NGC 7603 – the discordant redshift system

NGC 7603 (object 1 in Figure 1) system was included in Halton Arp’s Atlas of peculiar galaxies. Subsequently Arp (1971) noticed that the companion of NGC 7603, NGC 7603B (object 4 in figure 1), had much higher redshift than NGC 7603 and was connected to NGC 7603 by an anomalous arm. Unfortunately, not even the abstract of Arp (1971) seems to be available online. Fortunately, Arp (1972) summarizes the findings from the system:

The outer regions of the disk are very disturbed and from it leads a single, long arm or filament which terminates at a smaller companion galaxy. … The companion is the only conspicuous galaxy that is in the vicinity of NGC 7603, the single filament leads directly to the companion and terminates exactly at the position of the companion.

It turned out that the companion has a redshift of cz = 16800 km/s while the main galaxy has cz = 8700 km/s. The redshift difference seems to be clearly too big for them to be physically associated in the traditional thinking. Toomre & Toomre (1972) noted in passing:

NGC 7603 appears linked by a faint double “bridge” to a small companion which Arp (1971a) recently found to be redshifted 8000 km s-1 more than the main galaxy. Even if one imagines that those two galaxies superpose merely by chance, there remains the problem of why 7603 itself exhibits those and other wisps.

The double bridge issue refers to the fact that there seems to be another fainter bridge leading to the companion. This issue will be further discussed below. Arp (1974) also emphasized the disturbed appearance of the main galaxy by presenting an image that showed how the interior parts of NGC 7603 were disturbed. Arp presented the obvious question:

If the action of the high redshift companion has not distrubed the central galaxy, then what has?

Walker et al. (1974) used high-resolution electrograph imaging to study the possible connection between NGC 7603 and NGC 7603B. First they go over the previous studies on the issue and they give a brief description of Arp (1971) which is not freely available, so we’ll take a quote from Walker et al. about it:

Arp considers that the companion is not an accidental projection of a background galaxy onto the spiral arm of NGC 7603 because of the peculiar nature of both objects and because the halo surrounding the companion is brighter at the point where the luminous filament from NGC 7603 touches it, indicating physical interaction.

Walker et al. then proceeded to study the luminous connection and the halo brightening with high-resolution electrograph. They found that the companion’s luminosity profile is asymmetric in the side of the connecting bridge, but they were able to explain this asymmetry. The asymmetry arises because of the overlapping of the companion and the arm from NGC 7603 and there’s also a faint foreground star placed properly to enhance the asymmetry. They subtracted the luminosity of the NGC 7603 arm and the star, and the asymmetry vanished completely making them to conclude that they found no evidence for interaction between the companion and the connecting arm.

Arp (1975) provided new images of the system and noted that the arm to the companion continues beyond the companion:

This 48-inch Schmidt photograph shows not only this second fainter arm, but also a fainter continuation of the main arm that extends on beyond the companion.

He also mentioned that there were some unpublished photographs by Roger Lynds that showed the same thing.

Hoyle (1983) studied the system as part of his theoretical work on spiral galaxies and their halos. He mentioned five strange properties of the system that Arp had found, one being the fact that:

The light distribution of the satellite has an unusually sharp boundary.

Hoyle argued that as the apparently connecting material were centered on the companion but weren’t leading to any special place in the main galaxy, the situation had to be so that the companion was tearing material out of the main galaxy. According to Hoyle that would then suggest that the companion would be more massive than the main galaxy. He argued further that the main galaxy is not likely to be of low mass, leading to the suggestion that the companion has very high mass which lead Hoyle to suggest that the redshift of NGC 7603B might originate from gravitational redshift. He derived the needed mass for this and arrived to too large mass which would tear NGC 7603 apart if the companion would be so massive and so close to NGC 7603 as it appears, so Hoyle suggested that the companion must be in front or behind of NGC 7603. He estimated the proper distance for the companion for it not to tear the main galaxy apart and then calculated that:

The geometry is such that lines from the Galaxy to the satellite and to the Earth are inclined at about 5° if the satellite is in front, and at about 175° if the satellite is behind the Galaxy.

Sharp (1986) provided new images and spectra of the system. Sharp studied the hydrogen alpha images and found them to be largely featureless. He contrasted the situation with a reference system where companion didn’t have discordant redshift. In that system there were lot of hydrogen alpha emission. Sharp also noted that there were two knots in the arm of NGC 7603. Sharp also made surface brightness profiles for NGC 7603 and NGC 7603B and noted on NGC 7603B:

However, the structure as revealed by the profile is only really acceptable for a background object: companions to bright galaxies generally show either tidally stripped cores alone or a broad diffuse appearance (Wirth and Gallagher 1984).

But Sharp also noted that the profile not fitting to a companion galaxy doesn’t yet prove that NGC 7603 cannot be a companion. Sharp then proceeded to subtract smooth elliptical representation from the image of NGC 7603 to reveal faint features. Sharp confirmed the halo brightening that Arp originally noticed. Sharp also suggested that NGC 7603B had some spiral structure which enabled some distance indicator tests which resulted distances much further than NGC 7603 distance, but Sharp said the distance indicator results were only suggestive.

MacKenty (1990) presented new imaging of the system and concluded:

In agreement with Sharp (1986), the CCD images obtained here show two arms or tails crossing the “companion” galaxy and extending beyond it. The original assertion by Arp (1971) that the “arm” ended at the “companion” galaxy is not borne out by the deeper images.

López-Corredoira & Gutiérrez (2002) mentioned that Arp (1971) had already noted the two knots in the arm that Sharp (1986) also noted. López-Corredoira & Gutiérrez also mentioned that there had been previous unsuccesful attempts to get the spectra of the two knots. López-Corredoira & Gutiérrez succeeded in that. They measured the redshifts of the knots, which they said were very well centered in the arm. The resulting redshifts were z = 0.243 (object 3 in figure 1) and z = 0.391 (object 2 in figure 1). They also estimated the redshift of the arm itself to be z = 0.030 which agrees with the redshift of NGC 7603. They concluded:

We have clearly shown that two of the compact emission lines objects in the filament have redshifts very much greater than those of NGC7603 and its companion galaxy. Thus we have presented a very well known system with anomalous redshifts, NGC7603, to be an apparently much more anomalous than was previously thought. There are 4 objects with very different redshifts apparently connected by a filament associated with the lower redshift galaxy. This system is at present the most spectacular case that we know among the candidates for anomalous redshift.

López-Corredoira & Gutiérrez (2004) presented new observations (imaging and spectroscopy) of the NGC 7603 field. They noted that the halo of NGC 7603 is not symmetric with respect to NGC 7603. They also noted some additional filaments end extensions, one of which they suggested might be a counter-arm to the bright arm. They say about general appearance of NGC 7603 and its filaments:

Figures 1–3 show that NGC 7603 and its filament are apparently distorted by significant tidal interaction. The existence of the filament is also a possible sign of tidal interaction or a debris from satellite disruption (Johnston et al. 2001).

They then proceeded to study the neighborhood of NGC 7603. They performed a QSO search which resulted in 38 candidate objects but only 3 of them were considered as real QSO candidates and one of them they later showed spectroscopically not to be a QSO. They studied the two objects in the arm they found in 2002. They noted that they seem to be extended objects and they both seemed to have tails (one of the two pointing towards NGC 7603). They measured the spectrum of some of the objects in the field. Most of the objects they measured turned out to be HII-galaxies, including the two in the arm. They found some spectral features (H-alpha equivalent widths) that were quite exceptional, but would be normal if the objects would be assumed to be at NGC 7603’s distance. They confirmed the redshifts of the two objects in the arm. They also found three additional high redshift objects nearby (objects 8, 9, and 10 in Figure 1).

López-Corredoira & Gutiérrez (2004) also addressed the question of the perturber in the system, i.e. a candidate object that might cause the perturbed outlook of NGC 7603. They didn’t find very good candidates but they found at least one unlikely candidate 2.5 arcmin from NGC 7603 and in the direction of the arm. Couple of bigger galaxies were > 10 arcmin from NGC 7603.

They calculated probabilities in the system. They showed that if the two candidate QSOs in the field really would turn out to be QSOs, there still wouldn’t be an excess of QSOs in the field. They also calculated the probability for the three objects apparently connected to NGC 7603 by the arm. They used the area of the arm and calculated how likely it would be to find three objects within it. The result they got was “few times” 10-9. They also discussed some different variations of the probabilities in the system according to object types and such.

They considered some explanations for the system. Clusters at the same line of sight they found unlikely. Amplification of background objects by gravitational macrolensing wasn’t good explanation because it would call for a huge mass in NGC 7603. Microlensing they also found not likely because type of objects required for the lensing in this case were not likely to be present in such large numbers. Non-cosmological redshift they couldn’t reject but there also weren’t very good explanations for it. They also mentioned variable mass hypothesis which obviously doesn’t fail here because it has been developed with this kind of systems in mind. They considered the hypothesis of galaxies ejecting new matter also and thought that to fit the system very well. They noted that the two HII-galaxies in the arm have redshift counterparts in other objects they measured redshifts (z = 0.245 & z = 0.246 and z = 0.394 & z = 0.401) which would also support ejection hypothesis (as the closeness in redshift might suggest common origin) but it also could just suggest that there are two groups of objects in the field at those redshifts. They favor the ejection hypothesis because it explains the low probabilities in the system. They noted that the redshifts of objects were close to the Karlsson peak of z = 0.30. They also presented a sketch of the system assuming the ejection hypothesis.

López-Corredoira & Gutiérrez (2006) discussed the possible perturber in the system. They noted that there were two candidate galaxies at about 2 arcmin from NGC 7603 but that there were redshifts available for them and they had even higher redshift than NGC 7603B. They concluded that the perturbing galaxy doesn’t seem to be there if it is not NGC 7603B.


The spectrum of NGC 7603, like some other Seyfert galaxies, shows variability in monthly and yearly scales so that its spectral lines become weaker and stronger. This issue has been studied (among others) by Tohline & Osterbrock (1976) and Kollatschny et al. (2000).

Objects 6 (z = 0.077) and 7 (z = 0.129) in Figure 1 form a very well aligned line with the nucleus of NGC 7603.

Figure 1. The objects with measured redshifts near NGC 7603. Size of the image is 10 x 10 arcmin. Image is from Digitized Sky Survey (POSS2/UKSTU Blue).

Objects and their data

1 NGC 7603 SA(rs)b: pec 0.029524 14.04 0
2 NGC 7603:[LG2002] 3 galaxy 0.391000 21.4 (B) 0.622
3 NGC 7603:[LG2002] 2 galaxy 0.243000 21.4 (B) 0.859
4 NGC 7603B galaxy, Sy1 0.055742 16.4 (g) 0.983
5 SDSS J231855.52+001619.0 galaxy 0.076980 17.8 (g) 1.703
6 SDSS J231901.10+001651.8 galaxy 0.077111 17.3 (g) 2.493
7 SDSS J231906.07+001912.3 galaxy, NLAGN 0.128555 18.3 (g) 5.144
8 SDSS J231854.10+001411.6 galaxy 0.246 21.6 (g) 0.767
9 SDSS J231851.34+001410.6 galaxy 0.117 20.7 (g) 1.395
10 SDSS J231850.55+001413.0 galaxy 0.401 21.2 (g) 1.573

NED objects within 10′ from NHC 7603
SDSS image of the system.


Arp, 1971, ApL, 7, 221, “NGC 7603, a galaxy connected to a companion of much larger redshift”

Arp, 1972, IAUS, 44, 380, “Ejection of Small Compact Galaxies from Larger Galaxies”

Arp, 1974, IAUS, 63, 61, “Evidence for non-velocity redshifts”

Arp, 1975, PASP, 87, 545, “Photographs of 3C 120 and NGC 7603 Compared to Spectracon Images”

Hoyle, 1983, Ap&SS, 93, 55, “What is a spiral galaxy?”

Kollatschny et al., 2000, A&A, 361, 901, “Strong spectral variability in NGC 7603 over 20 years”

López-Corredoira & Gutiérrez, 2002, A&A, 390, 15, “Two emission line objects with z > 0.2 in the optical filament apparently connecting the Seyfert galaxy NGC 7603 to its companion”

López-Corredoira & Gutiérrez, 2004, A&A, 421, 407, “The field surrounding NGC 7603: Cosmological or non-cosmological redshifts?”

López-Corredoira & Gutiérrez, 2006, AIPC, 822, 75, “Research on candidates for non-cosmological redshifts”

MacKenty, 1990, ApJS, 72, 231, “Seyfert galaxies. I – Morphologies, magnitudes, and disks”

Sharp, 1986, ApJ, 302, 245, “Anomalous redshift companion galaxies – NGC 7603”

Tohline & Osterbrock, 1976, ApJ, 210, 117, “Variation of the spectrum of the Seyfert galaxy NGC 7603”

Toomre & Toomre, 1972, ApJ, 178, 623, “Galactic Bridges and Tails”

Walker et al., 1974, ApJ, 194, 125, “Direct electronographic observations of luminous connections between galaxies with discordant redshifts”

3 Responses

  1. Have you seen this letter?

    The Missing Goliath’s Slingshot: Massive Black Hole Recoil at M83

    The Astrophysical Journal Letters, Volume 717, Issue 1, pp. L42-L46 (2010).

    Horacio Dottori

  2. I hadn’t seen it but I’m looking at it now. Very interesting. I have to read it thoroughly tomorrow.

    Edited to add: here’s the link to the paper for interested.

  3. I looked at the paper more closely. There is (at least) one thing I don’t understand from the paper: why is there two redshifts (cz = 620 km/s and z = 0.018) measured for this object? Apparently your evidence suggests that this object might have been ejected from the nucleus of M83 and therefore could have large velocity explaining the discrepancy between redshifts of this object and M83, but that seems to leave the 620 km/s measurement unexplained.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: