Science Talk

What's The Matter?: Cold Dark Matter and the Milky Way's Missing Satellites

In this episode, Scientific American editor George Musser talks with Caltech Astronomer Josh Simon about dark matter, and about the efforts to try to locate the so-called missing satellites of the Milky Way--small galaxies that have yet to be found in the numbers that the cold dark matter theory predicts. Plus we'll test your knowledge of some recent science in the news. Websites mentioned on this episode include:;

Steve: Welcome to Science Talk, the weekly podcast of Scientific American for the seven days starting January 23rd, 2008. I'm Steve Mirsky. This week: a walk on the dark side. We're going to hear about dark matter from Caltech Astronomer, Josh Simon. Plus, we'll test your knowledge about some recent science in the news. The annual meeting of the American Astronomical Society took place earlier this month in Austin, Texas. Scientific American's expert astronomy editor, George Musser, attended and he got a chance to speak with Josh Simon, the Robert Millikan fellow—to oil drop a name—at the California Institute of Technology. You've got your regular matter and your dark matter, which physicists and astronomers think makes up most of the stuff in the universe. Of course, we're not even talking about the dark energy here. For a two-page primer on the whole dark matter–dark energy situation, you can check out our SciAm "Ask the Experts"feature we ran back in August 2006. It's at, or just jump right in with the upcoming interview. Anyway, the dark matter can't be seen—that's why they call it dark—but it can be divided into the hot dark matter and the cold dark matter. The hot dark matter would be particles traveling super fast and would explain how big clusters of galaxies could've formed, but because the particles go so fast it cannot explain how individual galaxies got a chance to form. For that you need the cold dark matter. This chilly dark matter would move slowly enough to allow individual galaxies to clump together, but the cold dark matter hypothesis faces some conflict with observations. George Musser spoke with Josh Simon about dark matter as well as what's its like to actually do astronomical research. The first voice you'll hear is Josh Simon.

Simon: We have a theory right now called cold dark matter, which describes the structure of the universe on large scales and how galaxies form within that structure; and so when you look at the universe, the distant universe, then the predictions of that theory agree very well with the observation, it's really remarkable. The problem is that when we actually zoom in and look at individual galaxies in detail, then we start to find some discrepancies between what the theory predicts the universe should look like and what we actually see, and so one of the more serious of these discrepancies is called the missing satellite problem. The prediction of the theory is basically that a galaxy like the Milky Way, which we live in, should be surrounded by hundreds of smaller galaxies, which we call dwarf galaxies; and when we actually go out and look for these dwarf galaxies around the Milky Way, until a couple of years ago, we had only identified 11 of them.

Musser: Can you give us a sense of the size of our Milky Way and what the size of one of the dwarf galaxies would be?

Simon: Sure. So, the Milky Way is, you know, tens of thousands of light years across. It has a total mass of approximately a trillion times the mass of the sun. Dwarf galaxies, there isn't a strict definition, but in general you could say a dwarf galaxy is something smaller than about a tenth the size of the Milky Way and then they go on down essentially as far as you want to go from there. So the dwarf galaxies that we know about tend to range from about one millionth the size of the Milky Way up to a tenth.

Musser: And do they orbit around the Milky Way like moon is around a planet?

Simon: Yeah, exactly! Now most of the moons around planets tend to be in circular orbits and the dwarf galaxies around the Milky Way usually are in much more elliptical orbits, so they can come much closer into the center and then go out to large distances. That's not always the case. Some of them are in relatively circular orbits.

Musser: Can we actually see one with our naked eye?

Simon: Yes, in the Southern Hemisphere are the two brightest and most massive dwarf galaxies around the Milky Way, the Large and Small Magellanic clouds.

Musser: So if the numbers of these galaxies were as the theory predicts, how many would there be?

Simon: Well, that depends on what mass you want to go down to, but I think the minimum number you'd expect is a hundred, and there could be as many as 500 expected in theory. So, the difference between what the theory predicts and what we actually see seemed to be at least a factor of 10, and possibly closer to a factor of a 100. There was a really, really serious disagreement.

Musser: Now for [some] people this doesn't call the whole dark matter paradigm into question?

Simon: Well, some people did think that that was exactly what it is, not dark matter itself, but cold dark matter at least. What I have been trying to do is investigate the problem in more detail and see whether there's any way to resolve the problem so that, you know, the theory could still be correct; or if there's just no way to bring the observations and the predictions of the theory into agreement, in which case, you know, we would be in trouble and we would have to scrap the theory and find something else.

Musser: So what are some of the ideas that are circulating around to try to solve the problem?

Simon: Well there have been a lot of suggestions for what could be going on. Basically they fall into three categories. So, one possibility is simply that the theory is wrong—dark matter is not cold or if you want to go even more radical, that dark matter doesn't exist, although there are not too many people who advocate that point of view. So if the cold dark matter theory is wrong, then there is no reason to expect its prediction that there should be hundreds of these dwarf galaxies to actually be satisfied by the real universe. A second possibility is that there really are hundreds of dwarf galaxies orbiting the Milky Way and we just hadn't managed to find most of them. So, I and various other people had tried to do searches for new dwarf galaxies to see, if maybe for some reason, we had just missed a large fraction of the population. And then the third possibility is that these hundreds of low-mass dark matter halos, these clumps of dark matter, really are out there orbiting the Milky Way, but for some reason, most of them happen not to form any stars, and so we wouldn't be able to see them.

Musser: So they just would be big puddles of dark matter?

Simon: Exactly! There would be these completely dark clouds of dark matter, and nobody has yet come up with a way of definitively proving that such objects are out there; but it's not completely impossible, yeah, I would say.

Musser: Where do people's preferences lie today? What's the, kind of, theoretical thinking?

Simon: Well, I think the theory has always leaned towards the final possibility that these small dark matter halos are all out there and they're just not visible because they didn't form enough stars for us to be able to see them and you know, they came up with a variety of explanations for why these halos might not have been able to form so many stars. There's been, I would say, a small community of people who lean towards the first possibility—that the cold dark matter theory is wrong, and that's certainly a sufficiently important possibility that we absolutely have to test it. But I would say that that was not where the majority of the community thought that the answer would lie.

Musser: So that leaves option number two?

Simon: (laughs) Well, so that seemed like an obvious possibility and in fact, surveys recently, most notably the Sloan Digital Sky Survey, have in fact uncovered a population of very faint dwarf galaxies, much fainter than the faintest objects that we had previously known about.

Musser: What was special? Why did Sloan find it, whereas it had gone missing before?

Simon: So Sloan is simply more sensitive to these sorts of galaxies than any other survey has been. It covered a large fraction of the sky and observed for a long time, so it's able to see very faint stars. And then because the survey was digital, all the data are very easy to search through in an automated fashion, which makes it much easier to go out and identify these dwarf galaxies. In the past, people had tried to look for dwarf galaxies by examining photographic plates by eye, and scanning across them for faint balls of stars that could be dwarf galaxies.

Musser: Going blind in the process, however, I guess?

Simon: (laughs) Possibly. I don't know of any actual cases where that happened, but certainly it's a bit of a strain on the eyes.

Musser: So what precisely was the result that you presented?

Simon: There was this set of new dwarf galaxies discovered by three different teams of researchers using the Sloan data. They have now found 12 new dwarf galaxies around the Milky Way. Remember previously there were only 11 dwarfs that we had identified in the previous, you know, whole of human history. So, this was a big new population and because, you know, people had been wondering for many years whether such a population might exist, we thought it was obviously of importance to follow up on these new discoveries and try to ascertain the true nature of these systems and what their connection to the missing satellite problem would be. So my collaborator, Marla Geha from Yale University, and I used the Keck telescope last February to observe stars in most of these newly discovered dwarf galaxies.

Musser: Tell us a bit about Keck. What was it like to use it and physically what's it like, the biggest telescope in the world?

Simon: Yeah! So Keck is actually two telescopes, they are two 10-meter telescopes, which are currently the largest optical telescopes in the world. They're located on the mountain Mauna Kea in Hawaii. And so we were using one of the telescopes, actually the second one—Keck 2; each telescope has its own suite of different instruments, and so they kind of specialize in different areas. So, you know, we flew out to Hawaii. I have to admit I've never actually been up to the top of the mountain where the telescope actually sits; the control room for the telescope is down in the town on Waimea, which is not on the mountain. You can kind of see the telescopes up on the top of the mountain when the weather is clear, but just because the air is so thin up at 14,000 feet where the telescopes are located, they decided that in general it wasn't a good idea to have the astronomers, who were trying to make important decisions during the night actually up there.

Musser: So you actually went out there, did the observations. How long were you there for; what did it involve?

Simon: This was a three-night observing run. And so we were out there for, I guess, about five days, one on either end; and so we would observe about a hundred stars at a time in each one of these dwarf galaxies. Now of course many of the stars that we looked at turned out to just be foreground contaminants, they're stars in the Milky Way that happened to vibrate in front of these dwarf galaxies; and just with the Sloan Digital Sky Survey data you can't precisely separate the Milky Way stars from the dwarf galaxy stars. So that was one of the things we had to do. So we obtained spectra of these stars; there were over eight hundred stars in total that we observed and the spectrum allows us to measure the velocity of the star very precisely. So stars in the Milky Way follow a relatively consistent pattern in terms of their velocities because Milky Way is a nice spiral galaxy, everything is rotating around the center of the galaxy.

Musser: Like a big pinwheel.

Simon: Exactly; and so when we look in any particular direction, we know about what the velocity of the Milky Way star should be, and then if we see a lot of stars that are at a different velocity and all that at a very similar velocities to each other, then we know that those must be associated with a dwarf galaxy.

Musser: And what did you find in the end?

Simon: So what we did was we measured the velocity dispersion of the stars.

Musser: Dispersion?

Simon: Yeah, so [we measured] the range of velocities that is present in each of these dwarf galaxies, and we found that they have typical velocity dispersions of about 5 kilometers per second. So if you pick any random star within these dwarf galaxies, there's a good chance that it has a velocity within about 5 kilometers per second of the mean velocity of the entire galaxy, and so what that allows us to do is measure the mass of each of these galaxies. Now originally, when these systems were discovered, it wasn't clear that they were galaxies at all; they could have been globular clusters of stars which do not contain any dark matter in them, and so the only way to distinguish between those possibilities was to measure the velocity dispersion and see if the velocity dispersion indicated that there was dark matter in these systems.

Musser: So you're using the star's motion to infer the unseen dark matter?

Simon: Yes. That's exactly what we're doing. The measurement of the velocity dispersion, being about 5 kilometers per second, told us that there must be extremely large quantities of dark matter in these galaxies. If they had been globular clusters without dark matter, then the velocity dispersion would have been perhaps a tenth of a kilometer per second, so a factor of 10 or 100 lower than what we actually measure. So, it's a very clear result that they do have dark matter in them and the dark matter is in every case at least a factor of a hundred more than the combined mass of all the stars.

Musser: How does that compare to our own galaxy?

Simon: Normal massive galaxies like the Milky Way tend to have a ratio of, you know, 10 or 20:1 of dark matter over normal matter, and integrated over the entire universe, there's about six times as much dark matter as there is normal matter, so these tiny dwarf galaxies turned out to be the most dark-matter dominated galaxies that have ever been found. So by measuring the masses of these galaxies, we were then able to study them in the context of the missing satellite problem because we could go down to a specific mass and say, we know exactly how many dwarf galaxies there are that have masses of at least a million or 10 million times the mass of the sun. And that's exactly what the quantity that is predicted by the cold dark matter theory and so then we could compare on a hopefully one-to-one basis, our observations of the dwarf galaxies around the Milky Way with what the theory predicts, and so what we found is that even though the population of the Milky Way dwarf galaxies has more than doubled in the past three years, the total number of dwarfs that is out there is still smaller than what is predicted by the theory, so the initial conclusion was, well there's still a problem there, we have not located all of the missing satellites.

Musser: You mean, because even 20 to 25 satellites is less than 500, which is what was expected.

Simon: Yes. The number that we found is still less than what the cold dark matter theory predicts. So then we started looking into several of the scenarios that had been proposed by theorists to explain why these low-mass dark matter halos might not have been able to form any stars, and so there is a large literature on various suggestions for why this could be the case and so we tested several of those possibilities and we found some of them didn't work and we found one explanation that actually brought the theoretical predictions into excellent agreement with the observed number of dwarf galaxies around the Milky Way.

Musser: And what was that?

Simon: That is that the effect of reionization, which occurred at a red-shift of approximately 10, you know, less than a billion years after the big bang was able to strongly suppress star formation in these very small dwarf galaxies.

Musser: So walk us through that a little bit more.

Simon: Well, so initially the universe was filled with neutral hydrogen. If you go back to the big bang, everything was ionized, then very shortly after the big bang, about 300,000 years that combined into hydrogen atoms and helium atoms and a very small fraction of everything else; and so then over the next hundreds of millions of years, the normal matter, the hydrogen and helium came together and began to form stars and galaxies and at some point in that process there was a huge amount of ultraviolet radiation released, probably by the first stars but perhaps also by massive black holes.

Musser: So the universe was totally dark and suddenly this radiation comes?

Simon: Yeah! So prior to that was what's been called the dark ages; there have been several recent articles, I think, in Scientific American about exactly that.

Musser: Oh no! Never.

Simon: (laughs) And so then the universe was suddenly or perhaps not so suddenly reionized and this reionization—let's say that you have a small clump of dark matter that could become a dwarf galaxy and it has a little bit of hydrogen and helium gas associated with it. As soon as it gets hit by this blast of ultraviolet radiation, that gas in that galaxy will become ionized, and it will get heated up; and it will not be able to then cool down and collapse and form stars, which is what ultimately begins the process of forming a galaxy that we can see today.

Musser: So this blast of radiation as you put it, basically burned off a lot of the useful gas that could have formed stars, leaving these clumps of dark matter star poor.

Simon: Yes. So, the radiation doesn't destroy the gas, but it heats it up so much that it's not able to cool over a useful time scale for forming stars.

Musser: And what, is that due to the whole missing satellite problem, then?

Simon: Well that means that there should be hundreds of these so called dark galaxies out there—these clumps of dark matter that don't have any stars in them—but that we shouldn't be able to see 500 dwarf galaxies around the Milky Way because most of them are simply weren't able to form stars before this blast of radiation and reionization hit them and removed their gas supply for forming stars in the future.

Musser: So with this, you find your stand correctly [corrected], that there is one Milky Way…

Simon: Yes.

Musser: …[a] couple of dozen or few more small galaxies buzzing around the Milky Way…

Simon: Yes.

Musser: … like bees around the hive. We can see them because they've got some stars in them, but they are mostly dark matter; and then there is an even larger number, maybe ten times as many, completely dark blobs of dark matter. Is that the picture I should have in my head?

Simon: Yes, that's what I would say is our current best guess for what the situation actually is and so now the challenge will be trying to get definitive evidence to confirm that, which could come from being able to actually identify some of these dark galaxies, to prove that they really not only can exist, but do exist.

Musser: I thought you did that.

Simon: We've studied the existing dwarf galaxies and we've shown that their properties are consistent with being the objects that did manage to form stars before reionization occurred and the natural assumption is that that leaves a much larger population of halos that did not form stars, but that does not constitute proof that those dark galaxies really exist.

Musser: How do we see them if they are dark?

Simon: (laughs) Well that's a problem that I've been struggling with for the last couple of years. One possibility is that even though they don't have stars in them, they could have gas in them and so that would make them visible via radio waves, probably from the hydrogen gas, which tends to be the easiest thing to see.

Musser: So even though they're not bright with stars, they are giving out a little bit of radio emission.

Simon: Yes. So that would provide one possible means for detecting them. Alternatively if you've got hundreds of these things orbiting around the Milky Way, some of them are going to actually hit the disc of the Milky Way and you might expect that they would have some observable effects on the matter in the Milky Way. Another possibility is that there are some dwarf galaxies right now—most notably the Sagittarius dwarf—that are being tidally disrupted by the Milky Way.

Musser: So torn to shreds, you're saying.

Simon: Yes exactly. So Sagittarius has this enormous stream of stars coming out from either ends of it, stretching across almost the entire sky. These are known as tidal streams, and again, if you have a population of hundreds of these small dark-matter halos orbiting around the Milky Way, some of them should hit the streams from the Sagittarius dwarf and again could distort the path that those streams are following, and we might hope to be able to detect that distortion. Oh! And one final possibility, I should mention, is that it's possible that dark matter is actually able to annihilate with itself, so when you have matter and antimatter and you put them together, of course, you get the particles annihilate[d] and you get a flood of gamma rays. So, depending on the exact properties of whatever kind of particles makes up dark matter, it may be that it can do exactly the same thing. And so one of the exciting possibilities in the next few years is that there have been several gamma ray telescopes that have recently been completed or will be very soon and they can go out and look for gamma rays coming from apparently blank regions of sky. So there has been considerable excitement in the theoretical physics community about the possibility of seeing gamma rays from these otherwise invisible clumps of dark matter.

Musser: So we would see what is dark in that case?

Simon: Yes. Even though there aren't any stars there.

Musser: Well, it has been great talking to you. I really appreciate you taking the time.

Simon: Thanks, it was my pleasure.

Steve: And my thanks to George and Josh. Again the SciAm dark matter premier is at and for more on Josh Simon, just Google "Josh Simon" and the word "astronomy". His Web site is the first thing that comes up. Lot's of good stuff on his side including many articles for lay audiences about dark matter as well as his peer-reviewed research publications.

Now it is time to play TOTALL……. Y BOGUS. Here are four science stories, but only three are true. See if you know which story is TOTALL……. Y BOGUS.

Story number 1: Japanese researchers are preparing a paper airplane that they say will survive a trip from the international space station down to the surface of the earth.

Story number 2: A survey of more than 250 kids in English hospitals found that none of them liked clowns, images of which often appear in kids' wards.

Story number 3: Lots of people have started wearing pedometers to encourage more walking. A new study finds that without also adopting a diet, people who just wear a pedometer as part of [the] walking program actually gained weight in the first three months, probably because they think the exercise allows them more food.

And story number 4: One of this year's Nobel Peace Prize–winning team of scientists had a talk about climate change canceled at a high school in Montana, because some members of community feared that the talk would be anti-agriculture.

Time is up.

Story number 4 is true. University of Montana, Steve Running, a lead author of the Nobel-winning IPCC climate report had a talk canceled at the local high school because some residents didn't like the subject matter and thought it might be anti-agriculture. If they think hearing about global warming is bad for agriculture, wait till they get a load of global warming.

Story 1 is true. University of Tokyo scientists along with members of the Japan Origami Airplane Association are developing a paper airplane that they think will make it from the space station down to earth in one piece. The origami glider will look like a miniature space shuttle and will be treated to withstand some of the frictional heat of re-entry, which should be much less than what the shuttle has to withstand as the light, small, paper plane would be moving far more slowly.

And Story 2 is true. The survey of kids did indeed find that they don't like clowns. In fact they tend to find clowns kind of creepy. For more, check out the January 17th edition of the Daily SciAm Podcast, 60-Second Science.

All of which means that story number 3 about pedometer wearers gaining weight is TOTALL……. Y BOGUS. Because people in a pedometer based walking program did lose weight even without any dietary changes. That's according to a study published in the Annals of Family Medicine. And the longer they stayed in the program, the more weight they lost. It wasn't a huge amount, only an average of about three pounds for the more than 300 study subjects, who averaged 16 weeks in the study, but it's better than gaining weight, which is what most people do slowly but surely and increases in activity have health benefits independent of the weight loss, including better glucose tolerance and lower blood pressure.

Well that's it for this edition of the weekly SciAm podcast. You can write to us at and check out numerous features at including the "Image Gallery," featuring the messenger spacecraft's recent flyby photo of Mercury. There's also the new Scientific American Community and the chance to sign up for Scientific American newsletters and alerts. For Science Talk, the weekly podcast of Scientific American, I'm Steve Mirsky. Thanks for clicking on us.

Rights & Permissions
Share this Article:


You must sign in or register as a member to submit a comment.

Email this Article