[graham.d (OP)]

Back in the 1970s QSOs (Quasi Stellar Objects or Quasars) were thought to be quite close (to explain their brightness) but having anomolous redshifts far in excess of what would be expected. They were also associated with galaxy clusters that were relatively close and having normal quite low redshifts. The association with the clusters was largely statistical but there was also some visual evidence of filaments between the bodies in many cases. It is now generally accepted that QSOs are very energetic emitters, probably due to the accretion disc of a super large black hole at the centre of a huge galaxy, at very great distances indeed. The association with relatively close galaxy clusters has been put down to inadvertently weighted sampling and gravitational lensing by the cluster of a quasar behind. However, I have never seen any explanations regarding the apparent linking filaments - clearly not possible over the distances involved. Also I have never seen a thorough statistical review to see if this stacks up.

While I appreciate the evidence supporting QSOs being very distant, can anyone point to reasonably convincing evidence to refute the connection between the galaxy clusters and QSOs, particularly the apparent filaments and the statistical association?

[PhysBang]

However, I have never seen any explanations regarding the apparent linking filaments - clearly not possible over the distances involved.

Linking "filaments" are the kind of thing that one should expect to see in some photographic plates where there are a number of images that are close together. So far the best evidence that there are not these filaments is not statistical, but that in every case where accurate photometry was able to bring about adequate optical resolution, the filaments are not present.

[graham.d (OP)]

It would be very problematic if the filaments were really there as it would mean QSOs were not very distant after all. You may be right PhysBang though on a few pics I've looked at the filaments look real, but then it could be a coincidence that the QSO was behind the filament. I have only looked at a few selected ones so no statistics to discount this as a random chance.

I did wonder whether the association with galaxy clusters was that the gravitational lensing is actually allowing QSOs to be seen that would otherwise go unnoticed, thereby weighting the stats in favour of a correlation between the two. The lensing effect may be more pronounced and have a greater effect than was first thought as a result of a wider and more even distribution of dark matter around clusters. Whilst there are some that are obvious lensing - seeing the same QSO as multiple images for example - there are others where the QSO is some way off the cluster but too close to be "not associated" with it (or so it was once thought).

[PhysBang]

It would be very problematic if the filaments were really there as it would mean QSOs were not very distant after all. You may be right PhysBang though on a few pics I've looked at the filaments look real, but then it could be a coincidence that the QSO was behind the filament. I have only looked at a few selected ones so no statistics to discount this as a random chance.

I don't think that anyone took the idea seriously enough to actually take the time to do a statistical analysis. At least, that's the impression I have received talking to astronomers about the matter while looking for such statistical analysis. The Sloan Digital Sky Survey is using images of galaxies and images of quasars in order to generate separate statistical analysis of the distribution of matter in the universe. One could use this study to look for a tell-tale statistical deviation between galaxy distribution and quasar distribution. So far, none has been reported.

The problem with the filament pictures is that the light from the quasar source is usually bound to help whatever filaments are extending away from a galaxy to appear brighter, since the light from the quasar may contribute to bringing the light hitting that part of the photographic plate (or CCD) up to the threshold of detection. Thus there may be many such filaments extending from a galaxy in question that are not captured by our equipment.

Whilst there are some that are obvious lensing - seeing the same QSO as multiple images for example - there are others where the QSO is some way off the cluster but too close to be "not associated" with it (or so it was once thought).

The problem with this reasoning is that is demands that there are two kinds of quasars, things that cannot be distinguished from each other, one kind with redshift that is really associated with distance and one kind that happens to have redshift from some unknown origin that mimics the other kind. It might be that there are these two kinds of quasars, but it is something beyond our verification and something that contradicts the results that we get on cosmology from a number of other investigations. This may be one reason why nobody took the idea seriously enough to follow up with specific statistical analysis.

Think about if someone said that some people were really lizard men dressed up in human suits. This theory would have to admit that some people were actually humans and we know this because humans around us get cut or get in accidents and have to have surgery so we know that they are really humans. The lizard man hypothesis also has to explain why the lizard men never get into these accidents. Similarly, the two-kinds-of-quasar hypothesis has to explain why every quasar for which we have a clear measurement of distance always has the appropriate redshift.

[graham.d (OP)]

I was not suggesting two types of QSO, but trying to explain how there may be a correlation between galaxy clusters (in the foreground) and some extra visibilty to quasars behind them resulting from the lensing. In this way it could explain why the two would be correlated although not in any way connected. Nothing to do with lizard men!! 

It was thought for some years that the redshift from QSOs was related to effects other than Hubble expansion and that the QSOs were associated with galaxy clusters. Now we believe that this is not the case (and I agree) so my question was related to how we can now ignore the evidence supporting a correlation. It may simply be that, in the early years, the only place QSOs were sort was close to clusters and there was no stats showing that there was, in fact, no preference for this association. As the view that clusters and QSOs were associated was widely held, I would be surprised if there were no studies to refute it. It would have been a nice little program for a PhD student.

If there is a correlation then it opens up some interesting ideas regarding dark matter distribution and gravitational lensing.