COMMISSIONS 27 AND 42 OF THE IAU
             INFORMATION BULLETIN ON VARIABLE STARS
                         Number 3970
 
                                               Konkoly Observatory
                                               Budapest
                                               4 January 1994

                                               HU ISSN 0374 - 0676


                NEW TIMES OF LIGHT MAXIMA FOR CY Aqr


 CY Aqr is an SX Phe type variable whose magnitude in V band varies from 10.42m to
11.16m. Its spectral type varies from A2 to A8. The star was one of the most extensively
observed SX Phe type stars which had detectable period variations in the last sixty years.
There are many different interpretations for its variations (Kamper, 1985, etc.). In order
to know the real situation, we observed it on September 28, 1984 and got a new moment
of light maximum. After that during October 19 to 22, 1993 we observed the star again
and obtained eight new times of light maxima. The data showed an obvious systematic
delay with respect to the times calculated from the formula given by Pena et al. (1987).
To check the reliability of the differences we performed new observations on December
7, 1993 which showed that these new times of light maxima were reliable. The ten new
times of light maxima for CY Aqr are as follows:

No    T_max
      (HJD 2400000.0+)
1     45972.1828
2     49280.0396
3     49281.0776
4     49281.1390
5     49282.0540
6     49282.1150
7     49282.1760
8     49283.0313
9     49283.0915
10    49328.9938

At first we used the times of light maxima got before 1993 including those published
in the literature (Zissel, 1991; Rodriguez et al., 1990, etc.) and the first new time listed
above to obtain a linear fit in the form of:
T_max=T_01+P_01*E                                                   (1)
and a quadratic fit as:
T_max=T_02+P_02* E+0.5beta * E^2                                    (2)
We got T_01, P_01, T_02, P_02 and beta of the variable star as below:

-------------------------------------------------------------------------------
star   T_01(HJD)    P_01(days)  T_02(HJD)    P_02(days)    beta(day/cycle)
-------------------------------------------------------------------------------
CY Aqr 2434308.4299 0.061038330 2434308.4280 0.061038410   -3.72 * 10^-13
                +-2         +-1          +-1         +-1   +-6.5*10^-15                                                       

                               [FIGURE 1]

 Figure 1. The O-C diagram of CY Aqr derived from the linear ephemeris (1) with 
 data before 1993.

                               [FIGURE 2]

 Figure 2. The O-C diagram of CY Aqr derived from the quadratic ephemeris (2) with 
 data before 1993.

                               [FIGURE 3]

 Figure 3. The O-C diagram of CY Aqr derived from the linear ephemeris (1) with 
 the whole data.

  Figure 1 shows the O-C diagram obtained by using Eq. (1). Figure 2 shows the O-C
diagram as obtained from the quadratic fit. The negative rate of period change, beta, means
that the period of light variations was decreasing. From Figure 2 we can suspect roughly
a quasi-periodic variation.
  Then our all new times of light maxima were combined with the old ones to get a
whole data set. The results of the linear fit are
  T_0=2434308.4296+-0.0002,
  P_0=0.061038333+-0.000000001 days. Figure 3 shows the O-C diagram using linear fit
with the whole data. From Figure 3 we can find a new variation.
  There are two possible interpretations for the new variation. One of them is a light-
time effect caused by the orbital motion of CY Aqr around the mass center of a binary
system with an unseen companion. The orbital period would be about 50 years. The
other explanation that the rate of period change had a jump from a negative value to
a positive value. If we cut the whole data at about 1970 (E~10^5) to obtain two parts,
their rates of period changes would be about -5.4 * 10^-13 day/cycle for the first part
and 1.05 * 10^-12 day/cycle for the second part according to the fits with parabolic curves.
Because our new times of light maxima cover rather short time span, any kind of further
fit cannot provide more reliable information about the real period variation. More regular
observations are needed to check which one is real.


                          FU JIAN-NING ^1.2
                          JIANG SHI-YANG ^1
                          LIU YAN-YING ^1

                          ^1 Beijing Astronomical Observatory,
                          Chinese Academy of Sciences,
                          Beijing, 100080, China
                          ^2 Department of Physics,
                          Nanchang University,
                          Nanchang, 330047, China

  Supported by the Chinese Natural Science Foundation.



References:

Kamper, B. C., 1985, IBVS, No. 2802
Pena, J. H. et al., 1987, Rev. Mexicana Astron. Astrof., 14, 429 [BIBCODE 1987RMxAA..14..429P ]
Rodriguez, E. et al., 1990, IBVS, No. 3428
Zissell, R. E., 1991, J. Am. Assoc. Variable Star Obs., 20, No. 2, 251 [BIBCODE 1991JAVSO..20..251Z ]