Arp and Quasar-Galaxy alignments proposed statistical study

For a very long time Halton Arp has reported finding physical
associations between large nearby spiral galaxies and quasars at much
higher redshifts. If Arp's observations are accepted as valid then it
cannot be disputed that the cosmological origin of redshift and
therefore the big bang theory are wrong. It seems therefore that the
majority of cosmologists who still accept the big bang theory do not
accept Arp's observations as valid interpretations.

This is a request for astronomers who hold to the cosmological
interpretation of redshift to explain how they deal with Arp's
observations, and what evidence (if any) would make them change their minds?

Some of these observations are listed:

1. The apparent physical connection in the form of bridges between
objects of very different red shift observed in visible and x-ray. In
some cases multiple additional redshifts are found for small objects
clearly withing the bridges. A good example of this is found in
http://www.haltonarp.com/Articles/PDF/Research With Fred.pdf
where NGC7603 at z=0.029 is connected by a bridge to an object at
z=0.057 and two additional objects at z=0.243 and z=0.391 are found in
symmetrical positions within the bridge. If the cosmologogical red shift
is accepted then one is forced to conclude that this is a chance
alignment (or that the objects are travelling at relativistic
velocities), but Arp has found far too many of these for such a conclusion.

2. The alignment of quasars and other similar objects with the rotation
axis and with matched pairs in terms of red shifts of quasars at
symmetrical positions on opposite sides of the axis. In addition there
is a pattern of redshifts with high values nearer the galaxy and
dropping to values nearer the galaxy redshift at greater distance. Arp
interprets this as the quasars having been ejected from the spiral
galaxy along the axis in the matched pairs at various time intervals.
For many examples of these properties see
http://www.haltonarp.com/Articles/PDF/moriond.pdf

3. The consistency of peaks in quasar redshifts to specific values,
namely z=.06, .30, .60, .96, 1.41, 1.96, 2.64, 3.48 which are found with
minor variations in many examples of quasars associated with galaxies.
There is a common ratio of close to 1.23 in the (1+z) values, that is in
the actual wavelengths. Again see the paper listed in 2 above and other
papers listed at http://www.haltonarp.com/?Page=Abstracts

It is acknowledged that Arp made an early statistical mistake in his
probability calculations in one paper, so he went to study statistics so
that his later papers would be faultless in this regard. Perhaps it is
the case that astronomers/cosmologists suspect that he has taken so many
pictures and studied so many galaxies that he is able to find any
pattern that he seeks. Is that what people think?

If a statistical study of the above claims was made with comprehensive
samples of galaxies and quasars by computer analysis and it yielded
extremely low probabilities (like say 10^-10) would cosmologists abandon
the big bang and cosmological red shift assumptions? If not, why not?

If someone wanted to do such a study, which galaxy and quasar surveys
(please give links to where the data can be downloaded) would you
recommend be used that have been built without any selection criterion
(except brightness, z limits or region of sky) built into them? The
galaxy surveys do not need to be to very high magnitude as the
associations are generally with very nearbt galaxies. The quasar survey
is perhaps more difficult and advise on this is sought.

It is proposed that the above associations, particularly number 2, be
tested. The samples should be of as much of the sky as possible but need
not be the whole sky, but should be to some magnitude within the region
studied. To act as a control and prove that the statistical calculations
are correct, it is proposed to do the analysis a number of extra times
with the galaxy sample rotated at an arbitrary angle in space to the
quasar sample. If the associations show up strongly in the real sample
but not in the rotated sample then Arp must surely be accepted as right.
If the results are similar then Arp can comfortably be dismissed. Is
this a fair test?

Ray Tomes