The main results of our long-slit spectroscopic survey of the FELs of TTSs are summarized in Table 2 (click here) and Fig. 22 (click here).
For each TTS studied Table 2 (click here) shows the spatial properties (offset
and width) and kinematical properties (radial velocity and velocity
dispersion) of the indiviual FELs on both sides of the outflow (if bipolar).
Furthermore, if possible, the spatial and kinematic properties are listed
separately for the LVC and HVC. However, in several cases it has not been
possible to resolve spectroscopically these two components mostly because
their internal velocity dispersion is larger than their separation in radial
velocity. Furthermore it is not clear in several cases whether we observe a
LVC or a HVC. It will always be difficult to distinguish a LVC from a HVC
resulting from a jet which is located close to the plane of the sky. In
these cases we have printed all entries of Table 2 (click here) between the
corresponding columns of the HVC and LVC (see also comments to individual
objects). We note that the [NII] 
line, if present, is
regarded as a reliable indicator that a HVC indeed exists since all
available [NII] intensity profiles suggest that this line probably never
exhibits a LVC. In addition it is known from Hartigan et al.
(1995) that many TTSs with small IR excesses (
)
probably have a LVC only. The objects printed in bold face in
Table 2 (click here) are those for which the most detailed information is
available and for which the outflow direction is known (from direct imaging
and/or from long-slit spectroscopy) and which are closeby (usually at
distances of 150 pc, except V536 Aql, which is at 200 pc). There are three
more objects listed in Table 2 (click here) for which the outflow direction is
known (FS Tau B, Bretz 4 and AS 353 A) but which are not printed in bold
face, because either they are more distant and hence a derivation of their
spatial properties makes little sense due to our limited spatial resolution
or the central star is too faint to provide a reliable spatial reference for
the determination of the offset and the spatial resolution (e.g. FS Tau B).
Figure 20: Position-velocity maps of the [OI] 
, [NII]
and [SII] 
lines of V536 Aql extracted from
data taken with the 4.2 m ISIS Cassegrain spectrograph in September 1993
(
). The stellar continuum has been subtracted. This
subtraction could not be carried out perfectly for the [NII]
line due to the presence of the strong 
line wing. Contours,
relative positions and velocities as in Fig. 1 (click here)
Figure 21: Spatial width (top part), spatial offset (middle part) and
spatially integrated intensity (lower part) of the [OI] ,
[NII] and [SII] line emission of V536 Aql as
a function of the radial velocity (). The dotted line
in the plots of the spatial width indicates the lower limit for which
changes in width with velocity can be reliably measured. The radial
velocity has been measured relative to the stellar velocity. The data are
extracted from the continuum-subtracted position-velocity maps displayed in
Fig. 20 (click here)
In Fig. 22 (click here) we show the profiles of the FEL profiles of all studied
TTSs. In all cases, the data with the highest spectral resolution available
are printed. These profiles are obtained by spatially integrating the data
on the long-slit spectrograms. For comparison we also show the
line, which is in most cases much broader than the FELs and often shows a
different line profile structure.
Many of the TTSs in Table 2 (click here) show spatially extended FEL
regions on a 1'' scale, especially in the HVC (see also below). This is
one of our main results which was not expected a decade ago. Of the 38 TTSs
listed in Table 2 (click here) we have derived information on the spatial
properties of the FELs in 33 cases. A total of 27 TTSs of this sample show
at least one FEL with an emission centroid which is spatialIy offset from
the stellar position. In the [SII] line the spatial offset
of the high-velocity gas can reach values up to a few arcsec. Typical values
are 0.3'' to 1.0''. For the spatial width the values range from the
smallest measurable values (
; depending on the seeing and the
S/N ratio) up to 2.9''. Typical values are 0.8'' to 1.5''.
In order to derive more detailed information on the spatial properties of
the FELs it is important to compare the spatial properties of the different
emission lines and the different properties between the high-velocity and
low-velocity gas. We will perform that comparison mainly on the basis of
those 12 TTSs studied in greatest detail (i.e. for those TTSs printed in
bold face in Table 2 (click here)). We will first compare the properties of
their HVCs. If the existence of a well separated HVC is not clearly evident
in the line profiles, e.g. if it is blended with a LVC, we will compare the
spatial width and offset at velocities typical for the HVC (
) or if the [NII] line is detected at the velocity
of this line. (As already mentioned above, the [NII] line
never shows a LVC but only forms a HVC). Comparing the [OI] ,
[NII] and the [SII] lines it turns out that
in general the spatial width and offset of the high-velocity gas is
largest in [NII] and smallest in [OI]. In the following we will give
more detailed information on this important result: In all 12 TTSs the
measured offsets in [SII] are larger by a typical factor of 2 - 4 than in
[OI]. However, the offsets in [NII] are larger than in [SII]. From the 9
cases for which both [SII] and [NII] data are available (counting each side
of a bipolar flow separately) 7 cases show offsets in [NII] clearly larger
than those in [SII] by a typical factor of 1.5 - 2, while 2 cases show
similar offsets.
With regard to the spatial widths it has been found also that the values are smallest in [OI] and largest in [NII] and [SII]. However, in [NII] and [SII] similar widths have been obtained in most cases. Due to the fact that the spatial width cannot be measured that accurately (see also Sect. 2) any small differences between [NII] and [SII] will not be detectable in our data. In 11 of the 13 cases with available data on the spatial width, the values in [SII] are larger by a factor of 1.5 - 3 than those in [OI] and similar values have been obtained in the remaining 2 cases. From 10 cases where both [NII] and [SII] data are available similar values have been measured in 5 cases. In 3 cases the spatial width is larger in [NII] than in [SII] (by a factor of 1.3 to 2.0) and in 2 cases it is smaller than in [SII] (by a factor of 2). Note again that each side of a bipolar flow is counted as an individual case.
A comparison of the spatial properties of the HVC and LVC is more difficult
since among the 12 most well studied TTSs only 3 stars (CW Tau, DG Tau and
V536 Aql) show sufficiently separated HVCs and LVCs which can be
individually studied with the medium resolution 3.5 m Cassegrain twin
spectra. In order to include also the remaining 9 TTSs in our study we
compare in these stars the spatial width and offset of the high-velocity gas
(with radial velocities of 
) with the
properties of the gas around the stellar velocity. For the sake of
simplicity we will designate in the following comparison the emission from
the gas at high and low velocity "HVC'' and "LVC'', respectively. Since in
[NII] no LVC is observed this comparison is only useful for [OI] and [SII].
As expected from previous studies (e.g. Hirth et al. 1994a)
in most cases the HVC shows a much larger offset and spatial width. In [OI]
the offset of the HVC is significantly larger by a factor 2 - 3 than that
of the LVC in 8 out of the 10 cases with available data. Similar values of
the offset have been obtained for the remaining 2 cases. In [SII] in all 9
cases with available data the offset in the HVC is larger than that in the
LVC (typically by a factor of 2 - 4).
Comparing the spatial properties of the LVC between the [OI] and [SII]
lines, the same trend as for the HVC is observed. In all 11 cases with
available data the offset in [SII] is larger than in [OI] by typically a
factor of 2. The typical spatial offsets of LVCs are 0.2'' in the [SII]
line. The spatial widths of the LVC are in [SII] also larger
than in [OI] by a factor of about 2.
These results on the spatial properties of the HVC and LVC of the studied
FELs are not surprising, since similar data have been obtained in previous
detailed studies on CW Tau (Hirth et al. 1994a) and DG Tau
(Solf & Böhm 1994). These authors show that the gas in the
HVC and LVC can have quite different excitation conditions and also
different spatial properties (see also Hamann 1994). However
our much larger sample shows that these differences in the spatial
properties of the HVC and LVC are a general property of TTSs and not just a
peculiarity of a few individual objects. Like in the above mentioned
studies of CW Tau and DG Tau, we interpret the HVC and LVC in the context of
a model proposed by Kwan
& Tademaru (1988, 1995). According to this model, the HVC forms in
a well-collimated high-velocity jet (
)
while the LVC forms in a more compact region with intrinsically much smaller
flow speeds which may result from a disk wind or a disk corona.
As shown more quantitatively by Hamann (1994) the gas in the
LVC is generally of higher density and lower excitation than the gas in the
HVC. This also explains why the [NII] 
, 6583 lines
never form a LVC but only a HVC. Why the spatial offset of the HVC is
largest in [NII] and smallest in [OI] can easily be explained if the density
in the jets is decreasing and the excitation is increasing with increasing
distance from the star. For a jet with diverging stream lines (e.g. with
constant opening angle; see also below) a decreasing density with increasing
distance is rather obvious, but why the electron temperature is increasing
is unclear (i.e. it is unclear why the shock velocities increase with
distance). Nevertheless if the jets from TTSs have such properties near the
source (i.e. at distances of the order of 100 AU to several 100 AU), it is
quite clear that the jets will appear most compact in [OI] which samples the
densest gas due to its high critical density (
), while the jets will appear much larger in [SII]
, 6731 since for these lines the critical density
is about 100 times smaller than in [OI]. Since the [NII]
line samples gas of even lower densities and higher excitation the jets are
expected to appear most extended in these lines.
We like to mention that the LVC of [SII] can be spatially resolved in many objects and has a small but significant offset (typically 0.2'') from the source. This is also evident in a detailed study of DG Tau by Solf & Böhm (1994) and CW Tau by Hirth et al. (1994a) where the offset of the LVC is increasing with increasing blueshift of the low-velocity gas. All these results have certainly important implications for any detailed model of the FEL gas in TTSs.
The bipolar jets which give rise to spatially resolvable FEL regions for
closeby TTSs (
) often display apparent lenghts
of only 100 - 500 AU. However, these lenghts probably do not represent the
true physical lenghts of the jets (which in some of the "classical''
outflow sources extend up to about 1 pc) but probably only the brightest
region in which forbidden line emission can be detected (see below). It
should be pointed out that previous studies have already shown that some
spatially extended jets, like the ones from HH 30 or HL Tau (see Mundt et
al. 1990), strongly decrease in their surface brightness with
increasing distance from the source. Such a behaviour can be reasonably
explained by the following considerations. Let us assume that the jet can be
approximated by a homogenously filled cone at a given distance r from the
star and 
, say 
. Since 
holds for a cone of constant opening angle and since the emission
measure per unit length of this cone is 
(for
) the brightness Ij(r) of a spatially
unresolved jet will be 
. This means, that if 
is constant and the brightness of a
cone-like jet with a constant opening angle should strongly decrease with
increasing distance from the source and therefore only the brightest regions
will be detectable.
Interestingly, only 12 out of a sample of 27 TTSs with spatially extended FEL regions are known to be associated with a reflection nebula. This means that the TTSs in our sample have on average much fainter and smaller reflection nebulae than the "classical'' outflow sources associated with HH objects and HH jets (cf. Mundt et al. 1987). In the latter sample, nearly all sources are associated with reflection nebulae detectable on the POSS or on continuum-light CCD images. These differences can probably be explained by evolutionary effects, with the classical HH object and jet sources representing a sample of on average younger stars than the CTTSs listed in Table 2 (click here) and therefore many of the HH jet sources are still associated with dense gas and dust from remnants of their parental molecular cloud.
Recent HST images of objects like HH 30 or HL Tau (Stapelfeldt et al.
1995; Ray et al. 1996; Burrows et al.
1996) have shown that YSOs can be associated with rather compact
reflection nebulae which appear stellar-like on ground-based images. If
there are such bright reflection nebulae with sizes < 1'' around many
CTTSs in our sample then scattering of the light from the FEL regions may
affect significantly the determined spatial and kinematic properties in some
cases (see also Noriega-Crespo et al. 1991). In HL Tau, for
example, the LVC has a similar spatial width as the HVC (1'' at FWHM). In
this case this result can probably be explained by the fact that one does
not observe in the optical the star itself but rather its very compact
reflection nebula only since the star itself is highly obscured by the
circumstellar dust. Therefore the position derived from the continuum
spectrum of the star will not be representative of the true stellar
position. Light scattering may also explain why some [OI]
line profiles have a broad red wing as observed e.g. for CW Tau and several
other CTTSs (Hirth et al. 1994a). This would be the case if
the scattering dust would "see'' the blueshifted jet as a receeding jet
(i.e. if the scattering dust would be located between the star and that
region producing most of the blueshifted forbidden line emission of the
jet). We note that these broad red wings have been interpreted by
Hirth et al. (1994a) in terms of rotation in the formation
region of the LVC. The light scattering considered here is a reasonable
alternative interpretation.
Several of our investigated sources are double stars (DD Tau, T Tau, XZ Tau,
UY Aur and V536 Aql) separated by less than 1'' (e.g. Leinert et al.
1993;
Ghez et al. 1993, Ageorges et al.
1994). This appears to be an important result and suggests that also
TTSs in double or multiple systems with small separations (
AU)
form jets. Since the subset of known binaries in that separation range is
very small we have not investigated any possible correlation between the
properties of the flow and the binary systems (e.g. between the position
angle of the binary and the outflow direction). Certainly future HST
observations would be helpful to answer various interesting questions (e.g.
which of the two components of the binary is responsible for the jet, or
do both components show individual outflows, or how well aligned are the
outflows from both components).
About 25% of the TTSs of our sample show only the [OI]
, 6363 lines in the investigated wavelength range
(
) for a detection limit of 
. In the [OI] line a clearly recognizable double-peaked
line profile has been observed in about 50% of the TTSs in our sample of 33
objects with high S/N [OI] line profiles. For comparison, in
the sample of Hartigan et al. (1995), which contains 32 stars
and which is of similar or even higher spectral resolution, this fraction
was about 43%. Note that our sample includes about half the objects of the
sample of Hartigan et al. (1995) and that both samples have
been more or less picked on a random basis. Whether the TTSs showing only a
single velocity component in their [OI] lines have different physical
conditions in their FEL regions is unclear. As already outlined above there
are several reasons why both components could be blended into one component,
even for a sufficiently large spectral resolution. For 3 stars in our sample
(FN Tau, DO Tau and RW Aur) the LVC was only observed in the [OI] lines but
not in the [SII] lines. The same is the case for a few stars in the sample
of Hartigan et al. (1995) and Hamann (1994),
namely CI Tau, DK Tau, DL Tau, HM Lup, AS 353 A and SCrA. Since in all these
stars the HVC is detected in both the [OI] lines and [SII] lines a much
higher density (and/or lower excitation) is indicated for the LVC as
compared to the HVC, at least in these stars, but probably in a large
fraction of the TTSs (see Hartigan et al. 1995).
For 40% of the investigated objects we have detected the [NII]
line. In the sample of Hartigan et al. (1995)
this fraction was 37%. Apparently this fraction does not depend much on the
detection limit since the detection limit of Hartigan et al.
(1995) for FELs has been about 3 times higher than in our study.
Table 2.
(See following pages) Spatial and kinematical properties of the
FEL regions of the studied TTSs. In the case of known bipolar outflows,
the values are
quoted separately for the blueshifted and redshifted part of the outflow.
The kinematical data listed are derived from the spectra with the
highest spectral resolution available (see also Fig. 22 (click here)).
Uncertain values are quoted in parentheses. Note that in several cases it
has not been possible to resolve the two velocity components or it is not
clear whether the data represent a LVC or a HVC with small radial
velocities. In these cases all entries are inserted in Table 2 (click here)
between the corresponding columns of the HVC and LVC (see also comments to
individual objects). Columns: (1) Object number in survey. (2) Number in
the Herbig & Bell (1988) catalogue. (3) Name of the object;
printed in bold face for closeby (
) and most well
studied objects with known outflow direction. (4) Number of campaign (see
Table 1 (click here)). (5) Investigated FEL. (6) Position angle (PA) of the
outflow derived either from direct imaging or long-slit spectroscopy. (7)
PA of the long-slit spectrogram from which the quoted spatial data have
been derived. The PA is given for the blueshifted forbidden line emission.
(8) and (9) Radial velocity of the HVC and LVC relative to the stellar
velocity. (Asterisks denote heliocentric velocities). (10) and (11)
Velocity dispersion (FWHM) of the HVC and LVC corrected for the
instrumental profile. (12) and (13) Spatial offset of the emission centroid
of the HVC and LVC relative to the stellar position. (14) and (15) Spatial
width (
) of the HVC and LVC corrected for the spatial
resolution (see Sect. 2). (16) and (17) Equivalent width of the HVC and LVC
derived from Gaussian fitting to these two components. (18) Total
equivalent width. Note that is some cases this value is not always the sum
of the individual values of the HVC and LVC due to the non Gaussian profile
of these components. (19) An asterisk indicates that remarks on the TTS are
available
Remarks to Table 2 (click here)
FM Tau: Only a single velocity component with
is visible. It is unclear whether this
low-velocity emission results from a true LVC or is emission from a
high-velocity jet seen edge-on (see also Fig. 22 (click here)a).
FN Tau: The LVC in the [SII] lines is only marginally indicated (see also Fig. 22 (click here)a).
CW Tau: All values are taken from Hirth et al.
(1994a) where this object is discussed in detail. The counterjet in
displays radial velocities between +50 und 
.
DD Tau: The values given in Table 2 (click here)
are from Neckel & Staude (1993). In the [OI]
line profile (see Fig. 22 (click here)b) the two components of the bipolar
outflow are not resolved but apparently only the blueshifted side of the
bipolar outflow has been detected in the [SII] and [NII]
lines.
CZ Tau: The low S/N ratio of the obtained
spectrogram does not allow detailed kinematical and spatial studies. The
possible presence of a blueshifted and redshifted emission component in the
[OI] line profile (see Fig. 22 (click here)b) has to be confirmed
by deeper spectra.
RY Tau: The LVC is visible only in [OI]
(see Fig. 22 (click here)c). The two possible components listed
in Table 2 (click here) for the [SII] line are derived from a
two-component Gaussian fit to the profile.
FS Tau B: The kinematic data are based on a long-slit spectrogram published by Eislöffel & Mundt (1997).
FS Tau A: Only a single velocity component
with 
has been detected (see also
Fig. 22 (click here)d). It is unclear whether this low-velocity emission
results from a true LVC or if it originates from a high-velocity jet seen
edge-on. However, the detection of [NII] emission strongly
argues in favour of the presence of a HVC, since we are not aware of any
star where a true LVC also shows [NII] emission. Recent studies of
Eislöffel & Mundt (1997) of the FS Tau A & B regions show
that within a few arcsec from FS Tau A and at the position of FS Tau A there
is extended HH emission. This emission probably results from a poorly
collimated wind from FS Tau B when this wind hits the edges of a cavity. It
is conceivable that at least part of the observed emission in FS Tau A is
actually resulting from that extended HH emission. This would also explain
the rather small line widths observed in the FELs of FS Tau A.
T Tau: All values are taken from Böhm & Solf (1994), where this rather complex object is discussed in detail.
DG Tau: All values are taken from Solf & Böhm (1993), where this object is discussed in detail.
Haro 6-10: This object is discussed in detail in the text; see Sect. 3.1.
DH Tau: Only a single velocity component with
is visible (see also Fig. 22 (click here)e). It
is unclear whether this low-velocity emission results from a true LVC or if
it is emission from a high-velocity jet seen edge-on. The small offset of
the [OI] emission of +0.2'' argues more in favour of a HVC.
HL Tau: This object has been studied in detail by Mundt et al. (1990). From HST observations of Stapelfeldt et al. (1995) and Ray et al. (1996) it is known that the star itself is not directly observable in the optical, but only its bright compact reflection nebula. Since all measured spatial quantities refer to the centroid of the registered continuum emission of the reflection nebula, the quoted offset values do not represent the offset with respect to the stellar position. Also the values for the spatial width are probably altered due to the presence of the reflection nebula.
XZ Tau: This object is discussed in detail in the text; see Sect. 3.2.
UZ Tau E: This object is discussed in detail in the text; see Sect. 3.3.
GH Tau : A broad emission line wing extending
up to 
has been detected in the [OI]
line (see Fig. 22 (click here)g). The value of 
for the
radial velocity of the HVC is derived from a Gaussian fit to this broad
blueshifted emission.
HN Tau: In none of the FEL profiles of this star are two components directly evident, but as discussed in the text (see Sect. 3.4) there are indirect arguments for the presence of both a LVC and a HVC derived from the observed [NII] line profile. The listed radial velocities and velocity dispersions for the HVC and LVC of the [OI] and [SII] lines are the result of a two-component Gaussian fit to the profiles of these two lines (see also Fig. 22 (click here)h).
AA Tau: Only a single velocity component with
is visible. It is unclear whether this
low-velocity emission results from a true LVC or originates from a
high-velocity jet seen edge-on. However, the presence of [NII]
emission at 
strongly argues
that at least part of the [OI] emission results from a HVC,
since we are not aware of any star where a pure LVC also shows [NII]
emission (see also Fig. 22 (click here)i).
DO Tau: This object is discussed in detail in the text; see Sect. 3.5.
DP Tau: This object is discussed in detail in the text; see Sect. 3.6.
DQ Tau: It is unclear, whether the observed
emission component seen in [OI] and [SII] at
about 
is a LVC or is a HVC of a jet seen nearly
edge-on (see also Fig. 22 (click here)j).
DR Tau: The values for the spatial
offset are near the detection limit. This is also the reason why it
is hard to determine the outflow direction from the obtained long-slit
spectrogram. Probably the outflow direction is towards the direction of the
HH object found by Mundt & Eislöffel (1997) along 
. There might be another emission component between the HVC and
the LVC of the [OI] line at 
(see also Fig. 22 (click here)k).
UY Aur: This object is discussed in detail in the text; see Sect. 3.7.
RW Aur: This object is discussed in detail in the text; see Sect. 3.8.
HS Ori: Due to the relatively high radial
velocity of the [OI] line of 
, we believe that most of the emission originates in a HVC (see
also Fig. 22 (click here)l).
VZ Ori: Only a single velocity component with
is visible (see also Fig. 22 (click here)l). It
is unclear whether this low-velocity emission results from a true LVC or if
the emission originates from a high-velocity jet seen edge-on. However, the
large K-L value of 1.66 (Cohen & Kuhi 1979) favours more
the interpretation of a HVC (for detailed arguments see Fig. 11 (click here) of
Hartigan et al. 1995).
PU Ori: A very weak [SII]
emission at 
might be present
(
; see also Fig. 22 (click here)m).
V510 Ori: Although only a single velocity
component at relatively small radial velocities is visible in [OI]
(
) and [SII] (
), we believe that a significant portion of this emission
results from a HVC (i.e. from a jet being nearly oriented in the plane of
the sky). The main arguments for that interpretation are the presence of
[NII] emission and the large line width (see also Fig. 22 (click here)m).
LkH
313: A very broad [OI]
emission (
) is observed
for this object. The emission peaks at 
(see also Fig. 22 (click here)n). Maybe we observe, like for DD Tau a bipolar
outflow whose two sides are not spectroscopically resolvable due to large
internal line widths.
Bretz 4: This object is the source of a HH jet
(HH 271-273) and has been first imaged in the [SII] ,
6731 lines by Carballo & Eiroa (1992). The [OI] line profile shows a LVC centered at 
and a very broad blueshifted wing of high-velocity emission
extending to nearly 
(see also Fig. 22 (click here)n).
This high-velocity emission is approximately centered at 
(
).
V481 Mon: Due to the faintness of the star, the
S/N ratio of the data obtained is very low. Only a single velocity component
with 
is visible (see also
Fig. 22 (click here)n). It is unclear whether the observed low-velocity
emission results from a true LVC or if it originates from a high-velocity
jet seen edge-on.
NX Mon: A very broad [OI]
emission (
) is observed for this object
(see also Fig. 22 (click here)o). The emission peaks at 
. Maybe we observe, like for DD Tau, a bipolar outflow whose
two sides are not spectroscopically resolvable due to large internal line
widths.
AS 353 A: The extended bipolar outflow from
this TTS traced by HH objects (HH 32 A-D) has been discussed in detail by
Mundt et al. (1983), Hartigan et al. (1986),
Solf et al. (1986) and recently by Davis et al.
(1996). A [OI] line profile of the star has been
published by Hartigan et al. (1995). In addition to a slightly
blueshifted LVC at 
a broad blueshifted high-velocity
emission is observed, which extends up to 
(see
Fig. 22 (click here)o). For the intermediate velocities at 
the emission centroid is shifted +0.3'' relative to the source and
the spatial width is about 1''. For the [SII]-lines no spectra have been
secured because these lines are extremely weak (
of [SII]
equals 0.08 Å, Hartigan et al. 1995).
V536 Aql: This object is discussed in detail in the text; see Sect. 3.9.
MacC sH15: The emission in [OI]
is only barely detectable (
) and the
listed kinematical data are relatively uncertain (see also
Fig. 22 (click here)p).
Acknowledgements
We are indepted to the staff of the Calar Alto Observatory and the William Herschel Telescope for assistance during the various observing runs.
Figure 22: Lines profiles of th [OI] ,
[NII] ,
[SII] lines and the
line of all TTSs from our survey extracted
from spectrograms taken with the 2.2 m coudé spectrograph (c), the 3.5 m Cassegrain twin spectrograph (T)
and the 4.2 m ISIS Cassegrain spectrograph (I). The radial velocities (in km s-1) are relative to the stellar
velocity except for HS Ori, Lk
313 and V481 Mon. For these stars heliocentric velocities are given.
For all FEL profiles the continuum has been subtraced. For both the
and FEL profiles the line intensity
is normalized to the continuum.
Figure 22: continued
Figure 22: continued
