6 Source distribution throughout the MCs

Fujimoto & Kumai (1990) have proposed a model of the collision between the LMC and SMC about 0.2 Gyr ago (see also Murai & Fujimoto 1980; Mathewson et al. 1987). In this model the SMC should have penetrated through the LMC in the area around RA(B1950)=05$\rm ^h$20$\rm ^m$ and Dec(B1950)=-70$^\circ$  where we now find the lowest intensity of the radio emission. This is also the area of the steepest observed non-thermal spectral index (Meinert 1992). The proposal is that the SMC swept away the thermal gas along its track thus forming the bridge between the Clouds and initiating the Magellanic Stream. Fujimoto & Kumai (1990) concluded that during this collision the gas in the LMC was forced to rotate around the intruder galaxy, probably in a plane different from that of the LMC. This may have resulted in the asymmetric distribution of both atomic and molecular gas which is found mainly south of the 30 Doradus complex. Another remnant of this collision may be the L-component, which was recently detected in HI gas (Luks & Rohlfs 1992). Probably this model can also account for self-propagating star formation in this area.

Consider the following model. If the rotation of the gas in the LMC were disrupted in the region of the collision, perhaps by shock processes mentioned above, the density would increase and eventually a Jeans instability would occur. This mass assembly could also stop the gas coming in later along the same track and, because the primary complex has already started forming stars, it gave rise to pressure against the infall. Eventually this may induce another new complex of star formation in the direction of the infalling gas stream. If this model is correct 30 Doradus may be the latest, already evolved, of those star- forming complexes - probably preceded by the Shapley III complex - and it may have triggered star formation in the 30 Doradus complex.

Taking into account the ram pressure of the galactic halo gas onto the LMC interstellar medium (Meurer et al. 1985), it may be possible to give an alternative explanation for this star-forming complex. Ram pressure may well have compressed the gas at the border, and since the LMC is rotating in a clockwise sense the 30 Doradus complex would have reached this shock front earlier than N 159. In favour of the previous model it should be noted that the large gas cloud lies almost perpendicular to the bow shock, and so one would expect star formation all along its eastern ridge, not only at its northern end.

In order to understand the overall structure of the MCs, we plot the positions of sources intrinsic to both Clouds.

6.1 Source 2-D distribution throughout the LMC

The source distribution throughout the LMC is shown in Fig. 11. All SNRs and SNR candidates (62) are plotted together with 148 Hii regions (and HII region candidates).

  

Figure 11: The distribution of the radio sources intrinsic to the LMC. Asterisks represent SNRs and SNR candidates and open circles represent HII regions and Hii region candidates

Smith et al. (1987) first gave the 2-D distribution of the LMC discrete sources but for a limited number of sources ($\sim$20). Here, with a significantly improved number of LMC sources (209), we found that the LMC sources follow patterns initially indicated by Shapley (1956) and Martin et al. (1976). These patterns are possible spiral arms, which were introduced by Feitzinger et al. (1987 and reference therein) and Schmidt-Kaler (1993 and reference therein). From Fig. 11 it can be seen that some regions throughout the LMC do not have strong or obvious sources. Three such a regions in the LMC were observed here:

1.

the region north from 30 Doradus region with centre at $\sim$ RA(B1950)=05$\rm ^h$38$\rm ^m$ and Dec(B1950)=-68$^\circ$20$^\prime$. This region extends approximately for a diameter of some 50$\hbox{$^\prime$}$ from this centre;

2.

the region south of the optical bar, between the LMC Complexes A2$\hbox{$^\prime$}$, A2, A1 and the LMC Complexes B3, B2$\hbox{$^\prime$}$, B21 and

3.

the region between Shapley Constellation III, Shapley Constellation IV and the LMC Complex B1.

Generally, the distribution of the intrinsic sources throughout the LMC indicates some kind of spiral structure with a possible centre in the region around or close to 30 Doradus. According to Feitzinger et al. (1987), this large-scale spiral pattern consists of as many as six or seven arm-like features. On the other hand it is difficult to explain why such a young complex as 30 Doradus is a centre of evolved spiral structure. Also, it is very difficult to understand why the radio continuum picture of the LMC is so different from the optical, where the optical bar is so dominant. The optical bar of the LMC is not directly related to the spiral pattern; it probably reflects an earlier period of evolution of the galaxy. These questions could be answered in the new high-resolution HI survey of the LMC (L. Staveley-Smith private communication 1997). Also, the sharp eastern end on all continuum images of the LMC is noted. Similar tendencies can be found in the ROSAT X-ray surveys (Pietsch et al. in preparation). We believe that this is caused by the ram pressure of the LMC motion toward us.

6.2 Source 2-D distribution throughout the SMC

The SMC is much less rich than the LMC in intrinsic discrete sources. Only 37 SMC sources can be seen from our surveys and they are plotted in Fig. 12. Two regions (main-body of the SMC and eastern wing of the SMC) are dominant. However, we cannot confirm the depth of these regions and the distribution of sources along the line of sight with our observations.

  

Figure 12: The distribution of the radio sources intrinsic to the SMC. Asterisks represent SNRs and SNR candidates and open circles represent HII regions and Hii region candidates

Here, we note the known SMC X-ray binary system (1E 0035.4-7230; Kahabka & Pietsch 1996) which is some $\sim$60$\hbox{$^\prime$}$ on the far west side from the rest of the SMC. Why such a source is so distant from the rest of the galaxy at present remains a mystery.