Archive for Milky Way

The Gaia Sausage

Posted in The Universe and Stuff with tags , , , , , on March 10, 2019 by telescoper

I had to undertake a top secret mission on Friday, which turned out to be much less exciting than I’d hoped, but at least it gave me an excuse to catch some of the Royal Astronomical Society Open Meeting followed by dinner at the RAS Club. I actually sat next to the Club Guest Michael Duff, the eminent theoretical physicist Michael Duff who gave a nice after-dinner speech.

An artist’s impression of the Gaia Sausage. The Gaia fork has not yet been proved to exist.

The last talk at the RAS Meeting was by Neil Wyn Evans of Cambridge University in the Midlands on the subject of the `Gaia Sausage‘ (which, as you can see, has its own Wikipedia page). The Gaia Sausage is so named because it is consists of a marked anisotropy of the velocity distribution of stars in Milk Way, which is elongated in the radial direction (like a sausage) indicating that many stars are on near-radial (i.e. low angular momentum orbits). This feature has been revealed by studying the second data release from Gaia.

The work Wyn described in his talk is covered by a nice press release from Cambridge University which links to no fewer than five articles on it and related topics, which can all be found on the arXiv here, here, here, here and here.

The most plausible explanation of the Gaia Sausage is that it is a consequence of a major collision between the Milky Way with a smaller galaxy containing about 109 stars about 8-10 billion years ago, as illustrated in this simulation.

I vote that this explanation of the velocity structure of the Milky Way should henceforth be called the Big Banger Theory.

Geddit?

I’ll get my coat.

Advertisements

Gravitational Redshift around the Black Hole at the Centre of the Milky Way

Posted in The Universe and Stuff with tags , , , , , , on July 26, 2018 by telescoper

I’ve just been catching up on the arXiv, and found this very exciting paper by the GRAVITY collaboration (see herefor background on the relevant instrumentation). The abstract of the new paper reads:

The highly elliptical, 16-year-period orbit of the star S2 around the massive black hole candidate Sgr A* is a sensitive probe of the gravitational field in the Galactic centre. Near pericentre at 120 AU, ~1400 Schwarzschild radii, the star has an orbital speed of ~7650 km/s, such that the first-order effects of Special and General Relativity have now become detectable with current capabilities. Over the past 26 years, we have monitored the radial velocity and motion on the sky of S2, mainly with the SINFONI and NACO adaptive optics instruments on the ESO Very Large Telescope, and since 2016 and leading up to the pericentre approach in May 2018, with the four-telescope interferometric beam-combiner instrument GRAVITY. From data up to and including pericentre, we robustly detect the combined gravitational redshift and relativistic transverse Doppler effect for S2 of z ~ 200 km/s / c with different statistical analysis methods. When parameterising the post-Newtonian contribution from these effects by a factor f, with f = 0 and f = 1 corresponding to the Newtonian and general relativistic limits, respectively, we find from posterior fitting with different weighting schemes f = 0.90 +/- 0.09 (stat) +\- 0.15 (sys). The S2 data are inconsistent with pure Newtonian dynamics.

Note the sentence beginning `Over the past 26 years…’!. Anyway, this remarkable study seems to have demonstrated that, although the star S2 has a perihelion over a thousand times the Schwarzschild radius of the central black hole, the extremely accurate measurements demonstrate departures from Newtonian gravity.

The European Southern Observatory has called a press conference at 14.00 CEST (13.00 in Ireland and UK) today to discuss this result.

A Ghost of a Jet?

Posted in Astrohype, The Universe and Stuff with tags , , on June 4, 2012 by telescoper

Last week an article in Nature News caught my eye. Ghostly jets seen streaming from Milky Way’s core was the headline. It’s based on a paper by Su & Finkbeiner recently submitted to the arXiv. There’s even a picture showing the jets in glorious technicolour:

Wow! Impressive stuff. If the jets look like that it’s amazing nobody ever saw them before!

Oh, hang on. The picture is an “artist’s conception”. In other words, it’s what the jets might look like if they actually existed, as imagined by a bloke with a box of crayons.

And how strong is the evidence that they do exist? Here’s the last paragraph of the Nature article (my emphasis):

Although the emissions are dim and the observations don’t have the statistical significance that astronomers require for proof, Baganoff says that several properties make them compelling evidence of jets. They are brighter at higher γ-ray energies and also brighter than the surrounding interstellar medium. They also seem to be long and thin, as would be expected of jets. “Taking all of the evidence together, it appears highly plausible that the features are jets emanating from the Galactic Centre,“ he says.

If they “don’t have the statistical significance that astronomers require for proof” then one wonders why they’re being given so much publicity. In any case the “ghostly jets seen streaming from the Milky Way’s core” can’t be said to have really been “seen” for certain. But they are “highly plausible”. In other words, the authors would like them to be there.

All I can say is that it must have been a slow news day at Nature.

Still, nice drawing.

Dark Matter: Dearth Evaded

Posted in Astrohype, The Universe and Stuff with tags , , , , , , on May 23, 2012 by telescoper

While I’m catching up on developments over the last week or so I thought I’d post an update on a story I blogged about a few weeks ago. This concerns the the topic of dark matter in the Solar Neighbourhood and in particular a paper on the arXiv by Moni Bidin et al. with the following abstract:

We measured the surface mass density of the Galactic disk at the solar position, up to 4 kpc from the plane, by means of the kinematics of ~400 thick disk stars. The results match the expectations for the visible mass only, and no dark matter is detected in the volume under analysis. The current models of dark matter halo are excluded with a significance higher than 5sigma, unless a highly prolate halo is assumed, very atypical in cold dark matter simulations. The resulting lack of dark matter at the solar position challenges the current models.

In my earlier post I remarked that this  study   makes a number of questionable assumptions about the shape of the Milky Way halo – they take it to be smooth and spherical – and the distribution of velocities within it is taken to have a very simple form.

Well, only last week a rebuttal paper by Bovy & Tremaine appeared on the arXiv. Here is its abstract:

An analysis of the kinematics of 412 stars at 1-4 kpc from the Galactic mid-plane by Moni Bidin et al. (2012) has claimed to derive a local density of dark matter that is an order of magnitude below standard expectations. We show that this result is incorrect and that it arises from the invalid assumption that the mean azimuthal velocity of the stellar tracers is independent of Galactocentric radius at all heights; the correct assumption—that is, the one supported by data—is that the circular speed is independent of radius in the mid-plane. We demonstrate that the assumption of constant mean azimuthal velocity is physically implausible by showing that it requires the circular velocity to drop more steeply than allowed by any plausible mass model, with or without dark matter, at large heights above the mid-plane. Using the correct approximation that the circular velocity curve is flat in the mid-plane, we find that the data imply a local dark-matter density of 0.008 +/- 0.002 Msun/pc^3= 0.3 +/- 0.1 Gev/cm^3, fully consistent with standard estimates of this quantity. This is the most robust direct measurement of the local dark-matter density to date.

So it seems reports of the dearth were greatly exaggerated..

Having read the paper I think this is a pretty solid refutation, and if you don’t want to take my word for it I’ll also add that Scott Tremaine is one of the undisputed world experts in the field of Galactic Dynamics. It will be interesting to see how Moni Bidin et al. respond.

This little episode raises the question that, if there was a problem with the assumed velocity distribution in the original paper (as many of us suspected), why wasn’t this spotted by the referee?

Of course to a scientist there’s nothing unusual about scientific results being subjected to independent scrutiny and analysis. That’s how science advances. There is a danger in all this, however, with regard to the public perception of science. The original claim – which will probably turn out to be wrong – was accompanied by a fanfare of publicity. The later analysis arrives at a much less spectacular conclusion,  so will probably attract much less attention. In the long run, though, it probably isn’t important if this is regarded as a disappointingly boring outcome. I hope what really matters for scientific progress is people doing things properly. Even if it  don’t make the headlines, good science will win out in the end. Maybe.

Milky Way Satellites and Dark Matter

Posted in Astrohype, Bad Statistics, The Universe and Stuff with tags , , , , on May 4, 2012 by telescoper

I found a strange paper on the ArXiv last week, and was interested to see that it had been deemed to merit a press release from the Royal Astronomical Society that had been picked up by various sites across the interwebs.

The paper, to appear in due course in Monthly Notices of the Royal Astronomical Society, describes a study of the positions and velocities of small satellite galaxies and other object around the Milky Way, which suggest the existence of a flattened structure orientated at right angles to the Galactic plane. They call this the “Vast Polar Structure”. There’s even a nifty video showing this arrangement:

They argue that this is is evidence that these structures have a tidal origin, having been thrown out   in the collision between two smaller galaxies during the formation of the Milky Way. One would naively expect a much more isotropic distribution of material around our Galaxy if matter had fallen into it in the relatively quiescent way envisaged by more standard theoretical models.

Definitely Quite Interesting.

However, I was rather taken aback by this quotation by one of the authors, Pavel Kroupa, which ends the press release.

Our model appears to rule out the presence of dark matter in the universe, threatening a central pillar of current cosmological theory. We see this as the beginning of a paradigm shift, one that will ultimately lead us to a new understanding of the universe we inhabit.

Hang on a minute!

One would infer from this rather bold statement that the paper concerned contained a systematic comparison between the observations – allowing for selection effects, such as incomplete sky coverage – and detailed theoretical calculations of what is predicted in the standard theory of galaxy formation involving dark matter.

But it doesn’t.

What it does contain is a simple statistical calculation of the probability that the observed distribution of satellite galaxies would have arisen in an exactly isotropic distribution function, which they conclude to be around 0.2 per cent.

However, we already know that galaxies like the Milky Way are not exactly isotropic, so this isn’t really a test of the dark matter hypothesis. It’s a test of an idealised unrealistic model. And even if it were a more general test of the dark matter hypothesis, the probability of this hypothesis being correct is not what has been calculated. The probability of a model given the data is not the same as the probability of the data given the model. To get that you need Bayes’ theorem.

What needs to be done is to calculate the degree of anisotropy expected in the dark matter theory and in the tidal theory and then do a proper (i.e. Bayesian) comparison with the observations to see which model gives the better account of the data. This is not any easy thing to do because it necessitates doing detailed dynamical calculations at very high resolution of what galaxy like the Milky Way should look like according to both theories.

Until that’s done, these observations by no means “rule out” the dark matter theory.

On the Dearth of Dark Matter in the Solar Neighbourhood

Posted in Astrohype, The Universe and Stuff with tags , , , , , , , , on April 22, 2012 by telescoper

I’m a bit late getting onto the topic of dark matter in the Solar Neighbourhood, but it has been generating quite a lot of news, blogposts and other discussion recently so I thought I’d have a bash this morning. The result in question is a paper on the arXiv by Moni Bidin et al. which has the following abstract:

We measured the surface mass density of the Galactic disk at the solar position, up to 4 kpc from the plane, by means of the kinematics of ~400 thick disk stars. The results match the expectations for the visible mass only, and no dark matter is detected in the volume under analysis. The current models of dark matter halo are excluded with a significance higher than 5sigma, unless a highly prolate halo is assumed, very atypical in cold dark matter simulations. The resulting lack of dark matter at the solar position challenges the current models.

As far as I’m aware, Oort (1932, 1960) was the first to perform an analysis of the vertical equilibrium of the stellar distribution in the solar neighbourhood. He argued that there is more mass in the galactic disk than can be accounted for by star counts. A reanalysis of this problem by Bahcall (1984) argued for the presence of a dark “disk” of a scale height of about 700 pc. This was called into question by Bienaymé et al. (1987), and by Kuijken & Gilmore in 1989. In a later analysis based on a sample of stars with HIPPARCOS distances and Coravel radial velocities, within 125 pc of the Sun. Crézé et al. (1998) found that there is no evidence for dark matter in the disk of the Milky Way, claiming that all the matter is accounted for by adding up the contributions of gas, young stars and old stars.

The lack of evidence for dark matter in the Solar Neighbourhood is not therefore a particularly new finding; there’s never been any strong evidence that it is present in significant quantities out in the suburbs of the Milky Way where we reside. Indeed, I remember a big bust-up about this at a Royal Society meeting I attended in 1985 as a fledgling graduate student. Interesting that it’s still so controversial 27 years later.

Of course the result doesn’t mean that the dark matter isn’t there. It just means that its effect is too small compared to that of the luminous matter, i.e. stars, for it to be detected. We know that the luminous matter has to be concentrated more centrally than the dark matter, so it’s possible that the dark component is there, but does not have a significant effect on stellar motions near the Sun.

The latest, and probably most accurate, study has again found no evidence for dark matter in the vicinity of the Sun. If true, this may mean that attempts to detect dark matter particles using experiments on Earth are unlikely to be successful.

The team in question used the MPG/ESO 2.2-metre telescope at ESO’s La Silla Observatory, along with other telescopes, to map the positions and motions of more than 400 stars with distances up to 13000 light-years from the Sun. From these new data they have estimated the mass of material in a volume four times larger than ever considered before but found that everything is well explained by the gravitational effects of stars, dust and gas with no need for a dark matter component.

The reason for postulating the existence of large quantities of dark matter in spiral galaxies like the Milky Way is the motion of material in the outer parts, far from the Solar Neighbourhood (which is a mere 30,000 light years from Galactic Centre). These measurements are clearly inconsistent with the distribution of visible matter if our understanding of gravity is correct. So either there’s some invisible matter that gravitates or we need to reconsider our theories of gravitation. The dark matter explanation also fits with circumstantial evidence from other contexts (e.g. galaxy clusters), so is favoured by most astronomers. In the standard theory the Milky Way is surrounded by am extended halo of dark matter which is much less concentrated than the luminous material by virtue of it not being able to dissipate energy because it consists of particles that only interact weakly and can’t radiate. Luminous matter therefore outweighs dark matter in the cores of galaxies, but the situation is reversed in the outskirts. In between there should be some contribution from dark matter, but since it could be relatively modest it is difficult to estimate.

The study by Moni Bidin et al. makes a number of questionable assumptions about the shape of the Milky Way halo – they take it to be smooth and spherical – and the distribution of velocities within it is taken to have a very simple form. These may well turn out to be untrue. In any case the measurements they needed are extremely difficult to make, so they’ll need to be checked by other teams. It’s quite possible that this controversy won’t be actually resolved until the European Space Agency’s forthcoming GAIA mission.

So my take on this is that it’s a very interesting challenge to the orthodox theory, but the dark matter interpretation is far from dead because it’s not obvious to me that these observations would have uncovered it even if it is there. Moreover, there are alternative analyses (e.g. this one) which find a significant amount of dark matter using an alternative modelling method which seems to be more robust. (I’m grateful to Andrew Pontzen for pointing that out to me.)

Anyway, this all just goes to show that absence of evidence is not necessarily evidence of absence…

Galaxies from the Past

Posted in The Universe and Stuff with tags , , , on January 27, 2012 by telescoper

If you were wondering where I got yesterday’s piece from, the answer is that I fired up my old laptop and found it among a lot of old papers there. And by “old laptop”, I mean really old laptop: I bought it in 1995! Anyway, since I haven’t got time to write anything today here is another piece I wrote a long time ago but have only recently unearthed. This one is about Galaxies. It’s a lot longer than yesterday’s effort, but like that one I can’t remember what it was for. Still, some of you might find it interesting. The piece ends with a reference to galaxies observed as they were in the distant past, rather like the article itself!

–0–

A galaxy is a collection of stars, held together by their mutual gravitational attraction and orbiting around their common centre.  Galaxies range in size from dwarf systems of perhaps a few million stars, to giants containing up to a thousand billion. The Sun and all the stars visible in the night sky to the naked eye belong to one such galaxy, our own Galaxy the Milky Way. Although principally recognized through the light given off by their component stars, galaxies also contain other material such as clouds of gas and dust, and significant quantities of dark matter whose nature is not yet understood.

Only stars inside our Galaxy can be resolved with the naked eye; these stars have been studied and catalogued since antiquity. Ancient astronomers  also knew of the existence of a diffuse band of light crossing the sky they could not resolve into individual stars; we now call this the Milky Way. The word galaxy is derived from the Greek galaktos, meaning “milk”. The existence of galaxies other than our own is a much more recent discovery. While even relatively nearby stars appear as point sources of light, the light from other galaxies appears as cloudy and diffuse much like small fragments of the Milky Way. The generic term for a such sources is nebula, the latin word for “mist”.

A Persian astronomer, al-Sufi, in the 10th century AD described such a faint patch of light in the constellation Andromeda which is now known to be another galaxy, but it was only in the 18th Century that a systematic catalogue of  nebulae was compiled, by the French astronomer Charles Messier. Not all the objects he found were other galaxies – some were clouds of dust and gas inside our own – but the Messier catalogue contained 32 objects that we now know to be galaxies, including al-Sufi’s object, which was number 31 in his list. The Andromeda nebula is known to this day as M31. With the increasing power of astronomical telescopes, the list of known nebulae grew to thousands even before the use of astronomical photography became widespread. William and Caroline Herschel and, later, their son John played a leading role in identifying and cataloguing such objects in the early 19th century.

While the existence of large numbers of these nebulae was well established by the start of the 20th Century, their nature remained controversial. Since their distances could not be directly measured, it was possible that they could be inside our own galaxy. Many astronomers believed that the spiral structure seen in some of them, for example M31, suggested that they represented the formative stages of planetary systems like our own Solar System inside the Milky Way. Others argued that the nebulae were very much more distant than that, and were “island universes” on a much larger scale. This debate was only resolved in the 1920s, when Edwin Hubble was able to measure the distances to some nebulae using variable stars called Cepheids. He found them to be far too distant to be inside the Milky Way. This discovery established galaxies as the basic building-blocks of the Universe and gave rise to the field of extragalactic astronomy. Astronomers now estimate that there are as many galaxies in our observable Universe as there are individual stars in our own galaxy, i.e. around a hundred billion.

Galaxies come in a rich variety of shapes and sizes, but there are three basic types: Galaxies come in three basic types: spiral (or disk), elliptical and irregular. Hubble proposed a morphological classification, or taxonomy, for galaxies in which he envisaged the three basic types (spiral, elliptical and irregular) as forming a sequence which in the past was often assumed to represent various evolutionary stages of a galaxy . Although it is now not thought the interpretation as an evolutionary sequence is correct, Hubble’s nomenclature is still commonly used.

Spiral galaxies account for more than half the galaxies observed  in our neighbourhood.  These contain a bright central nucleus surrounded by a flattened disk that sometimes contains beautiful spiral arms. Hubble divided these galaxies into classes labelled as normal (S) or barred (SB) depending on whether the prominent spiral arms emerge directly the nucleus, or originate at the ends of a luminous bar projecting symmetrically through it . Spirals often contain copious amounts of dust, and the spiral arms containing many young stars givin them a noticeably blue colour.  The normal and barred spirals S and SB are further subdivided into a, b or c depending on how tightly the spiral arms are wound up.

The elliptical galaxies (E), which account for only around 10% of observed bright galaxies, are elliptical in shape and have no discernible spiral structure. They are usually red in colour, have very little dust and show no signs of active star formation. The further classification of elliptical galaxies into En depends on the degree of elongation of the galaxy: E0 is nearly spherical; E7 is cigar-shaped. Ellipticals tend to occur in regions of space where there are many other galaxies, giving rise to the idea that they might originally have been spiral galaxies but have lost their spiral structure through mergers or interactions with other galaxies.

The shapes and colours of elliptical galaxies resemble the corresponding properties of spiral nuclei. Elliptical galaxies cover a broad range in mass, from a few hundred thousand to a thousand billion times the mass of the Sun. Spiral galaxies seem to have a smaller spread in mass, typically weighing in at about a hundred billion times the mass of the Sun.

Lenticular, or S0 galaxies, were added later by Hubble to bridge the gap between normal spirals and ellipticals. Around 20% of galaxies we see have this morphology. They are more elongated than elliptical galaxies but have neither bars nor spiral structure.

Irregular galaxies have no apparent structure. They are relatively rare, and are often faint and small so are consequently very hard to see. Their irregularity may stem from the fact that they are have such small masses that the material within them is relatively loosely bound and may have been disturbed by the environment in which they sit.

The classification of galaxies proposed by Hubble applies to “normal” galaxies whose light output is dominated by radiation their constituent population of stars. Stars predominantly emit visible light, which occupies a relatively narrow part of the spectrum of electromagnetic radiation. Spiral galaxies also contain dust which is heated by starlight and radiates in the infra-red. Active galaxies are characterized by the prodigious amounts of energy they emit in regions of the spectrum normal galaxies cannot reach, particularly in radio and X-rays. Much of the energy broadcast by active galaxies is associated with the relatively small nucleus of the galaxy, so the term Active Galactic Nuclei (AGN) is often used to describe these regions. Sometimes the central nucleus is accompanied by a jet of material being ejected at high velocity into the surrounding intergalactic medium. The different types of active galaxy include Seyfert galaxies, radio galaxies, BL Lac objects, and quasars.

Seyfert galaxies are usually spiral galaxies with no radio emission and no evidence of jets. They do, however, emit radiation over a continuous range of frequencies from infra-red to X-rays. Splitting their optical light up into its characteristic spectrum reveals the presence of strong and variable emission lines.  One can see such lines in ordinary stellar spectra and consequently in the spectra of normal galaxies, but they are much more prominent in active galaxies. Radio galaxies, on the other hand, are more commonly elliptical galaxies. These objects are extremely dramatic in their appearance, frequently having two lobes of  radio-emitting material extending far away from the central compact nucleus. There is also sometimes the appearance of a jet of material, extending from the core into the radio lobes. It appears that material is ejected from the nucleus along the jet, eventually being slowed  down by its interaction with the intergalactic medium and forming the radio lobes. The central parts of radio galaxies seem to have properties similar to those of Seyfert galaxies.

BL Lac objects have spectra with no emission lines, but they emit strongly in all wavebands from radio to X-ray frequencies. Their main characteristic, however, is their extremely strong and rapid variability. It is thought that a possible explanation for these objects is that the observer is seeing a jet of material travelling head-on at close to the velocity of light.

The first quasars to be found were detected by their strong radio emission, but they were found to be so small that, like stars but unlike other galaxies, they could not be resolved with optical telescopes. For this reason they became known as quasi-stellar radio sources, or quasars for short. Later on, other such objects were found which did not emit radio waves at all, so the name was changed to quasi-stellar object or QSO, but the name quasar has in any case stuck. It seems that only one in about two hundred quasars is actually radio-loud, but the quasars are still the most powerful of all the active galaxy types.

These different kinds of objects were discovered at different times by different people and were originally thought to be entirely different phenomena. Now, however, there is a unified model in which these structures are all interpreted as having basically similar structure but a different orientation to the observer’s line-of-sight. The engine which powers the activity is thought to be a supermassive black hole, with a mass up to about 100 million solar masses. This seems very large, but is actually just a small fraction of the mass of the host galaxy, which may be a thousand times larger. Material surrounding the black hole is attracted towards it and undergoes a process of accretion, gradually spiralling in and being swallowed. As it spirals in, it forms an accretion disk around the black hole. This disk can be very hot, producing the X-ray radiation frequently seen in AGN, but its presence prevents radiation being transmitted through it. Radiation tends therefore to be beamed out of the poles of the nucleus and does not appear from the equatorial regions which are obscured by the disk. When the beamed radiation interacts with material inside the host galaxy or in the surrounding medium, it forms jets or radio lobes. Depending on the thickness of the disk, the size of the `host’ galaxy ,the amount of gas and dust surrounding the nucleus and the orientation at which the whole system is viewed one can, at least qualitatively, account for the variety of properties listed above.

It is not known what fraction of normal galaxies undergoes activity at some stage in their careers. Although active galaxies are relatively uncommon in our neighbourhood, this may simply be because the active phase lasts for a very short time compared to the total life of a galaxy. For example, if activity only lasts only one-thousandth of the total lifetime, we would expect only to see one in a thousand galaxies at any one time displaying the symptoms. It is perfectly possible, therefore, that the kind of extreme activity displayed by these galaxies is merely a phase through which all galaxies pass. If so, this would suggest that all galaxies should possess a massive black hole at their centre, which is no longer powering an accretion disk because there is insufficient gas left in the surrounding regions. Recent studies using the ultra-high resolution available on the Hubble Space Telescope suggest that most normal galaxies may indeed have black holes in their centres.

A somewhat milder form of activity is displayed by starburst galaxies which, as their name suggests are galaxies undergoing a vigorous period of star formation. Such activity is not thought to involve an active galactic nucleus, but is probably triggered by a tidal interaction between two galaxies moving closely past each other.

The stars in a galaxy exert gravitational forces on each other. This not only holds the galaxy together, it also causes the stars to move. The internal dynamical properties of galaxies are extremely important because they allow astronomers to work out how much matter is there.

In spiral galaxies, the component stars orbit roughly in a plane about the central nucleus. It is this bulk rotation that is responsible for the flattened shape of these systems. Much the same state of affairs applies in the Solar System, with all the planets moving in roughly circular orbits about the Sun. In the case of a disk galaxy that lies edge-on to the observer, stars on one side will be approaching while those on the other will be receding. These motions cause a Doppler shift in the light from different parts of the disk: one side will have a spectrum that is shifted towards blue colours, while the other side will be shifted to the red. One can therefore use spectroscopic methods to plot a graph showing how the rotation speed of material  varies with distance from the centre of rotation. Such a curve is called a rotation curve.  These curves show that the matter in spiral galaxies has a roughly constant velocity out to tens of thousands of light years from the centre. This is surprising because the planets of the Solar System have orbital speeds that fall off quite rapidly with distance from the Sun. Most of the mass of the Solar System lies in the Sun, which is near the centre of motion. Most of the light produced in a galaxy is likewise produced in the central regions. If all the mass in a galaxy were where the stars are, i.e. in the middle, the rotation speed should fall off the further out from the centre one looked. The simplest interpretation of this strange behaviour is that galaxies contain a large amount of material that does not produce starlight and which is not as concentrated in the centre of the galaxy as the stars. To make this work requires galaxies to be embedded in a diffuse halo of dark matter that is about ten times as large as the luminous part of the disk and containing perhaps ten times as much matter.

Dynamical studies of elliptical galaxies are more complicated because the stellar motions within them are not those of simple rotation. Nevertheless, these objects too reveal evidence for dark matter in similar quantity to that in spiral galaxies.

It is thought that less than 10 per cent of the total mass of a galaxy is in visible stars, but the form of the mysterious dark matter is not at all understood. The best candidate at the moment is some form of exotic particle left over from the Big Bang, usually called a WIMP (Weakly Interacting Massive Particle), although no such particle has yet been directly detected.

Galaxies are the basic building blocks of the Universe. They are not, however, the largest structures one can see. They tend not to be isolated, but cluster together. The distribution of nebulae on the sky was thought to be non-uniform even in the days of the Herschels, but it is only in the 20th century that it has become possible to map their three-dimensional positions in a systematic fashion.

The technique used to explore the large-scale distribution of galaxies is based on the discovery of the expanding universe usually attributed to Edwin Hubble, who built  on earlier work by Vesto Slipher. Slipher had discovered that lines in the optical spectra of galaxies were systematically shifted towards the longer wavelength, red end of the electromagnetic spectrum. Hubble extended this study by looking at these redshifts in tandem with the distances he had estimated for the galaxies. He found, to his surprise, that the redshift of a galaxy came out to be proportional to its distance. Contrary to popular belief, Hubble never really interpreted this himself as the result of cosmic expansion but the empirical correlation between redshift and distance now known as Hubble’s Law is the cornerstone of the big-bang cosmology. It is now accepted that the redshift of the galaxies arises from their motion away from the observer, similar to the Doppler shift that causes a change of pitch in a receding police siren. While the accurate determination of extragalactic distances remains difficult, measuring redshifts is rather straightforward. Hubble’s law has been used to chart the pattern traced out by millions of individual galaxies from their spectral shifts.

The general term used to describe a physical  aggregation of many galaxies is a cluster of galaxies, or galaxy cluster. Clusters can be systems of greatly varying size and richness. Our galaxy, the Milky Way,  is a member of the Local Group of galaxies which is a rather small cluster of galaxies of which the only other large member is the Andromeda galaxy (M31). On the other extreme, there are the so-called rich clusters of galaxies, also known as Abell clusters, which contain many hundreds or even thousands of galaxies in a region just few million light years across: prominent nearby examples of such entities are the Virgo and Coma clusters. In between these two extremes, galaxies appear to be distributed in systems of varying density.

Individual galaxy clusters are not the largest structures in the Universe. The distribution of galaxies on scales larger than around 30 million light years also reveals a wealth of complexity. Galaxies are not simply distributed in blobs, like the Abell clusters, but often lie in extended linear structures called filaments, such as the Perseus-Pisces chain, or flattened sheet-like structures like the Great Wall. The latter object is roughly two-dimensional concentration of galaxies, discovered in 1988 by astronomers from the Harvard-Smithsonian Center for Astrophysics. This structure is at least 200 million light years by 600 million light years in size, but is less than 20 million light years thick. It contains many thousands of galaxies and has a mass of at least 1016 solar masses.  The interconnecting network of filaments and sheets is aptly called the “cosmic web”, with rich clusters appearing where the parts of the web join together.

Rich clusters are clustered into enormous loosely-bound agglomerations called superclusters, containing anything from around ten rich clusters to more than 50. The most prominent known supercluster is called the Shapley concentration, while the most nearby is the Local Supercluster, a flattened structure in the plane of which the Local Group is moving. Superclustering is known to exist on scales up to 300 million light years, and superclusters may contain as much as 1017 solar masses of material or more.

These overdense structures are complemented by vast underdense regions known as voids, many of which appear to be roughly spherical.  These regions containing very many fewer galaxies than average, or even no galaxies at all. Voids with density less than 10% of the average density on scales of up to 200 million light years have been found in large-scale redshift surveys.

The existence of galaxies, clusters of galaxies and the overall complexity of large-scale structure in the Universe around us must be contrasted with the extreme simplicity of the very early Universe. Observations of the cosmic microwave background, relic radiation left over from the early stages of the Big Bang, suggest that the initial state of the Universe was almost featureless, with variations in density from place to place of less than one part in a hundred thousand.

The process that is thought to have transformed these smooth beginnings into the clumpiness we see today is called gravitational instability. If the universe were initially exactly smooth, it would have remained so as it expanded and cooled. But if there were small initial variations in density, these would become amplified. A small patch of the Universe that was more dense than average would exert a slightly greater gravitational pull on its surroundings than an average patch. This would cause material to flow in, making it even denser. This, in turn, would make it pull even more than average. This starts a runaway process by which small initial ripples can turn into dense clumps.

This basic idea has been around since it was first suggested by Sir James Jeans more than a hundred years ago, but it is only in the last ten years or so that a convincing picture has been put together explaining how it works in the expanding Universe. According the modern theories, most of the matter in the Universe is in the form of exotic particles left over from the primordial fireball phase that was the Big Bang. These particles are thought to be very slow-moving and are consequently called Cold Dark Matter (CDM). These particles cluster together via the process of gravitational instability, first forming small objects with the mass of a very small dwarf galaxy (around one hundred thousand solar masses). These small seed objects then progressively merge into larger objects in a hierarchical fashion, eventually forming galaxy-sized and cluster-sized dark matter clumps. These form gravitational wells into which gaseous matter falls and becomes trapped. Stars  form as gas clouds cool and fragment in the dark matter clumps. All this happens within a continuous sequence of interaction, disruption and merging. The whole process is extremely complicated, but extensive computer simulations show that the structure produced is very similar to the cosmic web revealed by observations, at least in the essential details.

Further support for these theoretical ideas is provided by observations of galaxies so distant that it has taken their light a large fraction of the age of the Universe to reach us. Looking at such objects allows astronomers to see galaxies in the process of formation.