The Size and Shape of Our Galaxy:
A Point-Form Summary.
This section of the course notes, and the associated PowerPoint presentation, makes the following critical points:
the flattened band of light we know as the Milky Way suggests that we are situated somewhere in a flattened system of stars (and gas and dust)
early investigations were little more than speculative, and imagined us at the centre. Star counts by Herschel seemed to confirm this view
photographic techniques allowed us to extend the star counts to much deeper levels, and Kapteyn deduced from such work that we were in the middle of a rather small flattened system a few thousand light years across
it was disconcerting to discover that we were apparently 'right in the centre', since there is no reason that we should be singled out in this way. Shapley realized, however, that there was good evidence that we are not in the very centre -- namely, the fact that the globular clusters are not symmetrically disposed around us. There are many more in the southern skies than the north, and Shapley deduced that the centre of the galaxy lay off in that direction. He also deduced that the galaxy was very much larger than anyone had realised (in fact he somewhat over estimated the size)
the flattening of the Milky Way arises because of its systemic rotation, but this is so slow that it is not easy to demonstrate. Oort studied the motions of the stars near the sun and was able to show that they display a pattern which reveals that the ones nearer the centre are 'overaking us on the inside' while the Sun itself is passing 'on the inside' the stars farther out from the centre. (This is reminiscent of the way the separate planets move around the Solar System as they orbit the sun.) Oort deduced that the centre of the Galaxy lies in the direction of the constellation Sagittarius, just as Shapley had independently deduced from the globular clusters
many flattened galaxies display spiral structure, but our own galaxy is hard to study since we are within it and there is an intervening 'fog' of gas and dust. Since the spiral arms are delineated by bright stars and clouds of ionized hydrogen, we can look at how these 'tracers' are arranged around us, and see what seem to be segments of spiral arms. (This was first shown in 1952.) In visible light, however, we cannot see right through the galaxy
radio astronomy solves the problem through observations of the 21-cm wavelength radiation emitted by neutral hydrogen. This long wavelength passes unaffected through the gas and dust in the Milky Way, allowing us to see the distinct spiral structure which it possesses
the origin of spiral structure has two explanations. In the more massive galaxies, including our Milky Way, there can be persistent, long-lived density waves in the interstellar gas. These waves pass through the gas, compressing various regions in turn; where the compression is high, new stars form and the spiral arms are delineated by young bright blue stars and the gas clouds they ionize
less massive spiral galaxies arise because of stochastic (random) effects. One small episode of star formation can spark others when massive young stars explode as supernovae, sending out shock waves which compress other gas and precipitate new star formation. An extended region of star formation is then drawn out by the differential rotation of the galaxy to form a short-lived rather stubby spiral feature
the age of the Milky Way is known from the study of its globular clusters. These are the oldest known stellar constiutents of the galaxy. Moreover, their motions and the compositions of the stars within them both suggest that they formed as the galaxy itself was gravitationally contracting to its present flattened form. We deduce that the galaxy formed about 14 billion years ago
Associated Readings from the Text.Please look at: Chapter 19, pages 596-622. Chapter 21, pages 652-656.Observable Features and Obvious Questions.What we have known about the Milky Way for centuries is that it is a band of light across the sky, which immediate suggests something like a great disk of luminous material within which we are ourselves located. Eventually, Galileo showed that it consisted of many millions of stars, but could not yet answer some fairly obvious questions about the Milky Way as a whole. Where are we located within the Milky Way - at the center? Near one side? How large is the Milky Way? Why is the Milky Way flattened? What structure does the Milky Way possess? Does it, for instance, have a spiral form? When and how did the Milky Way form? Let us address these questions in turn.Our Place in the Milky Way.A couple of centuries ago, astronomers had views of the universe which were necessarily blinkered. One reason was that they had no way of recording images - no photographic plates, for instance - and could only sketch or report on what the eye saw as they peered through a telescope. Among other things, this limits the faintness of the stars which they could study because the eye does not accumulate light the way photographic films do (so that a longer exposure reveals ever fainter objects). This limited the kinds of astronomical research which could be done. Another important constraint was that the modern, technical science of astronomy was most developed in Europe (and later in North America), which meant that there were large parts of the sky, visible only from the Southern hemisphere, which were essentially unexplored. Remembering those biases, ask yourself how you might determine where we are located in what seems to be a big band of stars. Well, the analogy I developed in class was simple: I know that I am at the edge of the lecture hall because I see students ranged in front of me but none in behind me. A student in the very center of the lecture hall sees other students all around, in roughly equal numbers. Of course, if the student looks `up,' toward the ceiling, she sees no students up there; nor does she see any if she looks `down,' toward the floor. This tells her that she is in a flattened distribution of people, rather than a big ball-shaped distribution of them. To the unaided eye, the band of light making up the Milky Way looks comparably bright in varous directions (but not, of course, if we look away from the band, like the student looking `up' or `down.' This suggests that we may be near the center of the system of stars. The first to speculate on this matter in a modern way was Thomas Wright, who assumed (in 1750) that we were in the middle of what he described as a `grindstone' model of the Milky Way. This was not based on any serious observations, but was instead mostly conjecture.A Serious Objection.One serious objection could and should have been raised to Wright's model, even by astronomers in 1750. His model implied, once again, that we were at the center of things, in a special location. Since the time of Copernicus, this has been an unsatisfactory feature of several astronomical findings, and a neo-Copernican perspective would probably (and correctly) have led some scientists to express reservations over Wright's surprising conclusion. Instead, as we will see, others drew similar conclusions on apparently more stable scientific grounds.Herschel's Star Counts.William Herschel, around 1790, carried out the first really scientific exercise by doing star counts in various directions (again just by eye, but using a telescope). On this basis, he drew a picture of the Milky Way as he understood it to be. His conclusion was that we were located near the very center of a flattened system of modest size. Herschel even assumed that he could work out some of the detailed structure: for instance, he assumed that if you saw only a few stars in a particular direction, that is because there are not many stars to be found that way, and that you are peering right through them into the black void beyond. Consequently, he drew a sketch of a Milky Way with lots of "ins and outs" which we now know to be completely meaningless.The Kapteyn Universe.I digress for a moment to explain a small piece of nomenclature: in the early part of this century, astronomers sometimes used the word `universe' to refer to the Milky Way, partly because they believed that it was the unique large stellar system in all of creation. (This terminology persisted for a time: when recognized as such, the external galaxies were, for a time, called "island universes.") Modern astronomers do not use the term in this way: the universe is everything we can observe and have any knowledge of. When we refer to the `Kapteyn Universe,' we really mean the interpretation of the Milky Way which Kapteyn derived. How did he do this? Working in 1910, Kapteyn was the first astronomer to use the new scientific tool of photographic plates to study the stars. He took many photographs of the sky, looking in varied directions, and counted stars to try and get an idea of how uniform the distribution was. The answer, again, was that we seemed to be located near the center of a rather small stellar system, a few thousand light years across. We now know that this is wrong, but why did he get such an answer? The most important reason is that he and other astronomers were simply not aware of how pervasive the interstellar dust was, material which makes stars look too faint. When he looked in some directions, he saw few stars because the light from the more remote ones had been dimmed by interstellar dust. Since the gas and dust is more common near the center of the galaxy, as we now know, this gives a very strong systematic bias to the interpretation, and Kapteyn's answer was forgiveable given his assumptions about the transparency of space - forgiveable, but quite wrong.Shapley and the Globulars.The person who finally got the right answer was an astronomer named Harlow Shapley. He realised, form his interest in globular star clusters, that he could see more of these interesting objects from the Southern hemisphere than from the North. (Our galaxy contains about 150 globulars. When you are in Chile or Australia, there are literally dozens of them overhead at times, although to the unaided eye only one or two of them show up as tiny fuzzy patches. From Kingston, we can conveniently see only a couple of dozen of them.) Shapley reasoned that one hundred globular clusters, each containing thousands or millions of stars, are more significant than the single star called the sun. They are located in a huge `halo' which surrounds the Milky Way, and the point which marks the middle of their broad distribution is probably a better indicator of the center of the Milky Way than anything else. Shapley concluded that we are actually tens of thousands of light years from the center. He was correct, although in fact he somewhat overestimated the distance for a variety of reasons. But the real significance of his discovery was that it resolved the question of why we should, by chance, find ourselves at the very center of things. The answer is that we are not!The Flattening of the Milky Way.Why is the Milky Way flattened? Well, why is the Solar System flattened? Why do the planets all move in roughly the same plane? As we saw earlier, the flattening of the Solar System is a by-product of how it formed: it flattened out as it collapsed gravitationally, spinning faster thanks to the Conservation of Angular Momentum. Likewise, pizza dough flattens out as we spin it, and the most likely explanation for the flattening of the Milky Way is that it is in grand rotational motion, like a huge scaled-up version of the Solar System. There is one important distinction, by the way. In the solar system, the planets, comets, and asteroids orbit the single dominant lump of matter called the Sun. In the Milky Way, there is no huge central object (although there may be a fairly massive black hole or a rich star cluster right near the center). A star like the sun is moving around the galaxy thanks to the combined gravitational influence of many millions of stars, piles of gas and dust, and so on - not simply in response to the gravitational tug of a single identifiable central object. But how could we test this thinking, to see if the stars in the Galaxy are on the move in a way which explains the flattening? You may recall that we asked similar questions in Physics 015, trying to think of ways in which we could test whether or not the Earth actually moves around the sun. Those techniques will not work here - since the sun goes around the Milky Way once every couple of hundred million years, for instance, we cannot look for parallax effects which come and go thanks to our changing `back-and-forth' position in space. The answer came from an astronomer named Oort, who realised that the pattern of motions of the stars around us should give us the clue we need. To help you understand this, I considered the following situation in class: suppose the sun and all stars were emitting no light whatsoever, so that we saw no background of stars, no stellar parallax, or any of these things. (Of course, we would die from the lack of heat, but forget that for now!) Could we prove to our satisfaction that the Earth is moving in an orbital path around a massive lump of invisible matter? The answer is yes, and to prove it does not even require us to look outside the solar system. With radio telescopes and radar signals, we could detect the presence of the other planets - Mercury, Venus, Mars, etc - and work out that they are moving in space. (We don't need to see them to do this, any more than an air traffic controller needs to see the planes approaching the airport. An 'echo' on the radar screen is all that is needed.) More to the point, we could work out that Mercury moves faster than Venus, and that Mars moves slower still; and we could work out that Mercury and Venus overtake us `on the inside track' while we overtake Mars and the outer planets. Correcting for the continuous motion of the Earth would be difficult but feasible, and eventually we would correctly deduce the entire structure of the Solar System, including the presence of the massive central object which controls all the orbits. Oort did something similar, by looking at which stars are `catching up to us' and which ones we are `catching up to' as the Sun and its stellar neighbours orbit the center of the Milky Way galaxy. In essence, the stars closer to the center of the galaxy overtake us "on the inside track" while we overhaul those farther out. This systematic effect is called shear, and arises because the galaxy is in differential motion (the stars nearer the centre move faster, just as Mercury moves faster than the Earth in its orbit around the sun). Oort did this work in the 1920s, with an immediate interesting and convincing consequence. Although he did not identify any object which marked the very center of the Milky Way, he was able to say in what direction we would have to look to face the center. That direction turned out to be the same as Shapley had deduced from the globular clusters -- a persuasive finding.The Structure of the Milky Way.Is the Milky Way a spiral? This question is in fact a rather modern one, because even in the early part of this century it was not clear that the spiral `nebulae' were even galaxies at all! (Some people thought that they might be solar systems in formation.) But with the wisdom of hindsight we can see that there are lots of pinwheel-like galaxies out there in the universe, and ask if ours is like that. There are reasons for thinking it might be: our galaxy contains lots of gas and dust, and so do the other `pinwheels.' But the problem is that we are right inside it, and the interstellar medium prevents us seeing much of the structure. Really we are like a person who has been set down in the middle of downtown London, in the midst of an English fog, and asked to draw up a diagram which shows the grid of London streets. We can see only so far. The first success came with the recognition that the spiral arms in other galaxies are prominent mostly because they are the regions in which stars have most recently formed. This means that the arms are beautifully delineated by hot young stars, clouds of ionised gas, and so on. So in 1952 a group of astronomers plotted the positions of the young stars and gas clouds which we can see near the sun. What immediately shows up is that they are not distributed at random, but rather in a pattern which is clearly comparable to part of the spiral structure we see in other galaxies. The other technique is to rely on observations made with radio telescopes, studying the 21-cm radiation given out by neutral hydrogen. Since the gas is most abundant in the spiral arms, and since the photons can get right through the interstellar dust, we can study the structure of the galaxy on very large scales, even detecting the gas on the far side of the Milky Way. In this way, we have mapped out the details of the structure of the Milky Way, which is unquestionably a spiral. We also have a pretty good understanding of what gives rise to the spiral structure (see pages 617-619). We now understand that there are two kinds of spiral galaxies: 1 'grand design' spirals, of which our galaxy is one, are those which are created by persistent density waves that travel through the interstellar gas. Where the gas is compressed to higher density, star formation takes place, and the hot blue stars and conspicuous ionized gas clouds associated with the star formation show up like beads on a string, delineating the spiral structure beautifully. 2 'flocculent' spiral structure is seen in lower-luminosity galaxies. In these systems, an episode of localized star formation leads to the quick appearance and death of a number of quite massive stars (along with a host of smaller ones which will last much longer). The death of a massive star is marked by a supernova, and the expanding shell of gas can lead to a compression of other gas in the vicinity -- and a subsequent episode of yet more star formation. In this way, one random star-formation region can spawn a few others in quick succession in its vicinity, and the differential rotation of the galaxy can smear out this conspicuous region into a stubby, short-lived spiral arm. Such galaxies lack the overall symmetry and imposing appearance of the 'grand design' spirals, since maintaining a big density wave pattern requires a fairly large threshold mass.When and How Did the Milky Way Form?We know the age of the Solar System because we can determine the age of the oldest rocks we can find within it - meteors and lunar rocks - and we believe that they were formed at the time of the origin of the Solar System itself. Can we do something similar in the Milky Way? Well, one complication is that the sun and planets formed a long time after some of the other parts of the galaxy (like the much older globular clusters), so we have to look at other objects. Since we have no direct samples for radioactive age dating, we can only rely on studying the light from the stars and gas. What can we deduce? There are some pitfalls to be aware of, best understood with a simple analogy. Suppose you wanted to know how long the human race has existed. It would obviously be a mistake to conclude that we have been on the planet for exactly 130 years, say, merely because you find the single oldest person on Earth to be a 130-year-old Russian. We have to remember that there have been many earlier generations of people, none of whom are still alive, so we must rely on the historical, archaeological, and fossil records to understand the true story. In the astronomical context, we start by identifying stellar populations of varied ages. We know the age of the sun and the Solar System very directly, and we can work out the ages of various clusters of stars by looking at the 'turnoffs' in their HR diagrams. (Look back at section 13 to remind yourself how this works!) In this way, we find a variety of stellar objects that span a range of ages, with the globular clusters the oldest of all, having ages near 14 billion years. This might prove insufficient, because for all we know there were even earlier generations of stars which have long since evolved away completely to invisible neutron stars, black holes, and inconspicuous white dwarfs, leaving no distinctive luminous tracers in the present day. But the globular clusters behave in a special way which reassures us that they were formed at or just after the time of formation of the galaxy itself. This special behaviour has to do with their motions, which are quite distinct from the simple orbital motions of the stars which are moving in the flattened plane of the galaxy. The stars in the globular clusters are also distinctive in composition, in an equally telling respect. Here is how the reasoning goes: The stars in many of the oldest globular clusters were formed from a gas which was almost pure hydrogen and helium, with only a tiny amount of heavier elements. We learn this from a study of the spectra of the stars in the cluster. (Deep within the stars, hydrogen is being converted to helium, and small amounts of heavier elements are being formed, but this does not affect what we see at their surfaces.) This is consistent with these being `first-generation' stars which formed out of the material which was created in the Big Bang, the origin of the universe. Moreover, the globular clusters are moving in great `loopy' orbits around the galaxy itself, rather than sharing the motion of most of the stars in the flattened plane of the Milky Way. For reasons I will explain in a moment, this is evidence that the globular clusters formed as the galaxy itself was gravitationally collapsing. There is one other piece of evidence: there is independent evidence (from the expansion of the universe) suggesting that the Big Bang itself occurred about 15 billion years ago, in which case the galaxy and the globular clusters must have formed fairly promptly after that. So this all hangs together, and we have now developed a coherent picture of the formation of the Milky Way galaxy itself about 15 billion years ago. (The sun formed only about 4.6 billion years ago, long after successive generations of massive stars had formed and thrown out lots of the heavy elements which are found in our bodies and in the planets.)The Motions of the Globular Clusters.The huge gas cloud which was eventually to become the Milky Way galaxy began its gravitational contraction, and some stars and clusters formed here and there. As the collapse continued, atoms of gas ran into one another and were slowed down by something analogous to `air resistance', which, coupled with the rotational motion, is why the cloud flattened out into a plane. But the denser stars and clusters, having formed so promptly, were unaffected by the air resistance and continued along the paths they had been on as the collapse progressed. A simple analogy emphasises the importance of the air resistance for low-density objects and its irrelevance for denser objects. Imagine flinging handfuls of metal bolts and small scraps of paper across the room. The paper bits would be quickly slowed to a halt and flutter to the ground; the bolts would continue unimpeded across the room. In like fashion, any stars (or clusters of stars) that formed early on would retain their original motions and wind up in 'loopy' orbits, reminiscent of the complex disordered behaviour of the distended cloud of gas within which they formed. By contrast, the individual atoms of gas, through collisions with one another, would quickly flatten out into a rotating flattened plane - the precursor of the disk of the Milky Way. Later generations of stars, including the Sun, would later form in this disc-like system. That `fossil evidence' of their motions so many billion years ago persists to this day, and the motions of the globular clusters reminds us that they must have formed before there had been any significant flattening at all - truly at the very early stages in the life of the galaxy.Summary.The globular clusters are believed to date from the origin of the galaxy because: Their stars were formed from essentially pure hydrogen and helium. Their ages are comparable to the `expansion age' of the universe deduced from the Big Bang. Their motions in the galaxy tell us that they must have been formed just as the galaxy itself was. In other words, these beautiful star clusters are essential tools in helping us look back to the very origin of galaxies (and in my own research, I use them for this purpose in galaxies other than the Milky Way). Previous chapter:Next chapter0: Physics 016: The Course Notes, spring 2005. 1: The Properties of the Sun: 2: What Is The Sun Doing? 3: An Introduction to Thermonuclear Fusion. 4: Probing the Deep Interior of the Sun. 5: The Sun in More Detail. 6: An Introduction to the Stars. 7: Stars and Their Distances: 8: The HR Diagram: 9: Questions Arising from the HR Diagram: 10: The Importance of Binary Stars: 11: Implications from Stellar Masses: 12: Late in the Life of the Sun: 13: The Importance of Star Clusters in Understanding Stellar Evolution: 14: The Chandrasekhar Limit: 15: Supernovae: The Deaths of Massive Stars, 16: Pulsars: 17: Novae: 18: An Introduction to Black Holes: 19: Gravity as Geometry: 20: Finishing Off Black Holes: 21: Star Formation: 22: Dust in the Interstellar Medium: 23: Gas in the ISM: 24: The Size and Shape of Our Galaxy: 25: The Discovery of External Galaxies: 26: Galaxies of All Kinds: 27: The Expanding Universe: 28: Quasars and Active Galaxies: 29: The Hot Big Bang: 30: The Geometry of the Universe: 31: Closing Thoughts: Part 1:Part 2:Part 3: |
(Wednesday, 22 April, 2026.)
