In this proposals we will make an attempt to get an indication for
the supernovae rates in the HII regions of M33. We will use [FeII]
emission as an indicator for the supernovae rates.
   Two prime questions asked are, does the supernovae rate depend on
the galactocentric radius and is the supernovae rate directly connected
with the slope of the radio spectral index.
   A radio continuum study of HII regions in M33, Lahuis and van der
Hulst (in preparation) has shown that the spectra of HII regions are
often not thermal. On the contrary, the spectral indices range from
-0.1 to -0.8, in the latter case the spectra appear to be largely
non-thermal. The excess of 20cm flux over 6cm flux ranges from a few
tenth to a few tens of a mJy. There also appears to be a steepening
of the spectral index with increasing galactocentric radius.
   The origin of this non-thermal radiation is not quite clear. One of
the possibilities thought of is the supernova activity within the HII
regions. That this could give a great contribution is demonstrated
in a radio continuum study of known supernovea by Duric. He found that
the excess of 20cm flux over 6cm flux is of the order of a few tenth
of a mJy to a few mJy. To check this we need to know the supernovae
rates within individual HII regions.
   The use of [FeII] emission has previously been suggested as an
indicator for the supernovae rates in starburst galaxies (e.g. Lumsden
and Puxley, 1993). We would like to try this for some individual HII
regions using SWS on three [FeII] lines.
   Since several of the transitions between the low lying levels of FeII
lie within the SWS range we now have an ideal opportunity to do this.