COMMISSION 27 OF THE I.A.U.
              INFORMATION BULLETIN ON VARIABLE STARS
                           Number 1264

                                                    Konkoly Observatory
                                                    Budapest
                                                    1977 April 13


       THE ACCELERATION OF THE ROTATION OF DQ Her

In addition to irregular intensity variations on timescales
of minutes, DQ Her shows fluctuations of its 71-s-phase from
night to night and from month to month which make it difficult
to count the number of elapsed periods if the observing seasons
are shorter than the gaps between them. The published data as
well as observations made at the 60-cm-refractor and (1977 only)
the 1-m-reflector at Hamburg-Bergedorf lack the continuity which
would be necessary to evaluate the basic properties of this longterm 
phase noise, they are, however, sufficient to establish a
consistent period count from 1967 to 1977.
   Fig. 1 shows the residuals            .                  ..
    O-C= phi_0 + (t-t_0)_nu0 + (t-t_0)^2*nu_0/2 + (t-t_0)^3*nu_0/6 - N,
where t is the heliocentric ephemeris time of the light maximum,
                                              .     ..
phi_0 the phase constant, nu_0 the frequency, nu_0 (nu_0) its first (second) 
time derivative at t=t_0, and N the integer number of elapsed
periods. The elements used for plotting the observations are
given in Table 1, column "initial".
   Three least-squares fits to the residuals are shown in
Fig. 1 and Table 1:
    "Q 67-77" (short dashes): second order polynomial for
1967-1977,
    "C 56-77" (long dashes) : third order polyn. for 1956-1977,
    "Q 56-77" (dash-dots) : second order polyn. for 1956-1977.
Weight 1.0 has been assigned to observations of about two hours,
weight 3.0 to observations of six hours or more. For both second
                         ..
order solutions adding a nu-term did not reduce the residuals
significantly. Both the smaller standard deviation of C 56-77 and
the agreement with Q 67-77 within the internal error limits 
indicate that C 56-77 should be preferred to Q 56-77, but we 
cannot rule out Q 56-77 completely. Only another five years of 
continued observation or data between 1959 and 1967 can remove the
ambiguity.             .    ..
   The short timescale nu_0/nu_0 ~-50 years of C 56-77 compares
well with the time since the nova outburst (1934-1969=-35y) and
suggests that the acceleration of the rotation is a shortlived
phenomenon associated with this outburst.

                   Table 1: Elements

t_0=2440322.87022 JED_hel (Herbst, Hesser, and Ostriker 1974, ApJ 193, 679)

             initial    Q 67-77      C 56-77      Q 56-77

phi_0          0.0     -0.839+-11   -0.849+-11   -0.829+-15   (periods)
v_0         1215.779     .779039      .779068      .778909    (d^-1)
.                           +-20         +-11          +-7
nu_0/2         0.500    0.445+-10    0.453+-3     0.505+-2    (10^-6d^-2)
..
nu_0/6         0        0           -8.3+-1.1     0           (10^-12d^-3)
standard                0.049        0.053        0.069       (periods)
deviation               3.5          3.7          4.9         (s)


                               [FIGURE 1]


                                             ECKMAR LOHSEN

                                         Hamburger Sternwarte
                                          D-2050 Hamburg 80