COMMISSION 27 OF THE I. A. U.
           INFORMATION BULLETIN ON VARIABLE STARS
                       Number 2808

                                        Konkoly Observatory
                                        Budapest
                                        14 October 1985

                                        HU ISSN 0374 - 0676




 THE PERIODICITY OF THE PHOTOMETRIC VARIATIONS OF HR 7671 (HD 190390)



    HR 7671 (HD 190390) has been classified as a F-type giant by Cowley
(1976), but evidence is accumulating now that the colors and the spectrum
of this star are those of a supergiant (e.g. Olsen, 1983; Sasselov, 1985;
Eggen, 1985). Since its galactic latitude is rather high (b II =-21.5deg),
HD 190390 is then a rather uncommon object. Additionally, it has recently
been found that the star is a photometric variable. Olsen (1983) observed
a steady decrease of the brightness of HD 190390 in data covering five
nights and later on observed similar variations but with another range.
    Our interest in HD 190390 arose from the large scatter of its data
in the Geneva photometric system (Rufener, 1981). We have been collecting
new observations of this object in the Geneva system since late 1982, using
the Swiss photometric telescope at the European Southern Observatory. In
the present paper, we report on the discovery of a stable periodicity in
our data, that cover 1040 days, from September 1982 till July 1985.
    The visual brightness variations during September-October 1982 and
May-July 1985 are shown in Figure 1. The time is expressed in Julian Days
- 2440000. A time scale of the order of a month can clearly be recognized
in the variations. A similar timescale is present in the 45 measurements 
made in 1983 and in the 22 measurements made in 1984. Fourier analysis of
the whole set of 123 data points reveals a most significant period of 28.49
days. A least-square cosine fit with this period has an amplitude of 0.11
mag and accounts for 80% of the total variance. The number of cycles
covered is 36. The phase diagrams for the color indices [U-B] and [B-V] and
for the visual brightness mv are given in Figure 2.

    
It is apparent from Figure 1 that the amplitude and the shape of the
light curve vary from cycle to cycle. Indeed, the residual scatter around
the mean light curve in Figure 2, sigma(res) = 0.038 mag, is much larger than 
the estimated observational error, which is 0.004 mag. However, no reduction
with any other period can remove a significant fraction of the remaining
variance. The scatter thus reflects irregular cycle-to-cycle variations.
    The [U-B] and [B-V] variations are respectively in antiphase and in
phase with the light variations. Such color variations are reminiscent of
the behavior of variables in the Cepheid instability strip. The spectral
type of HD 190390 and the length of the period are two other arguments
which naturally lead us to discuss the properties of this variable in terms
of what is known about the Cepheids.


                            [FIGURE 1]

 Figure 1: Light variations of HD 190390 during 1982 and 1985.




                            [FIGURE 2]

 Figure 2: Phase diagram for the color and light variations of HD 190390
           during the period 1982-1985. 



    It is highly unlikely that HD 190390 is a bona fide population I
Cepheid. The light variations show irregularities and the colors are much
bluer than those of the known long period Cepheids. Third, the galactic
latitude of the star and the absolute magnitude it would have if the
period-luminosity relation were valid for it would imply a distance of
more than 1 kpc from the galactic plane, clearly uncompatible with the
known scaleheight of the population I Cepheids.
     A high vertical distance is not uncompatible with HD 190390 being a
population II Cepheid. Also, there are indications that it is a metal-weak
star (Sasselov, 1985). On the other hand, the kinematics of the star do
not indicate a pronounced population II character (Eggen, 1985). Moreover,
the irregularity and the shape of the light curve are typical even for
long-period population II Cepheids.
     A third possibility is that HD 190390 is similar to HD 161796 and to
HD 163506 (Burki et al., 1980), stars that belong to a group of variable
supergiants at high latitude, labeled "UU Her stars" by Sasselov (1984).
The exact evolutionary stage of these stars is unknown, but it might be
that they are in rapid evolutionary phase. HD 190390 would be the first
strictly periodic member of the group. The study of the secular variations
of its period is of high interest for the constraints it imposes on the 
evolutionary stage of the star and, incidentally, on that of the whole group.
     We therefore plan to continue monitoring this star. A program of
simultaneous radial velocity observations with CORAVEL has now been started.
The results of this program will be published together with a quantitative
discussion of all photometric data.



                       C. WAELKENS* and M. BURNET**
                      (*) Astronomisch Instituut
                          Celestijnenlaan 200B
                          B-3030 Heverlee (Belgium)
                      (**) Observatoire de Geneve
                          CH-1290 Sauverny (Switzerland)

References:


Burki, G., Mayor, M., Rufener, F.: 1980, Astron.Astrophys.Suppl.Ser. 42, 383. [BIBCODE 1980A&AS...42..383B ]
Cowley, A.P.: 1976, Publ.Astron.Soc.Pacific 88, 95. [BIBCODE 1976PASP...88...95C ]
Eggen, O.J.: 1985, Astron.J. 90, 1260. [BIBCODE 1985AJ.....90.1260E ]
Olsen, E.H.: 1983, Astron.Astrophys.Suppl.Ser. 54, 55. [BIBCODE 1983A&AS...54...55O ]
Rufener, F.: 1981, Astron.Astrophys.Suppl.Ser. 45, 207. [BIBCODE 1981A&AS...45..207R ]
Sasselov, D.D.: 1984, Astrophys.Space Sci. 102, 161. [BIBCODE 1984Ap&SS.102..161S ]
Sasselov, D.D.: 1985, Inf.Bull.Variable Stars No. 2776.