After inspection of the images produced, we found no traces of neutral gas
which could be related to the Gum nebula except for the lowest LSR velocity
channels. In Fig. 2 (click here) we show l,b maps between -2 and +20
, which include both greyscale and line contours. One of the
most conspicuous features is a bright, extended circular structure,
in radius, centered approximately at 
,
, and showing a hole in the HI distribution at the
center. This structure, appearing between -2 and +12 , has already been
discussed in Dubner et al. (1992), under the denomination of
"thick HI shell''. The distance to the thick HI shell was estimated to
be 
, its expansion velocity in 6 to 8 , and its
mass in more than 
. It has been proposed that the
thick HI shell is the remnant of an old SN explosion. It has also been
suggested that this shell may be related to the Gum nebula. We will come
back to this point in Sect. 4.
Figure 2: Greyscale l,b maps of the 21 cm
emission between -2 and 20 , given in brightness temperature. The
black contours, uniform in the whole velocity range, are at 4, 8, 12, 16 20,
30, 40 and 50 K. The white contours are at 70, 100 and 130 K. The
greyscale levels are shown on top of the first two images. The LSR velocity
is indicated
At v=+6 , a large disk 
in diameter, roughly centered at
, can be observed. This structure is also seen at
+8 and +10 ; at v=+12 it has almost disappeared. This large
disk will be called hereafter the si "giant HI disk'', and will be shown
to agree in position and size with the optical image of the Gum nebula. The
southern hemisphere of the giant HI disk is not well defined due to the
presence of another shell, 
in diameter, centered at
, which appears from -10 to +14 .
From v=+12 to v=+16 , a tail at
can be seen emerging northwards from the giant HI disk. At the
same position, Chanot & Sivan (1983) have found faint,
diffuse 
wisps extending from the main body of the Gum nebula.
The giant HI disk is probably also present at LSR velocities lower than 6 ,
but confusion with local gas makes its recognition more difficult. It
seems reasonable to consider this structure as extending from 0 to +10 .
If we assume that the giant HI disk is undergoing a symmetric expansion, we
can roughly estimate its systemic velocity as +5 with an
expansion velocity of about 5 . In Fig. 3 (click here), we show an integrated
map between 0 and +10 , in which the giant HI disk and the thick
HI shell can be easily distinguished. To the left of the disk (longitudes
higher than 
), the emission at the galactic plane is brighter
than to the right (longitudes lower than 
) due to the effects
of velocity crowding in the fourth galactic quadrant near 
.
Figure 4 (click here) shows an overlap of the 
optical image of the Gum
nebula with some contours of the HI emission. The 
digitalized
image used here was produced from the photograph taken by Sivan
(1974). The white stars indicate the positions of 
Velorum
(
) and 
Puppis (
).
Figure 3: HI integrated map between 0 and +10 .
Contours are given in brightness temperature, and increase in steps of 25
K in the interval 
, and in steps of 50 K from 100 K to 300
K. White contours at 400, 500 and 600 K are included. The greyscale is
indicated in the wedge on top of the image. The small (white) and the large
(black) circles show the approximate locations of the thick HI shell and the
giant HI disk, respectively
Figure 4: 
optical image (Sivan
1974) overlapped with the HI contours at 30, 100, 200, 350 and 500 K. The
white stars indicate the positions of the stars 
Puppis and
Velorum
We have estimated the mass of the giant HI disk in 
, where d is the distance in kpc. According
to the galactic rotation model of Fich et al. (1989),
the kinematical distance corresponding to the center of the giant HI disk
at a velocity of +5 is 
, with an uncertainty of about
20%. It should be kept in mind that kinematical distances are strongly
sensitive to the galactic longitude in this part of the galaxy.
Therefore, we can consider that, within the errors, both the giant HI disk
and the thick HI shell share a distance of about 
.
In this case, the linear radius corresponding to the giant HI disk turns
out to be 
. The mass of the giant HI disk can be estimated
to be 
, and the expansion kinetic energy, in
.
