[OII] Morphologies

The ELFIT2D results presented in Section can only be properly interpreted by looking at the morphologies of the [OII] emission. Clearly, V sin will not be a measurement of rotation if the [OII] emission is concentrated in the nucleus of the galaxy. This section provides comments on ELFIT2D results, broad-light and [OII] morphologies and broad-band disk scale lengths for each galaxy.

A2390-101033. This galaxy looks completely normal on deep SIS V and R images apart for what appears to be ringing on one side of the disk. The surface brightness profiles are all fitted with an exponential disk scale length of 2.5 h kpc (Figures , and ). There is a close companion at an apparent angular distance of 3which corresponds to 7.7 h kpc at a redshift of 0.2460. The SIS [OII] spectrum is striking (Figure ). It shows a strong [OII] emission superposed over a much fainter continuum. The continuum is extended whereas the [OII] emission is unresolved. All of the emission is concentrated in the nucleus of the galaxy. The [OII] scale length measured by ELFIT2D is 0.09 h kpc which is significantly smaller than the broad-band scale length. Moreover, there is an extension on one side of the [OII] emission. This extension remains when the main unresolved [OII] component is subtracted out. The extension could be the kinematical signature of a small companion galaxy in the process of merging with its host. If so, the extension would be direct evidence that minor mergers are responsible for enhanced star formation activity in some galaxies at intermediate redshifts. This galaxy is the prototype of a ``kinematically anomalous'' object. It resembles blue nucleated galaxies described by Schade et al. schade95.

A2390-100686. This galaxy looks normal in Gunn and and in I. The surface brightness profiles are barely resolved and have a scale length of 1.2 h kpc (Figures , and ). There are no close companions. The SIS [OII] spectrum shows a tilted, extended emission centered on a faint continuum as extended as the [OII] emission. The tilted emission appears ``puffed up''. It is probably a result of seeing the [OII] doublet components almost resolved by wavelength stretching from the cosmological expansion. This object was classified as kinematically normal.

A2390-350416. This galaxy looks normal in all three bandpasses. It is better resolved in Gunn than in Gunn as shown by the surface brightness profiles on the CNOC MOS images (Figures and ). The Gunn and Johnson I profiles are fitted with a disk scale length of 1.8 h kpc. There are no nearby companions. The SIS [OII] spectrum shows a tilted line as extended as the continuum light. The [OII] scale length is 1.9 which is equal to the I and Gunn scale lengths. The [OII] emission is centered on the continuum. This object was classified as kinematically normal.

A2390-350471. The I image suggests that this galaxy could be distorted, but the Gunn and images look smooth. The position angle of its isophotes varies substantially. However, the surface brightness profiles are very well fitted by an exponential profile. The I and Gunn profiles have a scale length of 1.4 h kpc (Figures and ). There are no close companions. The SIS [OII] spectrum shows a tilted line with no apparent distortion. The [OII] scale length is 2.2 . This object was classified as kinematically normal.

E1512-301037. This galaxy has faint distortions on the Gunn and CNOC MOS image. There are at least two faint tails. However, the Gunn and profiles are well-fitted by an exponential profile with a scale length of 3.8 h kpc (Figures and ). There are two faint close companions with an apparent angular distance of 5(16.1 h kpc at z = 0.3457). The SIS [OII] spectrum is by far the most intriguing one in the sample. It shows a very faint continuum, and the [OII] line is donut-shaped (Figure ). This line shape can be interpreted in two ways: either (1) the line is made of two rotation curves from two galaxies very close together, or (2) the line arises from an expanding supershell presumably driven by supernova winds from a massive starburst. Unfortunately, no deep SIS images are available for this object. If option (1) is true, then the separation between the two galaxies would be 0which corresponds to 2.5 h kpc at a redshift of 0.3457. Option (1) was retained for the ELFIT2D analysis. This is why there are two entries in Table . However, there is a problem with option (1). If this object is in fact a very close pair, then there should a systematic offset in wavelength between the two rotation curves due to the relative velocities of the pair members. No such offset is observed. If option (2) is true, then the supershell has a diameter of 2.5 h kpc and an expansion velocity of about 80 km/s. These characteristics are amazingly similar to those of supershells observed in local dwarf irregular galaxies [\protect\astronciteMartin1995][\protect\astronciteMarlowe et al.1995]. It is likely that this object is a so-called Doppler ellipse, the first one found at intermediate redshifts. This object was of course classified as kinematically anomalous.

E1512-101526. The Gunn image shows the galaxy has a close companion at an apparent distance of 2(6.8 h kpc at z = 0.4026). Gunn and profiles are well fitted with an exponential disk scale length of 3.1 h kpc (Figures and ). The SIS [OII] spectrum has a faint continuum, and the [OII] emission shows no sign of rotation and appears unresolved. The [OII] scale length measured by ELFIT2D is 0.1 h kpc which is significantly smaller than the broad-band scale lengths. This object was therefore classified as kinematically anomalous.

E1512-201429. This galaxy looks normal in all three bandpasses. It is resolved on the CNOC Gunn and the SIS I band images. The profiles can be fitted with a scale length of 1.8 h kpc. (Figure , and ) The Gunn image is not well resolved. The nearest companion is at an apparent angular distance of 5(16.4 h kpc at z = 0.4231). The SIS spectrum shows an extended continuum with a double tilted [OII] emission line (Figure ). The double line is in fact the two components of the [OII] doublet resolved as a result of wavelength stretching by the cosmological expansion. The blue component of the doublet clearly has a different brightness from the red component as a function of galactocentric radius. This indicates that the electron density in the interstellar medium of this galaxy varies as a function of distance from its center. This object was classified as kinematically normal.

E1621-100515. This galaxy looks normal in all three bandpasses. The profiles are all very well fitted by an exponential disk scale length of 1.5 h kpc (Figures , and ). There are no close companions. The SIS spectrum shows a tilted line with a faint, extended continuum. Unfortunately, the line received a string of cosmic rays hits which seem to have affected the ELFIT2D fit. This object was classified as kinematically normal, but it was not included in the final Tully-Fisher sample because of the effect of so many cosmic rays in the ELFIT2D fit.

A2390-100225. This galaxy looks normal in I and Gunn and . It is resolved on the SIS I and CNOC MOS Gunn images. The exponential profiles are well fitted with a scale length of 1.2 h kpc (Figures , and ). There are no close companions. The MOS spectrum shows a very faint, concentrated [OII] emission. ELFIT2D was not able to achieve a good fit of this [OII] distribution. The MOS ELFIT2D simulation (Figure ) shows that the [OII] flux received from this galaxy was simply too low for ELFIT2D to find meaningful parameter values.

A2390-101084. Deep SIS V and R images show that this distorted galaxy has at least four close companions within a radius of 4(9.8 h kpc). The Gunn images also show many faint companions. This galaxy is in fact the central cD galaxy of the cluster Abell 2390. This galaxy was observed serendipitously as part of one of the MOS multi-object exposures. The V and R surface profiles have a scale length of 2.5 h kpc (Figure ). The MOS spectrum shows a strong, unresolved [OII] emission line. The [OII] scale length is 0.4 h kpc which is significantly smaller than the broad-band scale length. This object was classified as kinematically anomalous. Since cD galaxies are believed to be the result of galaxy cannibalism, this suggests that kinematically anomalous galaxies may be related to merging events. The MOS ELFIT2D simulation (Figure ) shows no biases.

A2390-200928. This galaxy is not clearly resolved on the CNOC MOS images. There is no sign of distortions, and there are no close companions. The MOS spectrum shows a fairly concentrated [OII] emission. The [OII] scale length is 0.3 h kpc. This object was classified as kinematically normal. However, if the broad-band scale length has been seriously underestimated due to seeing effects, then this galaxy should have been classified as kinematically anomalous. The MOS ELFIT2D simulation (Figure ) shows no biases.

A2390-200802. This galaxy is barely resolved on the CNOC MOS images, and it looks more elongated in than in . The exponential profiles can be fitted with a scale length of about 1.2 h kpc (Figures , ). There are no signs of distortions, and there are no close companions. The MOS spectrum shows a compact, tilted line. The [OII] scale length is 1.0 h kpc which is identical to the broad-band scale length. This object was classified as kinematically normal. The MOS ELFIT2D simulation (Figure ) shows no r bias, but V sin appears to slightly overestimated. It is therefore not surprising to find this galaxy above the Tully-Fisher relation in Figure with a large error bar.

A2390-200372. The galaxy is fully resolved on the CNOC MOS Gunn and images, and it looks normal on both images. The exponential profiles are well fitted with a scale length of 1.7 h kpc (Figures and ). The nearest object is at an apparent angular distance of 5(17 h kpc at z = 0.3485). The MOS spectrum shows a nice tilted line over a faint, less extended continuum. The [OII] emission, which is centered on the continuum emission, has a scale length of 3.2 h kpc. This object was classified as kinematically normal. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-201845. The galaxy is fully resolved on the CNOC MOS Gunn and images, and it has a distorted morphology. There are two close companions within a radius of 4(12 h kpc at z = 0.3387). The profiles are not well fitted by an exponential. In fact, the profiles have the shape expected from a de Vaucouleurs profile in log space. If an exponential is used to fit the profiles, the resulting scale length is 4.8 h kpc (figures and ). The MOS spectrum shows a strong and slightly tilted [OII] line centered on a faint continuum. The [OII] scale length was 0.4 h kpc which is much smaller than the broad-band scale length. This early-type galaxy was classified as kinematically anomalous. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-201773. This galaxy looks distorted on the SIS I band and MOS Gunn images. There is a large extension on the west side. It may be a close pair in the process of merging. The surface brightness profiles on the CNOC MOS images are very well fitted with an exponential scale length of about 1.6 h kpc (Figures and ) whereas the SIS I band image yields a scale length of 1.2 h kpc (Figure ). The MOS spectrum shows a very strong, tilted line, but this line appears to be significantly distorted with respect to a pure rotation curve. This object was not included in the final Tully-Fisher sample. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-200730. The galaxy is fully resolved on the CNOC MOS Gunn and images, and it looks smoother on the image. There are no close companions. The exponential profiles are well fitted with a scale length of 1.5 h kpc (Figure and ). The MOS spectrum shows an unresolved [OII] emission. The [OII] scale length was 0.2 h kpc. This object was classified as kinematically anomalous. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-200334. The galaxy has a fully resolved exponential profile in Gunn with a scale length of 2.0 h kpc (Figure ), but it is not resolved in Gunn . The Gunn looks elongated, and there are no close companions. The MOS spectrum shows a tilted [OII] centered on a faint continuum. The [OII] scale length was 1.5 h kpc, and this object was classified as kinematically normal. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-200672. The galaxy has a fully resolved exponential profile in Gunn (scale length = 1.5 h kpc, Figure ), but it is not well resolved in Gunn . The Gunn image has a very strong nucleus. There are no close companions. The MOS spectrum shows a strong, unresolved [OII] emission centered on a faint continuum. The [OII] scale length was 0.3 h kpc, and this object was classified as kinematically anomalous. Considering the fact that many kinematically anomalous galaxies have close companions, it is somewhat surprising that this galaxy is isolated. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-201268. The galaxy is not well resolved in Gunn and , so the scale length of 1.4 h kpc (Figures and ) is probably a lower limit. There are no close companions. The and images are both elongated. The MOS spectrum shows an extended and tilted [OII] line centered on a more compact continuum. The [OII] scale length was 1.4 h kpc, and this object was classified as kinematically normal. This galaxy has low [OII] flux, so its V sin has a fairly large error bar in Figure , and the MOS ELFIT2D simulation (Figure ) shows that V sin may be slightly underestimated.

E1512-202096. This galaxy looks normal in all three bandpasses. The galaxy is best resolved in Gunn . It is best fitted with a scale length of 2.1 h kpc (Figure ). There are no close companions. The MOS spectrum shows a very interesting [OII] morphology. There is a faint rotational component symmetric about the faint continuum. Atop this rotational component, there is a relatively strong source about 0(2.6 h kpc at a redshift of 0.4252) away from the center of the continuum. This source, which is unresolved spatially and spectrally, could be a giant HII region. The [OII] scale length measured with ELFIT2D was 2.0 h kpc, which is identical to the broad-band scale length. Based upon its faint rotational component, this object was classified as kinematically normal. The MOS ELFIT2D simulation (Figure ) shows no biases.

E1512-201125. This galaxy is resolved in Gunn (scale length = 1.6 h kpc, Figure ), but it is not resolved in Gunn . The image looks distorted. There is an small extension on one side which could be a very close companion. The MOS spectrum shows a faint [OII] line which is not resolved. The [OII] scale length was 0.5 h kpc, and this object was classified as kinematically anomalous. Despite the very low [OII] flux, the MOS ELFIT2D simulation (Figure ) shows no systematic biases, but the scatter in V sin is large.

Seven galaxies (identified by the superscript n in Table ) have [O II] scale lengths significantly smaller (and consistent with being unresolved) than their broadband scale lengths. Two of these galaxies (identified by the superscript c in Table ) are serendipitously observed cluster galaxies. Such discrepancies indicate that the [OII] gas kinematics is decoupled from galaxy rotation in some galaxies. Two properties of kinematically anomalous galaxies have emerged so far: (1) All but one have close companions. This suggests that enhanced star formation activity may be the result of merger events; and (2) Some of these galaxies appears to be of early-types. Kinematically ``anomalous'' field galaxies make up 25 of the field sample. This percentage is similar to the fraction (20) of field galaxies up to z = 0.3 with properties between Seyfert 2 galaxies and LINERs [\protect\astronciteTresse et al.1994]. Unfortunately, the study of kinematical anomalies is not suited to long-slit or multi-object spectrographs. Slits provide spatial information along their axis. Only integral field spectrographs such as Fabry-Pérot interferometers and fibre bundles can fully uncover the complex kinematics expected in those cases.