When two objects (e.g. two galaxies) are sufficiently far apart, they join the Hubble flow and they get further away from each other. Normally, this would last "forever" (until, from the perspective of one of the galaxies, the other one "arrives" to the Hubble horizon and effectively disappears). But are there any cases where the galaxies would get further apart by the Hubble flow, but then, somehow, at some point, both of them would abandon the Hubble flow and would begin to approach each other getting closer and closer?

  • $\begingroup$ I'm not sure it's worth an answer, but there are a couple of misconceptions in the question. First, joining the Hubble flow isn't really a thing. Peculiar velocities tend to decrease over time if $H>0$, but there is no threshold at which an object starts being part of the Hubble flow. Second, there is no time at which a receding object crosses a horizon and disappears as seen by another object. It just gets more and more redshifted, like when falling into a black hole. And the horizon is the cosmological event horizon, not the Hubble distance (which isn't a horizon). $\endgroup$
    – benrg
    Commented Dec 17, 2022 at 8:45

1 Answer 1


Sufficiently large overdensities detach from the Hubble flow and collapse to form the structures we see in the Universe, i.e. galaxies, groups, clusters, and superclusters. The formation is "hierarchical", meaning that smaller structures form earlier, later merging to form larger structures. The largest virialized structures in the Universe are galaxy clusters.

So in this sense, the answer to your questions is "Yes" because matter initially follows the expansion, but at some later point turns over and collapses.

But even if two clumps of matter are receding, if their fate is to one day merge their recession will be somewhat retarded prior to this event. So you might say that they did not really follow the Hubble flow to begin with. So perhaps it would be more fair to interpret your question as "Can two objects which are not destined to merge nevertheless merge?"

For this to happen, something must cause at least one of the objects to accelerate toward the other at sufficient speed to "catch up" with the expansion that would otherwise cause them to recede forever.

Accelerating a galaxy is not easy, but in the process of virialization, as some objects lose energy and get gravitationally more bound, others may pick up enough energy to be ejected from the system. This could cause one galaxy, otherwise bound to merge with others, to escape and merge with other (nearby) galaxies.

If you consider not galaxies, but particles, then its more easy: Various physical processes accelerate particles to near-light speed. At any point in time in the history of the Universe there is a maximum distance that a photon — or, effectively, a particle traveling at 99.999999% the speed of light — may reach before the accelerated expansion makes it impossible to catch up with the Hubble flow. This distance defines the cosmic event horizon and the particle horizon.

Currently, this distance is 16.5 billion lightyears (Glyr). In other words, any particle, even a photon, emitted today from anything farther away than 16.5 Glyr from another object will no the able to reach that object, ever. Conversely, objects that are closer to each other than this distance may be able to reach each other. Of course, for speeds smaller that the speed of light, the max distance is also smaller.


You must log in to answer this question.

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