COMMISSION 27 OF THE I. A. U.
INFORMATION BULLETIN ON VARIABLE STARS
Number 1798
Konkoly Observatory
Budapest
1980 June 3
93 LEONIS: A NEW VARIABLE STAR
The bright (V=4.5m) star 93 Leonis is a long-period RS CVn binary with a
spectral type of A+G5IV-III according to Batten et al. (1978) and an orbital
period of 71.70d according to Cannon (1910). Young and Koniges (1977) quote a
spectral type of F8 IV but note that a secondary component in the spectrum,
probably the cooler star, is responsible for the moderately strong Ca II H & K
emission.
To look for the photometric distortion wave characteristic of so many other RS
CVn binaries, we observed 93 Leo photoelectrically in 1976 and in 1979. The
1976 observations, by Landis, revealed no variation larger than +-0.01m, as
reported by Heiser (1978), but the 1979 observations did show 93 Leo to be
variable, with an amplitude around 0.03m.
We observed 93 Leo differentially with respect to the comparison star 92 Leo
mostly in one color. Landis had observed on 14 nights in 1976 between JD
2,442,871.6 and 2,442,937.6. Louth, Montle, Skillman, and Vaucher observed on
44 nights in 1979 between JD 2,443,906.9 and 2,444,043.6, almost two cycles of
the 71.7d orbital period. Details of the observations at Dyer are given by
Vaucher (1979). The individual differential observations, generally three on
each night, were corrected for differential atmospheric extinction with nightly
extinction coefficients at Dyer but mean coefficients at the other
observatories. Then they were transformed to V of the UBV system with known
transformation coefficients and a mean color difference of Delta(B-V)=-0.41m,
where Delta is in the sense 93 minus 92 Leo. At Dyer observations were
obtained also in B of the UBV system, but they are not presented here.
On 8 nights in 1979 at Dyer the comparison star 92 Leo was observed 17 times
differentially with respect to the check star HR 4505. There was no indication
of significant variability of one respect to the other, the standard deviation
of a single observation from the mean being +-0.005m in V and +-0.009m in B.
Nightly means of the 1979 Delta V magnitudes are plotted in the Figure versus
phase computed with the ephemeris
JD(hel.)=2,418,017.865+71.70d n,
where the epoch is a time of conjunction (the A-type star behind) derived from
the time of periastron given by Cannon (1910) and the period is that used by
[FIGURE 1]
Cannon in his determination of the spectroscopic orbit. We used Fourier
analysis to express the light as
l=A_0+A_1cos +A_2cos +B_1sin ,
where unit light corresponds to Delta V=-0.750m. The Fourier fit appears as
the curve in the Figure. The resulting coefficients are given in the Table
below.
1976 1979
A_0 0.9739+-0.0022 1.0000+-0.0019
A_1 +0.0055+-0.0033 +0.0122+-0.0025
A_2 +0.0012+-0.0032 +0.0015+-0.0028
B_1 +0.0011+-0.0031 -0.0038+-0.0028
The variation we see in 1979 is probably a distortion wave, with a full
amplitude of a V=0.028m+-0.005m and a minimum at 0.452p+-0.034p. Seeing
this variability prompted us to look again at the 1976 photometry. Fourier
analysis of those observations resulted in the coefficients shown also in the
Table. The distortion wave was probably there in 1976 also because, although
its amplitude of Delta V=0.013m+-0.007m was smaller and significant at only
the 2 sigma level, its minimum at 0.532p+-0.087p was at the same phase,
within the uncertainties. A final curiosity is the smaller value of A_0 in
1976, indicating a mean light level almost 0.03m fainter than in 1979. Any
uncertainty in the transformation of Landis' data should have produced a
systematic error of only +-0.005m.
We can probably say with confidence that the differential reflection effect
cannot be producing this variation. For one reason, measurable reflection
would be unlikely in such a long-period binary. For a second reason, its
amplitude may be variable. And for a third reason, maximum light occurs
around phase 0.0p, when the A-type (presumably the hotter) star is behind. In
making this last statement we are assuming the uncertainty in Cannon's 71.70d
period is around +-0.01d, in which case phases in our Figure would be
uncertain by only +-0.05p. If the uncertainty is much larger, then we have
lost our fix on the phasing, because the vanishingly small A_2 coefficients
can provide no help.
D.S.H. is happy to acknowledge that part of this work was supported by
N.A.S.A. research grant NSG-7543.
DOUGLAS S. HALL
GREGORY W. HENRY
CHRISTOPHER A. VAUCHER
Dyer Observatory
Vanderbilt University
Nashville, Tennessee 37235
HOWARD J. LANDIS ^*
Price Road West
RFD No. 2, Box 44ED
Locust Grove, Georgia 30248
HOWARD LOUTH ^*
2199 Hathaway Road
Sedro Woolley, Washington 98284
ROBERT E. MONTLE
Veen Observatory
3308 Kissing Rock Road Southeast
Lowell, Michigan 49331
DAVID R. SKILLMAN ^*
9514 48th Avenue
College Park, Maryland 20740
References:
Batten, A.H., Fletcher, J.M., Mann, P.J. 1978, P.D.A.O. 15, 158. [BIBCODE 1978PDAO...15..121B ]
Cannon, J.B. 1910, J.R.A.S. Canada 4, 455. [BIBCODE 1910JRASC...4..452C ]
Heiser, A.M. 1978, B.A.A.S. 10, 98. [BIBCODE 1978BAAS...10...98H ]
Vaucher, C.A. 1979, M.A. Thesis, Vanderbilt University.
Young, A. and Koniges, A. 1977, Ap. J. 211, 836. [BIBCODE 1977ApJ...211..836Y ]
^* of the A.A.V.S.O.