We have analyzed photometric and spectroscopic data of the association LH47 in the superbubble N44. The photometry has been reduced in a standard way. The optical and UV spectra were used together with the photometric information to derive spectral classifications and temperatures for those stars. For all other stars the UBV photometry was transformed to the theoretical plane with the conversions given by Massey et al. (1995b). We found a large scatter of the bright stars along the main sequence in the U-B colour. This scatter, which is also present in the study of Oey & Massey (1995), leads to a broadening of the main sequence in the HRD and thus to an unclear age determination from the HRD.
From an isochrone fit to the CMD
we find an age for the youngest population of 5 to 6 Myr and a mean reddening
of mag.
We do not find strong evidence for an age difference between the interior
and the exterior of the shell.
However, this conclusion is based mainly on two young stars seen inside the
shell.
They could also be located at the border or outside of the shell and
only being projected in front of the centre.
Due to the scatter in the U-B colour, which is transfered to the HRD,
the HRD did not help to improve the age determination from the photometry.
However, the HRD and the spectroscopically investigated stars suggest a lower
age limit for the association of 4 - 5 Myr, in perfect agreement with
the age determination from the BV photometry.
From the transformation of our UBV photometry to and
we get to know an individual reddening for all stars with Q < -0.4 mag.
The distribution of these reddening values peaks at
mag,
thus confirming the mean reddening derived from the isochrone fit to the
BV CMD.
From the reddening values we construct a reddening chart of N44.
In the centre of the shell we find reddening values around the mean value.
South-west of the central shell, between N44B and N44C, we find a peak
of the reddening which may be associated with the higher gas content in this
region.
Chin (1995) found the peak of the CO emission very concentrated
compared to the H
and HI emission
and the CO map of Cohen et al. (1988).
Using our grid of astrometric reference stars, we find that the position
of the
CO emission given by Chin (1995) lies exactly in the centre
of the central shell of N44.
The N44 superbubble with LH47 inside is therefore at the front side
of a molecular cloud.
It appears probable, that the expansion of N44 will trigger more star
formation in that molecular cloud.
The molecular transitions studied by Chin (1995) showed radial velocities
of km s
, which is in good agreement with the
radial velocities we found for our spectroscopically investigated stars.
From the main sequence stars in the CMD we construct the luminosity function.
We find a slope of for a power law representation
of the luminosity function.
This agrees well with the slopes found in other associations.
To construct the mass function we use two methods.
First, we derive the masses of all main sequence stars in the CMD
from the relation of and mass from the best fitting isochrone.
A power law fit to the distribution yields a slope of
,
equal to the Salpeter (1955) value.
Second, we count all stars between the stellar evolutionary tracks in the HRD.
Since for the conversion of the photometry to the theoretical plane
the U magnitude is essential, the completeness in the HRD is worse
than in the CMD.
Thus the count rates between the tracks in the HRD are smaller than
in the comparable bins of the CMD.
Surprisingly, the slope of the resulting mass function is nearly the same.
A comparison of the mass function slope inside and outside the shell
shows only a slightly steeper slope in the outside area.
The slope we found for the IMF is very similar to the values found for
other associations and open clusters.
The idea of an universal shape of the IMF is thus again supported.
Acknowledgements
Many thanks go to Thomas Lehmann for providing his digital image for the correction of the shutter delay times and to Georges Meynet for the program to calculate isochrones from their tracks. We thank Hans-Joachim Tucholke for calculating the exact positions of our target stars for spectroscopy from the MACS and Klaas S. de Boer for a careful reading of the manuscript. We are indebted to You-Hua Chu for providing the photometry of Lee and to Sally Oey for providing us with digital versions of her photometry. We thank Kathy Eastwood for fruitful discussions. This research has made use of the Simbad database, operated at CDS, Strasbourg, France. JMW and DJB acknowledge support from the DFG Graduiertenkolleg "Magellanic Clouds". DJB is grateful for a Feodor-Lynen-Fellowship of the Alexander von Humboldt-Foundation.