4. Line profiles
In Fig. 1 we show some of the observed lines towards the IRAS source. The 12 CO(J =1-0), HCO (J =1-0) and CS(J =2-1) lines show a two-peaked profile with either the blue- or red-shifted peak being the brighter whilst the other lines show a single-peaked profile. The peaks of the rarer isotopes coincide with the dip of the main isotopic lines. Thus, the dips in these profiles are not due to two clouds at different radial velocities but are rather caused by absorption in a foreground material with a lower excitation temperature.
In addition to the lines presented here, the CS(J =3-2) and H2 CO(212 -111) profiles are double-peaked with the blue-shifted component being the brighter, and the DCO (J =1-0) line is skewed towards blue. A detailed modelling of all the line profiles is being undertaken using a non-LTE radiative transfer program (Lehtinen et al. 1996, in preparation).
4.1. 12 CO(J =1-0)
The 12CO(J=1-0) line profiles show broad wings extending from -2 to +8 km s-1. In Fig. 2 we show the velocity integrated line maps over the velocity intervals 0-1 (solid contours) and 6-7 km s-1 (dashed contours). The IRAS source at (0,0) drives an outflow, which is oriented approximately in the east-west direction. The outflow is more extended to the west than to the east. At the position ( )=(-200, -100) there is indication of another red-shifted outflow component.
4.2. CS(J =2-1) and CS(J =5-4)
The central part of the observed CS(J =2-1) map is shown in Fig. 3. The spectra on and near the IRAS position at (0,0) show a double-peaked profile with the blue-shifted peak brighter than the red one while further away the lines are single-peaked but skewed towards the blue. This behaviour is just what is expected for a collapsing cloud. The symmetric shape of the J =5-4 transition can be explained if it is optically thin. The J =5-4 transition is wider than the J =2-1 transition, as predicted by infall models (Zhou 1992).
4.3. HCO (J =1-0)
In Fig. 4 a map of the HCO (J =1-0) emission is shown. There is a symmetry axis, approximately in the north-south direction at . There is a reversal of line profiles east and west of this symmetry axis in the sense that the left-hand line component is stronger on the eastern and the right-hand component on the western side. This kind of asymmetry in the profiles is just what is expexted for a rotating cloud, with the symmetry axis as the axis of rotation (Adelson & Leung 1988). Thus the cloud DC 303.8-14.2 would have a rotation axis oriented in the north-south direction. Another possible cause for the observed velocity gradient is the outflow, which is also oriented in the east-west direction.
If the asymmetry in the line profiles were due to rotation only, then along the lines of sight towards the rotation axis the line profiles would not be affected by the cloud rotation. If the rotation axis were located at , the observed profiles (higher red- than blue-shifted component) at would indicate that there is radial expansion of the mass traced by HCO emission. This would be the opposite as determined from the CS profiles. However, without a detailed knowledge of the cloud structure, other explanations for the asymmetrical profiles, e.g. deviation of low density material from spherical symmetry, or an uneven contribution from red-shifted and blue-shifted outflow emission, can not be ruled out.
There are two problems which complicate the interpretation of the HCO profiles. Firstly, what is the real position of the IRAS source ? The IRAS Point Source Catalog gives for the positional error ellipse the semi-major and semi-minor axes of 20 and 6, respectively. The position angle from north to east of the major axis is 15. If the IRAS source were located at to the east, the HCO profiles would be symmetric towards it. This is, however, improbable given the positional uncertainty of the IRAS source. Secondly, is the IRAS source located at the center of the cloud, i.e. is it located on the possible rotation axis ? This would be expected if the star was very recently formed. On the other hand, if the star had a projected velocity of 1.0 km s-1 (Clark 1987), it would take yrs to travel an angular distance of 20. In comparison, the maximum ages of Class 0 and Class I objects are estimated to be and yrs, respectively (André 1994).
4.4. HCN(J =1-0)
In Fig. 5 we show HCN(J =1-0) spectra at the IRAS source position (0,0) and 20 east and west of it. Each one of the three hyperfine structure (hfs) components is seen to display the two velocity peaks seen in CS(J =2-1) and HCO (J =1-0) lines. In LTE, the optical thicknesses of the hfs components have the ratios 1:5:3 (F =0-1: F =2-1: F =1-1, from left to right). The observed line intensities are more equal, however, indicating substantial optical depths. The central and the rightmost hfs components, with the highest optical depths, have a behaviour similar to the HCO (J =1-0) line, i.e. at the (0,0) position the red-shifted peak is higher than the blue-shifted one, and at the eastern position the profiles are reversed. On the other hand, the leftmost hfs component, with the lowest optical depth, behaves similarly to the CS(J =2-1) line, i.e. the blue-shifted peak is the stronger one.
© European Southern Observatory (ESO) 1997