PHOTOMETRY, POLARIMETRY, AND INTERFEROMETRY OF B[e] STARS. VARIABILITY

Miroshnichenko Anatoly S.

Invited review at the Workshop on B[e] stars held at the Institute d'Astrophysique de Paris, 1997 June 9-12. Published in: B[e] stars. (eds.) Jaschek, C. and Hubert, A.M. Kluwer Academic Publishers, 1998. p.145-164 (Astrophysics & Space Science Library, v.233).

I. INTRODUCTION

• Spectral type -- mainly B
• Presence of emission lines
• Excess of near--infrared radiation

• Formation of planetary nebulae as a single event
• Interaction of an OB star with a late-type companion
• Direct ejection from a particularly massive Be or Oe star

During follow up studies 47 % of them have been recognized as the objects at well-understood evolutionary stages: pre-main-sequence (PMS) HAEBEs (7), LBVs (3), PNe (16), symbiotic systems (8), eclipsing binaries (2: RY Sct, GG Car). Other objects fall into the several categories:

1. Supergiants similar to those in the Magellanic Clouds found by Zickgraf et al. (1986) -- 6 stars (3 Pup, MWC 300, MWC 349, HD 87643, HD 316285, HD 326823)
2. Probable binary systems consisting of a hot and a cool star -- 6 (MWC 84, CPD-52 9243, MWC 623, CPD-57 2874, HD 89249, HD 327083)
3. Well-observed objects with puzzling nature -- 4 (FS CMa, HD 50138, MWC 342, CD -42 11721)
4. Poorly-studied stars with uncertain state -- 27 stars

Recently The et al. (1994) analysed observational characteristics of the early-type stars with evidence for dusty envelopes trying to separate those at the PMS evolutionary stage. They found no unique observational data set allowing to do this and compiled a new catalog of ``Herbig Ae/Be stellar group'' containing 287 currently known candidates to HAEBEs, B[e] stars, and, unstudied objects listed by Dong & Hu.

Photometric and polarimetric observations of B[e] stars have been obtained so far mostly at an occasional basis. There are only a few programs in which long-term photometric observations of a number of B[e] stars have been carried out: Kilkenny et al. (1985), Manfroid et al. (1991), Sterken et al. (1993, 1995), The et al. (1994), Bergner et al. (1995). An ongoing program including classical B[e] stars and recently found candidates is described by Kuratov et al. (this volume). Polarimetric observations were carried out by Barbier & Swings (1982), Zickgraf & Shulte-Ladbeck (1989), and reported in the papers quoted in Table 3. A lot of results are still unpublished, and I am very grateful to the owners of these observations for sending them to me for the preparation of this review.

II.PHOTOMETRIC PROPERTIES

Objects with light curves

Photometric history of the best observed object, FS CMa, has been recently summarized by de Winter & van den Ancker (1997) for a period since the beginning of the century. Its visual brightness varied between 7.4 and 6.9 untill early 60s and then declined to 8.8 in some 20 years. Since the minimum in 1981 the star continues brightening untill the present time reaching a level of 8.0. Multicolor photoelectric photometry obtained since late 50s showed that its optical color-indices changed a little on background of strong brightness changes. A trend can hardly be seen in B-V while dispersion of other color-indices veils it. Quasi-regular brightness variations with a period of 296.5 days have been found by Halbedel (1989) for a four-year data set. Our analysis of the observations obtained in 1975-1996 shows presence of significantly slower variations with a period of $\sim 1600 days (Fig.2). Their signature is also visible in the object's behavior before the decline. Variations in the J and H bands follow the optical changes even displaying some traces of the 1600-d cycles. Variability at longer wavelengths is not studied enough because of small number of observations. Sitko et al. (1994) comparing two SEDs obtained between 0.3 and 20 microns in 1980 and 1992 concluded that changes in the visual and mid-IR were opposite. Photometry in the 10-microns band obtained in late 60s shows that the object was at the 1992 level at that time. At the same time, mean brightness in a region between 2 and 8 microns is stable for more than 25 years. Grady et al. (1993) detected high-velocity gas moving toward the star in the IUE spectra which has also been found for a number of HAEBEs.

HD 50138, a very similar object with respect to FS CMa but having later spectral type, demonstrate the same photometric behavior. It was as faint as V=6.8 in 1982-1983 (Kilkenny et al. 1985) and since that time has been increasing visual brightness. At the moment it reached a level of 6.5 at which it was before 80s. Optical color-indices have not changed like in FS CMa. Near-IR brightness follows the optical changes with decreasing amplitude towards longer wavelength. This is expected for the case of formation of a dusty envelope with less optical depth than that of FS CMa. Periodicity can hardly be found because of smaller data set. Accretion phenomena have been found in the IUE spectra of the star (Grady et al. 1996) as well as in the optical region (Pogodin 1997).

A bright southern IR source, HD 87643, is known to have a spectrum similar to that of Eta Carinae and, thus, to be a candidate to B[e] supergiants (see Araujo et al., this volume). It declines almost linearly since 60s displaying variations around the mean track of the order of 0.5 mag. (Fig.1). The optical color--indices get redder with the brightness decline except U-B. The latter turned out to be bluer untill V = 9.0 and redder when the star declined further. We found a quasi--periodicity of 925.3 days by means of Fourier analysis of its visual light curve (Fig.2). The most recent observations (Torres et al. 1995 and our data obtained at the SAAO on 1997 July 10) show that now the star is as faint as V = 9.5. Near-IR observations of the object are rare but the data show no trend in the JHK bands and 20 % increase of brightness in the M--band between two observations obtained in early 70s and in 1988. If we compare its mean SEDs of 1983 (when the IRAS data were obtained) and of 1988, we can see the same effect as that in FS CMa: decline in the optics and brightening in the mid-IR. It is very important to continue monitoring of this object.

MWC 300, an object considered as a B[e] supergiant (Wolf & Stahl 1985) or a HAEBE star (Finkenzeller & Mundt 1984), shows no brightness trend over the last 20 years and no periodical variations. An amplitude in the V-band is nearly 0.35 mag., U-B and B-V do not depend on the brightness level. A very weak reddening with the brightness decrease is noticed only in V-R. If its luminosity is Mbol = -9.5 (Wolf & Stahl 1985) a distance toward it should be as large as 15 kpc.

Attention to MWC 342 has been attracted by Bergner et al. (1990) who detected a 132-day cycle most pronounced in the U-band. A cycle with a half of this period was also found in the R-band polarization. Since 1989 a group of V.S.Shevchenko obtained more than 450 UBVR photometric observations of this source. Their results show the presence of the 132-day cycle only in the 1989 set. Later it disappeared. In general, the star displays gradual brightening in the optical and near-IR region (see Fig.8). Closeness of a soft X-ray source to the star motivated Miroshnichenko (1991) to suggest that it is a Be/X-ray binary.

MWC 84 and MWC 930 have been observed by Bergner et al. (1995) and Shevchenko et al. (1993). Miroshnichenko (1995) found quasi-periodic variations in these data (11.7 and 58.4 respectively) and suggested that they are binaries because of anomalous Balmer decrements (see Fig.5), features of late-type stars in their spectra, and photometric behavior. MWC 930 was not included to the Allen & Swings' list because of small near-IR color-indices, however, it turned out to have a strong far-IR excess and emission lines including those of extremely strong Fe II and can be classified as a B[e] star.

MWC 623 was first suggested to be a binary system consisting of a hot and a cool star by Arkhipova & Ipatov (1982) from a spectrophotometric study (B0 V+ M1 III) and then by Zickgraf & Stahl (1989) by means of high-resolution spectroscopy (B2 V+ K2 II-III). The difference in spectral type estimates was partly due to the absence of the object's optical photometry by that time. Bergner et al. (1995) obtained nearly 30 UBVRIJHK observations, modeled the object's SED, and derived the B2 + K7 combination for the spectral types of the companions. The MWC 623 SED in the IRAS region is similar to those of late-type stars. No regular variability has been found in the photometric data.

MWC 349 is a well--known bright IR and radio source. It most likely belongs to the Cyg OB2 association which stars display a very strong interstellar reddening. Based on this, MWC 349 is considered to be a B[e] supergiant. It exhibits ~0.4 mag.variability in the VRIJHK-bands which is correlated in all these bands (Bergner et al. 1995). Analysis of this small data set allows to suggest a cyclicity with a period of ~ 1000 days. Gottlieb & Liller (1978) studied the photographic light-curve of MWC 349 based on Harvard plates obtained between 1900 and 1980 and also found evidences for slow cyclic brightness changes.

Thus, less than 20 % of B[e] stars were observed enough to analyze their light curves. The averaged amplitude of their optical variability is nearly 0.5 mag., which is significantly larger than that obtained for the Magellanic Clouds B[e] supergiants. However, we should keep in mind that these stars have been observed within quite short time intervals. More than a half of them, no matter are they suspected to be binaries or not, display quasi-periodic brightness changes which nature is not understood at the moment. Two stars, FS CMa and HD 87643, dislpay long-term high-amplitude variations of visual brightness which could be due to recent and probably recurrent dust formation events. Recently Arkhipova et al. (1996) using old photographic plates found another star, HDE 341617, showing similar photometric behavior. This approach being applied to other rarely observed B[e] stars could reveal new such objects allowing to study this process in more detail.

Spectral energy distributions

Analysis of their optical color-indices shows that in color-color diagrams they locate close to the mean reddening vector for the hottest stars or even above it. This means that they certainly contain a high-temperature star, as it is seen from spectroscopy, dominating the UV and blue spectral region. However, it turns out to be difficult to separate binaries containing a hot and a cool star and single stars. For comparison VV Cep-type binaries containing an OB-dwarf and an M-supergiant are shown in Fig 3. In many cases brightness ratio of the components makes objects very similar to highly reddened hot stars. However, VV Cep stars display smaller IR-excesses and steeper flux decrease toward longer wavelengths in the IRAS region than B[e] stars. It can be easily seen in the [25]-[12] ~ [60]-[25] diagram (Fig.4).

Several other B[e] stars (CPD-52 9243, CPD-57 2874, HD 327083, He2-139, M1-76, and probably AS 225 which 12 micron flux has been measured with a poor quality) also fall in this region. He2-139 was recently recognized as a symbiotic system. Both CPD stars and the HD star were found to show CO first overtone emissions at 2.3 micron (McGregor et al. 1988) which are observed in the late-type star spectra. Additionally, CPD-52 9243 displays anomalous Balmer decrement. M1-76 and AS 225 are not studied enough to make definite conclusions. Thus, there are several indications that all these stars are binaries containing a cool star besides the hot one. They have smallest IR-excesses among B[e] stars, however, higher than those of VV Cep stars. Six of them were observed by IRAS with its Low-Resolution Spectrometer. Only 3 Pup, which has a companion much fainter than the primary, shows silicate features at 10 and 18 micron in emission, other spectra are quite noisy and flat.

Recently we found an object exhibiting an intermediate IR-excess which is larger than those of possible B[e] binaries but still smaller than those of other B[e] stars. It is MWC 657 distinguished by Dong & Hu. We obtained its UBVRIK photometry and modeled its SED in the range of 0.3 - 60 micron (Miroshnichenko et al. 1997) under assumption that it is an early-B star surrounded by dusty envelope. As a result we obtained the envelope visual optical depth of 0.05, and the ratio of its outer to inner radius equal to 700. Density distribution of the circumstellar dust was found to be r**(-1.5) - r**(-2.0), where r is the radial distance. Therefore, as discussed above, possible B[e] binaries with very small IR-excess have extremely thin dusty envelopes, or their IR-excesses are due to dust around their cool companions.

Two high-luminosity stars, HDE 316285 and HDE 326823, display small IR-excesses in a wavelength region shortward of 12 micron which can be explained by free-free radiation of gaseous envelopes. However, ADDSCAN data show that both sources appear extended at the 25-100 micron IRAS bands. This could be due to radiation from dusty envelopes ejected from the stars in the past. Another suggested supergiant, MWC 349, shows IR-excess as strong as that of MWC 657, but almost featureless LRS spectrum probably indicating a deficit of silicates in its surrounding dust. It also appears extended at 25-100 micron. The detected cyclic variations do not exclude presence of a companion. MWC 349 has a higher temperature than HDE 316285 and HDE 326823, and could be simply younger forming its envelope more recently.

A group of other B[e] stars display very strong IR-excesses. It includes HD 45677, Ve2-27, MWC 300, MWC 922, and MWC 939. The former two display clear silicate feature in emission, MWC 300 has a flat LRS spectrum, while that of MWC 922 raises towards longer wavelengths. Quite a steep decrease of the SEDs of HD 45677 and MWC 300 points out to small extension of their dusty envelopes. They both are suggested to be pre-main-sequence stars although they are isolated from star forming regions. Ve2-27 and MWC 922 appearing slightly extended at 100 micron could have larger envelopes. They are rather faint in the optical region and detailed spectroscopy is needed to make a next step in their understanding.

Two other objects, MWC 342 and HD 50138, show almost the same decrease in their far-IR SEDs but smaller excess values which can be explained by smaller dust optical depths. Optical spectrum of MWC 939 was found to be similar to that of Eta Carinae by Arkhipova (1975). 12 UBV observations obtained at the Crimean Observatory by her group in 1988 - 1993 are very close to each other as well as to the optical brightness given by Allen & Swings (1976).

HD 87643 displays a medium IR-excess with silicate features in emission. AS 78 from the Dong & Hu's list was shown to be similar to HD 45677 from the spectroscopic point of view (Shejkina & Miroshnichenko, this volume). It is probably a main-sequence star with a dusty envelope of a rather small extension (ratio of the outer and inner radii is < 1000). He3-1191 has a very strange SED among B[e] stars containing a strong 3-micron bump and an LRS spectrum with a deep at 18 micron. de Winter et al. (1994) reported that the object displays very strong emission-line spectrum and have stellar temperature of at least 30000 K. They suggested that it is a pre-main-sequence star or a symbiotic system. More detailed information about B[e] SED modeling can be found in Ivezic et al. (this volume).

Other stars from Table 2 are poorly-studied. We certainly have to extend the sample of B[e] stars in order to make statistical analysis of the observed and physical characteristics possible. At the moment none of the objects of Tables 1 and 2 is clearly understood except several possible binaries and supergiants, for which some arguments in favor of this interpretation have been obtained. Nevetheless, evolutionary state of the latters is not well-determined. Another group with steep decrease of the IR-brightness probably formed their circumstellar envelopes quite recently. Perhaps they belong to the horizontal branch and are very young PNe. Otherwise they could pass an evolutionary stage near the main-sequence or after it (young B[e] supergiants) because HAEBEs usually loose their near-IR excess first. The descendants of the lower mass HAEBEs, $\beta$ Pictoris stars, retain their far-IR excesses for a long time already being at the main-sequence. At the same time we do not know much about evolution of more massive HAEBEs near the main-sequence, nor about evolution of massive stars just after leaving it. Not only distances towards B[e] stars are uncertain but stellar effective temperatures are also poorly known because of the photospheric line veiling by circumstellar emission. These objects, as well as others having complex structure, require comprehensive studies using all possible observational methods.

III. POLARIMETRY

Long-term polarimetric observations of FS CMa in 70s were carried out by Coyne & Vrba (1976). They found polarization to be variable with time and increasing towards longer wavelengths. All observed phenomena were interpreted as supporting the model of a dust ring surrounding several gaseous regions in the extended circumstellar disc around the star. Spectropolarimetry in the UV obtained with the WUPPE during the Astro-1 mission in 1990 showed a near-UV position angle flip by 90 degrees and a steep rise of polarization toward shorter wavelengths (Shulte-Ladbeck et al. 1992). This provides new evidence for a disc-like geometry of the circumstellar dust around FS CMa.

Another star has been observed by WUPPE during the Astro-2 mission in 1995 is HD 50138. It shows an almost flat polarization spectrum with a raise toward shorter wavelengths and without any position angle flip (Bjorkman et al. 1996). The UV polarization indicates the presence of the intrinsic component due to a non-spherical envelope while the absence of the flip can be due to a less dense circumstellar disk than those of FS CMa.

Zickgraf & Shulte-Ladbeck (1989) studying wavelength dependences of polarization of 8 faint B[e] stars concluded that data for two of them, MWC 342 and MWC 623, can be explained by Mie scattering, other two, MWC 645 and MWC 939, by Thomson scattering, while 3 stars: MWC 300, MWC 922, and MWC 1055, probably have no intrinsic polarization. It is followed from the above consideration that MWC 300 and MWC 922 show significant IR-excesses which can not exclude the presence of circumstellar polarization in these objects. Ivezic et al. (1998) found that all reddening of MWC 922 is circumstellar. This partially explains why polarization of stars in its vicinity (according to Zickgraf & Shulte-Ladbeck) is rather small (~ 1.5 %).

Comments to Table 3. AV is the overall extinction estimated from optical color--indices.

For the last star in the Zickgraf & Shulte-Ladbeck sample, MWC 349, some intrinsic polarization has been suspected. Recently Yudin (1996) studied photometric and polarimetric behavior of this object using quasi simultaneous observations obtained in 1985 - 1992 and found strong and rapid polarization variability. It varies between 4 and 11 % in the R-band. He estimated interstellar polarization in the object's vicinity and concluded that a significant part of the observed polarization has circumstellar origin (Pintrinsic ~ 11 % in the blue and visual regions). Yudin considers three mechanisms to be responsible for the polarization raise: Mie scattering in a dusty disk, Thomson scattering in high-density ionized clumps located inside the disk, and interstellar polarization. Spectropolarimetry of MWC 349 in the mid-IR (8-13 micron) was obtained by Aitken et al. (1990). They found that polarization peaks near 11 micron at 1.5 % and position angle also varied with wavelength. An upper limit for the interstellar polarization at this wavelength was found to be 0.5 % which is significantly lower than the observed level. The authors concluded that grains producing the polarization appear to lie in a dense rotating circumstellar disc. This is in agreement with radio data and speckle interferometric results.

Highly variable polarization has been detected earlier for MWC 342 (Bergner et al. 1990). Its changes were found to be quasi-regular with a period of 66 days(after subtraction the interstellar component) which is half a period of photometric changes in the U-band (Fig.8). Such a behavior is predicted for binary systems containing a compact object and a normal star (Dolginov et al. 1979). This prompted Miroshnichenko (1991) to suggest that MWC 342 was a Be/X-ray system.

Thus, the polarimetric results can be summarized as follows:

• All B[e] stars observed polarimetrically display a significant degree of polarization. In many cases it can not be explained by the interstellar mechanism only.
• Mie scattering, as well as Thomson scattering, can account for the observed wavelength dependence of polarization, so that neither gaseous nor dusty component can be ignored in analysis of observations.
• Polarization of the objects observed many times is found to be variable. This and spectropolarimetry in some cases confirms a hypothesis of non-spherical dust distribution around B[e] stars.

IV. INTERFEROMETRY

HD 45677 has been resolved at a wavelength of 4.8 micron with FWHM of 0.1 +/- 0.02 arcsec. Using distance measured by HIPPARCOS (350 +/- 50 pc, ESA 1997) one can estimate linear dimension of the source of 35 A.U. Dewarf & Dyck (1993) introduced a diagram displaying an IRAS color-index [12]-[25] versus circumstellar cloud diameter in which sources with redder color-indices have in general more extended envelopes. The authors' interpretation that sources with larger size circumstellar halos are in earlier stages of evolution while sources with bluer colors are in later stages is in agreement with predictions of evolutionary models by Adams et al. (1987).

V. CONCLUSIONS

• The majority of galactic B[e] stars locate outside of star forming regions and, hence, more likely are evolved stars rather than pre-main-sequence ones. Lack of their far-IR radiation with respect to that of near-IR could be explained as a sign of recent dust formation in their envelopes as well as of the absence of distant remnants of protostellar phase usually seen around young stars.regarding all objects which can be classified as B[e] stars.
• Basically, mechanisms of dust envelope formation suggested by the discoverers have been confirmed by follow up studies. Observations obtained over the last 20 years made it possible to study these processes in more detail, however additional efforts are required to determine physical parameters of the objects using up-to-date models.
• There is a lack of high-quality observations of galactic B[e] stars. Variability of their various characteristics requires coordinated long-term observations using different observational techniques. It is extremely important to obtain long-term mid- and far-IR observations which could tell us more about circumstellar envelope evolution.
• A large reservoir of new B[e] star candidates is a list first proposed by Dong & Hu and examined in a catalogue by The et al. (1994). Optical and near-IR photometry and at least medium-resolution spectroscopy is needed to preliminary classify these objects.

Aitken, D.K., Smith, C.H., Roche, P.F., & Wright, C.M. (1990). MNRAS, 247, 466

Allen, D.A., Swings, J.P. (1976). A&A, 47, 293

Arkhipova, V.P. (1975). Variable stars, 18, 289

Arkhipova, V.P., Ipatov, A.P. (1982). Sov.Astr.Let., 8, 554

Arkhipova, V.P., Ikonnikova, N.P., Esipov, V.F., Noskova, R.I. (1996). Pis'ma v Astron. Zhurn., 22, 526

Barbier, R., Swings, J.P. (1982). Proc. IAU Symp. N 98, 241

Bergner, Yu.K., Miroshnichenko, A.S., Sudnik, I.S., et al. (1990). Astrofizika, 32, 203

Bergner, Yu.K., Miroshnichenko, A.S., Yudin, R.V., et al. (1995). A&AS, 112, 221

Bjorkman, K.S. et al. (1996). ASP Conf.Ser., 97, 197

Carlson, E., Henize, K.G. (1979). Vistas in Astronomy, 23 , 213

Coyne, G. & Vrba, F.J. (1976). ApJ, 207, 790

Dewarf, L.E. & Dyck, H.M (1993). AJ, 105, 2211

de Winter, D., The, P.S., & Perez, M.R. (1994). ASP Conf.Ser., 62, 413

de Winter, D., van den Ancker, M.E. (1997). A&AS, 121, 275

Dolginov, A.Z., Gnedin, Yu.N., Silant'ev, N.A. (1979). Propagation and polarization of radiation in space. Nauka, Moscow

Dong, Y.S., Hu, J.Y. (1991). Chin. A&A, 15, 275

Drew, J.E. et al. (1997). MNRAS, 286, 538

ESA (1997). The Hipparcos and Tycho Catalogues, ESA SP-1200

Finkenzeller, U. & Mundt, R. (1984). A&AS, 55, 109

Gottlieb, E.W., Liller, W. (1978). ApJ, 225, 488

Gnedin, Yu.N. et al. (1992). Sov. Astr. Let., 18, 454

Grady, C.A., et al. (1993). ApJ, 415, L39

Grady, C.A., et al. (1996). A&AS, 120, 157

Halbedel, E.M. (1989). PASP, 101, 999

Hutchingson, M.G., Albinson, J.S., Barrett, P., et al. (1994). A&A, 285, 883

Ivezic, Z., Miroshnichenko, A.S., Elitzur, M. (1998). In B[e] stars. (eds.) C.Jaschek and A.M.Hubert, Ap&SS Library, 233, 227

Kazarovets, E.V. & Samus, N.N. (1997). IBVS N 4471

Kilkenny, D., Whittet, D.C.B., Davies, J.K., et al. (1985). SAAO Circ., 9, 55

Leinert, C. (1986). A&A, 155, L6

Manfroid, J., Sterken, C., Bruch, A., et al. (1991). A&AS, 87, 481

McGregor, P.G., Hyland, A.R., Hillier, J. (1988). ApJ, 324, 1071

Miroshnichenko, A.S. (1991). Infrared spectral region of stars. Proc. Int. colloq. held in Montpellier. Jaschek, C., Andrillat, Y. (eds.), p. 163

Miroshnichenko, A.S. (1994). ASP Conf.Ser., 62, 134

Miroshnichenko, A.S. (1995). A&ATr, 6, 251

Miroshnichenko, A.S., Kuratov, K.S., Ivezic, Z., \& Elitzur, M. (1997). IBVS N 4506

Parthasarathy, M. & Pottasch, S.R. (1989). A&A, 225, 521

Pogodin, M.A. (1997). A&A, 317, 185

Shulte-Ladbeck, R.E. et al. (1992). ApJ, 401, L105

Sterken, C., Manfroid, J., Anton, K., et al. (1993). A&AS, 102, 79

Sterken, C., Manfroid, J., Beele, D., et al. (1995). A&AS, 113, 31

The, P.S., de Winter, D., & Perez, M.R. (1994). A&AS, 104, 315

Torres, C.A.O. et al. (1995). AJ, 109, 2146

van der Veen, W.E.C.J. & Habing, H.J. (1988). A&A, 194, 125

Wackerling, L.M. (1970). Mem.Roy.Astron.Soc., 73, 153

Wolf, B. & Stahl, O. (1985). A&A, 148, 412

Yudin, R.V. (1988) Sov.Astron. 32, 652

Yudin, R.V. (1996). A&A, 312, 234

Yudin, R.V. & Evans, A. (1998). A&AS, 131, 401

Zickgraf, F.-J., Wlof, B., Leitherer, C., Appenzeller, I., Stahl, O. (1986). A&A, 163, 119

Zickgraf, F.-J. & Shulte-Ladbeck, R.E. (1989). A&A, 214, 274

Zickgraf, F.-J. & Stahl, O. (1989). A&A, 223, 165

Completed 1997 September 29
Last updated 2000 July 07