The actual number:

How far apart are galaxies on average?

An attempt to visualize such a thing:

If galaxies were the size of peas, how many would be in a cubic meter?


4 Answers 4


The simple answer is that the average galaxy spacing is around a few megaparsecs, while the biggest galaxies are around 0.1 megaparsecs in size. So the average spacing is somewhere in the range of 10 - 100 times the size of the biggest galaxies. The peas I had for lunch today were (at a guess - I didn't measure them!) 5mm in diameter so the interpea spacing would be 5 - 50cm, or between 8 and 8,000 per cubic metre.

But this is a very misleading statistic. Galaxies are not distributed uniformly, but instead are grouped into clusters, which are themselves grouped into superclusters. Also galaxies vary enormously in size, with dwarf galaxies around a thousand times smaller than the biggest galaxies.

I would resist the temptation to assign any significance to my figures above. However there is a take home message i.e. galaxies are much, much, much closer relative to their size than stars are. That's why galaxy collisions are quite frequent while stellar collisions are rare to the point of non-existence.

  • 1
    $\begingroup$ +1 for including a note on significance of statements! I was thinking of a homogeneity scale, so (super)clusters are an important detail. $\endgroup$
    – user12345
    Commented Mar 31, 2013 at 14:25
  • 7
    $\begingroup$ +1 for commentary on the fact that galaxies are much closer relative to their size than star systems. I did not know that and is very informative! $\endgroup$ Commented Dec 18, 2016 at 23:19

Assuming (in 2018) we have about two trillion galaxies in an observable universe 46.5 billion light years in radius, we can compute the average volume per galaxy and take the cube root of that volume to obtain a flawed but useful notion of average distance between galaxies:

$$\sqrt[3]{\frac{ \frac{4 \pi}{3} \ (4.65\cdot10^{10})^3}{2\cdot10^{12}}} \approx 5,949,391 \approx \text{six million light years}$$

While this is a notional average, it is not the typical distance between nearest-neighbor galaxies. It would be akin to computing the average distance between grains of sand on the beach by assuming that the grains of sand are spread uniformly over the whole earth. (Bear with me as the illustration may be useful for comparison.)

Assuming $7.5 \cdot 10^{18}$ grains of sand on an earth with a radius of $6,378$ kilometers, the square root of the average area occupied by a grain of sand is:

$$\sqrt{\frac{4 \pi \ (6.378 \cdot 10^6)^2}{7.5 \cdot 10^{18}}} \approx 0.00826 \approx \text{about a centimeter}$$

...yielding about a centimeter of separation between grains of sand on the beach, which is as misleading as the notion of six million light years between galaxies.

So, the best way to answer your question might be to look at our own Local Group of galaxies. Our Local Group has two major players, Andromeda (M31) and the Milky Way, which are separated by about 2.5 million light years. Around each of these is a clump of much smaller orbiting satellite galaxies. For the Milky Way, our satellite galaxies (a few dozen of them) vary in distance from our galactic center from about $50,000$ light years to over a million light years.

Finally, to answer your question informally, one might say that judging by our own neighborhood, galactic nearest neighbors tend to be on the order of about $100,000$ light years apart, but galaxies tend to travel in clusters that are separated by tens to hundreds of times that much.


Simple question requires a simple answer! There are 9,900,000 light years on average between galaxies.

How we come to that number: The average distance between galaxies is a few megaparsecs. A megaparsec is 1 million parsecs and there are roughly 3.3 light years to a parsec, so that means there are on average, sticking with threes, we'll say a few equals 3, 9,900,000 light years between galaxies in a galaxy cluster that is.

  • $\begingroup$ This answer doesn't really provide anything not covered by John's answer. $\endgroup$
    – pela
    Commented Feb 23, 2017 at 11:48

For study and visualization of 3D structure and distribution of 4673 galaxies less than 200 Mpc from Milky Way, we (Douglas and I) have developed an open source web simulation with many tools to aid analysis: riteshsingh.github.io/galaxies/

The code is open source under the MIT license at github.com/RiteshSingh/galaxies.

The associated publication is available here: doi:10.1093/astrogeo/ataa061 and papers.ssrn.com/sol3/papers.cfm?abstract_id=3668727.

  • 1
    $\begingroup$ Can you edit to clarify whether by "we" you mean "the scientific community" or "I and my colleagues"? I'm 99% certain it's the latter, based on name similarity, but it's always nice to be crystal-clear when making these sorts of disclosures. $\endgroup$
    – rob
    Commented Aug 26, 2020 at 22:24
  • $\begingroup$ Done sir, thank you for helping me improve the answer 😊 $\endgroup$ Commented Aug 27, 2020 at 4:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.