COMMISSION 27 OF THE I. A. U.
         INFORMATION BULLETIN ON VARIABLE STARS
                     Number 2576

                                      Konkoly Observatory
                                      Budapest
                                      28 August 1984

                                      HU ISSN 0374 - 0676

            SPECTROSCOPIC OBSERVATIONS OF PU Vul

  The peculiar nova-like variable PU Vul was observed on July 3,5,9,1984
using the grating spectrograph at the Nasmyth focus of the 60/90-cm Schmidt
telescope of Beijing Observatory. The emulsion used was Kodak 103aF. The
spectral range was 3500-6700 A. The reciprocal dispersion was 167 A/mm. The
spectral features were almost the same as those on December 11,1983 (Liu
Zongli, Hao Xiangliang, IBVS, No. 2466, 1984). The main absorption lines
were H gamma-H10, Ca II H,K, Fe II and Ti II. The H beta was filled with
emission. The H alpha emission was quite strong still.
  Since 1982 the H alpha emission has remained for two years. Since 1981 PU Vul
has been always at maximum light according to Kolotilov et al.(IBVS, No.
2097, 1982), Purgathofer et al.(IBVS, No.2291, 1983) and our photographic
observations.
  Besides our observations, we have reviewed the overall behaviour of this
object. All phenomena observed for this star led us to suggest that this
object should be an exceptionally slow nova. And it may be a binary which
consists of an M giant and a hot component. Due to the mass exchange and
accretion of the hot component the temperature of the hot component increased. 
Many mini-eruptions happened and a great deal of material was ejected.
Thus there is a lot of circumstellar matter around the hot component. This
matter produces a large amount of extinction at the short wavelength and
changes its distribution of radiation. So this binary slows the spectral
type of the M giant in general. And it shows the spectral type of the A or
F giant only during the maximum light. This model might explain most of the
observational results, such as the spectral type of M4 on August 30, 1958
(C.B. Stephenson, IAUC, No.3356, 1979), the spectral type of M4 on
September 6.59, 1978 (K. Ishida, IAUC, No.3350, 1979), the spectral type
A or F during the maxima of 1979 (Gershberg R.E. et al, Soviet Astron.J.,
1982, 59, No 1, 6-14) and 1981-1984, and the spectral type of M at the
phase of the rapid decrease of brightness and at the minimum in 1980
(Gershberg R.E. et al., Soviet Astron. J., 1982, 59, No 1, 6-14). The
circumstellar matter makes the Balmer decrement so large that only the
H alpha emission could be seen in general. The H beta emission could only
sometimes be seen. The diluted radiation from a hot star is sufficient to
excite the D1,2 emissions, Balmer emissions and the nebular lines of [N II] and
[O III]. The orbital inclination of PU Vul may possibly be small, so no
indication of orbital motion could be seen.



                   LIU ZONGLI

             Beijing Astronomical Observatory
             Academia Sinica
             Beijing, China