by Isaac Asimov
That was the situation in the early 1920s: the known universe was less than 200,000 light-years in diameter and consisted of our galaxy and its two neigh bors. The question then arose whether anything existed outside that.
Suspicion rested upon certain small patches of luminous fog, called nebulae (from the Greek word for “cloud”), which astronomers had long noted, The French astronomer Charles Messier had catalogued 103 of them in 1781. (Many are still known by the numbers he gave them, preceded by the letter M for Messier.)
Were these nebulosities merely the clouds they seemed? Some, such as the Orion Nebula (first discovered in 1656 by the Dutch astronomer Christian Huygens), seemed to be just that: a cloud of gas and dust, equal in mass to about 500 suns like ours, and illuminated by hot stars within. Other nebulosities, on the other hand, turned out to be globular clusters—huge assemblages of stars.
But there remained patches of luminous cloud that seemed to contain no stars at all. Why, then, were they luminous? In 1845, the British astronomer William Parsons (third Earl of Rosse), using a 72-inch telescope he had spent his life building, had ascertained that some of these patches had a spiral structure, which gave them the name “spiral nebulae” but did not help explain the source of the luminosity.
The most spectacular of these patches, known as M-31, or the Andromeda Nebula (because it is in the constellation Andromeda), was first studied in 1612 hy the German astronomer Simon Marius. It is an elongated oval of dim light about half the size of the full moon. Could it be composed of stars so distant that they could not be made out separately even in large telescopes? If so, the Andromeda Nebula must be incredibly far away and incredibly large to be visible at all at such a distance. (As long ago as 1755, the German philosopher Immanuel Kant had speculated on the existence of such far distant star collections: island universes, he called them.)
In the 1910s, there was a strong dispute over the matter. The Dutch-American astronomer Adriaan Van Maanen had reported that the Andromeda Nebula was rotating at a measurable rate. To do so, it had to be fairly close to us, If it were beyond the galaxy, it would be too far away to display any perceptible motion. Shapley, a good friend of Van Maanen, used his results to argue that the Andromeda Nebula was part of the galaxy.
Arguing against this assumption was the American astronomer Heber Doust Curtis. Although no stars were visible in the Andromeda Nebula, every once in a while an exceedingly faint star would make its appearance. Curtis felt this to be a nova, a star that suddenly brightens several thousand fold. In our galaxy, such stars end up being quite bright for a short while before fading again; but in the Andromeda Nebula, they were just barely visible, even at their brightest. Curtis reasoned that the novas were exceedingly dim because the Andromeda Nebula was exceedingly far away. Ordinary stars in the Andromeda Nebula were altogether too dim to be made out, but just melted together in a kind of faintly luminous fog.
On 26 April 1920, Curtis and Shapley held a well-publicized debate on the matter. On the whole, it was a standoff, although Curtis turned out to be a surprisingly good speaker and presented an impressive defense of his position.
Within a few years, however, it was clear that Curtis was in the right. For one thing, Van Maanen’s figures turned out to be wrong. The reason is uncertain, but even the best can make errors, and Van Maanen had apparently done so. Then, in 1924, the American astronomer Edwin Powell Hubble turned the new 100-inch telescope at Mount Wilson in California on the Andromeda Nebula. (It was called the Hooker telescope after John B. Hooker who had provided the funds for its construction.) This powerful instrument resolved portions of the nebula’s outer edge into individual stars, thus showing at once that the Andromeda Nebula, or at least parts of it, resembled the Milky Way and that there might be something to this “island universe” notion.
Among the stars at the edge of the Andromeda Nebula are cepheid variables. Using these measuring rods, Hubble decided that the nebula was nearly a million light-years away! So the Andromeda Nebula was far, far outside our galaxy. Allowing for its distance, its apparent size showed that it must be a huge conglomeration of stars, almost rivaling our own galaxy.
Other nebulosities, too, turned out to be conglomerations of stars, even farther away than the Andromeda Nebula. These extra-galactic nebulae all had to be recognized as galaxies—new universes that reduced our own to just one of the many in space. Once again the universe had expanded. It was larger than ever—not merely hundreds of thousands, but perhaps hundreds of millions, of light-years across.
SPIRAL GALAXIES
Through the 1930s, astronomers wrestled with several nagging puzzles about these galaxies. For one thing, on the basis of their assumed distances. all of them were apparently much smaller than our own. It seemed an odd coincidence that we should be inhabiting the largest galaxy in existence. For another thing, globular clusters surrounding the Andromeda galaxy seemed to be only one-half or one-third as luminous as those of our own galaxy. (Andromeda is about as rich in globular clusters as our own galaxy, and its clusters, are spherically arranged about Andromeda’s center. This finding seems to show that Shapley’s assumption that our own clusters are so arranged was reasonable. Some galaxies are amazingly rich in globular clusters. The galaxy M-87, in Virgo, possesses at least 1,000.)
The most serious problem was that the distances of the galaxies seemed to imply that the universe was only about 2 billion years old (for reasons I shall discuss later in this chapter). This was puzzling, for the earth itself was considered by geologists to be older than that, on what was thought to be the very best kind of evidence. The beginning of an answer came during the Second World War, when the German-born American astronomer Walter Baade discovered that the yardstick by which the galaxies’ distances had been measured was wrong.
In 1942, Baade took advantage of the wartime blackout of Los Angeles, which cleared the night sky at Mount Wilson, to make a detailed study of the Andromeda galaxy with the 100-inch telescope. With the improved visibility, he was able to resolve some of the stars in the inner regions of the galaxy. He immediately noted some striking differences between these stars and those in outskirts of the galaxy. The brightest stars in the interior were reddish, whereas those of the outskirts were bluish. Moreover, the red giants of the interior were not nearly so bright as the blue giants of the outskirts: the latter had up to 100,000 times the luminosity of our sun, whereas the internal red giants had only up to 1,000 times that luminosity. Finally, the outskirts of the Andromeda galaxy, where the bright blue stars were found, was loaded with dust; whereas the interior, with its somewhat less bright red stars, was free of dust.
To Baade, it seemed that here were two sets of stars with different structure and history. He called the bluish stars of the outskirts Population I and the reddish stars of the interior, Population II. Population I stars, it turns out, are relatively young, with high metal content, and follow nearly circular orbits about the galactic center in the median plane of the galaxy. Population II stars relatively old, with low metal content, with orbits that are markedly and with considerable inclination to the median plane of the galaxy. populations have been broken down into finer subgroups since Baade’s discovery.
When the new 200-inch Hale telescope (named for the American astronomer, George Ellery Hale, who supervised its construction) was set up on Palomar Mountain after the war, Baade continued his investigations. He found certain regularities in the distribution of the two populations, and these depended on the nature of the galaxies involved. Galaxies of the class called elliptical (systems with the shape of an ellipse and with rather uniform internal structure) apparently were made up mainly of Population II stars, as were globular clusters in any galaxy. On the other hand, in spiral galaxies (galaxies with arms that make them look like a pinwheel) the spiral arms were composed of Population I, set against a Population II background.
It is estimated that only about 2 percent of the stars in the universe are of the Population I type. But our own su
n and the familiar stars in our neighborhood fall into this class. From this fact alone, we can deduce that ours is a spiral galaxy, and that we lie in one of the spiral arms. (Hence, the many dust clouds, both light and dark, in our neighborhood: the spiral arms of a galaxy are clogged with dust.) Photographs show that the Andromeda galaxy also is of the spiral type.
Now to get back to the yardstick. Baade began to compare the cepheid stars in globular clusters (Population II) with those found in our spiral arm (Population I). It turned out that the cepheids in the two populations were really of two different types, as far as the relation between period and luminosity was concerned. Cepheids of Population II followed the period-luminosity curve set up by Leavitt and Shapley. With this yardstick, Shapley had measured the distances to the globular clusters and the size of our galaxy with reasonable accuracy. But the cepheids of Population I, it now developed, were a different yardstick altogether! A Population-I cepheid was four or five times as luminous as a Population-II cepheid of the same period. Hence, use of the Leavitt scale would result in miscalculation of the absolute magnitude of a Population-I cepheid from its period. And if the absolute magnitude was wrong, the calculation of distance must be wrong: the star would actually be much farther away than the calculation indicated.
Hubble had gauged the distance of the Andromeda galaxy from the cepheids (of Population I) in its outskirts—the only ones that could be resolved at the time. Now, with the revised yardstick, the galaxy proved to be about 2.5 million light-years away, instead of less than a million. And other galaxies had to be moved out in proportion. (The Andromeda galaxy is still a close neighbor, however. The average distance between galaxies is estimated to be some thing like 20 million light-years.)
At one stroke, the size of the known universe was more than doubled, and the problems that had plagued the 1930s were solved. Our galaxy was no longer larger than all the others; the Andromeda galaxy, for instance, was definitely more massive than ours. Second, it now appeared that the Andromeda galaxy’s globular clusters were as luminous as ours; they had seemed less bright only because of the misjudgment of their distance. Finally, for reasons I will explain later, the new scale of distances allowed the universe to be considered much older, bringing it into line with the geologists’ estimates of the age of the earth.
CLUSTERS OF GALAXIES
Doubling the distance of the galaxies does not end the problem of size. We must now consider the possibility of still larger systems—of clusters of galaxies and clusters of clusters. Actually, modern telescopes have shown that clusters of galaxies do exist. For instance, in the constellation of Coma Berenices there is a large, ellipsoidal cluster of galaxies about 8 million light-years in diameter. The Coma cluster contains about 11 ,000 galaxies, separated by an average distance of Oil II 300,000 light-years (as compared with an average of something like 3 million light-years between galaxies in our own vicinity).
Our own galaxy seems to be part of a local group that includes the Magellanic Clouds, the Andromeda galaxy, and three small satellite galaxies near it, plus some other galaxies; a total of nineteen members altogether. Two of these, called Maffei One and Maffei Two (for Paolo Maffei, the Italian astronomer, who first reported them), were discovered only in 1971. The lateness of the discovery was owing to the fact that they can only be detected through dust clouds that lie between them and ourselves.
Of the local group, only our own galaxy, Andromeda, and the two Maffeis are giants, whereas the rest are dwarfs. One of the dwarfs, IC 1613, may contain only 60 million stars; hence it is scarcely more than a large globular cluster. Among galaxies, as among stars, dwarfs far outnumber giants.
If galaxies do form clusters and clusters of clusters, does that mean that the universe goes on forever and that space is infinite? Or is there some end, both to the universe and to space? Well, astronomers can make out objects up to an estimated 10 billion light-years away, and there they seem to be reaching a limit. To see why, I must now shift the direction of discussion a bit. Having considered space, let us next consider time.
Tbe Birth of the Universe
Mythmakers have invented many fanciful creations of the universe (usually concentrating on the earth itself, with all the rest dismissed quickly as the “sky” or the “heavens”), Generally, the time of creation is set not very far in the past (although we should remember that, to people in the preliterate stage, a time of a thousand years was even more impressive than a billion years is today).
The creation story with which we are most familiar is, of course, that given in the first chapters of Genesis, which, some people hold, is an adaptation of Babylonian myths, intensified in poetic beauty and elevated in moral grandeur.
Various attempts have been made to work out the date of the Creation on the basis of the data given in the Bible (the reigns of the various kings, the time from the Exodus to the dedication of Solomon’s temple, the ages of the patriarchs both before and after the flood). Medieval Jewish scholars put the of the Creation at 3760 B.C, and the Jewish calendar still counts its years from that date. In 1658 A.D., Archbishop James Ussher of the Anglican Church calculated the date of the Creation to be 4004 B.C.; while others following his lead placed it exactly at 8 P.M. on 22 October of that year. Some theologians of the Greek Orthodox Church put Creation as far back as 5508 B.C.
Even as late as the eighteenth century, the Biblical version was accepted by the learned world, and the age of the universe was considered to be only 6,000 or 7,000 years at most. This view received its first major blow in 1785 in the form of a book entitled Theory of the Earth, by a Scotch naturalist named James Hutton. Hutton started with the proposition that the slow processes working on the surface of the earth (mountain building and the cutting of river channels, and so on) had been working at about the same rate throughout the earth’s history. This uniformitarian principle implied that the processes must have been working for a stupendously long time to produce the observed phenomena. Therefore the earth must be not thousands but many millions of years old.
Hutton’s views were immediately derided. But the ferment worked. In the early 1830’s, the British geologist Charles Lyell reaffirmed Hutton’s views and, in a three-volume work entitled Principles of Geology, presented the evidence with such clarity and force that the world of science was won over. The modern scince of geology can be dated from that work.
THE AGE OF THE EARTH
Attempts were made to calculate the age of the earth on the basis of the uniformitarian principle. For instance, if one knew the amount of sediment laid down by the action of water each year (a modern estimate is 1 foot in 880 years), one could calculate the age of a layer of sedimentary rock from its thickness. It soon became obvious that this approach could not accurately determine the earth’s age, because the record of the rocks was obscured by erosion, crumbling, upheavals, and other forces. Nevertheless, even the incomplete evidence indicated that the earth must be at least 500 million years old.
Another way of measuring the age of the earth was to estimate the rate of accumulation of salt by the oceans, a suggestion first advanced by Edmund Halley as long ago as 1715. Rivers steadily washed salt into the sea; since only fresh water left it by evaporation, the salt concentration rose. The assumption was that the ocean had started as fresh water; hence, the time necessary for the rivers to have endowed the oceans with their salt content of over 3 percent could have been as long as a billion years.
This great age was very agreeable to the biologists, who, during the latter half of the nineteenth century, were trying to trace the slow development of living organisms from primitive one-celled creatures to the complex higher animals. They needed long eons for the development to take place, and a billion years gave them sufficient time.
However, by the mid-nineteenth century, astronomical considerations brought sudden complications. For instance, the principle of the conservation of energy raised an interesting problem with respect to the sun. The sun was pouring out energy in coloss
al quantities and had been doing so throughout recorded history. If the earth had existed for countless eons, where had all this energy come from? It could not have come from the usual familiar sources. If the sun had started as solid coal burning in an atmosphere of oxygen, it would have been converted to carbon dioxide (at the rate it was delivering energy) in the space of about 2,500 years.
The German physicist Hermann Ludwig Ferdinand von Helmholtz, one of the first to enunciate the law of conservation of energy, was particularly interested in the problem of the sun. In 1854, he pointed out that if the sun were contracting, its mass would gain energy as it fell toward its center of gravity, just as a rock gains energy when it falls. This energy could be converted into radiation. Helmholtz calculated that a contraction of the sun by a mere 1/10,000 of its radius could provide it with a 2,000-year supply of energy.
The British physicist William Thomson (later Lord Kelvin) did more work on the subject and decided that, on this basis, the earth could not be more than 50 million years old; for at the rate the sun had spent energy, it must have contracted from a gigantic size, originally as large as the earth’s orbit around the sun. (This assumption meant, of course, that Venus must be younger than the earth and Mercury still younger.) Lord Kelvin went on to estimate that if the earth itself had started as a molten mass, the time needed to cool to its present temperature, and therefore its age, would be about 20 million years.
By the 1890s, the battlelines were drawn between two apparently invincible armies. The physicists seemed to have shown conclusively that the earth could not have been solid for more than a few million years, while geologists and biologists seemed to have proved just as conclusively that the earth must have been solid for not less than a billion years.