Comparisons with Other Galaxies

It is useful to compare the estimated numbers of WR stars in NGC 1569 with those for similar galaxies as a function of galaxy properties such as luminosity, total mass (dynamical), and ionized gas mass. This we do in Figures 7-9. The data used in the production of these figures came from several sources. References where we obtained the number of WR stars in the various galaxies were the following: [Massey et al.(1992),Vacca & Conti(1992),van der Hucht(1996),Izotov et al.(1997),Massey & Johnson(1998b),Bransford et al.(1999),Breysacher, Azzopardi, & Testor(1999),Drissen et al.(1999),van der Hucht(1999),Guseva et al.(2000)]. Absolute magnitudes were readily available in several papers [de Vaucouleurs & Pence(1978),Thuan & Martin(1981),Bergvall & Olofsson(1986),Thronson et al.(1990),Kobulnicky et al.(1995),Guseva et al.(2000)] and the NASA/IPAC Extragalactic Database. However, limited data could only be found on the total (dynamical) masses [de Vaucouleurs(1960),Burns & Roberts(1971),Cottrell(1976),Loiseau & Bajaja(1981),Thuan & Martin(1981),Jackson et al.(1987),Dufour & Hester(1990),Taylor et al.(1993),Kobulnicky et al.(1995),Stil & Israel(1998),Wilkinson & Evans(1999)] and ionized hydrogen masses [Viallefond & Goss(1986),Dufour & Hester(1990),Sivan et al.(1990),Thronson et al.(1990),van den Broek et al.(1991),Waller(1991),Vacca & Conti(1992),Walterbros & Braun(1994),Kennicutt et al.(1995),Mendez et al.(1999),Guseva et al.(2000)]. Since there are so many sources and different combinations of data for the ordinate and abscissa values, we have not labeled the various data points with different symbols. Because we have two values (detected and estimated total) of the number of WR stars in NGC 1569, we have plotted both in Figures 7-9. The enlarged, solid NGC 1569 datum point is the detected number of WR stars, and the enlarged, hollow NGC 1569 datum point is the estimated total number of WR stars.

Figure 7 plots the number of WR stars detected versus the absolute blue (or photographic) magnitudes of Wolf-Rayet, starburst, and spiral galaxies. The ordinate error bars on the NGC 1569 data points indicate our estimate of the uncertainty due to complications of the nebular He II detected in the galaxy and the estimation procedure for the total number of WR stars as explained in the previous section. The ordinate error bars for a majority of the points comes from using the [Guseva et al.(2000)] data because they find the number of WNL stars with various emission lines. NGC 2403's ordinate error bars are because [Drissen et al.(1999)] report 25-40 individual WR stars. The abscissa error bars of a majority of the points come from using the absolute photographic magnitude (M$_{\rm pg}$) from [Guseva et al.(2000)]. We are assuming that their M$_{\rm pg}$ is $\pm1$ of M$_{\rm B}$. From the figure, one finds a well defined linear correlation between the logarithm of the WR stellar count and absolute blue magnitude for a majority of the Wolf-Rayet galaxies. This is not surprising, for one would expect a correlation between the blue luminosity and the number of WR stars, since WR stars and their O progenitors dominate the blue luminosity of star-forming galaxies. However, both NGC 1569 points are found below the general trend, which could be due to either (a) NGC 1569 is a post-starburst galaxy for which most of the recently formed massive stars have already evolved through the WR phase, or (b) our estimated number of WR stars in NGC 1569 is too low (i.e., no one has detected WC stars in NGC 1569). We suspect that (a) is the more likely reason, and that the SMC might be a similar post-starburst system since the numbers of WR stars in it are rather accurately known. Such might not be the case for estimates of WR stars in the Milky Way and NGC 2403 spirals, which are galaxies likely to have a more uniform recent star formation rate than starbursting irregulars.

Figure 8 is a plot of the number of WR stars versus total mass of galaxies, using a more limited number of good data available in the literature, and Figure 7 ordinate error bars are used here for the correlated data. This graph, albeit quite sparse on good data for the galaxy masses, suggests that total galaxy mass is not a good tracer of the observable number of WR stars in galaxies. This is not surprising since the total galaxy mass includes dark matter, which does not relate directly to recent star formation, and the number of WR stars seen in a galaxy depends critically on the recent star formation efficiency. Star formation efficiency would possibly make a better correlator because several of the small galaxies plotted here are producing similar numbers of WR stars compared to NGC 1569 and the LMC. Five galaxies (SMC, IC 10, NGC 2403, Mrk 178, and Mrk 209) all seem to have a relatively low level of recent star formation efficiency.

Finally, Figure 9 is a plot of the number of WR stars versus the ionized hydrogen (H II) mass, and the Figure 7 ordinate error bars are again used here for the correlated data. This graph, though again sparse on good data for other WR galaxies, shows a positive relation between the number of Wolf-Rayet stars and the mass of H II (given such a conclusion is strongly weighted on the SMC and IC 10). Such is not surprising, given that the young clusters containing WR stars also contain prodigious numbers of O-stars which ionize the surrounding gas. The abscissa lower limits derived from spectral properties in [Guseva et al.(2000)] also follow this trend. Several of the lower limits fall within the good data and could signify that the majority of the ionized hydrogen mass fell within the slit. The two lower limits at the right were probably slit locations over larger Markarian galaxies.

The comparisons between NGC 1569 and other WR and nearby galaxies shown in Figures 7-9 further support the idea that NGC 1569 is seen to be in a ``post-starburst'' phase, but has had a major, even eruptive, very recent starburst. While the mass of NGC 1569 is $\sim$30 times less than the LMC, it has approximately similar numbers of WR stars, similar blue luminosity, and a larger H II mass. By comparison with the SMC and IC 10, which have similar to slightly higher dynamical masses, they have fewer WR stars compared to NGC 1569. This, coupled with the extensive system of ionized filaments seen in NGC 1569 [Heckman et al.(1995)], suggests that the magnitude of the very recent ($\sim$10 Myr) starburst in NGC 1569 was exceptional. Today, the numbers of WR stars and H II mass found for NGC 1569 reflect the relatively recent end of this starburst, compared to other nearby, star-forming Im galaxies of comparable mass like the SMC and IC 10. In this respect, NGC 1569 still retains the properties of starbursting blue-compact dwarf galaxies compared to typical irregular (Im) systems. Possibly in another 50 Myr or so, NGC 1569 will have properties more similar to the SMC, IC 10, and other irregular systems of similar mass.