Globular Cluster
A Hubble Space Telescope image of the typical globular cluster Messier 80, an object made up of hundreds of thousands of stars and located in the direction of the constellation of Scorpius. The Milky Way galaxy has an estimated 160 globular clusters of which one quarter are thought to be ‘alien’. Image: NASA / The Hubble Heritage Team / STScI / AURA. Click for hi-resolution version.

milky way

Alien Star Clusters Are Invading the Milky Way

23 Feb , 2010 by


We’re being invaded! About one-fourth of the star clusters in our galaxy are actually invaders from other galaxies, according to a new paper. Research from a team of scientists from Swinburne University of Technology in Australia shows that that many of our galaxy’s globular star clusters are actually foreigners – having been born elsewhere and then migrated to our Milky Way. “It turns out that many of the stars and globular star clusters we see when we look into the night sky are not natives, but aliens from other galaxies,” said Duncan Forbes. “They have made their way into our galaxy over the last few billion years.”

Previously astronomers had suspected that some globular star clusters, which each contain between 10000 and several million stars were foreign to our galaxy, but it was difficult to positively identify which ones.

Using Hubble Space Telescope data, Forbes, along with his Canadian colleague Professor Terry Bridges, examined globular star clusters within the Milky Way galaxy.

They then compiled the largest ever high-quality database to record the age and chemical properties of each of these clusters.

“Using this database we were able to identify key signatures in many of the globular star clusters that gave us tell-tale clues as to their external origin,” Forbes said.

“We determined that these foreign-born globular star clusters actually make up about one quarter of our Milky Way globular star cluster system. That implies tens of millions of accreted stars – those that have joined and grown our galaxy – from globular star clusters alone.”

The researchers’ work also suggests that the Milky Way may have swallowed up more dwarf galaxies than was previously thought.

“We found that many of the foreign clusters originally existed within dwarf galaxies – that is ‘mini’ galaxies of up to 100 million stars that sit within our larger Milky Way,” said Forbes. “Our work shows that there are more of these accreted dwarf galaxies in our Milky Way than was thought. Astronomers had been able to confirm the existence of two accreted dwarf galaxies in our Milky Way – but our research suggests that there might be as many as six yet to be discovered.”

“Although the dwarf galaxies are broken-up and their stars assimilated into the Milky Way, the globular star clusters of the dwarf galaxy remain intact and survive the accretion process,” Forbes continued. “This will have to be explored further, but it is a very exciting prospect that will help us to better understand the history of our own galaxy.”

Read the team’s paper.
Source: Royal Astronomical Society

By  -        
Nancy Atkinson is currently Universe Today's Contributing Editor. Previously she served as UT's Senior Editor and lead writer, and has worked with Astronomy Cast and 365 Days of Astronomy. Nancy is also a NASA/JPL Solar System Ambassador.

17 Responses

  1. Jon Hanford says:

    A free eprint of this fascinating study is available here:

    My goodness, they have several groups of clusters listed supposedly related to infalling galactic systems. A clever bit of stellar archeology.

  2. Aqua says:

    Globular Cluster M54 in Sagittarius, near the handle of the ‘Tea Pot’ asterism, has long been thought to be ‘probably extragalactic’ and originating from the Sagittarius dwarf galaxy. I’ve pointed out this Globular during many summer star parties as being as much, which always brought comments of wonder and skepticism from fellow observers. Looks like the skeptics will have to bow out, now…

    My question would be.. which of these ‘interlopers’ weren’t originally GC’s but instead are the ‘stripped’ remnants of dwarf galaxy cores?

  3. Aqua says:

    I guess the real question would be: How could one tell the difference? What is the criteria there? Number of stars? Existence of a black hole in the middle? Size of the black hole? I mean really… don’t GC’s look like mini-elliptical galaxies?

  4. Aqua says:

    Jon Hanford – Thanks for the link.. the conclusion somewhat clears up my question(s), but raises others!

  5. Jon Hanford says:

    Aqua, yeah, some globular clusters exhibit different epochs of star formation. What’s the mechanism behind that? Orbital passages through the thick disk of the MWG? It’s an open question. And then factor in Dark Matter. We’ve got a lot of questions to nail down, but the future looks bright.

  6. Lawrence B. Crowell says:

    I have not read the paper yet, but a question does come to mind. If GC’s can enter a galaxy from the outside then are there GC’s in intergalactic space? I have thought of GCs as gravitationally bound to a galaxy.


  7. Astrofiend says:

    I thought that the article implied that the ‘extragalactic’ GCs joined our galaxy through the Milky Way cannibalising other, smaller dwarf galaxies. If this is the case, there is no need for them to be intergalactic travelers…

  8. Jon Hanford says:

    Globular clusters seem inextricably linked to galaxies and their formation mechanisms, but I see no reason why a globular cluster may eventually escape the gravity well of a galaxy (intergalactic globular clusters are known to exist in some galaxy clusters, for instance).

  9. Hon. Salacious B. Crumb says:

    Jon said;

    Globular clusters seem inextricably linked to galaxies and their formation mechanisms, but I see no reason why a globular cluster may eventually escape the gravity well of a galaxy (intergalactic globular clusters are known to exist in some galaxy clusters, for instance).

    The answer is actually very simple. The problem is the same as for ejections for multiple stars and ejection of stars in open star clusters. Globulars orbit galaxies, just like the planets in the solar system orbit the sun. However, unlike the solar system, there are other gravitational sources which potentially influence the orbit of the body on collection of stars. For a globulars to leave the Milky Way/ galaxy, it must have energy which gives the globular an escape velocity exceeding 525 km.sec^-1. (Being root 2 (1.414) time the orbital around the galactic centre. I.e. The Sun moves around the galaxy about 300 km.s^-1. For the sun to escape this is the 525 value.)

    There are actually many reasons why globulars can be accelerated to escape – mostly likely encounters with other galaxies, companions or multiple objects, which significantly influence the galactic orbit sufficiently so the globular can move into intergalactic space.

    The question boils down to kinetic energy (angular momentum) of the object moving in an orbit. If the energy is sufficient enough it is unceremoniously ‘turffed-out”, if not it continues to orbit in an elliptical orbit.

    Note: I continue to be greatly concerned and staggered with the apparent lack of knowledge of even basic stellar evolution and gravitation, when they probably remains the best and well understood knowledge we have in astrophysics.

    Please. If UT wants to remain relevant, please highlight articles on these subjects and give some general background on evolution theory. Make me wonder of the level of understanding in how gravity works. It is not mystical nor due to some unknown magical source, it is basically how the universe works. Really.

  10. Hon. Salacious B. Crumb says:

    Oh… and the exact opposite is also true. Of the globular losses energy, it can be captured by the galaxy in question.

  11. Hon. Salacious B. Crumb says:

    My previous statement should read;

    Oh… and the exact opposite is also true. If the globular losses energy, it can be captured by the galaxy in question.

  12. Hon. Salacious B. Crumb says:

    Jon said;

    “Aqua, yeah, some globular clusters exhibit different epochs of star formation.”

    Yes they do, but what is the timeframe you are talking about?

    Star formation is globulars likely began during the formation of the galaxies, but last perhaps a few hundreds of million of years. If the universe is 13.7 billion years old, the variation is simply a drop in the ocean.

    How do we know the globulars have periods of star formation? Simply from the colour-magnitude diagrams, which often show double main sequences. Most globulars these evolutionary differences are fairly minor, and suggest that star formation was not all at once but over a short period – possibly with intermittent burst of star formation depending on the gas collapse of the nebulosity that forms the stars in the globular. (Open cluster do exactly the same things – as do star formation with the Association and its centrally attached Open Clusters.)

    The other question is why globular survive and dwarf galaxies don’t. This is mostly because of the gravitation potential of these very large collections of stars. The potential energy of globulars is held by the core binaries (hard and soft), which act gravitationally to hold much of the angular momentum. Due to the size of globular are between 5 and 300 parsecs, and averaging perhaps some 50pc., these core binaries act as a glue mostly holding the cluster together.

    For dwarf galaxies they have no such binding energies, and so are easily striped apart by the strong gravity of the Milky Way.

    It is as Harlow Shapley states in his book “Star Clusters” pg.208 in 1930!!!
    “Time seems to leave its marks, however, on the clusters along the Milky Way. Some of the globular clusters maybe affected in freedom, form, and eventual survival by contacts with galaxies or other clusters.”

    He also says on pg.200;

    “The existing galactic clusters are necessarily in the process of dispersion through encounters; but it is not equally evident that existing globular clusters are doomed to assimilation by the Galaxy and subsequent transformation.”

    Hence, globular survive losing only few of their members, dwarf galaxies stars are all greedily consumed and come part of the Milky Way as streams of high-velocity stars.

    We an intergalactic globular is captured, it loses energy, and forms an highly elliptical orbit (probably the most important key to being captured), as most globular clusters have highly elliptical orbits, the assumption has been they were captured. (Harlow Shapley even released this even in the1940s!! ).

    Duncan Forbes, in this article says;

    “This will have to be explored further, but it is a very exciting prospect that will help us to better understand the history of our own galaxy.”

    It has been known for absolute ages; from high velocity stars, the anomalies in the galactic rotation, the shape of the Milky Way, the interactions of the Magellanic Clouds, and even the Milky Way’s own collection of dwarf galaxies, etc. – that other dwarf galaxies have been consumed.

    We see globulars in many nearby dwarf galaxies I.e The Fornax Dwarf Galaxy, so therefore it not logical that the Milky Way has taken their globular and made them apart of our own galaxy?

    Again, even the earliest theories of the Galaxy’s origin, without the basis of modern knowledge of the Milky Way, Shapley (1930) pg. 193 said;

    “In the discussions of galactic origin and behavior that grew out of the earlier work on clusters and the local star system,1 it was suggested that the discoidal galactic system, originating from the combination of independent star clouds and clusters, has long been growing, by assimilating such groups, to its present relatively enormous size.

    This is just again self-promoted propaganda to support their funded research. Goes to show that the ol’ biblical adage from Ecclesiastes “There is nothing new under the sun” still holds so true!

  13. Jon Hanford says:

    Two papers recently were posted on arXiv exploring multiple stages of star formation in the IC 1613 dIG and the Cetus dG:

    “The ACS LCID project. III. The star formation history of the Cetus dSph galaxy: a post-reionization fossil” @

    “The ACS LCID Project. II. Faint Variable Stars in the Isolated Dwarf Irregular Galaxy IC1613” @ (http) ://

  14. Jon Hanford says:

    These 2 papers were published in the 21st century, BTW :)

  15. Jon Hanford says:

    HSB Crumb,

    I think the notion of multiple star forming periods in some globular clusters is well established, no matter the length of time involved. Detailed star formation epochs are being delineated in a number of nearby globular clusters, as a literature search easily shows. The challenge to astrophysicists is to explain these star forming events and find a way to plausibly model their (occasional) occurrence.

  16. Hon. Salacious B. Crumb says:

    @ Jon

    “I think the notion of multiple star forming periods in some globular clusters is well established,”

    That is not exactly my point. Multiple stars, like the so called trapezia, are known to eject components, often rearranging the dynamics and structure of the multiple. I.e. Form a closer binary with a third more distant component after eject the four star. Open star clusters also eject members by interactions between member.

    A similar process occurs with globulars which are either absorbing or lost them into interstellar space.

    (This answers your own original question;

    “Globular clusters seem inextricably linked to galaxies and their formation mechanisms, but I see no reason why a globular cluster may eventually escape the gravity well of a galaxy (intergalactic globular clusters are known to exist in some galaxy clusters, for instance).”

    This is the process of dynamics of exchanges of bodies in large gravitational systems. My issue has absolutely nothing to do with nature of multiple stars within globulars.

  17. Jon Hanford says:

    Certainly, my apologies for misinterpreting your statements in your posts. I thank you for your time and advice in this matter, for I am but a novice in matters such as these.

Comments are closed.