COMMISSION 27 OF THE I. A. U.
            INFORMATION BULLETIN ON VARIABLE STARS
                           NUMBER 41


                                          Konkoly Observatory
                                          Budapest
                                          12 February 1964


                       NOTE ON Z CIRCINI


      While the first edition of Kukarkin and Parenago's "General
Catalogue of Variable Stars" listed Z Cir as a probable eclipsing
binary, the second edition mentions the eclipsing and the semiregular
types of variability as equally possible.
      An investigation, just carried out at Louvain, proves that
the star cannot be an eclipsing variable, but that it is almost certainly
an R Coronae Borealis star with a strong tendency towards periodicity
and with a short characteristic interval between the minima. Both
properties make it a really interesting object for observers in the
southern hemisphere.
      The brightness of Z Cir was estimated on 326 Metcalf plates,
taken by the writer and by several other observers at the Boyden

                          [FIGURE 1]

Observatory between 1960,6 and 1963,8. The results of the estimates
are shown in the Figure. Magnitudes were derived from star-counts
and the tables in Groningen Publication No. 43.
     As can be seen, the estimates confirm the existence of the
long-living flat maximum which had been noticed earlier by the
Harvard observers and had led them to presume variability of the
Algol type.+/
      The beginning of the descending branch could be well located
on three occasions, namely on JD 2437409, 2437829 and 2438197. The
interval between the two outer data covers two unequal cycles of 394
days mean length. If this trial period is corrected to 394,6 days, it
accounts fairly well for our own minima and for all those communicated
in Harvard Bulletin No. 883+/.
      But this characteristic interval leaves no possibility that the
star could be an eclipsing binary because:
      1/ either 394d.6 is the orbital period (the observations require
an eccentric orbit), but then we are forced to the conclusion that the
eclipse lasts for about half a period, which is absurd. The descending
branch is indeed seen to last for more than 90 days and the never
observed ascending one should have approximately the same duration
in case of an eclipse, hence the foregoing conclusion;
      2/ or 394d.6 is only half the orbital period. But this assumption 
is excluded at once by the fact that both the odd and the even
minima have amplitudes greater than 2m.5.
      Hence our general conclusion: it appears most probable that
Z Cir is a variable of the R CrB type, and a very interesting one,
because of the trend to periodicity in the occurrence of the minima and
of the rather high frequency (1 in 13 months) of the latter.

                                   A. VAN HOOF
                              Astronomical Institute
                               University of Louvain
                                     (Belgium)

     +\ H. H. Swope, Harvard Bulletin No. 883, 1931. [BIBCODE 1931BHarO.883...23S ]