Has the Milky Way ever collided with another galaxy? Or will the collision with Andromeda be our first?

  • $\begingroup$ As discussed in this related answer to the same question about Andromeda, most large galaxies (such as MW) have gone through of the order of one major merger through its life. In addition, there are many minor mergers as well as smooth accretion of matter. $\endgroup$
    – pela
    Commented Dec 14, 2017 at 11:56
  • $\begingroup$ @pela Where the definition of minor vs. major merger is arbitrary and thus not physically meaningful, I'd argue. $\endgroup$ Commented Dec 14, 2017 at 16:32
  • $\begingroup$ @AtmosphericPrisonEscape I agree that there's no exact definition, but I don't think they're arbitrary. A major merger severely perturbs the system, induces extra star formation, etc., whereas a minor merger mostly just adds mass to the system. To be able to do the former, the mass ratio between the galaxies must be of the order of 1, while if it's much less ($\lesssim0.1$), it doesn't disturb the system much $\endgroup$
    – pela
    Commented Dec 15, 2017 at 15:53

2 Answers 2


According to our current understanding of physical cosmology (as contrasted with cosmologies of various cultural references) the Milky Way must necessarily have had collisions in a process called hierarchical clustering.
How we're certain of this is the following:

  • No matter what exactly happened at the big bang, we know the state of the universe at roughly the age of ~380.000 years, where the universe becomes transparent (more or less instantaneous, but let's omit that discussion). At this moment, matter emits light that teaches us the distribution of matter in the young universe. This is the Cosmic Microwave Background Radiation.
  • In the CMB we see that matter is remarkably homogeneous, apart from very small fluctuations, that help us build the cosmological standard model. This means at the time of the CMB emission, the universe had nearly the same matter density everywhere. Which is the contrary of clustered matter blobs (galaxies) that we see today.

CMB fluctuations
Fig. 1: CMB temperature fluctuations as seen by WMAP.
Those variations can be translated into density fluctuations, given a physical model. To our current understanding the form the seeds of galaxy growth.

  • Now we take those small fluctuations, add dark matter and evolve a model of the universe over time, in large-scale compuational facilities. Then we find that galaxies are a natural outcome of the small CMB fluctuations that become gravitationally unstable. Those unstable blobs of matter amplify the initial fluctuations, until whole pockets of matter collapse into dense, rotating objects.
  • Those objects would be dwarf galaxies, and those collide and merge and thus grow into larger, mature galaxies. This is the process of hierarchical clustering, I was referring to above. This process can be simplified to "small things collapse first", as opposed to "large things collapse first". In the latter case the Milky Way would not have had collisions before.

Yes, our galaxy had collisions before, and they were necessary to grow to it's current size. Scientists are searching for the relics of those collisions in a process called galactic archaeology (the link goes to a personal website, which I find neatly made).


Yes. We can see "stellar streams" surround the Milky Way, long streams of stars from globular clusters and dwarf galaxies that have been disrupted by the galaxy and are now slowly merging with it. There is an upper limit (based on the amount of blue stars in the halo) of about 60 past such mergers.

It does not look like the Milky Way formed from a big merger. Had it done so it would have a much thicker and fuzzier disk (lots of "heating" by having stars in random orbits). The Andromeda collision will turn the galaxies into an elliptical galaxy, which has even less structure.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .