COMMISSION 27 OF THE I. A. U. INFORMATION BULLETIN ON VARIABLE STARS Number 3123 Konkoly Observatory Budapest 23 December 1987 HU ISSN 0374-0676 GAMMA LUPI DOES NOT APPEAR TO BE A Be STAR * Ghosh et al. (1987) recently reported the discovery of the bright B2 IV star gamma Lupi (= HR 5776 = HD 138690) as a new Be star. This bright (V = 2.8 mag) star has not previously been seen to display emission lines, and the detection of a non-emission line B star that suddenly becomes a Be star would be very important indeed as Ghosh et al. (1987) point out. Even more important is, however, that such inferences are based on reliable observational facts (cf., e.g., the similar case of HD 91948 as discussed by Bolton and Lyons 1984). My inspection of nearly 70 references kindly supplied by the Centre de Donnees Stellaires, Strasbourg, for gamma Lup and the period 1950 to 1987 did not uncover any indication of circumstellar material in excess of what can be expected for non-emission line B stars. Jaschek et al. (1969) did not find the H-alpha emission looked for, nor did any other spectral classification work (e.g., de Vaucouleurs 1957, Slettebak 1968, Hiltner et al. 1969) discover bright spectral lines. Oegerle and Polidan (1982) searched Copernicus scans for circumstellar OI lambda 1304 A without detecting it at the 5 mA level. Likewise, Snow et al. (1979) only detected the two very strongest of the numerous FeIII lines often seen in Be stars. Therefore, there was probably no significant amount of circumstellar matter along the line of sight. The intrinsic linear polarization as measured by Serkowski (1970) is consistent with there also being no circumstellar matter in a nonspherical configuration. Barker et al. (1985) measured a longitudinal magnetic field strength of 45+/-74 Gauss, i.e. the star may as well have no magnetic field. Ghosh et al. mention 320 km/s as the v sin i of gamma Lupi. This appears to be the value given by Huang (1953). On the revised scale, Slettebak et al. (1975) derive only 210 km/s. Also, Slettebak's (1968) spectral classification is B2 IV, not B2 IVn, so that gamma Lupi at least is not an extreme Bn star. The star is known to be a close visual binary consisting of two components of nearly equal brightness with an angular separation of * Based on observations obtained at the European Southern Observatory, La Silla, Chile. less than 1 arcsec and an orbital period of about 150 years (Heintz 1956). Van Albada and Sher (1968) measured radial velocity variations with a full range of about 80 km*s and suggest that one of the two visual components is a spectroscopic binary with a possible period of 2.8+/-0.1 day. My re-analysis of their data and the more recent radial velocities by Levato et al. (1987) does not contradict this conclusion. However, van Albada and Sher's measurement of April 25.019, 1966 is rather discrepant. This problem is somewhat alleviated, but not completely removed, with a period near 0.739 day which is a 1 c/d alias and also roughly one-quarter of the period found by van Albada and Sher so that the radial velocity variations may also be non-orbital in origin. With this period, the radial velocity curve is very non-sinusoidal. A new series of multiple observations per night seems necessary for the conclusive resolution of this ambiguity. Independent of the result, it is almost certain already that gamma Lupi does contain an intrinsically variable star as is suggested by the two HeI 6678 A line profiles of Fig. 1 which were observed on two consecutive nights and display the typical signature of intermediate-order (m ~ 8-10) nonradial pulsation. On the other hand, in a search for new beta Cephei stars, gamma Lupi was tested by Jakate (1979) for rapid variability, but no variations were detected at the 0.01 mag level. Snow (1981) found a slight asymmetry in the UV resonance lines from which he deduced a mass loss rate of 9.65 x 10^-11 M/yr. He states that of the 22 stars in his sample only three are non-emission line stars (this statement was also carried over to Snow 1982), but y Lupi is not among the three exceptions mentioned. Since no other evidence of the star being a Be star is given and Snow et al. (1979) explicitly say that emission lines are not known from this star, gamma Lupi probably was just accidentally not included among the pure B stars in Snow (1981). All available evidence therefore suggests that gamma Lupi in fact did not so far harbour a Be star. The evidence that Ghosh et al. offer for gamma Lupi recently having turned into a Be star is based on four photographic spectra with a reciprocal dispersion of 16 A/mm (unfortunately the widening of the spectra is not given). The Halpha tracings each show two or three faint (Ghosh et al. do not provide a flux scale but it can indirectly be deduced with the help of Fig. 2 discussed below) spikes which however vary in position and strength. None of these features exceeds the amplitude of the noise in the adjacent continuum as is to be expected if they, too, are just typical spikes in the noise. At this level the discrimination between Be and B stars clearly becomes doubtful, and ultimately the latter category may even vanish. The reality of these features is also for another reason doubtful since general experience (e.g., Dachs et al. 1986) shows that the separation of the Ha emission peaks in Be stars is the larger the weaker the emission is (see Dachs et al. 1986 for the explanation in the framework of the standard equatorial disk model). By comparison, the features of Ghosh et al. appear too closely spaced. Purely by coincidence I have obtained a single Halpha spectrum of gamma Lupi with the Coude Echelle Spectrometer (CES) plus Reticon detector of the European Southern Observatory at La Silla on April 12, 1987, i.e. just 18-21 days after Ghosh et al. (1987). The characteristics of this spectrum and the reduction procedure applied to it are the same as described in Baade et al. (1988) except for possible low-amplitude large-scale distortions which have not been attempted to correct. The profile is reproduced in Fig. 2; at a S/N of better than 600 there is no trace of an emission component. [FIGURE 1] Figure 1: Two HeI lambda 6678 A profiles of gamma Lupi obtained on January 22 (top) and 23 (bottom), 1985. Note the wiggles which are probably indicative of intermediate order (m~8-10) nonradial oscillations. The flux scale is provided by the vertical bar; the offset between the two spectra corresponds to 2% of the adjacent continuum flux. [FIGURE 2] Figure 2: Halpha profile of gamma Lupi observed on April 14, 1987. The vertical bar gives the flux scale (in units of the adjacent continuum); it also indicates the order of magnitude of the fainter of the features reported by Ghosh et al. (1987). The numerous sharp absorption features are of telluric origin. The profile is not perfectly symmetric because low spatial frequency distortions due to flatfielding, etc. have not specifically been corrected for (cf. also Smith et al. 1987 and Baade at al. 1988). There is no indication whatever of an emission component. Observations by Baade et al. (1988) have shown that Be stars may undergo small outbursts which produce faint Ha line emission that fades on a timescale comparable to the time elapsed between the observations by Ghosh et al. and the one in Fig. 2. In so far the new data do not prove that gamma Lupi has never been a Be star, especially not because the occurrence of emission may also depend on the phase of the short-term variability. But the assumption that gamma Lupi is not a Be star still seems well justified. DIETRICH BAADE Space Telescope-European Coordinating Facility European Southern Observatory Karl-Schwarzschild-Str. 2 D-8046 Garching bei Munchen West Germany References: Baade, D., Dachs, J., van de Weygaert, R., Steeman, F.: 1988, Astron. Astrophys., in press (= ESO Scient. Prepr. No. 550) [BIBCODE 1988A&A...198..211B ] Barker, P.K., Landstreet, J.D., Marlborough, J.M., Thompson, I.B.: 1985, Astrophys. J. 288, 741 [BIBCODE 1985ApJ...288..741B ] Bolton, C.T., Lyons, R.W.: 1984, Inf. Bull. Var. Stars, No. 2580 Dachs, J., Hanuschik, R., Kaiser, D., Rohe, D.: 1986, Astron. Astrophys. 159, 276 [BIBCODE 1986A&A...159..276D ] de Vaucouleurs, A.: 1957, Mon. Not. R. astron. Soc. 117, 449 [BIBCODE 1957MNRAS.117..449D ] Ghosh, K.K., Kuppuswamy, K., Jaykumar, K., Rosario, M.J., Velu, C.: 1987, Inf. Bull. Var. Stars, No. 3057 Heintz, W.D.: 1956, Astron. Nachr. 283, 145 [BIBCODE 1956AN....283..145H ] Hiltner, W.A., Garrison, R.F., Schild, R.E.: 1969, Astrophys. J. 157, 313 [BIBCODE 1969ApJ...157..313H ] Huang, S.-S..: 1953, Astrophys. J. 118, 285 [BIBCODE 1953ApJ...118..285H ] Jakate, S.M.: 1979, Astronom. J., 84, 552 [BIBCODE 1979AJ.....84..552J ] Jaschek, C., Jaschek, M., Kucewicz, B.: 1964, Zeitschr. Astrophys. 59, 108 [BIBCODE 1964ZA.....59..108J ] Levato, H., Malaroda, S., Morrell, N., Solivella, G.: 1987, Astrophys. J. Suppl. Ser. 64, 487 [BIBCODE 1987ApJS...64..487L ] Oegerle, W.R., Polidan, R.S.: 1982, Publ. Astron. Soc. Pacific, 94, 997 [BIBCODE 1982PASP...94..997O ] Serkowski, K.: 1970, Astrophys. J. 160, 1083 [BIBCODE 1970ApJ...160.1083S ] Slettebak, A.: 1968, Astrophys. J. 151, 1043 [BIBCODE 1968ApJ...151.1043S ] Slettebak, A., Collins, G.W., Boyce, P.B., White, N.M., Parkinson, T.D.: 1975, Astrophys. J. Suppl. Ser. 29, 137 [BIBCODE 1975ApJS...29..137S ] Smith, M.A., Graves, J.E., Jaksha, D.B., Plymate, C.L., Ramsey, L.W.: 1987, Publ. Astr. Soc. Pacific 99, 654 [BIBCODE 1987PASP...99..654S ] Snow, T.P.: 1981, Astrophys. J. 251, 139 [BIBCODE 1981ApJ...251..139S ] Snow, T.P.: 1982, Astrophys. J. (Letters) 253, L39 [BIBCODE 1982ApJ...253L..39S ] Snow, T.P., Peters, G.J., Mathieu, R.D.: 1979, Astrophys. J. Suppl. Ser. 38, 227 [BIBCODE 1979ApJS...39..359S ] van Albada, T.S., Sher, D.: 1968, Bull. Astron. Inst. Netherlands 20, 204 [BIBCODE 1969BAN....20..204V ]