Spectroscopic studies of the forbidden emission lines (FELs) of
T Tauri stars (TTSs), like the [OI] 
, 6363 and
the [SII] 
, 6731 lines, have shown that these lines
serve as powerful probes of the outflow activity of young stellar objects
(YSOs) (see e.g. Appenzeller & Mundt 1989 or Edwards et al.
1993 for a review). Particularly for more evolved YSOs which are no
longer deeply embedded in their parental molecular cloud and are therefore
probably more evolved, the FELs often serve as the only diagnostic mean to
investigate their outflows, since these YSOs are usually not associated
with Herbig-Haro (HH) objects or HH jets.
FELs occur relatively frequently in YSOs, e.g.
about 30% of all TTSs show strong FELs with 
for the [OI] 
line (Cohen & Kuhi 1979), while
the [SII] , 6731 lines are far less often detected
probably because of their typically 5 times smaller flux.
Line profile data of the FELs for about 45 TTSs are presently
available in the literature (e.g.
Appenzeller et al.
1984; Edwards et al. 1987; Hamann 1994;
Hartigan et al. 1995). In many cases the FELs are blueshifted
and furthermore often double-peaked profiles are observed. The so-called
high-velocity component (HVC) of this double-peaked profile has typical
radial velocities between -50 and 
while the so
called low-velocity component (LVC) has values around -5 to 
. Already Jankovics et al. (1983)
concluded from the similar velocity distributions of HH objects and the
blueshift of the FELs that these lines form in anisotropic bipolar outflows.
According to this scenario, which is still basic to all models of FEL
formation, an optically thick circumstellar disk, obscures the majority of
the receeding part of the bipolar outflow. Therefore in most cases only
blueshifted FELs are observed. However, it should be noted that in the
higher S/N ratio data nowadays available more and more cases have been found
where both sides of the bipolar outflow have been detected.
The nature of the HVC and LVC has been heavily discussed in the literature. In most cases it has been tried to interpret the two components by a single flow component viewed under favourable inclination angles (e.g. Edwards et al. 1987; Hartmann & Raymond 1989; Gómez de Castro & Pudritz 1993; Ouyed & Pudritz 1993, 1994). However, from the available line profile data and particular from various long-slit spectroscopic studies (e.g. Solf 1989; Solf & Böhm 1993; Böhm & Solf 1994; Hirth et al. 1994a; Hirth et al. 1994b) a different picture emerged. It is most likely from these data that the HVC and LVC represent two physically distinct flow components, with the LVC being spatially much more compact, of lower excitation and usually of higher density. Furthermore the long-slit data provide now very strong arguments that the HVC forms in a jet (Hirth et al. 1994a), as first proposed by Kwan & Tademaru (1988), while the nature of the LVC is still a matter of debate. Kwan & Tademaru (1988, 1995) suggested that the LVC is formed in a disk wind or disk corona. It is also conceivable that this component originates in the boundary layer in the innermost part of the jet. This might also explain, why the LVC becomes faster and more tenuous with increasing distance from the source as observed in DG Tau (Solf & Böhm 1993) and CW Tau (Hirth et al. 1994a).
In order to understand better the rather complex spatial, kinematic and
excitation structure in the FELs of TTSs we have carried out a long-slit
spectroscopic survey of 38 TTSs. Our study significantly increases the
number of TTSs with available long-slit data and allows statistically more
significant conclusions on the spatial properties of the FEL regions. As
previously pointed out by one of us (e.g. Solf 1989) long-slit
spectroscopy is a rather powerful method to obtain information on the
small-scale (
) structure of the FELs even on subarcsec
(sub-seeing) scales because it is a differential method using the bright
stellar continuum as a spatial reference (see also below).
Our paper is structured as follows: In Sect. 2 we outline the observations and the data reduction. In Sect. 3 we describe the results of the 9 most detailly investigated TTSs. In Sect. 4 the results of our survey are summarized and discussed.