Non-Periodic Phenomena in Variable Stars 

                                 IAU Colloquium, Budapest, 1968



                  FLARE STARS NEAR NGC 7023 

              L. V. MIRZOYAN and E. S. PARSAMIAN
         Byurakan Astrophysical Observatory, Armenia, USSR


  The distribution of flare stars in T-associations can give some definite 
information about the early stages of stellar evolution. In order to determine 
this distribution the small T-association near NGC 7023 has been selected. Being 
very compact and comparatively close to us it is suitable for the mentioned 
purpose. The photographic observations of the NGC 7023 region began in 1962. 
The 40" Schmidt telescope of the Byurakan Astrophysical Observatory 
of the Academy of Sciences of Armenian SSR was used. 92 plates with 479 
five-minutes serial exposures embracing an effective observational time of 
40h were obtained. The limiting magnitude on our plates is 18.5m, consequently 
flare stars with absolute magnitude brighter than +11m (for the distance 
of the T-association equal to 280 pc, Weston 1953) can be detected on these 
plates. Nine flare stars (Mirzoyan et al. 1968 a, b), two of which flared twice 
during our exposures, have been found. Another possible flare star in this 
region has been announced by Rosino and Romano (1962). Thus, the number 
of known flare stars in this region is equal to 10.
  This number can be used for a rough estimation of the total number 
of flare stars in this region brighter than M ~ + 11m.
  Assuming, that the probability P_k of flare appearance obeys the Poisson's 
law:
               P_K = e^-nu*t (nu*t)^K/ K!

where t denotes the duration of the observations, nu the frequency of flares, 
K the number of flares during t, respectively, and using for the mathematical 
expectation of the number of stars flared K times, the expression

               N_K=NP_K,

one can calculate the number N of all probable flare stars in this region. With 
our small statistics (N_l = 8, N_2 = 2) we obtain N = 27. From this data 
also follows that the mean time interval between flares is 3.3d. It is difficult
to discover photographically flares with amplitudes less than 0.5m. Therefore, 
these numbers correspond to flare stars with amplitudes larger than 0.5m. 
Since the duration of the flare-maximum can be shorter than 5 min, i.e., 
shorter than our exposure time and the starting point of the exposures may 
correspond to different moments of the flares, the estimated amplitudes are 
in fact always less than the real ones. Therefore, the real number of flares 
with amplitudes exceeding 0.5m must be larger and the mean time interval
between flares correspondingly smaller than estimated. Naturally, our 
calculations refer only to stars brighter than M = +11. The number of 
corresponding stars in the Pleiades (r = 126 pc), is equal to 20 (Haro 1968). 
The total duration of the corresponding observations in Pleiades is about 189^h. 
The calculations similar to that brought above give N = 112 for the total number 
of flare stars in the Pleiades. Therefore, assuming that the luminosity function 
of flare stars in the region of NGC 7023 is close to that of the Pleiades, one 
can conclude that the number of all flare stars in the region of NGC 7023 
differs not much from the corresponding number in the Pleiades.




                             Fig. 1

   Of particular interest is the distribution of the known flare stars in the 
region of NGC 7023 presented in Fig. 1. It shows that eight flare stars out 
of ten are situated far from the centre of the T-association, probably farther 
than the limits of the association showed by the dotted circle having a radius 
of 1 deg (Kholopov, 1959). The remaining two stars have probably come to the 
central part of the cluster through the effect of projection, in reality they 
may be far from the centre in space. Thus, it may be suggested that the flare 
stars surround the T-association. It must be noted, we suppose, that all 
detected flare stars in the region of NGC 7023 are in a distance equal to the 
distance of the T-association. The opposite assumption, i.e. that they are 
field stars, is less probable, because of the very high frequency of flares of 
field stars needed in this case.
  If one assumes that all of them have been originated in the central 
region of the association, then the distribution of flare stars confirms Haro's 
(1965) hypothesis, according to which the flare stage of a star follows the 
T Tauri stage.
  On the other hand if we suppose Haro's hypothesis to be correct, the 
observed distribution of flare stars around the T-association NGC 7023 can be 
interpreted as a confirmation of the expansion of this association. However, 
our statistical sample is not rich enough for a final conclusion.
  The authors thank to Prof. Ambarzumian for valuable advices.


                 REFERENCES

Haro, G., Chavira, E., 1965, Vistas in Astronomy, vol. 8, 89. 
Haro, G., 1968, Stars and Stellar Systems, vol. VII. 147.
Kholopov, P. P., 1959, Astron. Zu. 36, 295.
Mirzoyan, L. V., Parsamian, E. S., Chavushian, O. S., 1968a, Soobshch. Byurakan Obs.
   39, 3.
Mirzoyan, L. V., Parsamian, E. S., Kalloglian, N. L. 1968b, unpublished. 
Rosin, L., Romano, G., 1962, Asiago Contr. N. 127.
Weston, E. B. 1953, Astr. J. 58, 48.


                  DISCUSSION

Rosino: I should like to know whether flares have been observed in known 
       nebular variables of NGC 7023 or whether the flares are usually 
       observed in stars which are not nebular variables.
Parsamian: No one of these flare stars coincide with known nebular variable 
       stars in this region.