by Isaac Asimov
The whole conception of changes in heavenly bodies is modern. The ancient Greek philosophers—Aristotle, in particular—believed the heavens to be perfect and unchangeable. All change, corruption, and decay were confined to the imperfect regions that lay below the nethermost sphere—the moon. This seemed only common sense, for certainly, from generation to generation and from century to century, there was no important change in the heavens. To be sure, mysterious comets occasionally materialized out of nowhere—erratic in their comings and goings, ghostlike as they shrouded stars with a thin veil, baleful in appearance, for the filmy tail looks like the streaming hair of a distraught creature prophesying evil. About twenty-five of these objects are visible to the naked eye each century. (Comets will be discussed in more detail in the next chapter.)
Aristotle tried to reconcile these apparitions with the perfection of the heavens by insisting that they belonged to the atmosphere of the corrupt and changing earth. This view prevailed until late in the sixteenth century. But, in 1577 (before the days of the telescope), the Danish astronomer Tycho Brahe attempted to measure the parallax of a bright comet and discovered that it could not be measured. Since the moon’s parallax was measurable, Tycho Brahe was forced to conclude that the comet lay far beyond the moon and that there was change and imperfection in the heavens. (The Roman philosopher Seneca had suspected such change in the first century A.D.)
Actually, changes even in the stars had been noticed much earlier but apparently had aroused no great curiosity. For instance, there are the variable stars that change noticeably in brightness from night to night, even to the naked eye. No Greek astronomer made any reference to variations in the brightness of any star. It may be that we have lost the records of such references; on the other hand, perhaps the Greek astronomers simply chose not to see these phenomena. One interesting case in point is Algol, the second brightest star in the constellation Perseus, which loses two-thirds of its brightness, then regains it, and does this regularly every 69 hours. (We know now, thanks to Goodricke and Vogel, that Algol has a dim companion star that eclipses it and diminishes its light at 69-hour intervals.) The Greek astronomers made no mention of the dimming of Algol, nor did the Arab astronomers of the Middle Ages. Nevertheless, the Greeks placed the star in the head of Medusa, the demon who turned men to stone; and the very name Algol, which in Arabic, means “ghoul,” is suggestive. Clearly, the ancients felt uneasy about this strange star.
A star in the constellation Cetus, called Omicron Ceti, varies irregularlv. Sometimes it is as bright as the Pole Star; sometimes it vanishes from sight. Neither the Greeks nor the Arabs said a word about it, and the first man to report it was a Dutch astronomer, David Fabricius, in 1596. It was later named Mira (Latin for “wonderful”), astronomers having grown less frightened of heavenly change by then.
NOVAE AND SUPERNOVAE
Even more remarkable was the sudden appearance of new stars in the heavens, the Greeks could not altogether ignore. Hipparchus is said to have been so impressed by the sighting of such a new star, in the constellation Scorpio in 134 B.C., that he designed the first star map, in order that future new stars might be more easily detected.
In 1054 A.D., in the constellation Taurus, another new star was sighted—a phenomenally bright one. It surpassed Venus in brightness and for weeks was visible in broad daylight. Chinese and Japanese astronomers recorded its position accurately, and their records have come down to us. In the Western world, however, the state of astronomy was so low at the time that no European record of this remarkable occurrence has survived, probably because none was kept.
It was different in 1572, when a new star as bright as that of 1054 appeared in the constellation Cassiopeia. European astronomy was reviving from its long sleep, The young Tycho Brahe carefully observed the new star and wrote a hook entitled De Nova Stella. It is from the title of that book that the word nova was adopted for any new star.
In 1604, still another remarkable nova appeared, in the constellation Serpens, It was not quite as bright as that of 1572, but it was bright enough to outshine Mars. Johannes Kepler observed this one, and he too wrote a book about the subject.
After the invention of the telescope, novae became less mysterious. They were not new stars at all, of course, but faint stars that had suddenly brightened to visibility.
Increasing numbers of novae were discovered with time. They would brighten many thousandfold, sometimes within the space of a few days, and then dim slowly over a period of months to their previous obscurity. Novae showed up at the average rate of twenty per year per galaxy (including our own).
From an investigation of the Doppler-Fizeau shifts that took place during nova formation and from certain other fine details of their spectra, it became plain that the novae were exploding stars. In some cases, the star material blown into space could be seen as a shell of expanding gas, illuminated by the remains of the star.
On the whole, the novae that have appeared in modern times have not been particularly bright. The brightest, Nova Aquilae, appeared in June 1918 in the constellation Aquila. This nova was, at its peak, nearly as bright as the star Sirius, which is itself the brightest in the sky. No novae, however, have appeared to rival the bright planets Jupiter and Venus, as the novae observed by Tycho and by Kepler did.
The most remarkable nova discovered since the beginning of the telescope was not recognized as such. The German astronomer Ernst Hartwig noted it in 1885; hut even at its peak, it reached only the seventh magnitude and was never visihle to the unaided eye.
It appeared in what was then called the Andromeda nebula and, at its peak, was one-tenth as bright as the nebula. At the time, no one realized how distant Andromeda nebula was, or understood that it was actually a galaxy made of several hundred billion stars, so the apparent brightness of the nova occasioned no particular excitement.
After Curtis and Hubble worked out the distance of the Andromeda galaxy (as it then came to be called), the brilliance of that nova of 1885 suddenly staggered astronomers. The dozens of novae discovered in the Andromeda galaxy by Curtis and Hubble were far dimmer than that remarkably (for the distance) bright one.
In 1934, the Swiss astronomer Fritz Zwicky began a systematic search of distant galaxies for novae of unusual brightness. Any nova that blazed up in similar fashion to the one of 1885 in the Andromeda would be visible, for such novae are almost as bright as entire galaxies, so that if the galaxy can be seen, the nova can be as well. By 1938, he had located no fewer than twelve of such galaxy-bright novae. He called these extraordinarily bright novae supernovae. As a result, the 1885 nova was named at last—S Andromedae, the S standing for “supernova.”
Whereas ordinary novae attain an absolute magnitude of, on the average, −8 (they would be 25 times as bright as Venus, if they were seen at a distance of 10 parsecs), a supernova could have an absolute magnitude of as much as −17. Such a supernova would be 4,000 times as bright as an ordinary nova, or nearly 1,000,000,000 times as bright as the sun. At least, it would be that bright at its temporary. peak.
Looking back now, we realize that the novae of 1054, 1572, and 1604 were also supernovae. What is more, they must have flared up in our own galaxy, to account for their extreme brightness.
A number of novae recorded by the meticulous Chinese astronomers of ancient and medieval times must also have been supernovae. One such was reported as early as A.D. 185; and a supernova in the far southern constellation of Lupus in 1006 must have been brighter than any that have appeared in historic times. It may, at its peak, have been 200 times as bright as Venus and one-tenth as bright as the full moon.
Astronomers, judging from remnants left behind, suspect that an even brighter supernova (one that may actually have rivaled the full moon) appeared in the far southern constellation Vela 11,000 years ago, when there were no astronomers to watch, and the art of writing had not yet been invented. (II is possible, however, that certain prehistoric pictograms may have been drawn that refer to thi
s nova.)
Supernovae are quite different in physical behavior from ordinary novae, and astronomers are eager to study their spectra in detail. The main difficulty is their rarity. About 1 per 50 years is the average for any one galaxy. Although astronomers have managed to spot more than 50 so far, all these are in distant galaxies and cannot be studied in detail. The 1885 supernova of Andromeda, the closest to us in the last 350 years, appeared a couple of decades before photography in astronomy had been fully developed; consequently, no permanent record of its spectrum exists.
However, the distribution of supernovae in time is random. In one galaxy recently, 3 supernovae were detected in just 17 years. Astronomers on earth may yet prove lucky. Indeed, one particular star is now attracting attention. Eta Carinae is clearly unstable and has been brightening and dimming for quite a while. In 1840, it brightened to the point where, for a time, it was the second brightest star in the sky. There are indications that make it appear as though it may be on the point of exploding into a supernova. One trouble, though, is that, to astronomers, “on the point of” can mean tomorrow or ten thousand years from now.
Besides, the constellation Carina, in which Eta Carinae is found, is like the constellations Vela and Lupus, so far south that the supernova, when and if it occurs, will not be visible from Europe or from most of the United States.
But what causes stars to brighten with explosive violence, and why do some become novae and some supernovae? The answer to this question requires a digression.
As early as 1834, Bessel (the astronomer who was later the first to measure the parallax of a star) noticed that Sirius and Procyon shifted position very slightly from year to year in a manner that did not seem related to the motion of the earth. Their motions were not in a straight line but wavy, and Bessel decided that each must actually be moving in an orbit around something.
From the manner in which Sirius and Procyon were moving in these orbits, the “something” in each case had to have a powerful gravitational attraction that could belong to nothing less than a star. Sirius’s companion, in particular, had to be as massive as our own sun to account for the bright star’s motions. So the companions were judged to be stars; but since they were invisible in the telescopes of the time, they were referred to as dark companions. They were believed to be old stars growing dim with time.
Then, in 1862, the American instrument maker Alvan Clark, testing a new telescope, sighted a dim star near Sirius; and, sure enough, on further observation, this turned out to be the companion. Sirius and the dim star circled about a mutual center of gravity in a period of about fifty years. The companion of Sirius (Sirius B, it is now called, with Sirius itself being Sirius A) has an absolute magnitude of only 11.2 and so is only about 1/400 as bright as our sun, although it is just as massive.
Sirius B seemed to be a dying star. But, in 1914, the American astronomer Walter Sydney Adams, after studying the spectrum of Sirius B, decided that the star had to be as hot as Sirius A itself and hotter than our sun. The atomic vibrations that gave rise to the particular absorption lines found in its spectrum could be taking place only at very high temperatures. But if Sirius B was so hot, why was its light so faint? The only possible answer was that it was considerably smaller than our sun. Being hotter, it radiated more light per unit of surface; but to account for the small total amount of light, its total surface had to be small. In fact, we now know that the star cannot be more than 6,900 miles in diameter; it is smaller than the earth in volume, even though it has a mass equal to that of our sun! With all that mass squeezed into so small a volume, the star’s average density would have to be about 130,000 times that of platinum.
Here was nothing less than a completely new state of matter. Fortunately, by this time physicists had no trouble in suggesting the answer. They knew that in ordinary matter the atoms are composed of very tiny particles, so tiny that most of the volume of an atom is “empty” space. Under extreme pressure, the subatomic particles can be forced together into a superdense mass. Yet even in superdense Sirius B, the subatomic particles are far enough apart to move about freely so that the far-denser-than-platinum substance still acts as a gas. The English physicist Ralph Howard Fowler suggested in 1925 that this be called a degenerate gas, and the Soviet physicist Lev Davidovich Landau pointed out in the 1930s that even ordinary stars such as our own sun ought to consist of degenerate gas at the center. The companion of Procyon (Procyon B), first detected in 1896 by J. M. Schaberle at Lick Observatory in California, was also found to be a super-dense star although only five-eighths as massive as Sirius B; and, as the years passed, more examples were found. These stars are called white dwarfs, because they combine small size with high temperature and white light. White dwarfs are probably numerous and may make up as much as 3 percent of all stars. However, because of their small size and dimness, only those in our own neighborhood are likely to be discovered in the foreseeable future. (There are also red dwarfs, considerably smaller than our sun, but not as small as white dwarfs. Red dwarfs are cool and of ordinary density. They are the most common of all stars—making up three-fourths of the total—but, because of their dimness, are as difficult to detect as white dwarfs. A pair of red dwarfs, a mere six light-years distant from us, was only discovered in 1948. Of the thirty-six stars known to be within fourteen light-years of the sun, twenty-one are red dwarfs, and three are white dwarfs. There are no giants among them, and only two, Sirius and Procyon, are distinctly brighter than our sun.)
The year after Sirius B was found to have its astonishing properties, Albert Einstein presented his general theory of relativity, which was mainly concerned with new ways of looking at gravity. Einstein’s views of gravity led to the prediction that light emitted by a source possessing a very strong gravitational field should be displaced toward the red (the Einstein shift). Adams, fascinated by the white dwarfs he had discovered, carried out careful studies of the spectrum of Sirius B and found that there was indeed the red shift predicted by Einstein. This was a point in favor not only of Einstein’s theory but also of the superdensity of Sirius B; for in an ordinary star such as our sun, the red-shift effect would be only one-thirtieth as great. Nevertheless, in the early 1960s this very small Einstein shift produced by our sun was detected, and the general theory of relativity was further confirmed.
But what have white dwarfs to do with supernovae, the subject that prompted this discussion? To work toward an answer, let us go back to the supernova of 1054. In 1844, the Earl of Rosse, investigating the location in Taurus where the Oriental astronomers had reported finding the 1054 supernova, studied a small cloudy object. Because of its irregularity and its c1awlike projections, he named the object the Crab Nebula. Continued observation over decades showed that the patch of gas was slowly expanding. The actual rate of expansion could be calculated from the Doppler-Fizeau effect, which, combined with the apparent rate of expansion, made it possible to compute the distance of the Crab Nebula as 3,500 light-years from us. From the expansion rate it was abo determined that the gas had started its expansion from a central explosion point nearly 900 years ago, which agrees well with the date 1054. So there can be little doubt that the Crab Nebula, which now spreads over a volume of space some 5 light-years in diameter, represents the remnants of the 1054 supernova.
No similar region of turbulent gas has been observed at the reported sites of the supernovae of Tycho and Kepler, although small spots of nebulosity have been observed close to each site. There are some 150 planetary nebulae, however, in which doughnut-shaped rings of gas may represent large stellar explosions. A particularly extended and thin gas cloud, the Veil Nebula in Cygnus, may be what is left of a supernova explosion 30,000 years ago. It must have been even closer and brighter than the supernova of 1054—but no civilization existed on earth to record the spectacle.
There are even suggestions that a very faint nebulosity enveloping the constellation Orion may be what is left of a still older supernova.
In all these cas
es, though, what happened to the stars that exploded? Have they simply vanished in one enormous puff of gas? Is the Crab Nebula, for instance, all that is left of the 1054 supernova, and will this simply spread out until all visible sign of the star is forever gone? Or is some remnant left that is still a star but too small and too dim to be detected? Is there, in other words, a white dwarf left behind (or something even more extreme), and are white dwarfs, so to speak, the corpses of stars that were once like our sun? These queries lead us into the problem of the evolution of stars.
EVOLUTION OF THE STARS
Of the stars near us, the bright ones seem to be hot and the dim ones cooler, according to a fairly regular brightness-temperature scale. If the surface temperatures of various stars are plotted against their absolute magnitudes, most of the familiar stars fall within a narrow band, increasing steadily from dim coolness to bright hotness. This band is called the main sequence. It was first plotted in 1913 by the American astronomer Henry Norris Russell, following work along similar lines by Hertzsprung (the astronomer who first determined the absolute magnitudes of the cepheids). A graph showing the main sequence is therefore called a Hertzsprung-Russell diagram, or H-R diagram (figure 2.5)
Figure 2.5. The Hertzsprung-Russell diagram. The dotted line indicates the evolution of a star. The relative size of the stars are given only schematically, not according to scale.
Not all stars belong in the main sequence. There are some red stars that, despite their rather low surface temperature, have large absolute magnitudes, because their substance is spread out in rarefied fashion into tremendous size, and the sparse heat per unit area is multiplied over the enormous surface to a huge total. Among these red giants, the best-known are Betelgeuse and Antares. They are so cool (it was discovered in 1964) that many have atmospheres rich in water vapor, which would decompose to hydrogen and oxygen at the higher temperatures of our own sun. The high-temperature white dwarfs also fall outside the main sequence.