COMMISSIONS 27 AND 42 OF THE IAU
                INFORMATION BULLETIN ON VARIABLE STARS
                             Number 4091

                                                   Konkoly Observatory
                                                   Budapest
                                                   20 September 1994

                                                   HU ISSN 0374 - 0676


             IS GLIESE 410 OR BD+23d2297 THE VARIABLE STAR?


  The dM1 star, Gliese 410, was reported as a photometric variable star by Bopp
et al. (1983). It apparently displayed V-band modulation with an amplitude of 0.1
magnitudes and a period of 2.935 days. Such a short period in a star with a deep
convection zone is likely to give rise to magnetic activity in its outer atmosphere, leading
to chromospheric and coronal emission. Subsequent spectroscopy has shown that Gliese
410 is mildly active. It has a chromospherically filled Halpha line (although it is not in
emission) and has modest emission lines at Ca II H and K (Panagi & Mathioudakis
1993). We have compared Gliese 410 with other dwarfs that have similar known periods
and spectral types. Gliese 388 (AD Leo - P=2.7 days), Gliese 490A (BF CVn - P=3.17
days) and Gliese 803 (AU Mic - P=4.87 days). These all show Halpha in emission, have
other chromospheric lines that are up to an order of magnitude stronger than in Gliese
410, and are considered to be amongst the most active late-type stars in the solar
neighbourhood. If Gliese 410 does have a 2.935 day period, then it is anomalously
under-active.
  Bopp et al. (1983) used the comparison star BD+23d2297 in their investigation.
The V and B-V magnitudes of this star are given as 8.70 +/- 0.35 and 0.72 +/- 0.13 by
the Hipparcos Input Catalogue (Turon et al. 1992), in reasonable agreement with the
spectral class of G5 (given in the SAO catalogue), if the star is a dwarf. This star
lies within the error circle of a ROSAT Wide Field Camera (WFC) EUV source in
the most recent WFC catalogue (Pye et al. 1994). The EUV source is designated 2RE
J1101+223 in IAU nomenclature, and BD+23d2297 has been positively identified as the
optical counterpart, on the basis of chromospheric emission at Ca II H and K. A rough
calculation of the EUV to bolometric luminosity yields a value of ~ 2 x 10^-4, which
is characteristic of the most magnetically active stars - usually young, rapid rotators
or close binary systems (Jeffries et al. 1991). BD+23d2297 is therefore very likely to
show short-period photometric variability caused by starspots rotating across the visible
surface, and may be the true source of the variability seen by Bopp et al. (1983).


                            [FIGURE 1]

     Figure 1: A spectrum of BD +23 2297 in the H alpha region.

                            [FIGURE 2]

     Figure 2: The radial velocity curve folded with the ephemeris
   HJD 2449055.132 + 1.528E. Errors are generally smaller than 2 km/s.


  BD+23d2297 has been observed spectroscopically at the 2.5 m Isaac Newton
Telescope, with an intermediate dispersion spectrograph and a CCD camera. The spec-
tra obtained were in the red and had a resolution of 0.441A. BD+23d2297 is clearly a
short-period, double-lined spectroscopic binary and has a Halpha feature that appears to be
chromospherically filled in (see Figure 1). Analysis of a radial velocity curve taken over
6 nights (Figure 2) yields the following orbital elements. Period = (1.528 +/- 0.006) days,
K1 = K2 = (110 +/- 2) km s^-1, gamma = 3.9 +/- 2.3 km s^-1 and zero eccentricity. If the stars
are tidally locked, as they usually are in short-period systems, then starspot modulation
with the orbital period might be expected. A consideration of the frequencies involved
quickly reveals that aliases with a one day period would produce a signal at a period
of 2.89 +/- 0.02 days, very close to the period obtained by Bopp et. al. (1983). Bopp et
al. do not present their raw data, so we cannot make a detailed examination of this
hypothesis. Nevertheless, the aliased period is sufficiently close that spot migration or
differential rotation could explain the remaining discrepancy.

   In summary, we have found that BD+23d2297, a star used as a comparison to
derive the rotation period of Gliese 410, is likely to be variable itself. If Gliese 410 is
the true variable, it is decidedly under-active for its rotation rate. BD+23d2297 is a
short-period SB2, and the orbital period aliased with a 1 day^-1 frequency could produce
the modulation period found for Gliese 410. Further photometry of both stars is needed
to resolve the issue.


              R. D. JEFFRIES, D. BERTRAM & B. R. SPURGEON,
              School of Physics and Space Research,
              University of Birmingham,
              Edgbaston,
              Birmingham B15 2TT. UK.
              E-Mail rdj@xun4.sr.bham.ac.uk



References


Bopp, B. W., Africano, J. L., Stencel, R. E., Noah, P. V., Klimke, A.
  1983, Astrophys. J., 275, 691. [BIBCODE 1983ApJ...275..691B ]
Jeffries, R. D. et al. 1991, Vistas Astron., 34, 267. [BIBCODE 1991VA.....34..267J ]
Panagi, P. M., Mathioudakis, M. 1993, Astron. Astr. Suppl., 100, 343. [BIBCODE 1993A&AS..100..343P ]
Pye, J. P. et al. 1994, Mon. Not. R. astr. Soc., submitted. [BIBCODE 1995MNRAS.274.1165P ]
Turon, C. et al. 1992, The Hipparcos Input Catalogue, ESA SP-1136,
  ESA Publications, Noordwijk. [BIBCODE 1992ESASP1136.....T ]