According to numerical simulations and to observations of barred galaxies,
the interaction of the bar and the disk creates mildly shocked zones
that enhance the formation of Hii regions, mainly at the ends and along the
leading edges of the bar. This also produces a net flow of disk gas and dust
towards the central regions of the galaxy, where the
interaction with the bulge and nuclear dynamics often enhances the
circumnuclear star formation later in the evolution of these systems.
In Fig. , the scarcity of star forming regions in the bar
(both, at the ends and along the leading front of the bar), and the presence
of the circumnuclear ring of star formation, argues for a relatively old
dynamical age of the bar and associated systems in NGC 6951. Other imaging
and spectroscopic results give further support to this scenario.
When we consider all the information available (cf. sections 3 and 4),
a pattern seems to emerge in the circumnuclear region in which:
(a) There is a widespread population of supergiant stars that dominate the
light along and inwards of the circumnuclear ring, between about 6 arcsec
and 1.5 arcsec radius. These are most clearly seen in the K dominated
J-K map,
together with the large CaT equivalent widths measured in this same
zone (cf. section 4.2).
(b) The correspondence between the regions more luminous in
B and in
H1#1
is only good for the two brightest ones, but there are two regions
bright in H1#1
with apparently no corresponding enhanced B that must be
understood in terms of extinction; and there is widespread enhanced B along
and inwards of the ring with little or no corresponding H1#1
emission,
that indicates star formation in the post-nebular phase. It is interesting to
notice that B is enhanced both locally in a knotty structure along the ring
and more diffuse along and inwards of the ring; this might be interpreted as a
signature of two modes of star formation coexisting here: continuous and
bursting. It is possible that a continuous star formation process with
ocasional bursts takes place in a region such as this one, where a more
or less steady supply of gas has been going on for a long time, as supported by the
large fraction of molecular to total mass within the region (cf. section 4.3 above).
(c) There is some indication of a time sequence in the star formation history
along the ring, more apparent in the W-N quadrant. First, the strong K knot at
position angle 13#13250. Second, two knots bright both in B and K
with very little diffuse H1#1
emission at PA=270 and 300; this is
where Kohno et al. (1999) find the leading secondary maximum of HCN distribution,
and also the location of the leading secondary maximum of the 6 cm radio emission
(Saikia et al. 1994). Third, the Hii region to the north, PA=342,
that is brightest in B, H1#1,
HCN and 6 cm. Finally, the region just behind
this one along the ring, partly associated with the CO peak from Kohno et al.,
and that might be predicted to be the next in the sequence of star formation.
A similar, albeit somewhat less well defined, trend is observed in the south-eastern
side.
In conclusion, a qualitative look at the colours suggests that there may be two coexisting modes of star formation, continuous and bursting, and that the bursting mode may be sequential from the spearhead backwards along the entrained material as it is shocked in the circumnuclear ring.
Some of the ideas proposed here are only suggested by the wealth of information indicated by the data, and a detailed quantitative analysis of the stellar populations will be necessary to elucidate the validity of some of them. We are gathering spectroscopic data at other wavelengths to address this future analysis.