Added 6/16/2005, modified 6/19/05



R. Rydin comments


A series of long-distance measurements of galaxy distributions has been made, called the deep red-shift North-South (N-S) Galactic pencil Surveys [1]. The data extend out to 5 billion light years in either direction from the plane of the Milky Way, and exhibit the remarkable structure of an almost symmetric damped sinusoid with an apparent period of about 400 million light years measured in today's distances, as shown in the accompanying Figures. A second survey was done at an angle of 45 degrees to the plane of the Milky Way that shows remarkable similarities to the N-S survey. The correlation coefficient of the peak separations is greater than 0.9, meaning that the period is real and the data are correlated. The distribution of galaxies has a common cause!

The actual N-S data are presented below, where the data represent the total count of galaxies in a space interval in a small cone versus distance, as converted from redshift using Hubble's Law.

The data were analysed using a "Pairs Correlation". Since the North and South surveys are really independent, this is a combination of an autocorrelation function and also a cross correlation of the two sets of data! The strong spatial periodicity was confirmed by doing a Fourier Transform spectral analysis of the data.


The 45- degree data are presented below.

 Figures are from Koo, et al, 1993, ASP 51, pp17-25 (Ref 1)

The radial autocorrelation function of the N-S galaxy distribution data given in Figure 2 is shown in Figure 3. For a purely random distribution of galaxies, the autocorrelation should be a Dirac delta function, that is, there should be a single tall narrow peak at 0 with a magnitude of zero everywhere else. For a correlated distribution, such as a Big Bang expansion from a single original region, the peak at 0 should be much lower, and there should also be an intersecting curve with a smooth drop off because the correlation would lessen with distance as motion randomized the distribution. In both cases, the area under the curve would be exactly the same.

The actual autocorrelation function exhibits a very low peak at 0, with most of the area under the curve spreading out slowly with distance, indicating a very high degree of correlation over the entire set of data. Furthermore, a sinusoidal distribution with a fixed period of 410 million light years is superimposed on the slowly decreasing portion, visible until the statistics of the data mask it!

In fact, since the North and the South sets are independent observations, the autocorrelation of the complete set contains the crosscorrelation of North and South!

The unnormalized radial autocorrelation function of the 45 degree pencil survey was computed in Mathcad using the original 10 Mpc bins, and then repeated using 25 Mpc bins to make the data comparable to the N-S data. The autocorrelation function of the N-S survey was repeated, and the two sets of data were crosscorrelated radially. The results are shown below, where it is seen that both sets of data exhibit the same periodicity with respect to the Earth, and both sets are completely interconnected by a common cause that is radial and not planar!

In addition to the deep pencil beam surveys, data also exist for wider but shallower surveys of quasars and radio galaxies, peculiar velocity information, Lyman-alpha break galaxies, optical and IRAS galaxies and clusters of galaxies, X-ray clusters of galaxies, and the supercluster void distribution. "Different types of observational data [2] point to the existence of a characteristic scale of about 130h-1 Mpc (400 million light years) in the large scale structure, and moreover to a regular density distribution of luminous matter in the universe. It was shown that a random structure could not explain the observed distribution. Statistical analysis of the deviations from periodicity showed that, even for a perfectly regular structure, a somewhat favored direction and/or location within the structure may be required. The presence of the observed periodicity up to a great distance and in different directions seems rather amazing. No generally accepted theory of structure formation yields a regular structure. The presence of the observed periodicity up to a great distance and in different directions, and its physical origin, is still an open question. Thus, the observations and their theoretical analysis present growing evidence for the regularity of the supercluster-void network: different objects trace the same high-density regions at large scales and suggest the existence of a typical large scale matter distribution of the universe. Rich superclusters and voids form a quasi-rectangular lattice with a mean separation of rich superclusters across voids 120 - 130h-1 Mpc. This scale is significantly larger than predicted by standard models of structure formation by gravitational instability. We consider the possibility that density fluctuations required to explain the present cosmological largest scale structures of the universal texture may have arisen in a different way from the standard way, and they may result from a completely different mechanism not necessarily with gravitational origin."



Mock pencil-beam redshift surveys from very large cosmological N-body simulations of two Cold Dark Matter (CDM) model universes have been created and compared [3] with the apparent periodicity observed in the data of Broadhurst et al. (BEKS, 1990). "Although the simulated redshift histograms frequently display regularly-spaced clumps, the spacing of these clumps varies between catalogues and there is no 'preferred' period over the many realizations. The power spectrum analysis alone shows that the BEKS data are significantly more periodic than the models. The supercluster statistic gives a two per cent probability of finding a structure as regular as the BEKS data. Restricting to a length scale ~100-150 h-1 Mpc, however, the number of samples which show the kind of periodicity seen in the BEKS data is extremely small for each of these statistics. Overall no sample is more regular than the BEKS data for all three statistics for a single period."

"The two popular CDM models studied here are apparently unsuccessful in reproducing the observed periodicity. From this result, together with the fact that the statistical results appear to be insensitive to the choice of the bias model, we conclude that CDM models conflict with the BEKS observation. Either the models need additional physics, or the data are a fluke or are somehow biased."

"Various possible physical explanations have been proposed, such as coherent peculiar velocities (Hill, Steinhardt and Turner 1991) oscillations in the Hubble parameter (Morikawa 1991) or baryonic features in the power spectrum (Eisenstein et al. 1998), but all of them seem to require either additional mechanisms with fine tuning beyond the standard theory or cosmological parameters significantly different from currently favoured values. Follow-up observations to BEKS by Koo et al. (1993) did not show a strong regularity in two other directions, although around the Galactic Pole the regularity was found to be further strengthened. Our results [3] give the a priori probability for such apparent periodicity in CDM models. Several more deep surveys might suffice to judge whether the discrepancy with BEKS reflects a major inconsistency."

A well-respected theorist, Paul Steinhardt, told me in an e-mail that the periodic correlation is only seen in the numerous 1D pencil surveys, begins to disappear in 2D, and washes out in 3D. He takes that as evidence that there is no periodicity at all! The 2D survey that is referred to is a medium-deep minislice [4], 4 degrees wide by 49 minutes broad, taken close to the north galactic pole. It is incomplete in that only 40% of the galaxies identified had their redshifts measured at the time of publication. Nevertheless, these data are not at all inconsistent with the radial periodicity shown in the other surveys and do not suggest a transverse periodicity. In fact, the minislice data were crosscorrelated to the N-S data and 45 degree data, and exhibit roughly the same 400 million light year radial periodicity, as seen below.

3D surveys would be shallow in the radial direction and broad in the transverse directions. In fact, that is a characteristic of the "Great Wall"; it is a radial wall and does not have corrugations in it in either transverse direction. Hence, if the correlation only existed in the insufficiently-sampled radial direction, 3D correlations would indeed fail to show periodicity. A more likely explanation is that the periodicity is truly a 1D radial depth effect and does not appear in the transverse directions at all.

Actually, there is a question here about the meaning of 3D correlations for galactic surveys. As radial distance increases, more and more volume is included in a narrow pencil cone. What geometry should we use? If it is Cartesian geometry, somewhat akin to a uniform expansion, then we have to correct each data point to make sure that it fits in a given cubical box in the big picture, especially if the r direction does not go along one of the sides of the box. Then we do the correlations across the boxes. If it is spherical, r-theta-phi, geometry, then the volume elements will not be of the same size with distance. The radial correlations can use a constant radial distance interval, but this allows the volume to increase with distance. One-D radial correlations are fair, subject to a correction for cone spread. But how do we treat the transverse correlations in a fair manner? At a short radius, we may have less than a 400 million light year period in the transverse directions, while at a larger radius we may have several of these periods. The only fair way to do this analysis is to do the transverse correlation separately for each radius and then see if the transverse directions line up. And then, what do we mean by line up? Should they line up in a transverse sense or in a Cartesian sense? The bottom line is that any 3D correlations that may have been done need to be carefully interpreted.



Furthermore, when the actual galaxy count data shown in figure 2 are divided by the one-over-r-squared distance from the Earth, to correct the data for the area spread of the cone, then the relative radial distribution drops off very rapidly in all directions from the Earth! The use of charge-coupled detectors allowed observation of galaxies out to greater than 20th magnitude. It is possible that the observations may have missed counting galaxies as distance increased due to decreasing brightness, thus accounting for some of the attenuation. In fact, astronomers assume that a strong falloff in brightness exists because they expect the universe to be homogeneous, and such an effect is called a luminosity function in the experimental papers. Nonetheless, it is hard to believe that these very clear patterns would be seen at all if there was a strong statistical effect due to brightness counting losses!

 My own feeling is that brightness losses account for only a fraction of the apparent attenuation with distance. The density of galaxies in the universe is not uniform at all, but decreases with distance, automatically resolving Olber's paradox!

  Logarithmic Scale of Relative Galaxy Density vs Distance from the Milky Way



Charles Sven presented detailed NASA and other data in his May 2005 paper at the Natural Philosophy Alliance meeting at the University of Connecticut [5]. Sven's contention is that the most recent observational data place the Earth near the center (Sven's estimate is 30 million light years) of a spherically uniform distribution of galaxies, quasars, gamma ray bursts, CMB, etc. Some of the data that he gathered from government websites is presented below. The Virgo Cluster is about 55 million light years away.

Note: For pie slice diagrams, the theta spread is accurately seen, but the data as a function of radius have been summed (integrated) over the phi angle out of the plane. This means that for a truly uniform density distribution, the brightness of the count data should increase directly with radius r from the origin. On the contrary, if the brightness appears constant, then the actual density is decreasing as 1/r !

The rather complete distribution of galaxies is shown below. Again, the distribution is quite symmetric with respect to Earth, and suggests an Origin somewhat to the North. We visibly see the Great Wall, and other shells or walls as radius increases. But the distribution does drop off with distance, despite the fact that the volume sampled increases strongly with distance, so this again supports the contention that the density of the universe is not constant but decreases strongly from the Origin.

A somewhat different representation of the data is shown below, where the spidery nature of the periodic concentric shells seems even more obvious!

The distribution of known quasars is shown below. The distribution is remarkably symmetric relative to the Earth. There are no quasars at all for 2 billion light years, and then the distribution apparently increases with distance. The quasar data plot actually represents much more sampled volume as distance increases, which is the similar to the 1-over-r-squared effect noted for the pencil surveys. Hence, the density of quasars actually drops off strongly with distance!

Furthermore, there is a visual hint of radial periodicity in this data!

In fact, the data are periodic, as seen below in a figure from Arp's book, Seeing Red [6]. Arp insisted on quantizing this data using a type of Bode's Law, but there is a rather clear periodicity with a step of dz = 0.3, as shown by red tick marks. Since there is some question whether quasars obey the same Hubble distance law as galaxies, it is somewhat problematical to expect to see the same 400 million light year period cited above.

Nonetheless, Arp comments on the pencil surveys, "To obtain the primary data, one had to read off the preferred redshifts from their graphs. It is clear that their main peaks were around z = .06 {essentially zero} and z = 0.30 with some fine structure. This is extraordinarily interesting because this coincides with the first two peaks for the redshifts for quasars."

There is also extensive data on Gamma Ray Bursts. Sven points out that the distribution is remarkably uniform with respect to the Earth, again placing us near the Origin.

Finally, Sven presents the CMB data, but without the usual false coloration normally used to emphasize certain patterns that Cosmologists use to imply Primordial fluctuations that support their Big Bang theory. Again, he concludes that the CMB is remarkably uniform and spherically symmetric with respect to the Earth.



The final problem that frustrates Cosmologists is the disturbing observance of large-scale-drift in our local galactic clusters [7]. The Earth has a dipole-anisotropy with respect to the reference frame of the CMB, which appears to be centered near the super-cluster Virgo! The Earth appears to be moving towards Virgo at a deviation of about 30° and at a relative velocity of 600 km/s. But the cosmologists expected it to be moving the other way!

Indeed, a group of astronomers nicknamed the "Seven Samurai" has resolved the motion of these galaxies into proper Hubble motion related to the uniform expansion of space, and peculiar local motion caused by gravitational attraction [8]. The Hubble data exhibit a strange S-shaped deviation from a straight line at the Virgo distance! They have demonstrated that there is a net drift towards Perseus-Pisces, which is about 600 million light years from Earth. This region has been dubbed the Great Attractor, although there is no visible concentration of mass in this direction. The cause of this behavior is still unexplained.

There are other peculiarities with regard to Virgo. According to Arp [6], the only 6 galaxies having measured blueshifts are in this cluster, and there is a curious almost half periodicity relative to the Great Walls measured up to, but not beyond, Virgo that is also unexplained! Arp concludes, "Another key aspect of General Relativity is that all reference frames should be equivalent. Work by Franco Selleri [Athens conference 1997 (Open Questions in Relativistic Physics, Apeiron, 1998), and Foundation of Physics Letters, 10, 73, 1997] and others, however, shows that under the most general transformations of coordinates, the classical Sagnac experiment can only be reconciled if there is a primary reference frame. I feel that this result was logically almost forced by the discovery of the cosmic microwave background. This radiation, supposedly pervading all space, must form a unique reference frame in spite of the fact that arguments have been advanced that it does not contradict general relativity."

Finally, there was a "star-less galaxy" discovered in 2005 in the vicinity of Virgo, the first such object ever detected.



 1) D.C. Koo, N. Ellman, R.G. Kron, J.A. Munn, A.S. Szalay, T.J. Broadhurst, and R.S. Ellis, "Deep Pencil-Beam Redshift Surveys as Probes of Large Scale Structures", Astronomical Society of the Pacific, Conference Series, Vol. 51, 1993, and S.R. Majewski, class notes, Department of Astronomy, University of Virginia, March 1996.

  2) D.P. Kirilova and M.V. Chizhov, "Large Scale Structure and Baryogenesis", XXIst Moriond Astrophysics Meeting - Les Arcs, March 10-17, 2001.

  3) N. Yoshida, J. Colberg, S. D. M. White, A. E. Evrard, T. J. MacFarland, H. M. P. Couchman, A. Jenkins, C. S. Frenk, F. R. Pearce, G. Efstathiou, J. A. Peacock, and P. A. Thomas (The Virgo Consortium), "Simulations of deep pencil-beam redshift surveys", Mon. Not. R. Astron. Soc., 13 Mar 2001.

  4) C.N.A. Willmer, D.C. Koo, A.S. Szalay and M.J. Kurtz, "A Medium-Deep Redshift Survey of a Minislice at the North Galactic Pole", APJ 437, pp 560-563, 1994.

  5) Charles Sven, "The Big Bang Explained", 12th Annual Conference of the Natural Philosophy Alliance (NPA), University of Conecticut, Storrs CT, 23 - 27 May 2005.

  6) Halton Arp, Seeing Red, Aperion Press, 1998.

  7)  Lawrence Krauss, Quintessence, the Mystery of Missing Mass in the Universe, Basic Books, Parts II-IV, New York, NY, 2000.

  8) Alan Dressler, Voyage to the Great Attractor, Vintage Books, Random House, 1995.