Astronomical Discovery Part 8

ACCIDENTAL DISCOVERIES

[Sidenote: The Oxford new star found during work on Astrographic Chart.]

In reviewing various types of astronomical discovery I have laid some stress upon the fact that they are, generally speaking, far from being accidental in character. A new planet does not "swim into our ken," at any rate not usually, but is found only after diligent search, and then only by an investigator of acute vision, or other special qualifications. But this is, of course, not always the case. Some discoveries are made by the merest accident, as we have had occasion to remark incidentally in the case of the minor planets; and for the sake of completeness it is desirable to include among our types at least one case of such accidental discovery. As, however, the selection is a little invidious, I may perhaps be pardoned for taking the instance from my own experience, which happens to include a case where one of those remarkable objects called "new stars"

walked deliberately into a net spread for totally different objects. There is the further reason for choosing this instance: that it will afford me the opportunity of saying something about the special research in which we were actually engaged, the work of mapping out the heavens by photography, or, as it has been called, the Astrographic Chart--a great scheme of international co-operation by which it is hoped to leave as a legacy for future centuries a record of the state of the sky in our age.

Such a record cannot be complete; for however faint the stars included, we know that there are fainter stars which might have been included had we given longer exposures to the plates. Nor can it be in other ways final or perfect; however large the scale, for instance, on which the map is made, we can imagine the scale doubled or increased many-fold. But the map will be a great advance on anything that has. .h.i.therto been made, and some account of it will therefore no doubt be of interest.

[Sidenote: Origin of the chart.]

We may perhaps begin with a brief historical account of the enterprise.

Photographs of the stars were taken many years ago, but only by a few enthusiasts, and with no serious hope of competing with eye observations of the sky. The old wet-plate photography was, in fact, somewhat unsuited to astronomical purposes; to photograph faint objects a long exposure is necessary, and the wet plate may dry up before the exposure is concluded--nay, even before it is commenced, if the observer has to wait for pa.s.sing clouds--and therefore it may be said that the successful application of photography to astronomy dates from the time when the dry plate was invented; when it became possible to expose a plate in the telescope for hours, or by acc.u.mulation even for days. The dry plate remains sensitive for a long period, and if it is desired to extend an exposure beyond the limits of one night, it is quite easy to close up the telescope and return to the operations again on the next fine night; and so on, if not perhaps indefinitely, at any rate so long as to transcend the limits of human patience up to the present.

[Ill.u.s.tration:

VII.--GREAT COMET OF NOV. 7TH, 1882 (_From a photograph taken at the Royal Observatory, Cape of Good Hope._)]

[Sidenote: Comet of 1882.]

[Sidenote: Stars shown on the pictures.]

But to consider our particular project. We may a.s.sign, perhaps, the date 1882 as that in which it first began to take shape. In that year there was a magnificent bright comet, the last really large comet which we, in the Northern Hemisphere, have had the good fortune to see. Some of us, of course, were not born at that time, and perhaps others who were alive may nevertheless not have seen that comet; for it kept somewhat uncomfortably early morning hours, and I can well remember myself feeling rather more resentment than grat.i.tude to the man who waked me up about four o'clock to see it. Many observations were of course made of this interesting visitor, and what specially concerns us is that at the Cape of Good Hope some enterprising photographers tried to photograph it. They tried in the first instance with ordinary cameras, and soon found--what any astronomer could have told them--that the movement of the earth, causing an apparent movement of the comet and the stars in the opposite direction, frustrated their efforts. The difficulties of obtaining pictures of moving objects are familiar to all photographers. A "snap-shot" might have met the difficulty, but the comet was scarcely bright enough to affect the plate with a short exposure. Ultimately Dr. David Gill, the astronomer at the Cape Observatory, invited one of the photographers to strap his camera to one of the telescopes at the Observatory, a telescope which could be carried round by clockwork in the usual way, so as to counteract the earth's motion, and in effect to keep the comet steadily in view, as though it were at rest. As a consequence, some very beautiful and successful pictures of the comet were obtained, and on them a large number of stars were also shown. They were, as I have said, not by any means the first pictures of stars obtained by photography, but they represented in facility and in success so great an advance upon what had been formerly obtained that they attracted considerable attention. They were sent to Europe and stimulated various workers to further experiments.

[Sidenote: The brothers Henry begin work.]

[Sidenote: Conference of 1887.]

The late Dr. Common in England, an amateur astronomer, began that magnificent pioneer work in astronomical photography which soon brought him the Gold Medal of the Royal Astronomical Society for his photographs of nebulae. But the most important result for our purpose was produced in France. There had been started many years before by the French astronomer Chacornac a series of star maps round the Zodiac similar in intention to the Berlin maps which figured in the history of the discovery of Neptune.

Chacornac died before his enterprise was very far advanced, and the work was taken up by two brothers, Paul and Prosper Henry, who followed Chacornac in adopting for the work the laborious method of eye observation of each individual star. They proceeded patiently with the work on these lines; but when they came to the region where the Zodiac is crossed by the Milky Way, and the number of stars in a given area increases enormously, they found the labour so great as to be practically prohibitive, and were in doubt how to deal with the difficulty. It was at this critical moment that these comet photographs, showing the stars so beautifully, suggested the alternative of mapping the stars photographically. They immediately set to work with a trial lens, and obtained such encouraging results that they proceeded themselves to make a larger lens of the same type; this again was satisfactory, and the idea naturally arose of extending to the whole heavens the scheme which they had hitherto intended only for the Zodiac, a mere belt of the heavens. But this rendered the enterprise too large for a single observatory. It became necessary to obtain the co-operation of other observatories, and with this end in view an International Conference was summoned to meet in Paris in 1887 to consider the whole project. There were delegates from, if not all nations, at any rate a considerable number:--

France 20 British Empire 8 Germany 6 Russia 3 Holland 3 U.S. America 3 Austria 2 Sweden 2 Denmark 2 Belgium 1 Italy 1 Spain 1 Switzerland 1 Portugal 1 Brazil 1 Argentine Republic 1

[Sidenote: Choice of instrument.]

[Sidenote: Expense of "doublet."]

[Sidenote: Advantages of reflector.]

[Sidenote: Refractor chosen.]

The Conference had a number of very important questions to discuss, for knowledge of the photographic method and its possibilities was at that time in its infancy. There was, for instance, the question whether all the instruments need be of the same pattern, and if so what that pattern should be. The first of these questions was settled in the affirmative, as we might expect; in the interests of uniformity it was desirable that the maps should be as nearly similar as possible. The second question was not so easy; there were at least three different types of instruments which might be used. First of all, there was the photographic lens, such as is familiar to all who have used an ordinary camera, consisting of two lenses with a s.p.a.ce between; though since each of these lenses is itself made up of two, we should more correctly say four lenses in all. It was with a lens of this kind that the comet pictures had been taken at the Cape of Good Hope, and it might seem the safest plan to adopt what had been shown to be capable of such good work. But there was this difficulty; the pictures of the comet were on a very small scale, and taken with a small lens; a much larger lens was required for the scheme now under contemplation, and when there are four separate lenses to be made, each with two surfaces to polish, and each requiring a perfectly sound clear piece of gla.s.s, it will be obvious that the difficulties of making a large compound lens of this kind are much greater, and the expense much more serious than in the case of a single lens, or even a pair. It was this question of expense which had led the brothers Henry to experiment with a different kind of instrument, in which only one pair of lenses was used instead of two. Their instrument was, in fact, very similar to the ordinary telescope, excepting that they were bound to make their lenses somewhat different in shape in order to bring to focus the rays of light suitable for photography, which are not the same as those suitable for eye observation with the ordinary telescope. Dr. Common, again, had used a third kind of instrument, mainly with the view of reducing the necessary expense still further, or, perhaps, of increasing the size of the instrument for the same expense. His telescope had no lens at all, but a curved mirror instead, the mirror being made of gla.s.s silvered on the face (not on the back as in the ordinary looking-gla.s.s). In this case there is only one surface to polish instead of four, as in the Henrys' telescope, or eight, as in the case of the photographic doublet; and, moreover, since the rays of light are reflected from the surface of the gla.s.s, and do not pa.s.s _through_ it at all, the internal structure of the gla.s.s is not so strictly important as in the other cases. Hence the reflector is a very cheap instrument, and it is, moreover, quite free from some difficulties attached to the other instruments. No correction for rays of light of different colours is required, since all rays of whatever colour come to the same focus automatically. These advantages of the reflector were so considerable as to almost outweigh one well-known disadvantage, which is, however, not very easily expressed in words. The reflector might be described as an instrument with a temper; sometimes it gives excellent results, but at others _something_ seems to be wrong, though the worried observer does not exactly know what. Long experience and patience are requisite to humour the instrument and get the best results from it, and it was felt that this uncertainty was sufficient to disqualify the instrument for the serious piece of routine work contemplated in mapping the heavens. Accordingly the handier and more amiable instrument with which the brothers Henry had done such good work was selected as the pattern to be adopted.

[Sidenote: Doublet would have been better.]

It is curious that at the Conference of 1887 nothing at all was said about the type of instrument first mentioned (the "doublet lens"), although a letter was written in its favour by Professor Pickering of Harvard College Observatory. Since that time we have learnt much of its advantages, and it is probable that if the Conference were to meet now they might arrive at a different decision; but at that time they were, to put it briefly, somewhat afraid of an instrument which seemed to promise, if anything, too well, especially in one respect. With the reflector and the refractor it had been found that the field of good images was strictly limited. The Henrys' telescope would not photograph an area of the sky greater in extent than 2 in diameter at any one time, and the reflector was more limited still; within this area the images of the stars were good, and it had been found that their places were accurately represented. Now the "doublet" seemed to be able to show much larger areas than this with accuracy, but no one had been able to test the accuracy to see whether it was sufficient for astronomical purposes; and although no such feeling was openly expressed or is on record, I think there is no doubt that a feeling existed of general mistrust of an instrument which seemed to offer such specious promises. Whatever the reason, its claims were pa.s.sed over in silence at the Conference, and the safer line (as it was then thought) of adopting as the type the Henrys' instrument, was taken.

[Sidenote: The eighteen observatories.]

This was perhaps the most important question settled at the Conference, and the answers to many of the others naturally followed. The size of the plates, for instance, was settled automatically. The question down to what degree of faintness should stars be included, resolved itself into the equivalent question, What should be the length of time during which the plates were exposed? Then, again, the question, What observatories should take part in the work? became simply this: What observatories could afford to acquire the instruments of this new pattern and get other funds for carrying out the work specified? It was ultimately found that eighteen observatories were able to obtain the apparatus and funds, though unfortunately three of the eighteen have since found it impossible to proceed. The following is the original list, and in brackets are added the names of three other observatories which in 1900 undertook to fill the places of the defaulters.

OBSERVATORIES CO-OPERATING FOR THE ASTROGRAPHIC CHART.

+----------------------+------------+----------+ Observatory. Zones of Number Declination. of Plates. +----------------------+------------+----------+ Greenwich +90 to +65 1149 Rome +64 " +55 1140 Catania +54 " +47 1008 Helsingfors +46 " +40 1008 Potsdam +39 " +32 1232 Oxford +31 " +25 1180 Paris +24 " +18 1260 Bordeaux +17 " +11 1260 Toulouse +10 " + 5 1080 Algiers + 4 " - 2 1260 San Fernando - 3 " - 9 1260 Tacubaya -10 " -16 1260 Santiago (Monte Video) -17 " -23 1260 La Plata (Cordoba) -24 " -31 1360 Rio (Perth, Australia) -32 " -40 1376 Cape of Good Hope -41 " -51 1512 Sydney -52 " -64 1400 Melbourne -65 " -90 1149 +----------------------+------------+----------+

[Sidenote: Sky covered twice.]

In the list is also shown the total number of plates that were to be taken by each observatory. When once the size of the plates had been settled, it was a straightforward matter to divide up the sky into the proper number of regions necessary to cover it completely, not only without gaps between the plates, but with actually a small overlap of contiguous plates. And more than this, it was decided that the whole sky should be completely covered _twice over_. It was conceivable that a question might arise whether an apparent star image on a plate was, on the one hand, a dust speck, or, on the other hand, a planet, or perhaps a variable or new star. By taking two different plates at slightly different times, questions of this kind could be settled; and to make the check more independent it was decided that the plates should not be exactly repeated on the same portion of sky, but that in the second series the centre of a plate should occupy the point a.s.signed to the corner of a plate in the first series.

[Sidenote: Times of exposure.]

Then there came the important question of time of exposure, which involved a long debate between those who desired the most modest programme possible consistent with efficiency, and those enthusiasts who were anxious to strain the programme to the utmost limits attainable. Ultimately it was resolved to take two series of plates; one series of long exposure which was set in the first instance at 10 minutes, then became 15, then 30, then 40, and has by some enterprising observers been extended to 1-1/2 hours; the other a series of short exposures which have been generally fixed at 6 minutes. Thus instead of covering the sky twice, it was decided to cover it in all four times, and the number of plates a.s.signed to each observatory in the above list must be regarded as doubled by this new decision. And further still, on the series of short-exposure plates it was decided to add to the exposure of six minutes another one of three minutes, having slightly s.h.i.+fted the telescope between the two so that they should not be superimposed; and later still, a third exposure of twenty seconds was added to these. It would take too long to explain here the reasons for these details, but it will be clear that the general result of the discussion was to extend the original programme considerably, and render the work even more laborious than it had appeared at the outset.

[Sidenote: Measurement of plates.]

[Sidenote: The reseau.]

[Sidenote: The microscope.]

[Sidenote: Reversal of plates.]

[Sidenote: Personal equation.]

When all these plates have been taken, the work is by no means finished; indeed, it is only just commencing. There remains the task of measuring accurately on each of the short-exposure plates the positions of the stars which it represents, numbering on the average some 300 or 400; so that for instance at Oxford the total number of stars measured on the twelve hundred plates is nearly half a million. These are not all separate stars; for the sky is represented twice over, and there is also the slight overlap of contiguous plates; but the number of actual separate stars measured at this one observatory is not far short of a quarter of a million, and it has taken nearly ten years to make the measurements, with the help of three or four measurers trained for the purpose. To render the measures easy, a network or reseau of cross lines is photographed on each plate by artificial light after it has been exposed to the stars, so that on development these cross lines and the stars both appear. We can see at a glance the approximate position of a star by counting the number of the s.p.a.ce from left to right and from top to bottom in which it occurs; and we can also estimate the fraction of a s.p.a.ce in addition to the whole number; but it is necessary for astronomical purposes to estimate this fraction with the greatest exactness. The whole numbers are already given with great exactness by the careful ruling of the cross lines, which can be s.p.a.ced with extraordinary perfection. To measure the fraction, we place the plate under a microscope in the eye-piece of which there is a finally divided cross scale; the centre of the cross is placed over a star image, and then it is noted where the lines of the reseau cut the cross scale. In this way the position of the image of a star is read off with accuracy, and after a little practice with considerable rapidity. It has been found at Oxford that under favourable conditions the places of nearly 200 stars per hour can be recorded in this way by a single measurer, if he has some one to write down for him the numbers he calls out. This is only one form of measuring apparatus; there are others in which, instead of a scale in the eye-piece, micrometer screws are used to measure the fractions; but the general principle in all these instruments is much the same, and the rate of work is not very different; while to the minor advantages and disadvantages of the different types there seems no need here to refer.

One particular point, however, is worth noting. After a plate has been measured, it is turned round completely, so that left is now right, and top is now bottom, and the measurements are repeated. This repet.i.tion has the advantage first of all of checking any mistakes. When a long piece of measuring or numerical work of any kind is undertaken there are invariably moments when the attention seems to wander, and some small error is the result. But there are also certain errors of a systematic character similar to those denoted by the term "personal equation," which has found its way into other walks of life. In the operation of placing a cross exactly over the image of a star, different observers would show slight differences of habit; one might place it a little more to the right than another. But when the plate is turned round the effect of this habit on the measure is exactly reversed, and hence if we take the mean of the two measures any personal habit of this kind is eliminated. It has been found by experience that such personal habits are much smaller for measures of this kind than for those to which we have long been accustomed in observations made by eye on the stars themselves. The troubles from "personal equation" have been much diminished by the photographic method, and certain peculiarities of the former method have been clearly exhibited by the comparison. For instance, it has gradually become clear that with eye observations personal equation is not a constant quant.i.ty, but is different for stars of different brightness. When observing the transit of a bright star the observer apparently records an instant definitely earlier than in recording the transit of a faint one; and this peculiarity seems to be common to the large majority of observers, which is perhaps the reason why it was not noticed earlier. But when positions of the stars determined in this way are compared with their positions measured on the photographic plates, the peculiarity is made clearly manifest. For example, at Oxford, our first business after making measurements is to compare them with visual observations on a limited number of the brighter stars made at Cambridge about twenty years ago. (About 14,000 stars were observed at Cambridge, and we are dealing with ten times that number.) The comparison shows that the Cambridge observations are affected with the following systematic errors:--

If stars of magnitude 10 are observed correctly, then " " 9 " 0.10 secs. too early " " 8 " 0.16 "

" " 7 " 0.19 "

" " 6 " 0.21 "

" " 5 " 0.23 "

[Sidenote: Main object of the work.]

This may serve as an ill.u.s.tration of various incidental results which are already flowing from the enormous and laborious piece of work which, as far as the University Observatory at Oxford is concerned, we have just completed, though some of the other colleagues are not so far advanced.

But the main results will not appear just yet. The work must be repeated, and the positions of the stars just obtained must be compared with those which they will be found to occupy at some future date, in order to see what kind of changes are going on in the heavens. Whether this future date shall be one hundred years hence, or fifty, or ten, or whether we should begin immediately to repeat what has been done, is a matter not yet decided, and one which requires some little consideration.

[Sidenote: The concluding year.]

I have said perhaps enough to give you a general idea of the work on which we have been engaged at Oxford for the last ten years. Ten years ago it seemed to stretch out in front of us rather hopelessly; the pace we were able to make seemed so slow in view of the distance to be covered. We felt rather like the schoolboy who has just returned to school and sees the next holidays as a very remote prospect, and we solaced ourselves much in the same way as he does, by making a diagram representing the total number of plates to be dealt with and crossing off each one as it was finished, just as he sometimes crosses off the days still remaining between him and the prospective holidays. It was pleasant to watch the growth of the number of crosses on this diagram, and by the end of the year 1902 we had the satisfaction of seeing very little blank s.p.a.ce remaining. Now, up to this point it had not much mattered whether any particular plate was secured in any particular year, or in a subsequent year, so long as there were always sufficient plates to keep us occupied in measuring them. But it then became a matter of importance to secure each plate at the proper time of year; for the sun, as we know, travels round the Zodiac among the stars, obliterating by his radiance a large section of the sky for a period of some months, and in this way a particular region of the heavens is apt to "run into daylight," as the observatory phrase goes, and ceases to be available for photography during several months, until the sun is again far enough away to allow of the particular region being seen at night.

[Sidenote: A disappointment.]

[Sidenote: A curious plate.]

[Sidenote: A strange object.]

[Sidenote: A new star?]

Roughly speaking then, if a plate which should be taken in February is not secured in this month owing to bad weather, the proper time for taking it will not occur again until the following February; and when there was a fair prospect of finis.h.i.+ng our work in 1903, it became important to secure each plate at the proper time in that year. Hence we were making special efforts to utilise to the full any fine night that Providence sent in our way, and on such occasions it is clearly an economy, if not exactly to "make hay while the sun s.h.i.+nes," at any rate to take plates vigorously while the sun is _not_ s.h.i.+ning and the night is fine; leaving the development of them until the daytime. There is, of course, the risk that the whole night's work may in this way be lost owing to some fault in the plates, which might have been detected if some of them were immediately developed. Perhaps in the early days of our work it would have been reckless or foolish to neglect this little precaution; but we had for years been accustomed to rely upon the excellence of the plates without finding our trust betrayed; and the sensitiveness of the plates had increased rather than diminished as time went on. Hence it will be readily understood that when one fatal morning we developed a series of some thirty plates, and found that owing to some unexplained lack of sensitiveness they were all unsuitable for our purpose, it came as a most unwelcome and startling surprise. It was, of course, necessary to make certain that there was no oversight, that the developer was not at fault, and that the weather had not been treacherous. All such possibilities were carefully considered before communication with the makers of the plates, but it ultimately became clear that there had been some unfortunate failure in sensitiveness, and that it would be necessary to repeat the work with opportunities restricted by the intervening lapse of time.

However, disappointments from this or similar causes are not unknown in astronomical work; and we set about this repet.i.tion with as little loss of time and cheerfulness as was possible. Under the circ.u.mstances, however, it seemed desirable to examine carefully whether anything could be saved from the wreck--whether any of the plates could be admitted as _just_ coming up to the minimum requirements. And I devoted a morning to this inquiry. In the course of it I came across one plate which certainly seemed worth an inclusion among our series from the point of view of the number of stars shown upon it. It seemed quite rich in stars, perhaps even a little richer than might have been expected. On inquiry I was told that this was not one of the originally condemned plates, but one which had been taken since the failure in sensitiveness of the plates had been detected; was from a new and specially sensitive batch with which the courteous makers had supplied us; but though there were certainly a sufficient number of stars upon the plate, owing to some unexplained cause the telescope had been erroneously pointed, and the region taken did not correspond to the region required. To investigate the cause of the discrepancy I thereupon took down from our store of plates the other one of the same region which had been rejected for insufficiency of stars, and on comparing the two it was at once evident that there was a strange object on the plate taken later of the two, a bright star or other heavenly body, which was not on the former plate. I have explained that by repeating the exposure more than once, it is easily possible to recognise whether a mark upon the plate is really a celestial body or is an accidental blot or dust speck, and there was no doubt that this was the image of some strange celestial body. It might, of course, be a new planet, or even an old one which had wandered into the region; but a few measures soon showed that it was not in movement. The measures consisted in comparing the separation of the three exposures with the separation of the corresponding exposures of obvious stars, for the exposures were not, of course, simultaneous, and if the body were a planet and had moved in the interval between them, this would be made manifest on measuring the separations. No such movements could be detected; and the possibilities were thus restricted to two. So far as we knew the object was a star, but might be either a star of the cla.s.s known as _variable_ or of that known as _new_. In the former case it would become bright and faint at more or less regular intervals, and might possibly have been already catalogued; for the number of these bodies already known amounts to some hundreds.