are also hinted: the calm discipline of the crew of the Victoria as she sank from the ram of the Camperdown, the yet nobler devotion of the men of the Birkenhead, as they formed up in line on deck and cheered the boats that bore away the women and children to safety, whilst they themselves went down with the ship into the shark-crowded sea.

CHAPTER VII

THE TRANSIT AND CIRCLE DEPARTMENTS

The determination of time is a duty the importance of which readily commends itself to the general public. It is easy to see that in any civilized country it is very necessary to have an accurate standard of time. Our railways and telegraphs make it quite impossible for us to be content with the rough-and-ready sun-dial which satisfied our forefathers. But it should be remembered that it was neither to establish a 'longitude nought,' nor to create a system of standard time, that Greenwich Observatory was founded in 1675. It was for 'The Rectifying the Tables of the Motions of the Heavens and the Places of the Fixed Stars, in order to find out the so-much-desired Longitude at Sea for the perfecting the Art of Navigation.'

The two related departments, therefore, those of the Transit and the Circle, which are concerned in the work of making star-catalogues, come next in order to the Time Department. Though both departments deal with the same instrument, the transit circle, they are at present placed at opposite ends of the Observatory domain; the Circle Department being lodged in the upper computing room of the old building; the Transit Department in the south wing of the New Observatory in the south ground.

It may be asked why, if this were the purpose of the Observatory at its foundation, two and a quarter centuries ago; if, as was the case, the work was set on foot from the beginning and was carried out with every possible care, how comes it that it is still the fundamental work of the Observatory, and, instead of being completed, has assumed greater proportions at the present day than ever before?

The answer to this inquiry may be found in a special application of the old proverb, originally directed against the discontent of man: 'The more he has, the more he wants.' For, however paradoxical it may seem, it is true that the fuller a star-catalogue is, and the more accurate the places of the stars that it contains, the greater is the need for a yet fuller catalogue, with places more accurate still.

It is worth while following up this paradox in some detail, as it affords a very instructive example of the way in which a work started on purely utilitarian grounds extends itself till it crosses the undefined boundary and enters the region of pure science.

We have no idea who made the earliest census of the sky. It is written for us in no book; it is not even engraved on any monument. And yet no small portion of it is in our hands to-day, and, strangest of all, we are able to fix fairly closely the time at which it was made, and the latitude in which its compiler lived. The catalogue is very unlike our star-catalogues of to-day. The places of the stars are very roughly indicated; and yet this catalogue has left a more enduring mark than all those that have succeeded it. The catalogue simply consists of the star names.

An old lady who had attended a University Extension lecture on astronomy was heard to exclaim: 'What wonderful men these astronomers are! I can understand how they can find out how far off the stars are, how big they are, and what they weigh—that is all easy enough; and I think I can see how they find out what they are made of. But there is one thing that I can't understand—I don't know how they can find out what are their names!' This same difficulty, though with a much deeper meaning than the old lady in her simplicity was able to grasp, has occurred to many students of astronomy. Many have wished to know what was the meaning of, and whence were derived, the sonorous names which are found attached to all the brighter stars on our celestial globes: Adhara, Alderamin, Betelgeuse, Denebola, Schedar, Zubeneschamal, and many more. The explanation lies here. Some 5000 years ago, a man, or college of men, living in latitude 40° north, in order that they might better remember the stars, associated certain groups of them with certain fancied figures, and the individual star names are simply Arabic words designed to indicate whereabouts in its peculiar figure or constellation that special star was situated. Thus Adhara means 'back,' and is the name of the bright star in the back of the great Dog. Alderamin means 'right arm,' and is the brightest star in the right arm of Cepheus, the king. Betelgeuse is 'giant's shoulder,' the giant being Orion; Denebola is 'lion's tail.' Schedar is the star on the 'breast' of Cassiopeia, and Zubeneschamal is 'northern claw,' that is, of the Scorpion. So far is clear enough. The names of the stars for the most part explain themselves; but whence the constellations derived their names, how it was that so many snakes and fishes and centaurs were pictured out in the sky, is a much more difficult problem, and one which does not concern us here.

One point, however, these old constellations do tell us, and tell us plainly. They show that the axis of the earth, which, as the earth travels round the sun, moves parallel with itself, yet, in the course of ages, itself rotates so as in a period of some 26,000 years to trace out a circle amongst the stars. This is the cause of what is called 'precession,' and explains how it is that the star we call the pole-star to-day was not always the pole-star, nor will always remain so. We learn this fact from the circumstance that the old constellations do not cover the entire celestial sphere. They leave a great circular space of 40° radius unmapped in the southern heavens. This simply means that the originators of the constellations lived in 40° north latitude, and stars within 40° of their south pole never rose above their horizon, and consequently were never seen, and could not be mapped, by them. In like manner, the star census taken at Greenwich Observatory does not include the whole sky, but leaves a space some 52° in radius round our south pole. Since the latitude of Greenwich is nearly 52° north, stars within that distance of the south pole do not rise above our horizon, and are never seen here. But if we compare the vacant space left by the old original constellations with the vacant space left by a Greenwich catalogue of to-day, we see that the centre of the first space, which must have been the south pole of that time, is a long way from the centre of the second space—our south pole of to-day. The difference tells us how far the pole has moved since those old forgotten astronomers did their work. We know the rate at which the pole appears to move, by comparing our more modern catalogues one with another; and so we are able to fix pretty nearly the time when lived those old first census-takers of the stars, whose names have perished so completely, but whose work has proved so immortal.

These old workers gave us the constellation groupings and names which still remain to us, and are still in common, every-day use. Their work affords us the most striking illustration of the result of precession, but precession itself was not recognized till nearly 3000 years after their day, when a marvellous genius, Hipparchus, established the fact, and 'built himself an everlasting name' by the creation of a catalogue of over 1000 stars prepared on modern principles. That catalogue formed the basis of one which survives to us at the present time, and was made some 1750 years ago by Claudius Ptolemy, the great astronomer of Alexandria, whose work, which still bears the proud name of Almagest, 'The Greatest,' remained for fourteen centuries the one universal astronomical text-book.

A modern catalogue contains, like that of Ptolemy, four columns of entry. The first gives the star's designation; the second an indication of its brightness; the third and fourth the determinations of its place. These are expressed in two directions, which, in modern catalogues, not in Ptolemy's, correspond on the celestial sphere to longitude and latitude on the terrestrial. Distance north or south of the celestial equator is termed 'declination,' corresponding to terrestrial latitude. Distance in a direction parallel to the equator is termed 'right ascension,' corresponding to terrestrial longitude. For geographical purposes we conceive the earth to be encircled by two imaginary lines at right angles to each other—the one, the equator, marked out for us by the earth itself; the other, 'longitude nought,' the meridian of Greenwich, fixed for us by general consent, after the lapse of centuries, by a kind of historical evolution. On the celestial globe in like manner we have two fundamental lines—one, the celestial equator, marked out for us by nature; the other at right angles to it, and passing through the poles of the sky, adopted as a matter of convenience. But a difficulty at once confronts us—Where can we fix our 'right ascension nought'? What star has the right to be considered the Greenwich of the sky?

The difficulty is met in the following manner: For six months of the year, the summer months, the sun is north of the celestial equator; for the other six months of the year, the months of winter, it is south of it. It crosses the equator, therefore, twice in the year—once when moving northward at the spring equinox; once when moving southward at the equinox of autumn. The point where it crosses the equator at the first of these times is taken as the fundamental point of the heavens, and the first sign of the zodiac, Aries the Ram, is said to begin here, and it is called, therefore, 'the first point of Aries.'

One of the very first facts noticed in the very early days of astronomy was that, as the stars seemed to move across the sky night by night, they seemed to move in one solid piece, as if they were lamps rigidly fixed in one and the same solid vault. Of course it has long been perfectly understood that this apparent movement was not in the least due to any motion of the stars, but simply to the rotation of the earth on its axis. This rotation is the smoothest, most constant, and regular movement of which we know. It follows, therefore, that the interval of time between the passage of one star across the meridian of Greenwich and that of any other given star is always the same. This interval of time is simply the difference of their right ascension. If we are able, then, to turn our transit instrument to the sun, and to a number of stars, each in its proper turn, and by pressing the tapping-piece on the instrument as the sun or star comes up to each of the ten wires in succession, to record the times of its transit on the chronograph, we shall have practically determined their right ascensions—one of the two elements of their places.

The other element, that of declination, is found in a different manner. The celestial equator, like the terrestrial, is 90° from the pole. The bright star Polaris is not exactly at the north pole, but describes a small circle round it. Twice in the twenty-four hours it transits across the meridian—once when going from east to west it passes above the pole, once when going from west to east below the pole. The mean between these two altitudes of Polaris above the horizon gives the position of the true pole.

A complete transit observation of a star consists therefore of two operations. The observer, as we have already described, sees a star entering the field of the telescope, and as it swims forward, he presses the galvanic button, which sends a signal to the chronograph as the star comes up to each of the ten vertical wires in succession. But, beside the ten wires, there are others. Two vertical wires lie outside the ten of which we have already spoken, and there is also a horizontal wire. The latter can be moved by a graduated screw-head just above the eye-piece, and as the star comes in succession to these two vertical wires, this horizontal wire is moved by the screw-head, so as to meet the star at the moment it is crossing the vertical wire, and the observer presses a second little button, which records the position of the horizontal wire on a small paper-covered drum. Then, the transit over, the observer leaves the telescope and comes round to the outside of the west pier. Here he finds seven large microscopes, which pierce the whole thickness of the pier, and are directed towards the circumference of a large wheel which is rigidly attached to the telescope and revolves with it. This wheel is six feet in diameter, and has a silver circle upon both faces. Each circle is divided extremely carefully into 4320 divisions—these divisions, therefore, being about the one-twentieth of an inch apart. There are, therefore, twelve divisions to every degree (12 × 360 = 4320), and each division equals five minutes of arc. The lowest microscope is the least powerful, and shows a large part of the circle, enabling the observer to see at once to what degree and division of a degree the microscope is pointing. The other six microscopes are very carefully placed 60° apart—as equally placed as they possibly can be. These microscopes are all fitted with movable wires—wires moved by a very fine and delicate screw; the screw-head having divisions upon it so that the exact amount of its movement can be told. Each of the six screw-heads will read to the one five-thousandth part of a division of the circle; in other words, to the one hundred thousandth part of an inch. Using all six microscopes, and taking their mean, we are able to read to the one-hundredth of a second of arc. If, therefore, the observations could be made with perfect certainty down to the extremest nicety of reading which the instrument supplies, we should be able to read the declination of a star to this degree of refinement. It may be added that a halfpenny, at a distance of three miles, appears to be one second of arc in diameter; at three hundred miles it would be one-hundredth of a second. It need scarcely be said that we cannot observe with quite such refinement of exactness as this would indicate. Nevertheless, this exactness is one after which the observer is constantly striving, and tenths, even hundredths, of seconds of arc are quantities which the astronomer cannot now neglect.

The observer has then to read the heads of all these seven microscopes on the pier side, and also two positions of the horizontal wire on the screw-head at the eye-piece. The following morning he will also read off from the chronograph-sheet the times when he made the ten taps as the star passed each of the ten vertical wires. There are, therefore, nine entries to make for one position of a star in declination, and ten for one position of a star in right ascension. The observer will also have to read the barometer to get the pressure of the air at the time of observation, and one thermometer inside the transit room, and another outside, to get the temperature of the air. In some cases thermometers at different heights in the room are also read. A complete observation of a single star means, therefore, the entry of two-and-twenty different numbers.

It may be asked, What is the use of reading the barometer and thermometer? The answer to the question can only be given by contradicting a statement made above, that the true pole lay midway between the position of the telescope when pointing to the pole-star at its upper transit, and its position when pointing to it at its lower transit. The pole being very high in the heavens in this country, there are a great number of stars that, like the pole-star, cross the meridian twice in the twenty-four hours—once when they pass above the pole, moving from east to west, once when they pass below it, moving from west to east As the real distance of a star from the true pole does not alter, it follows that we ought to get the position of the pole from the mean of the two transits of any of these stars, and they ought all to exactly agree with each other. But they do not. So, too, I said that the stars all appeared to move as in a single piece. If, then, we constructed an instrument with its axis parallel to the axis of the earth, and fixed a telescope to it, pointing to any particular star, if we turn the telescope round as fast from east to west as the earth itself is turning from west to east—if we built an equatorial, that is to say—we ought to find that the star once in the centre of the field would remain there. As a matter of fact, when the star got near the horizon it would soon be a long way from the centre of the field.

Sir George Airy, the seventh Astronomer Royal, makes, with reference to this very point, the following remarks:

'Perhaps you may be surprised to hear me say the rule is established as true, and yet there is a departure from it. This is the way we go on in science, as in everything else; we have to make out that something is true, then we find out under certain circumstances that it is not quite true; and then we have to consider and find out how the departure can be explained.'

In this particular case, the disturbing cause is found in the action of our own atmosphere. The rays of light from the star are bent out of a perfectly straight course as they pass through the various layers of that atmosphere, layers which necessarily become denser the closer we get to the actual surface of the earth. Every celestial body therefore appears to be a little higher in the sky than it really is. This action is most noticeable at the horizon, where it amounts to about half a degree. As both sun and moon are about half a degree in diameter, it follows that when they have really just entirely sunk below the horizon they appear to be just entirely above it. It happens, in consequence, on rare occasions, that an eclipse of the moon will take place when both sun and moon are together seen above the horizon.

It was a great matter to discover this effect of refraction. It was soon seen that it was not constant, that it varied with both temperature and pressure. It is, indeed, the most troublesome of all the hindrances to exact observation with which the astronomer has to contend; partly because of its large amount—half a degree, as has been already said, in the extreme case—and partly because it is difficult in many cases to determine its exact effect.

The double observation with the transit circle gives us, then, the place in the sky where the star appeared to be at the moment of observation, not its true place; to find that true place we have to calculate how much refraction had displaced the star at the particular height in the sky, and at the particular temperature and atmospheric pressure at which the observation was made.