Five Reasons We Think Dark Matter Exists (Synopsis)

Whether you've been coming around to Starts With A Bang for years or whether you just discovered us a few weeks ago, chances are you've heard us take on the issue of dark matter -- whether it exists and, if so, what its properties are -- and how we think we know that.

Image credit: CMB pattern for a universe with normal matter only compared do our own, which includes dark matter and dark energy. Generated by Amanda Yoho on the Planck CMB simulator at Image credit: CMB pattern for a universe with normal matter only compared do our own, which includes dark matter and dark energy. Generated by Amanda Yoho on the Planck CMB simulator at

And while every professional in the field has the same information at their disposal, each one pieces it together differently in their own minds. For some of you, my take on things may not resonate as well as the take of another, and so today I'm incredibly pleased to share with you this article by Amanda Yoho, theoretical and computational cosmologist, where she goes through the five strongest separate pieces of evidence that, when taken all together, make a near-impossible-to-escape case for dark matter.

Image credit: “Sloan Digital Sky Survey 1.25 Declination Slice 2013 Data” by M. Blanton and the Sloan Digital Sky Survey. Image credit: “Sloan Digital Sky Survey 1.25 Declination Slice 2013 Data” by M. Blanton and the Sloan Digital Sky Survey.

Go read the whole thing, and then come back here and leave your comments!

More like this

"All the evidence, experimental and even a little theoretical, seems to indicate that it is the energy content which is involved in gravitation, and therefore, since matter and antimatter both represent positive energies, gravitation makes no distinction." -Richard Feynman It was a big last week at…
"Scientific discovery and scientific knowledge have been achieved only by those who have gone in pursuit of it without any practical purpose whatsoever in view." -Max Planck Tomorrow morning, at 8 AM my time, the press conference that cosmologists have spent the past decade waiting for will finally…
"A new scientific truth does not triumph by convincing its opponents and making them see the light, but rather because its opponents eventually die, and a new generation grows up that is familiar with it." -Max Planck (For Alan L., from the comments on this post.) When you look out at the night sky…
"What makes the universe so hard to comprehend is that there's nothing to compare it with." -Anonymous If I were brand new to theoretical cosmology, I might be skeptical of a whole bunch of "dark" things that I'd heard of. "Dark matter?" "Dark energy?" Come on; you've got to be kidding me! You're…

We just concluded the 2014 SLAC Summer Institute ( entitled "Shining Light on Dark Matter."

The institute covered in great detail not only the many parallel threads of evidence for dark matter, but also our current and planned efforts at measuring it, directly (via interactions with underground detectors or by producing it in high energy particle colliders) and indirectly (via high resolution/high cadence astronomical observation and by detection of astrophysical products of dark matter interactions).

By Michael Kelsey (not verified) on 19 Aug 2014 #permalink

@ Michael

are there any recordings of talks on youtube?

By Sinisa Lazarek (not verified) on 19 Aug 2014 #permalink

@Sinisa #2: Unfortunately not; we'll a bit behind the curve when it comes to technology :-/ However, the detailed agenda (under "Program" on the main page's sidebar) does have links to most of the speakers' slides. Not all speakers provided PDFs to the organizers, though.

By Michael Kelsey (not verified) on 19 Aug 2014 #permalink

In Yoho’s evidence #2, we read “…the velocity of stars remained approximately constant, regardless of how far they were from the galactic center.” Does this mean galaxies effectively have no gravitational center?

Also, “If instead a large fraction of the galaxy’s mass resided in a diffuse dark matter ‘halo’ that extended well beyond the edges of the luminous matter, the observed galactic rotation curves could be explained.”
What is meant by 'halo'? Does it mean like a donut - with no dark matter in the center of the galaxy and none outside the galaxy, but just a ring around the galaxy? [Perhaps a 3-D ring or hollow globe?]

In #3, “The existence of dark matter leaves a characteristic imprint on CMB observations, as it clumps into dense regions and contributes to the gravitational collapse of matter, but is unaffected by the pressure from photons.”

I thought the Standard Cosmological Model (i.e. The Big Bang Theory) was based on homogeneity, i.e. no clumps in the universe. How can you have clumps and still hold to the SCM?
If dark matter contributes to gravitational collapse of matter, why doesn’t the matter collapse completely, into black holes or into many “points of singularity”?

In point #4, “a.) Dark matter interacts with its surroundings significantly less frequently than ordinary matter.” This seems contrary to point #2, where dark matter supposedly has a CONSTANT effect on the velocity of stars, keeping the speeds constant.

Is anyone working on direct detection of the cause of the universal constants having their precise and necessary settings?

By See Noevo (not verified) on 19 Aug 2014 #permalink

Michael Kelsey,
Any idea on when the SLACers might expect some results? What’s the budget for these detection experiments?
I’m sorry to hear a Stanford group such as SLAC is behind the curve technologically.

By See Noevo (not verified) on 19 Aug 2014 #permalink

I have asked Prof. Krauss on his facebook, but he didn't answer me. I watched his speech in Echoes From The Beginning,…

As he said empty space is not really empty, thus could we say that there is no such thing called nothing, or in other words, Existence is absolute and it's an 'a priori', a everlasting thing? the real eternity? I'm not religious,but philosophical.

If so,and if this unknown stuff breaks the second law of thermodynamics,thus unfortunately,it seems it indicates there is indeed the possibility of the emergence of a very strong Being(in whatever name and degree of power you like),just like the possibility of ET's existence in the normal and common matter-energy state.You said we're almost ignorant of that vast stuff,so we can't exclude the possibility that there is something evolved through a way of continuous trial and error in that 'timeless' Existence. the evolution universe perspective is from Clement Vidal's paper The Beginning and the End: The Meaning of Life in a Cosmological Perspective(he published it in book! this year)

Best wishes!
Anticipating your reply!

See Noevo,

Perhaps someone else can elaborate on my comments, but I'll give your questions a shot:

"Does this mean galaxies effectively have no gravitational centers?"

I don't think so. Galaxies still have a gravitational center, usually a massive black hole. The objects in a typical galaxy still revolve around this center. What it does mean is that these objects do not behave as we would expect from our observations of other gravitationally bound systems, such as our solar system. In our solar system, for instance, planets such as Neptune revolve around the sun at much lower speed than do planets such as Mercury. The farther from the sun a planet is, the slower its speed, and this occurs in a mathematically predictable way. When that math is applied to galaxies, it no longer fits the observed data. That's why the MOND theories were developed - the math of the MOND theories DOES fit the observed data for galactic rotational speeds. It does not explain the other phenomena that were pointed out in this article, though, which is why dark matter is a better explanation. (As for the actual distribution of dark matter halos goes, I am not qualified to answer this. Maybe someone more familiar with the details could help.)

"How can you have clumps and still hold on to the SCM?"

I believe you may have misunderstood the assumptions of the SCM. Obviously, the universe is clumpy. Any realistic model of the universe cannot assume a completely homogenous universe. However, the early universe, despite what this article seems to say, WAS homogenous to a large degree. When, for instance, you look at the pictures showing the fluctuations in the CMB, keep in mind that these fluctuations are measured in microkelvins. Considering that the CMB itself has a temperature just under 3K, these fluctuations are very small. Thus, it's still reasonable to claim that the early universe was homogenous, just not perfectly so.

"If dark matter contributes to the collapse of matter, why doesn't matter collapse completely, into black holes or into many 'points of singularity'?"

Ethan just posted an article regarding this point. I would suggest you look it up for a more complete and competent answer. The short answer, though, is that the collapse of matter is inhibited by the expansion of the universe.

"Dark matter interacts with its surroundings significantly less frequently than ordinary matter. This seems contrary to point #2, where dark matter supposedly has a CONSTANT effect on the velocity of stars, keeping the speeds constant."

Dark matter interacts GRAVITATIONALLY very readily. As far as we know, it does not interact in any other way to any significant degree with either itself or ordinary matter. The constant speeds are a result of gravitational interaction. Other interactions with dark matter just do not seem to occur.

"Is anyone working on direct detection of the cause of the universal constants having their precise and necessary settings?"

Certainly, that's a question of interest. However, it's also a very difficult thing to investigate. It's possible that the multiverse idea might lead to an answer. In such a scenario, inflation occurs at different locations in the pre-inflationary universe. These regions are causally disconnected from each other and can have differing values of the fundamental constants. Given enough regions, the probability of one of them having fundamental constants leading to the development of a universe capable of harboring intelligent life is increased. All of this seems to be conjecture at this point, however. Ethan has also written about this in the past, and one of the major problems with this line of inquiry is that inflation wipes out information about the universe as it was pre-inflation, which obviously makes investigations into the matter quite difficult.

If so,and if this unknown stuff breaks the second law of thermodynamics

I am not sure what the 2LOT has to do with quantum fluctuations. I think you mean the 1LOT, and you're asking whether the spontaneous production of particle pairs violates conservation of mass and energy. It doesn't. :)

To answer your bigger question, the spontaneous production of some much larger, more complex Being (like an eternal God) is forbidden by QM because larger the fluctuaton, the shorter the time it lasts. You can't get eternal beings out of quantum fluctuations - you can't even get human-scale beings lasting human-scale fractions of seconds.

@See Noevo #5: Sarcasm. The comment was a reply to Sinisa's question about whether we had recorded video of the SSI lectures. Sorry you didn't get it.

By Michael Kelsey (not verified) on 20 Aug 2014 #permalink

@See Noevo #4:

1) No. It means that the visible disk-like galaxy cannot be modelled like the solar system, with all of the mass at a central core, and the stars orbiting about that core. See more below.

2) By "halo," in this context we mean a large "spherical" (or at least, spherically symmetric) volume, completely filled with dark matter. The visible galaxy is embedded in that halo. With a little calculus, and knowing Gauss's Law, you should be able to work out that such a distribution leads to a fairly flat rotation curve.

3) Sean T #7 already addressed this. First, the obvious clumpiness we observe today is a _consequence_ of the growth of structure; it is also of small enough amplitude that the FLRW metric (which assumes homogeneity and isotropy) is a good approximation. Second, the fluctuations we see in the CMB are of amplitudes around one part per million or less (microkelvins, vs. 2.73 K for the overall value), which are also small enough to make the FLRW metric a good approximation even in the early universe.

4) By "interaction" in this context, we (obviously? I guess not) mean "interaction other than gravity." If the dark matter had enough other interactions with normal matter that we could detect it now, for example, then those interactions would lead to an effective "friction", slowing down galaxies in a way which is not observed.

By Michael Kelsey (not verified) on 20 Aug 2014 #permalink

I don't know how seriously to take this, but I have read that there are speculations out there that gravity is such a weak force because there are more than 3 spatial dimensions and that the gravitational force somehow "leaks" into other dimensions whereas the other fundamental forces do not. If such a thing is reasonable, would it be possible that matter that exists in such additional dimensions could be the explanation of dark matter? After all, under this idea, gravity is the only force that interacts with the additional dimensions beyond the three familiar ones. This would account for the main property we know about dark matter, namely that it seems to only interact gravitationally with normal matter.

Obviously, there's not really any independent evidence for additional spatial dimensions, so I realize that this explanation would not be favored. However, does it make any sense or is there something out there that's already observed that would falsify such an idea?

Dark Matter is inferred because gravitation.

Dark Matter infuses the universe.

Dark Matter should be denser in the vicinity of concentrations of ordinary matter because gravitation.

The estimates of the mass of local bodies, sun and planets, arrived at by summing the masses of constituent atoms and molecules match closely the masses as determined by Newtonian mechanics.

Protons are 85% dark matter?

By Roy Lofquist (not verified) on 20 Aug 2014 #permalink

Very interesting article!!! One thing I didn't see in the article that I would have liked to would have been the actual real time that it takes for galaxies to collapse into each other? I do think dark matter exists, and it will be interesting to see how we approach future research into it!

@Roy Lofquist #12: "Dark Matter should be denser in the vicinity of concentrations of ordinary matter because [of] gravitation." That is true enough, but requires a _quantitative_ statement (i.e., science instead of philosophy) in order to draw any conclusions.

In order to resolve the observational constraints (virial velocities in galaxy clusters, graviational lensing, galactic rotation curves), the amount of dark matter needs to be about five times that of normal, baryonic matter, when measured on scales of galaxies or larger.

The interstellar medium has a density of about 100 atoms/molecules per cm^3 (the range is 10^-4 up to 10^6 in cold molecular clouds). So that suggests a dark matter density around the solar system of maybe 500-ish GeV per cm^3 (I'm quoting in mass/energy units because we don't know the mass of a DM particle). For a WIMP DM candidate (m ~ 10 GeV), that translates to maybe 50 DM particles per cm^3!

The Sun has a volume of 1.4 x 10^18 km^3, or 1.4 x 10^33 cm^3, and a mass of 2 x 10^30 kg (known to better than 1 part in 10,000!), or 1.1 x 10^57 GeV. The total "extra" mass of DM in the Sun, using the values I estimated above, would be 7 x 10^35 GeV, or something like 10^-23 solar masses.

That would have absolute NO MEASURABLE EFFECT on the solar system, and indeed is immesurably small.

Your last question, "Protons are 85% dark matter." Nope, sorry. Quantum chromodynamics (the modern, proper field theory of the strong interaction) allows us to do calculations of composite particle masses (like protons), and what we find is that the stored energy of the gluons binding the quarks together into a proton is sufficient to explain the entire ~GeV mass of the proton. We don't need any mysterious "other" to get there.

By Michael Kelsey (not verified) on 20 Aug 2014 #permalink

@Sean T #11: I've read a bit about such models, but I am not a theorist, and don't feel competent to adequately criticize them.

There's certainly no reason in principle why mass in the bulk (i.e., mass out in those extra dimensions) shouldn't interact graviationally with our brane.

What I don't know (perhaps a Google Scholar or arXive search could get you more information?) is whether such a model could reproduce the clumpiness (including filaments, haloes, subhaloes, etc.) we observe, which existing cosmological simulations with conventional DM do reproduce.

By Michael Kelsey (not verified) on 20 Aug 2014 #permalink

@Michael Kelsey #14

Lots of big scary numbers there, Michael. However, they don't square with the assertion that the total mass of dark matter is five times the mass of baryonic matter. Are you saying that there is no dark matter in the solar system?

By Roy Lofquist (not verified) on 20 Aug 2014 #permalink

Michael Kelsey,
You wrote in #10 above, “First, the obvious clumpiness we observe today is a _consequence_ of the growth of structure; it is also of small enough amplitude that the FLRW metric (which assumes homogeneity and isotropy) is a good approximation.”

What do you mean by “small enough”? Is the recent discovery of the largest structure in the universe, a structure that would take 4 billion light years to cross, still “small enough” that we can assume homogeneity?

By See Noevo (not verified) on 20 Aug 2014 #permalink

Re: #17

When he says small enough, I think he means in regard to the variation in CMB, not spatial size of a particular cosmic structure.

Also, I think it can be argued that even the LQG disappears on a large enough scale.

By John Murray (not verified) on 20 Aug 2014 #permalink

@ Michael #3

thank you. unfortunately, have a hard time following ppt's without commentary. am not that skilled sadly :)

let's hope by next year someone gets a webcam ;)

By Sinisa Lazarek (not verified) on 20 Aug 2014 #permalink

@Roy Lofquist #16
Lots of big scary numbers there, Michael. However, they don’t square with the assertion that the total mass of dark matter is five times the mass of baryonic matter. Are you saying that there is no dark matter in the solar system?

No he isn't, just that there is expected to be so little it has no measureable effect. There is no requirement for the distribution of dark matter to closeley match that of baryonic matter. If there were, wouldn't dark matter amount to nothing more than a revision of the value of the gravitational constant by a factor of six?

David L:

There is no requirement for the distribution of dark matter to closeley match that of baryonic matter. If there were, wouldn’t dark matter amount to nothing more than a revision of the value of the gravitational constant by a factor of six?

I'm not disagreeing, but it's an interesting topic because it brings up questions of DM kinetics and thermodynamics. Maybe someone can correct me, but AFAIK we have very little idea just how lumpy or smooth (on, say, the scale of a few light years) DM is. AIUI, what we know is just basically what you stated - that it can't be very tightly correlated with baryonic matter, because if it was, that would observationally be no different from a higher gravitational constant.

I guess because of the Bullet Cluster we can give a partial answer to the kinetics question, in that we can observe one case where the DM and baryonic matter has remained uncoupled for a couple hundred million years. But still, AIUI there is lots of room for discovery here.

that it can’t be very tightly correlated with baryonic matter, because if it was, that would observationally be no different from a higher gravitational constant.

I carefully avoided which direction the value of the gravitational constant would change. i.e. if measured mass doubles, then gravitational constant halves. Thinking about if further, they could only remain tightly correlated if the mass of the antimatter and the mass of the matter moved in the same path. therefore no change in the gravitational constant..

@eric #22, David L #21, Roy Lofquist #16: There are properties of DM which we can infer, quantitatively, from its gravitational behaviour, *even though* we do not know what kind of particle(s) it is in detail.

I've already (in other comments) mentioned the fact that we know DM doesn't interact significantly with baryonic matter (if it did, we would have already seen those interactions). We also know that DM doesn't interact significantly with itself!

Why? If it did, then there would be a mechanism for DM to dissipate kinetic energy. Then large spherical volumes of DM would collapse to smaller, high density systems, their angular momentum would increase, and they would become disk-shaped. That's what happens with baryonic gas clouds (which *do* have such dissipative interactions).

If DM collapsed the same way baryonic matter does, then we would NOT SEE the same gravitation effects -- we wouldn't see flat rotation curves, we wouldn't see bullet-cluster gravitational lensing offsets, etc.

The lack of disipative interactions means that the DM in haloes must be relatively uniform on scales of kiloparsecs, because it doesn't have a way to lose energy and "cool down" into clumps that small.

By Michael Kelsey (not verified) on 21 Aug 2014 #permalink

@Michael Kelsey #25,

So Dark Matter is not only dark but also very strange. Its constituents interact with baryonic matter via gravity but not with other "particles" of DM. Some how I think even Einstein would have trouble following that train.

By Roy Lofquist (not verified) on 21 Aug 2014 #permalink

@Roy Lofquist #26: I think you've got it. The empirical evidence for dark matter (as described both in this article and many, many research papers) is extremely good. Whatever it is, it interacts gravitationally but not electromagnetically nor through the strong (nuclear) interaction. We certainly have examples of other matter with those properties -- neutrinos, for one.

In fact, the case for DM today is rather similar to the case for neutrinos back in the late '30s when beta decay was being studied. At the time, they were completely undetectable (even in principle!). They were introduced specifically to "rescue" energy and momentum conservation from their apparent violation by the continuous electron spectrum. Eventually, the evidence for neutrinos, in a variety of contexts, became strong enough for a scientific consensus to form, and an understanding of their behaviour was sufficient to design and successfully operate neutrino detectors.

Dark matter was "invented" as a hypothesis to "rescue" graviational theory (both GR and simple Newtonian gravity) from its apparent violation by galaxy clusters and galactic rotation. Now, there are so many independent lines of evidence supporting DM (again, see this article) that we have a scientific consensus around its existence, and can begin to inquire about how its properties could lead to direct detection.

By Michael Kelsey (not verified) on 21 Aug 2014 #permalink

@Roy Lofquist #24: Unfortunately, that model quite explicitly does not reproduce the wide range of obsevations which independently support the dark matter hypothesis.

Neither, by the way, does MOND by itself (MOND requires an additional contribution from dark matter in order to reproduce the graviational lensing data).

The basic, trivial test of any unconventional theory is whether it can (a) successfully reproduce existing data, and (b) make testable predictions for new data. Plasma cosmology fails (a).

By Michael Kelsey (not verified) on 21 Aug 2014 #permalink

Sean @11, you're correct in your memory, there is such a theory. It may also "explain" Dark energy, as these dimensions are scrolled up tighter and tighter as the universe expands, leaving less and less room for interaction to take place.

However, "explain" may be the wrong word, since details of what shape the interaction would take wasn't (at the time I remember the idea) limited by any observable characteristics. I.e. it was all free parameters, therefore could explain *anything*.

I like the idea as it is a neat encapsulation of two problems.

But that means it has to be pretty precisely worked out to see if it would become a reliable theory.

@Wow #29: Extra dimensions scaling with a(t)? That's a seriously cool way to address the dimensionality issue. Can you point me toward a good reference? Preferably one that a poor experimentalist can understand?

As for the lack of "explaining", I thought the idea of gravity in the bulk was just to extend GR into N dimensions, keeping the form of the Einstein equation the same. That "shouldn't" (but, as an experimentalist, what do I know?) involve a bunch of free parameters.

I also rather like the idea: It takes an already existing theoretical construct (string theory), and nicely resolves two quite different apparent paradoxes in our observations. That sort of solution (qualitatively) has been a hallmark of really good physics for the last few centuries.

By Michael Kelsey (not verified) on 22 Aug 2014 #permalink

I had a flash that dark matter is like oil which is a fuel for illumination.

By Shmuel Wahli (not verified) on 26 Aug 2014 #permalink

why call it dark matter when it is obviously transparent?

@kat #32: It was called "dark matter" because it doesn't interact with light at all. Ordinary transparent matter does interact with light -- it has an index of refraction, and light can be bent (as you see looking into a swimming pool), focused (as with a lens), or dispersed in different directions by frequency (as with a prism).

By contrast, dark matter doesn't seem to interact with light at all: as far as light is concerned, it might as well be vacuum.

By Michael Kelsey (not verified) on 27 Aug 2014 #permalink

@Wow #29: Extra dimensions scaling with a(t)? That’s a seriously cool way to address the dimensionality issue. Can you point me toward a good reference?

I'll do my best. This was in 1990-1, so I've a shedload of remembering to do before I can get something relevant into google scholar to turn it up.

I suppose, in the meantime, you can always fantasise about what the theory could be.

Heck, that'll be half of what I'll be doing to get the memory back in place...

@Wow #34: Ah! I thought this was something just out. I'll do an arXiv and Google Scholar search with that time frame, and see what I come up with. Thanks much!

By Michael Kelsey (not verified) on 29 Aug 2014 #permalink

All these a priori assumptions just strike me as the same sort of thing that went on with Geocentrism.

"How do you know the Earth isn't moving around the sun?"
"Look around you, idiot, anyone can see it isn't moving. Why would you even question such a thing?"

How do you know these galaxies aren't coming apart? You're assuming that they're orbiting the center and made all your mass calculations based on Kepler's laws. What foundation do you have for believing the stars in the spiral are gravitationally locked?

By Donald Airey (not verified) on 03 Sep 2014 #permalink


Alternate ideas are fine, but one has to work out the consequences of them and subject them to the same rigors as the accepted theory. In your case, let's suppose for the sake of argument that the stars in the spiral arms are not gravitationally bound to the center. In that case, why would these stars be moving faster relative to the center than what would be predicted by standard gravitational models (I'm not sure if GR is necessary or if Newtonian gravity is acceptable as an approximation)? Why would these stars be orbiting the center of the galaxy at all if they are not gravitationally bound? You may be right, but what's your mechanism for explaining the gravitational rotation curve?

Think about it this way. Suppose as a thought experiment that the planet Neptune suddenly became gravitationally unbound from the solar system. Would Neptune's speed increase as a result? Would Neptune continue to orbit the sun? No, Neptune would continue at the same speed at which it is currently orbiting and would follow a straight line path tangent to its orbit. If we observed Neptune following an orbit around the sun at a higher speed, we would need a mechanism to explain why that occurred; becoming gravitationally unbound would not be a valid explanation.

Sean, I'm happy to work withing the outline you suggested, but let's use the sun as an example. The sun became bound to the galactic core billions of years ago and was at one time gravitationally bound in an orbit that obeyed Kepler's laws. Under this scenario, you may apply gravity to the situation to figure out the mass based on the distance from the core and the velocity.

Several billion years ago, however, metric expansion began to tear the galaxy apart. The sun is no longer bound to the core and on a spiral that will eventually lead to it's ejection from the galaxy along with every other object in the spiral. Kepler's laws don't apply and so there is no missing mass for which an accounting is required.

By Donald Airey (not verified) on 04 Sep 2014 #permalink

@Donald Airey #38: You don't seem to quite understand how the cosmological expansion works (although Ethan has put up numerous posts describing it, and there are plenty of reputable sources from which you can learn more).

The metric expansion, because it is so small (70 km/s per megaparsec), does not affect graviationally bound structures like solar systems or galaxies. Can you please point us to some reputable source (peer-reviewed journal articles or secondary source which cites peer-reviewed research) which describes the structure of the galaxy in the terms you use? Your description is unsupported by available research and observation.

By Michael Kelsey (not verified) on 04 Sep 2014 #permalink

Michael, your metric - 70 km/s - is based on a linear model of expansion. We know the expansion is accelerating, so what possible basis do you have for using a metric based on a linear model when it obviously doesn't apply to the observed universe?

By Donald Airey (not verified) on 04 Sep 2014 #permalink


You are completely missing the point. Your idea that stars are gravitationally unbound from the center does NOT explain why stars rotate more rapidly than expected about that center. In fact, it cannot explain why stars rotate about the center AT ALL.

In order to maintain an orbit around a central mass, a body must be subjected to a centripetal force. In all large scale astronomical structures, that force comes from gravitational attraction between the object in orbit and the central mass. If the object is unbound gravitationally, where is the centripetal force? In a gravitationally unbound system, orbits are not possible; orbits are evidence that the system is bounded.

The situation is precisely the opposite of what one would expect if the stars in the arms of galaxies became unbounded gravitationally. The stars are in fact too strongly held. The situation is precisely analogous to the situation in astronomy during the mid to late 19th century. At that time, a new planet, Uranus was discovered and its orbit predicted from Newton's laws. This planet was found to move in a way that was anomalous. At the same time, the orbit of Mercury also could not be accounted for using Newton's laws.

There were two possibilities: either Newton's laws were incomplete or the mass distribution within the solar system was not completely known. Historically, the Uranus problem was solved first. It was proposed that the mass distribution of the solar system was different from what was known at the time, namely that there existed another planet further from the sun than Uranus. The orbit of this planet was predicted. Once that was done, it was off to the telescopes to look for this new planet, and of course it was found and named Neptune.

This led to the perfectly reasonable idea that the orbital anomalies of Mercury could be similarly explained. It was proposed that a new planet (referred to as "Vulcan" by most astronomers) existed closer to the sun than Mercury. Again, the orbit that this planet would need to have was calculated and off to the telescopes. Of course, when you learned the names of the planets in grade school, you did not learn about "Vulcan". That's because this time there was no planet there to be found. An explanation had to wait until Einstein developed General Relativity, which nicely explained Mercury's orbit.

Hopefully, you can see the parallel. We now observe stars in galaxies that don't move as we predict from Newtonian gravity and/or GR. We are faced with the same two choices: either our gravitational theories are incomplete or we don't know the mass distribution of these galaxies accurately. Dark matter is the answer that involves a new mass distribution. The answer that involves incomplete gravitational theory is known as MOND (Modified Newtonian Dynamics).

Both of these theories can account for the galactic rotation anomaly. However, dark matter can account for observations that are unrelated to galactic rotation, such as large scale structures in the universe and irregularities in the cosmic microwave background. MOND cannot account for these other observations. Given two theories, one that can account for observations beyond what the theory was developed to explain is to be preferred over a theory that can only explain the problem that it was developed to explain. Thus, dark matter is the preferred theory. Your "gravitationally unbound stars" idea does not even account for the galactic rotation curve, much less any other observations, and thus should be rejected.

Sean, please read my answer above. The stars of the galaxy were bound at one time and obeyed Kepler's laws. Under those conditions, you can derive the mass from the velocity and the distance from the center. In the last billion years, they have become unbound due to metric expansion. These stars are spiraling away from the galactic core and so Kepler's laws don't apply and you can't infer anything about the mass from the velocity.

By Donald Airey (not verified) on 04 Sep 2014 #permalink

@Donald Airey #42: Clearly you don't understand basic mechanics. If you start with a system which is bound, with centripetal force causing circular motion, and then turn off that force, you don't get "spiraling out." What you get is a straight motion, tangent to the last orbit.

Or are you proposing that, magically, galaxies are "expanding" just fast enough that the stars stay in orbit (contradicting your plaintive "gravitationally unbound"), but those orbits are increasing in size?

How does your model explain the morphological consistency between galaxies today (and we have _many_ nearby examples) and galaxies more than "a billion" light years away? If your unboundedness really did turn on at some specific point, then we should see clear differences in galaxy morphology, which we don't.

By Michael Kelsey (not verified) on 04 Sep 2014 #permalink

@Michael #43, Gravitationally Bound means that it follows Kepler's laws. We are not turning off the force like a switch, we are simply unbinding it. Take, for example, the sun, in a bound orbit around the center of the galaxy at roughly 27 Light Years. Now, every year, add 10,000 km to that distance through metric expansion. What happens to our sun over the course of, say, 7 billion years?

That's right, it no longer obeys Kepler's laws. You can no longer draw any inference about it's mass from it's velocity.

By Donald Airey (not verified) on 04 Sep 2014 #permalink

Michael @43, Oh, BTW, you simply need to study the history of the Universe a little more in depth. Spiral Galaxies are a relatively recent event. You don't find Spiral Galaxies in the early universe (except by the freak collision that might generate one). And there is considerable evidence today that elliptical galaxies have little or no "Dark Mass" or, peculiar velocities attributed to "Dark Matter".

By Donald Airey (not verified) on 04 Sep 2014 #permalink


I am inclined to believe that Michael Kelsey is right and that you lack a basic understanding of mechanics. There is NO way that a gravitationally unbound object can orbit the center of mass of the galaxy. Being in orbit IMPLIES that the object is gravitationally bound. As I amply pointed out above, there are two possible explanations for deviations from theoretical orbital speed, but being gravitationally unbound cannot explain why an object orbits faster than theory predicts. Being gravitationally unbound implies that there is no orbit at all!

As I mentioned above, being gravitationally unbound simply means that the bodies don't obey Kepler's laws. There's copious evidence that they don't. One possible resolution is to imagine a mass that doesn't exist. Another possibility is to assume there's nothing wrong with Kepler's laws. I'm examining the latter.

Michael, at least, understands that the issue is one of metric expansion. While we disagree on the amount of metric expansion available in the universe in the current epoch, but he's got a grasp of the main issue.

Thanks very much for your time.

By Donald Airey (not verified) on 04 Sep 2014 #permalink

@Donald Airey #43 et seq.: I'm sorry, Donald, but you are misusing well-defined and well-understood terminology. "Gravitationally unbound" means NOT GRAVITATIONALLY BOUND. It does NOT man "non-Keplerian orbits."

The galaxies in a typical cluster do not, in any way, follow Kepler's laws, because there is no "central mass" about which they orbit; but they are most definitely gravitationally bound together.

The stars in a galaxy do not follow Keplerian orbits, and would not follow such orbits even in the absence of dark matter. Why not? Because the "central mass" of a typical galaxy is comparable to the mass of the stars in the disk or elliptical structure. That means, in the absence of dark matter, outer stars see _more mass_ about which they orbit than inner stars do. Consequently, even without dark matter, stellar velocities would fall off with distance much more slowly than a naive application of Kepler's law would predict.

Your assertions about spiral vs. elliptical galaxies are simply wrong. The oldest observed spiral galaxy (BX442) is eleven billion years old. Ellipticals are not "older" or "more primitive" than sprials. Rather, they are the _result_ of mergers of spiral galaxies.

The galaxies which appear to be deficient in dark matter are one specific class of ellipticals, the "dwarf spheroidals." THe larger ellipticals have the same dark matter fraction (about 80%) that spiral galaxies do, based on their rotation curves or virial velocity measurements.

By Michael Kelsey (not verified) on 05 Sep 2014 #permalink

@Michael #48, you are correct. Well, to be specific, a gravitationally bound object is a closed void where an unbound object has a hyperbolic path.

Regarding the age of spirals, I stand by my assertion. BX442 is an anomaly. The reason this galaxy is in the news is that it isn't supposed to be there. Spiral galaxies are a relatively late addition to galactic formations.

The point is this. Dark Matter is based on the assumption that the stars are in orbit around a central mass (meaning the path is closed). I find no basis for this assumption. Please let me know what law of nature allows us to look at the instantaneous velocity, mass and distance from the galactic core and conclude that the orbit is closed.

If there is no such law (I suspect there isn't), then we are dealing with an assumption. In this case, I can posit that the stars are not in a closed orbit and hypothesis that the velocity profiles are the result of metric expansion ripping the galaxies apart.

By Donald Airey (not verified) on 05 Sep 2014 #permalink