The Discovery of External Galaxies:
A Point-Form Summary.
This section of the course notes, and the associated PowerPoint presentation, makes the following critical points:
a little over a century ago, there was uncertainty about the nature of the spiral nebulae which we now know to be galaxies like our own. Some astronomers believed them to be objects like solar systems in formation; others believed them to be rather remote
the problem would be easily resolved if the distances could be measured, but this was not easily done. The nebulae are too far away for direct parallax measurements, so astronomers had to use other methods. The obvious one is to compare the brightness of a star in the nebula to that of a nearby star, but this has its own problems. In the first place, the nebulae are so remote that only the very brightest giant stars would show up at all, at the very limits of the observing techniques of the time; moreover, how can you be sure that the dot of light you detect is a star at all, and not (say) some tiny planet in formation?
the answer came with the Cepheid variable stars, which behave in a distinctive way which tell us that they are stars for certain
Cepheid variables pulsate in and out, like beating hearts, and consequently vary in brightness as they do so. (Their surface temperatures also change as they pulsate.) Cepheids are rare, both because they are massive stars (which are less common than low-mass stars) and because they are in short-lived late stages of their lives
the rarity of Cepheids means that they are spread out in the Milky Way, even though there are many of them, and thus it is hard to intercompare their properties. But there are thousands of them in the nearby Large and Small Magellanic Clouds (small satellites of the Milky Way). In the Clouds, the Cepheids are all at roughly the same distance from us, and they can be readily intercompared
a study by Henrietta Leavitt revealed that the Cepheids in the Clouds obey a Period-Luminosity law: the long-period (slower-pulsating) Cepheids are brighter than their more rapidly pulsating brethren
in external galaxies, therefore, Cepheids show up by (a) being bright supergiants, and (b) by drawing attention to themselves by varying in a conspicuous way. A coordinated study then allows us to determine their periods, and tells us whether we have found an extermely bright long-period Cepheid or one which varies a bit more quickly and which is therefore a bit less luminous. We can then ask how far away these Cepheids must be to look as faint as they do
after Henrietta Leavitt discovered the P-L law, it fell to Hubble to make the crucial discovery of Cepheid variables in some of the nearby nebulae (including M31, the Andromeda galaxy). This proved that the nebulae were millions of light years away, and consequently huge -- as large as our own Milky Way
this discovery was regretted by Shapley, who had shown that our own galaxy was enormous but who believed that the spiral nebulae were insignificant by comparison. Part of his resistance, however, was due to the fact that he and Hubble were very different, and did not particularly like each other. Hubble was a very interesting character about whom much has been written, although the adulatory stories coming from his widow should be viewed with great skepticism. Still, his astronomical contributions were enormous
Associated Readings from the Text.Please look at: Chapter 20, pages 626-636 in particular.What Are They?By the early years of this century, astronomers were able to take photographs of any number of nebulae (= clouds, in Latin) in the universe. The big question was `What are they?' Speculation ranged widely: some thought them to be galaxies comparable to our own, while others thought they were small objects within which a central star was in the process of formation, surrounded by a small blobs of light which might be young planets. The way to resolve the problem, of course, is to determine the distance of at least one (and preferably a lot of) these nebulae. If they turn out to be very far away, perhaps millions of light years - as we now know they are - then they must be very large objects, and extremely bright to show up at all. (Galaxies are hundreds of thousands of light years in diameter and emit the combined light of billions of stars.) But if they are nearby, they are small and may not be much brighter than a typical star. So how do we deduce the distance? These objects, whatever they are, proved to be much too far away for ordinary parallax measurements. What we need to do is identify something within a nebula, preferably a star of some kind which can be directly compared to a nearby counterpart of known distance. The ideal, for instance, would be to be able to point to a spot of light within a nebula and say: ``There! That object is exactly like the sun in every respect, except that it is fainter. We can now deduce how far away it must be to look so faint.'' This is simple in principal but difficult in practice, for a couple of reasons: The nebulae are so fantastically far away that even today the Hubble Space Telescope cannot even detect stars as bright as the sun inside the Andromeda galaxy (the nearest big spiral galaxy). So at the very least we will have to try to identify and study stars which are intrinsically much brighter. Fortunately, there are many giant and supergiant stars which are bright enough, although they were at the bare limits of detectability for astronomers early in this century, given their primitive photographic techniques. Even if you can point to a spot of light in a nebula and say ``That's a star,'' nothing prevents a competitor astronomer from saying, ``No, I believe that that is a small planet in the process of formation.'' You need proof of the nature of the object, but how will you get it? Ideally, this would come in the form of a spectrum which would give you some notion of the nature of the thing you are looking at - is it a gas cloud like the Orion Nebula? Is it a star? But again technology defeats you: in the early 1900s, it was scarcely possible to get a direct image of a remote nebula, a picture in which all the light is brought to a common focus. If the light from a single dot of light within a nebula were to be spread out into its separate wavelengths to make a spectrum, it would be much too thinly spread to be detectable. So the ideal is to recognize, within a nearby galaxy like the Andromeda spiral, a point of light which is demonstrably a star but which you recognize as such without getting a spectrum. Moreover, you can only consider stars which are intrinsically very bright. Are there targets which allow this? Remarkably enough, there are: the Cepheid variable stars.Cepheid Variables.Cepheid variables are ideal for our purposes in the following ways: They are very bright, so can be seen from a long way off. They vary in brightness in a way which draws our attention to them (so we don't waste time on other objects). The way in which they vary tells us something about how bright they are intrinsically, as I will explain, so they are useful distance indicators. On the other hand, Cepheids are massive stars, which means that they are few in total number. (Remember that massive stars are rare, while low-mass stars much more common. For every million sun-like stars in the galaxy, there may be only one massive enough to become a Cepheid.) For this reason, even the closest Cepheid to us is likely to be far away, and they are hard to study in great detail. Massive stars become Cepheids only for a short time, near the ends of their lives as they are starting to run out of fuel in their centres. This means that the chance of seeing a star of the right mass in this particular phase of its life is very slim, further reducing the total number of Cepheids that are around at any given time. (I used an analogy in class: if you take a picture of a crowd at a football game, what is the likelihood that you will have caught someone in the middle of a sneeze?) On the positive side, when we look at the Andromeda nebula we are looking at a collection of many billions of stars, some fraction of which are Cepheids which are bright enough to show up even in our photographs. They vary in brightness, so drawing attention to themselves when we intercompare photographs taken at different times. And their behaviour and brightness allows us to determine how far away they are, as I will explain in a moment. It was in this way that Edwin Hubble first proved that the Andromeda spiral, and other galaxies, were comparable in size to our own and at vast distances.What Are Cepheids Doing, and Why?Cepheid variable stars are actually pulsating, growing alternately bigger and smaller rather like `beating hearts.' As their sizes change (they get bigger and smaller by about 10 percent in radius), their surface areas change, allowing more light to flow off the star, and their temperatures also change somewhat. We understand why this happens - it has to do with the atmosphere of the star, not any turning on and off of the nuclear reactions deep inside it - but the important feature for our purposes is that they behave in a very characteristic way. A typical Cepheid gets brighter fairly briskly, but then fades away somewhat more slowly, giving a recognizable `light curve.' Any given Cepheid behaves very predictably, repeating the cycle over and over, but different Cepheids have different periods. Some vary up and down every few days; some take a couple of months to do so, so are much more ponderous and deliberate in their cycle. What determines these differences?The Period-Luminosity Relationship.When we find Cepheid variables in our own galaxy, it is typically very hard to determine their distances in any direct way. They are rare, as I noted, so even the closest ones are too far away to study by parallax methods. Moreover, their remoteness means that we often see them behind great amounts of interstellar gas and dust, which confuses any estimates of their properties (like their true brightnesses). If one Cepheid looks brighter than another, it may be because it really is intrinsically brighter, because it is closer, or because the other one is hidden behind a lot of interstellar smog - or any combination of these! Clearly, this state of affairs makes it challenging to work out the fundamental properties of Cepheids. Or does it? Not necessarily! Suppose we had a sample of many Cepheids, all at a common distance from us and all seen through roughly the same amount of intervening smog. Surely then you could make some reasonable intercomparisons: for instance, the Cepheids which look the brightest really are. (This is precisely the point I made earlier about the importance of using star clusters in the study of stellar evolution.) It was with this simplification in mind that Henrietta Leavitt studied a whole series of photographic plates of the southern galaxies the Large Magellanic Cloud (LMC) and the Small Magellanic Cloud (SMC). (Within each of these galaxies, by the way, we have now studied more than one thousand Cepheids.) Studying these plates, Leavitt was able to determine the periods of a whole bunch of Cepheids - some vary every 10 days, some every 25 days, and so on - and the average brightness of each one. (Of course, a given Cepheid goes up and down in brightness, but one can still work out a good average level.) When she intercompared these numbers, she discovered the Period-Luminosity Law which is shown in your text on page 505, one which states that the long-period Cepheids are brighter than the short-period Cepheids. Their predictability and reliability make them useful distance indicators.Determining Distances with Cepheids: Mixed Blessings.If I had been in charge, Cepheids would have been made even simpler and more useful! In the perfect world, all the Cepheids would be exactly the same intrinsic brightness, regardless of the periods. This would have made our job much easier. In a remote galaxy, we would merely need to look for the variation which allows us to recognize a Cepheid (it `waves' at you by flashing away in distinct fashion). We would then compare its observed brightness to the known absolute brightness of the Cepheids close to us (those whose distances can be deduced by other means) to derive its distance. The actual Period-Luminosity law is more complicated, but still straightforward. It works as follows: We find a Cepheid in a remote galaxy by noticing the variation on a series of plates or images taken with electronic detectors. (These observations have to be spread out over many weeks or months.) The characteristic up-and-down in the brightness tells us it's a Cepheid for sure. We study the light curve to decide whether it is a long-period Cepheid (which immediately tells us that it is a brighter than average Cepheid) or a short-period Cepheid (a relatively faint one). In other words, the period alone tells us what kind of Cepheid this is. We then, in effect, compare the average brightness of the new-found Cepheid to that of a nearby example of the same kind, and derive a distance. People are often perplexed by the use of Period-Luminosity law, so let me restate the critical elements: The fact that we can detect variable stars in external galaxies is important because this gives us reassurance that we know what we are studying. (If the dots of light were actually planets in the process of formation, they would not be varying in ways which, by chance, mimic the characteristic light curves of Cepheids!) The variability also draws our attention to the most useful objects for careful scrutiny and measurement. There exist Cepheids with a wide range of periods. They could, in principle, also have had a wide range of intrinsic brightnesses, perhaps with no rhyme or reason. (For instance, there might have been one Cepheid with a period of one hundred days which is a thousand times as bright as another Cepheid with the same period.) The Period-Luminosity law tells us that this is not the case: Nature treated us more kindly than that! When you find a Cepheid with a period of one hundred days, it is bound to be exactly the same brightness as every other Cepheid with that period. If we know the distance to a local Cepheid with a period of one hundred days, then we can intercompare brightnesses to find the distance of the newly-discovered one in the remote target galaxy. Similar reasoning holds for Cepheids of every period: we simply have to calibrate the local relationship (find out how bright Cepheids of different periods are) and then intercompare like with like whenever we find new Cepheids in remote galaxies.The Great Debate.The discovery of the Period-Luminosity Law for Cepheids was in fact used by Shapley, who used Cepheid variables in a way I did not go into when he was trying to figure out how far we were from the center of our own Galaxy. But Shapley did not know about the Cepheids in external galaxies, which were only discovered some years later (his work on the Milky Way dates from about 1918; Hubble's discovery of the Cepheids in Andromeda from about 1925). In fact, despite the fact that Shapley believed in and indeed proved the notion of the large size of our own galaxy, he did not believe the external galaxies were comparable. In fact Shapley thought the spiral nebulae were little objects, small satellites of our own Milky Way, at some modest distance and considerably smaller than our galaxy. (He was right about the LMC and SMC: these really are satellite galaxies of the Milky Way.) There was even a famous public debate (which has been much exaggerated in importance over the years, however) between Shapley and an astronomer named Curtis, a debate which Shapley seemed to win at the time. In the long run, however, Shapley was proven wrong by Hubble.Shapley versus Hubble.There is a lot of fascinating history in the story of these two men, who deeply disliked each other. Shapley was a `country hick' from small-town America, with no pretensions. Hubble was the exact opposite, in many respects. For instance: Hubble was a Rhodes Scholar at Oxford, and thereafter adopted a phony British accent which disgusted Shapley. Hubble got a job at Mt. Wilson Observatory (in Los Angeles) after he graduated, but he turned it down to go to war (WW I) where he won a medal. Hubble was -- according to his wife, who almost literally worshipped him -- a great athlete who at one time even entertained thoughts of working towards the heavyweight boxing championship of the world (if you can trust her adulatory biography of Hubble) When Hubble worked in Los Angeles, he hob-nobbed with many Hollywood personalities. He was said to be so ruggedly handsome that many male movie stars would avoid being photographed with him. Given this background, you can understand that Shapley was particularly distressed to learn that Hubble had discovered Cepheid variable stars in some of the nearby galaxies, proving once and for all that they were immense star systems comparable to our own Milky Way - and that Shapley had been wrong. 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.)
