Full text of DOI:10.1046/j.1365-8711.2000.03687.x

Mon. Not. R. Astron. Soc. 317, 105±110 (2000)
A spectroscopic study of NGC 6251 and its companion galaxies
P. N. Werner,w D. M. Worrall and M. Birkinshaw
Department of Physics, University of Bristol, Tyndall Avenue, Bristol BS8 1TL
Accepted 2000 March 24. Received 2000 March 21; in original form 1999 November 30
A B S T R AC T
Measurements of the velocities of galaxies thought to be associated with the giant radio
galaxy NGC 6251 confirm the presence of a poor cluster with a systemic redshift of zꢀ ˆ
0:0244 ^ 0:0004 and a line-of-sight velocity dispersion of sz ˆ 283ꢀ1109; 252† km s21
:
This suggests a cluster atmosphere temperature of T ˆ 0:7ꢀ10:6; 20:2† keV; which is not
enough to confine the radio jet by gas pressure. The core of NGC 6251 shows strong
emission lines of [O iii] and Ha1[N ii], but there is no evidence for line emission from the
jet (detected in optical continuum by Keel).
Key words: galaxies: clusters: individual: Zw 1609.018212 ± galaxies: distances and
redshifts ± galaxies: individual: NGC 6251.
flux and coincident with the radio core. The remaining ,10 per
cent of the flux appeared to originate from an extended emission
1
I NTRO DUC TION
NGC 6251 is an mB ˆ 13:6 mag (de Vaucouleurs et al. 1991) giant
E2 radio galaxy known for its remarkable radio jet, which is
aligned (within a few degrees) from pc to Mpc scales (see
Readhead, Cohen & Blandford 1978, Perley, Bridle & Willis 1984
and Jones et al. 1986 for pc-scale observations, kpc-scale obser-
vations, and both, respectively). Keel (1988) detected the radio jet
in optical continuum emission and found that the spectrum was
unusually steep. Arguments that the galaxy contains a super-
massive black hole with mass ,109 M( (Young et al. 1979) were
recently supported by HST observations of the nuclear gas and
dust disc on a scale of a few Â100 pc (Ferrarese & Ford 1999).
NGC 6251 has been described as relatively isolated, but loosely
associated with the Zwicky cluster Zw 1609.018212.
This cluster, which Zwicky (1968) described as `medium
compact' and `near' with a population of 260, was found by
Gregory & Connolly (1973) to have two main concentrations. One
of them is A2247, listed as a separate cluster by Abell (1958).
Like NGC 6251, it lies on the eastern boundary of Zw
1609.018212. Its systemic velocity is 11 670 km s21 (Sargent,
Readhead & de Bruyn 1982, unpublished), and its interest lies
mainly in the fact that it contains a striking chain of galaxies. The
western part of the Zwicky cluster surrounds IC 1143. The three
galaxies which Gregory & Connolly (1973) ascribed to this cluster
segment have velocities close to 6500 km s21. NGC 6251 appears
to depart from this velocity by ,1000 km s21, and Prestage &
Peacock (1988) concluded that the local galaxy density is so low
that any atmosphere would more likely be associated with the
galaxy than the cluster.
component, probably a gaseous atmosphere with a scale of
,3 arcmin (130 kpc)1 FWHM. They obtained a satisfactory fit to
the PSPC radial profile by superposing a point source and an
isothermal b-model. The limited number of counts from the
modelled extended emission did not allow a useful gas tem-
perature measurement to be made. However, Birkinshaw &
Worrall showed that for the outer parts of the kpc-scale radio jet to
be confined by atmospheric pressure, a gas temperature of T $
5 keV is required. Feature A (the brightest part of the jet, only 10±
40 arcsec or ,7±29 kpc from the nucleus) can be confined by a
gas temperature of T $ 2 keV: The galaxy's internal stellar
velocity dispersion of sz ˆ 293 ^ 20 km s21 measured by Heck-
man, Carty & Bothun (1985a) suggests that the stellar gravita-
tional field cannot hold an atmosphere this hot. In this paper we
demonstrate the existence of a cluster associated more strongly
with NGC 6251 than Zw 1609.018212 based on velocities for 13
galaxies including IC 1143, and we use the velocity dispersion of
this cluster to estimate the temperature of the cluster atmosphere.
We discuss the consequences for pressure confinement of NGC
6251's jet.
2
O B S E RVAT I O NS A N D DATA R E DU CT I O N
Observations of the galaxies listed in Table 1 were carried out on
the nights of 1992 September 22 and 23, using the Red Channel
spectrograph on the Multiple Mirror Telescope (MMT). All
images were obtained on a 800 Â 800 CCD (binned 2/1) with
(15 mm)2 pixels using a long slit of dimensions 180 Â 1:5 arcsec2:
Low-resolution spectra were taken with a 300 line mm21 grating,
Birkinshaw & Worrall (1993) observed NGC 6251 with the
ROSAT PSPC and detected a spatially unresolved X-ray
concentration, containing about 90 per cent of the 0.1±2 keV
Ê
Ê
which had a useable range of 2560 A centred on ,6000 A, and
Ê
had spatial and wavelength dispersions of 0.5 arcsec and 3.2 A per
w E-mail: p.werner@bristol.ac.uk
1 In this paper we adopt a Hubble constant H0 ˆ 50 km s21 Mpc21
:
q 2000 RAS

106
P. N. Werner, D. M. Worrall and M. Birkinshaw
Table 1. Observational details for the low-resolution spectra.
Object
Name
Coordinates (2000)
R.A.
Exp.(s) Date observed
Dec.
0
s
1
2
3
4
5
6a
6b
7
8
9
10
11a
11b
12
13
NGC 6251
±
16h32m34:8 8283201700
1800
900
600
900
450
450
00
600
600
600
600
22 Sept. 1992
00
s
16h32m47:7 8283302500
s
16h32m43:3 8283403600
00
23 Sept. 1992
NGC 6252
Zw 367-010
UGC 10581
UGC 10464a
UGC 10464b
Zw 367-015
UGC 10431
UGC 10604
Zw 367-016
UGC 10471a
UGC 10471b
IC 1143
s
16h22m46:1 8282304500
s
16h43m11:6 8283705100
00
s
16h28m34:6 82853 1600
00
00
00
00
00
00
00
00
00
00
00
00
0
s
16h37m05:5 8280003700
s
16h25m18:3 8184705700
s
16h47m50:2 8185005000
s
16h38m48:5 8184102800
s
16h30m28:5 81833 4400
600
00
300
300
00
00
0
s
15h30m48:5 8282703400
s
IC 1139
15h29m18:3 82835 1500
Table 2. Observational details for the high-resolution spectra of NGC 6251.
Slit Position
Angle (8)
l range
Ê
(A)
Exp. time Date observed
(s)
Comments
115.3
115.3
25.3
4874±5526
6274±6926
4874±5526
6274±6926
1800
1800
1800
1800
22 Sept. 1992
along jet
Hb,[O iii], Mgb
along jet
00
Ha, [N ii]
23 Sept. 1992
perpendicular to jet
Hb,[O iii], Mgb
perpendicular to jet
Ha, [N ii]
00
25.3
Ê
pixel, respectively. A resolution of ,10 A (FWHM) was obtained
in this instrument configuration.
The galaxies listed in Table 1 are numbered in order of
increasing distance from NGC 6251 (which is object 1). Our
sample (which is not complete or magnitude-limited) includes two
galaxies with double cores, objects 6 and 11. Although in both
cases the two nuclei are listed separately (and labelled a and b),
the average redshifts were used as single measurements in
subsequent cluster-related calculations, as they are clearly bound
systems.
In addition, high-resolution spectra of NGC 6251 were obtained
(see Table 2) with a 1200 line mm21 grating, resulting in a
Ê
Ê
wavelength range of 652 A and a dispersion of 0.9 A per pixel.
Ê
Spectra centred on 5200 and 6600 A were obtained along the radio
jet (at a position angle of 1158: 3) and perpendicular to it ꢀP:A: ˆ
Ê
258: 3†: The resolution of these spectra was 2±3 A (FWHM). The
objective of these observations was to search for emission lines in
the optical knots detected in continuum by Keel (1988), and to
measure the rotation velocity and direction for the stars and gas in
the galaxy.
Figure 1. Raw (unfluxed) spectrum (in MMT counts) of the core of NGC
6251 in the observed frame. Note the strong [O iii], [O i] and Ha ‡ ‰N iiŠ
Ê
lines, as well as stellar absorption features. The resolution is ,10 A
(FWHM).
The reduction and analysis of our data were carried out using
the Image Reduction and Analysis Facility (iraf). The CCD
images were de-biased, trimmed and flat-fielded using the
ccdproc task. Cosmic ray hits and bad lines or columns on the
CCD were removed by interpolating across the smallest dimen-
sion of the affected regions with the fixpix routine. For the low-
resolution case, where only a nuclear one-dimensional spectrum
was required, the spectra were sky-subtracted using the back-
ground task, then wavelength-calibrated using HeNeAr compari-
son spectra, traced and extracted with the apextract package.
The data reduction in the high-resolution case was somewhat
complicated by the fact that information in the spatial direction
was to be preserved, so that the core spectrum could not simply be
traced along the geometrically distorted two-dimensional image.
A good alignment (subsequently found to be accurate to typically
a few tenths of a pixel) of the dispersion axis and spatial axis with
the CCD's columns and lines, respectively, was carried out using
the longslit package. HeNeAr calibration lines and the trace of
the galaxy nucleus were used to fit a two-dimensional geometrical
rectification function to the data. This correction was then applied
q 2000 RAS, MNRAS 317, 105±110

NGC 6251 and its companion galaxies
107
Table 3. Redshifts obtained from emission lines.
Object
Emission lines used
V(em)
(km s21
)
1
4
6a
7
8
10
11a
[O iii] 4958, [O iii] 5007, [O i] 6300 [N ii] 6584
[O iii] 4958, [O iii] 5007, [N ii] 6563 Ha, [N ii] 6584
[O iii] 4958, [O iii] 5007, [N ii] 6563 Ha, [N ii] 6584
[O iii] 4958, [O iii] 5007, Ha, [N ii] 6584, [S ii] 6716, [S ii] 6731
[N ii] 6563, Ha, [N ii] 6584
7276 ^ 75
6930 ^ 75
7723 ^ 47
4141 ^ 41
7635 ^ 63
16648 ^ 51
7128 ^ 54
Ha, [N ii] 6584
Hb, Ha, [N ii] 6584
Table 4. Results of the cross-correlations and peculiar velocities relative to the systemic velocity of
the cluster.
Object
V(abs)
z(abs)
Vpec
(km s21
Vpec=sz
Distance from NGC 6251
(arcmin)
(km s21
)
)
(Mpc)
1
2
3
4
5
6a
6b
7
8
9
10
11a
11b
12
13
7459 ^ 22
6913 ^ 56
7240 ^ 46
7001 ^ 38
7223 ^ 44
7797 ^ 43
7748 ^ 43
0:0249 ^ 0:0001
260
0.67
20.70
0.11
0
0
0:0231 ^ 0:0002 2273
0:0242 ^ 0:0002 46
0:0234 ^ 0:0001 2187
0:0241 ^ 0:0001
0:0260 ^ 0:0001
0:0258 ^ 0:0001
1.21
2.33
21.1
21.3
22.3
00
33.0
46.8
51.8
52.4
58.7
00
0.052
0.101
0.911
0.919
0.964
00
1.42
2.02
2.24
2.26
2.54
00
20.48
0.08
1.46
30
567
00
a
a
b
d
a
a
00
b
4224 ^ 140 0:0138 ^ 0:0004
±
±
1.11
c
c
d
±
7554 ^ 56
±
0:0252 ^ 0:0002
432
353
±
236
00
0.91
±
20.09
c
c
b,d
a
b,d
±
7158 ^ 102 0:0239 ^ 0:0003
±
a
a
a
00
7154 ^ 52
6548 ^ 40
6616 ^ 71
0:0239 ^ 0:0002
0:0218 ^ 0:0001 2629
0:0221 ^ 0:0002 2563
21.62
21.45
121
123
5.21
5.29
a average velocity of bound binary system used in velocity dispersion calculation.
b not cluster members ꢀVpec . 3000 km s21†:
c The weak continuum did not allow a good cross-correlation to be carried out.
d The peculiar velocity is calculated from the emission velocity.
to the image, and one-dimensional spectra at varying offsets from
the core of the galaxy were subsequently extracted.
Fig. 1 is an unfluxed spectrum of the core of NGC 6251 from
our low-resolution data.
September 22 observing run. NGC 7619 is a 12.1 mag E2 galaxy
with a velocity of V ˆ 3747 ^ 20 km s21 (Huchra et al. 1983).
More recent velocities for NGC 7619 (e.g. de Vaucouleurs et al.
1991; Huchra et al. 1995) are in reasonable agreement, so that any
systematic error introduced into our results by the choice of
template velocity is likely to be less than a few tens of km s21
.
The cross-correlation velocities of the galaxies in our sample
are shown in Table 4.
3
R EDS HI FTS
3.1 Measurements
Seven of the 15 spectra showed emission lines which enabled
redshifts to be determined. Table 3 shows which lines were used,
and the resultant heliocentric velocities. The error bars take
account of the scatter of the velocities calculated from the various
lines, as well as the errors on the individual line wavelengths. Due
to a slight defocusing effect which is particularly noticeable in the
emission lines, an error of 10 per cent of the FWHM of the line is
included in these latter errors.
3.2 Comparison of emission and absorption redshifts
Comparing the velocities obtained from emission lines (Table 3)
and absorption features (Table 4), it is evident that, although the
results are all consistent within the errors, there is a systematic
Vꢀabs† 2 Vꢀem† ˆ 65 ^ 9 km s21 (leaving out NGC 6251, which
will be considered separately). This could be due to the slight
defocusing effect described in Section 3.1. Alternatively, it could
indicate that the adopted velocity of the template galaxy NGC
7619 is a slight overestimate (see Section 2). Making a correction
of 65 km s21 to the V(em) of object 8 (the only emission velocity
used in the velocity dispersion calculation) has a negligible effect
on the systemic velocity and on sz.
Part of the unusually high discrepancy between the emission
and absorption redshifts for NGC 6251 can be ascribed to this
systemic error. The remaining ,100 km s21 difference is in
agreement with the results of Ferrarese & Ford (1999), who also
find the emission lines to be at a significantly lower velocity than
the absorption features. Adding up the Gaussian constituents of
We measure the absorption redshifts of the galaxies by the
standard method of cross-correlating their stellar spectra with the
stellar absorption-line spectrum of an object of known redshift
(Tonry & Davis 1979). The cross-correlation was carried out over
Ê
the wavelength ranges 5000±5450 and 5950±6200 A (including
the Mgb and Na D features while avoiding poorly subtracted night
Ê
sky lines around 5600 A) using the iraf task fxcor in the rv
package.
The template used for the cross-correlation was a very high
signal-to-noise spectrum of NGC 7619 obtained from a 600-s
exposure (with the 300 line mm21 grating) during the 1992
q 2000 RAS, MNRAS 317, 105±110

108
P. N. Werner, D. M. Worrall and M. Birkinshaw
Table 5. Comparison of measured velocities with the
literature
The mean velocity of the cluster was found to be czꢀ ˆ
7193 ^ 128 km s21; equivalent to a redshift zꢀ ˆ 0:0240 ^ 0:0004:
The one-dimensional line-of-sight velocity dispersion sz ˆ
Referencea
388ꢀ1128; 265† km s21
; yields a three-dimensional physical
Object
1
This paper
km s21
Published values
km s21
velocity dispersion of s ˆ 672ꢀ1239; 2145† km s21
:
7459 ^ 22
7400 ^ 22
6900 ^ 42
7380 ^ 100
7294 ^ 60
6900 ^ 600
6510 ^ 100
6840 ^ 100
7789 ^ 24
7710 ^ 100
17010 ^ 100
1774 ^ ±
7320 ^ 100
20208 ^ 184
7590 ^ 100c
6930 ^ 100
6395 ^ 143
Zcat 95
RC3 91
SRdB 82
SO 81
MO 79
SRdB 82
SRdB 82
SHDYFT 92
SRdB 82
SRdB 82
RH 91
SRdB 82
GC 73
SRdB 82
SRdB 82
GC 73
Such a low velocity dispersion suggests that the cluster is poor.
A comparison with the correlation between s and Abell richness
class R (e.g. Danese et al. 1980) indicates R # 0; corresponding to
a population #30.
As the results in Table 4 indicate, NGC 6251 has a peculiar
velocity of 260 ^ 130 km s21 ˆ 0:67 ^ 0:33 sz within the cluster.
It is generally believed that giant elliptical radio galaxies
dominating clusters are more or less at rest relative to the
cluster's systemic velocity. Although this is consistent at the two-
sigma level with the data above, there is a reason that the peculiar
velocity of NGC 6251 given in Table 4 may be an overestimate.
Considering the projected distances of objects 12 and 13 from the
rest of the cluster (.5 Mpc), their comparatively large peculiar
velocities ($1.45sz), and their closeness to one another in
position and velocity, it seems likely that they are part of the
physically distinct IC 1143 group, and that they should be
removed from the velocity dispersion calculations. This leads to a
significantly higher systemic velocity for the cluster, czꢀ ˆ 7328 ^
105 km s21 ꢀzꢀ ˆ 0:0244 ^ 0:0004† with sz ˆ 283ꢀ1109; 252†
km s21. In this case, NGC 6251 is one of the slowest-moving
members of the cluster, with a peculiar velocity of 128 ^
3
4
6
7240 ^ 46
7001 ^ 38
7773 ^ 30b
7
8
9
4224 ^ 140
7621 ^ 61
7554 ^ 56
10
11
12
16644 ^ 6
7156 ^ 57b
6548 ^ 40
a references: Zcat 95 Huchra et al. 1995; SHDYFT 92 Strauss
et al. 1992; RC3 91 de Vaucouleurs et al. 1991; RH 91
Richter & Huchtmeier 1991; SRdB 82 Sargent et al. 1982,
unpublished; SO 81 Shuder & Osterbrock 1981; MO 79
Miley & Osterbrock 1979; GC 73 Gregory & Connolly 1973.
b mean velocity of two core components, as previous authors
did not take account of the double nucleus.
c
This value is placed between brackets in the SRdB 82
paper, indicating that the authors are not entirely confident
about it.
107 km s21
:
their fits to the Ha 1 ‰N iiŠ system over all their (0.1-arcsec) HST
apertures produces an emission-line spectrum which is remarkably
similar in shape and position to our nuclear (1.8-arcsec aperture)
spectrum.
5
T HE DYNAM IC S O F N GC 6 2 51
Heckman et al. (1985b) measured the rotation curves of several
radio-loud and radio-quiet elliptical galaxies to test theories that
the rotation axis might coincide with the radio axis in powerful
radio galaxies ꢀlog P178 MHz $ 24 W Hz21†: They present rotation
curves for NGC 6251 in position angles of 278 and 1248, and
conclude that the galaxy rotates about an axis at P.A. ,858 ^ 288
3.3 Comparison of velocities with the literature
The redshifts of several of the galaxies observed here have also
been measured by other authors. They are compared with our
results in Table 5. There is considerable discord among the
velocities, with the most severe disagreement arising when
comparing our values to those published in the 1970s and early
1980s.
with a velocity of V , 47 ^ 16 km s21
:
We measured rotation curves for two position angles, 258: 3 and
1158: 3 (Table 2), the latter being aligned with the jet. The results
of the slit in P:A: ˆ 258:3 suggest a slow rotation of the order of
50 km s21 about the jet, consistent with Heckman et al.'s result in
P:A: ˆ 278: The velocities along the jet were inconclusive. Deeper
observations are required to obtain a sufficient signal-to-noise
ratio in the outer parts of the galaxy (at a radius $8 arcsec or
,6 kpc) if more precise information on the rotation axis or
velocity than that found by Heckman et al. is to be obtained.
Of the five galaxies showing serious discrepancies with
published values, one (source 7) has consistent absorption and
emission redshifts in this paper. We have a very high-quality
spectrum for object 3, and the cross-correlation with the template
galaxy yielded a high Tonry & Davis `R' factor (an indication of
the reliability of the resulting redshift). The coordinates given by
Richter & Huchtmeier (1991) for object 8 are 2 arcmin away from
the correct position, and Sargent et al. (1982, unpublished)
indicate that they are unsure of the accuracy of their velocity
measurement for source 10; in both of these cases, we suspect that
target misidentification in the earlier observations may be the
reason for the different redshift values.
6
L INE E MISSION ALO NG T HE RA DIO J ET?
The jet of the powerful radio galaxy M87 has been detected at
optical wavelengths, and many authors report imaging and
spectroscopic studies (see Keel 1988, and references therein).
Keel subsequently detected optical continuum emission from the
NGC 6251 jet, and his B-, V- and R-band fluxes of feature A (10±
40 arcsec or ,7±29 kpc from the nucleus) imply a spectrum much
steeper than M87 or other similar jets.
4 CLUS TE R SYS TE M IC REDSH I FT A ND
VE L OCI TY D I SP ERSI ON
Our observations were not deep enough to reveal significant
continuum emission from feature A, but we searched for line
emission from [O iii], Ha and [N ii]. No emission lines in the jet
We used the cross-correlation results in Table 4 to compute the
systemic redshift and the velocity dispersion of the cluster of
galaxies using the method of Danese, De Zotti & di Tullio (1980)
and as described by Werner, Worrall & Birkinshaw (1999).
were found, at an upper limit (for Ha) of 10217
±
Ê
10216 erg cm22 s21 for a range of linewidths of 300±1000 A,
q 2000 RAS, MNRAS 317, 105±110

NGC 6251 and its companion galaxies
109
82 30
00
81 30
00
17 00
16 50
40
30
20
10
00
15 50
40
RIGHT ASCENSION (J2000)
Figure 2. Second Digitized Sky Survey (DSS2) image, with our targets indicated by small circles. The ,18-diameter circle in the south-eastern corner of the
field is the outline of Abell 2247. The large circle indicates the bounds of Zwicky cluster Zw 1609.018212.
respectively. Poor weather prevented us from obtaining the
planned, much deeper, exposures.
corresponds to an X-ray temperature of T ˆ 1:2ꢀ10:8; 20:3† keV:
The alternative velocity dispersion of sz ˆ 283ꢀ1109; 252†
km s21 derived for the cluster without IC 1139 and 1143 (which
we argue are part of a separate group) suggests a lower gas
temperature of T ˆ 0:7ꢀ10:6; 20:2† keV:
7
D I SCUS S ION
Comparing this temperature of 0.7 keV with the extrapolated
LX 2 T relation of Arnaud & Evrard (1999), we find that it
7.1 NGC 6251 and Zw 1609.018212
corresponds to
2:4ꢀ111:6; 21:5†  1042 erg s21
a
bolometric X-ray luminosity LXꢀbol†
.
Considering our redshifts in the context of the two subsystems that
make up cluster Zw 1609.018212 (see Section 1), it is clear that
our sample does not include any members of the subcluster
separately listed as A2247 ꢀczꢀ ˆ 11 670 km s21†; and situated in
the eastern part of the Zwicky cluster, South-East of NGC 6251.
Objects 12 and 13 appear to belong to a slightly closer group or
small cluster on the western edge of Zw 1609.018212 and we
have accordingly argued for their removal from the velocity
dispersion calculations in Section 4. Objects 7 and 10 have
redshifts unlike those of other objects in the sample, and appear to
be more or less isolated field galaxies. This leaves us with eight
companion galaxies (two with double nuclei) near the velocity of
NGC 6251, and these seem to form a third subcluster within the
limits of Zw 1609.018212. It is clear that not all the brightest
galaxies within Zwicky's boundary are physically associated.
Fig. 2 shows the positions of our sources, as well as rough
outlines of clusters A2247 and Zw 1609.018212.
;
or LX . 2:2ꢀ110:4; 21:4† Â
1042 erg s21 in the band 0.2±1.9 keV. Worrall & Birkinshaw
(1994) estimate LXꢀ0:2±1:9 keV† ˆ 8  1041 erg s21 for gas within a 3-
arcmin (130-kpc) radius. Given the parameters of their b-model
fit, this corresponds to a total luminosity of LꢀX0:2±1:9 keV† . 2 Â
1042 erg s21 for the whole cluster, which is in good agreement with
our velocity dispersion via the sz 2 T and LX 2 T relations.
The predicted temperature for the velocity dispersion also
agrees with two-component spectral fits based on ROSAT PSPC
data (Birkinshaw & Worrall 1993), within the rather large
uncertainties. It is clear that the gas that can be confined in this
cluster is too cool for static pressure confinement of the radio jet at
distances between 5 and 200 kpc from the core.
8
C ONC LUSION S
In this paper we have shown that:
(i) there is a poor cluster of galaxies associated with NGC 6251;
it is one of at least three distinct and not necessarily related
subclusters of Zw 1609.018212;
(ii) the temperature of the cluster atmosphere inferred from the
velocity dispersion measurement is not high enough to provide
static pressure confinement of any part of the kpc-scale radio jet
7.2 Velocity dispersion and X-ray temperature
The one-dimensional velocity dispersion of a cluster of galaxies is
related to the X-ray gas temperature of the cluster atmosphere
(Lubin & Bahcall 1993; Bird, Mushotzky & Metzler 1995; Girardi
et al. 1995). The higher velocity dispersion reported in Section 4
q 2000 RAS, MNRAS 317, 105±110

110
P. N. Werner, D. M. Worrall and M. Birkinshaw
Â
G., Fouque P., 1991, Third Reference Catalogue of Bright Galaxies.
(at distances of 5 to 200 kpc from the core) according to the lower
limits on jet pressure calculated by Birkinshaw & Worrall (1993);
(iii) a tentative rotation curve perpendicular to the direction of
the jet agrees with that measured by Heckman et al. (1985b), and
(iv) there is no evidence for [O iii], Ha, or [N ii] line emission
at feature A on the jet, where Keel (1988) found optical continuum
emission.
Springer-Verlag, New York
Ferrarese L., Ford H. C., 1999, ApJ, 515, 583
Girardi M., Fadda D., Giuricin G., Mardirossian F., Mezzetti M., Biviano
A., 1996, ApJ, 457, 61
Gregory S. A., Connolly L. P., 1973, ApJ, 182, 351
Heckman T. M., Carty T. J., Bothun G. D., 1985a, ApJ, 288, 122
Heckman T. M., Illingworth G. D., Miley G. K., van Breugel W. J. M.,
1985b, ApJ, 299, 41
Better optical spectra of NGC 6251 are needed to probe deeper
for line emission and to confirm the tentative dynamical results
presented in Section 6. Forthcoming X-ray observations should
measure the temperature of the cluster atmosphere, predicted to be
0.5±2.0 keV, based on our measurements of the cluster velocity
dispersion.
Huchra J., Davis M., Latham D., Tonry J., 1983, ApJS, 52, 89
Huchra J., Geller M., Clemens C., Tokarz S., Michel A., 1995, The CfA
Redshift Catalogue, Version June 1995, Harvard-Smithsonian Center
for Astrophysics
Jones D. L. et al., 1986, ApJ, 305, 684
Keel W. C., 1988, ApJ, 329, 532
Lubin L. M., Bahcall N. A., 1993, ApJ, 415, L17
Miley G. K., Osterbrock D. E., 1979, PASP, 91, 257
Perley R. A., Bridle A. H., Willis A. G., 1984, ApJS, 54, 291
Prestage R. M., Peacock J. A., 1988, MNRAS, 230, 131
Readhead A. C. S., Cohen M. H., Blandford R. D., 1978, Nat, 272, 131
Richter O.-G., Huchtmeier W. K., 1991, A&AS, 87, 425
Shuder J. M., Osterbrock D. E., 1981, ApJ, 250, 55
Strauss M. A., Huchra J. P., Davis M., Yahil A., Fisher K. B., Tonry J.,
1992, ApJS, 83, 29
ACK NOWL E DGM ENT S
Observations reported in this paper were obtained at the Multiple
Mirror Telescope Observatory, a facility operated jointly by the
University of Arizona and the Smithsonian Institution.
The Second Digitized Sky Survey (DSS2) image was obtained
from the ESO/ST-ECF Archive (http://archive.eso.org/dss/dss).
Tonry J., Davis M., 1979, AJ, 84, 1511
Werner P. N., Worrall D. M., Birkinshaw M., 1999, MNRAS, 307, 722
Worrall D. M., Birkinshaw M., 1994, ApJ, 427, 134
Young P. J., Sargent W. L. W., Kristian J., Westphal J. A., 1979, ApJ, 234,
76
REFER ENC ES
Abell G. O., 1958, ApJS, 3, 211
Arnaud M., Evrard A. E., 1999, MNRAS, 305, 631
Bird C. M., Mushotzky R. F., Metzler C. A., 1995, ApJ, 453, 40
Birkinshaw M., Worrall D. M., 1993, ApJ, 412, 568
Danese L., De Zotti G., di Tullio G., 1980, A&A, 82, 322
de Vaucouleurs G., de Vaucouleurs A., Corwin H. G. Jr., Buta R. J., Paturel
Zwicky F. E., 1968, Catalogue of Galaxies and of Clusters of Galaxies,
Vol. VI. California Institute of Technology, Pasadena
A
This paper has been typeset from a T X/LT X file prepared by the author.
E
E
q 2000 RAS, MNRAS 317, 105±110