COMMISSION 27 OF THE I. A. U.
           INFORMATION BULLETIN ON VARIABLE STARS
                        NUMBER 617

                                       Konkoly Observatory
                                       Budapest
                                       1972 January 24


           SIX VARIABLE STARS IN SAGITTARIUS


Recent results (Table) and finder charts (Figure 1, approximately 10'x10')
are given for three new and three rediscovered variable stars in Sagittarius.

No. Name       R.A.(1900)   Dec.	   Max.   Min.  Type   J.D.   Period

1              18 07 12-25   48.4   11.5   15:    R CRB
2              18 23 10-24   37.6   14.2  [16     M           26120    320
3              18 24 02-23   19.3   14.5   15.8   M           26560    210
4  IU Sgr      18 28 58-31   22.4   13.5:  17.    M           41095    269
5  V1938 Sgr   18 41 43-27   44.0   13.8  [16.3   M           41180    254
6  V963 Sgr    18 43 25-31   40.5   11.0  [17.0   M           40480    561

Examining Nantucket plates with the Rodman blink microscope, Esther Hu found a
new R Coronae Borealis type star (Figure 2). Deepest observed minima occurred
in July 1936, July 1944 and May 1971. Notified about this discovery Dr. George
Herbig at Lick Observatory kindly reports that two low dispersion spectra
obtained by Harlan at the Crossley reflector on September 20 and 21, 1971
(during ascending light) show "an absorption-line spectrum without G-band that
is a good match for R CrB at the same dispersion."

The second and third stars in the Table, I discovered on Harvard plates. A
twelfth magnitude star is approximately one minute of arc north following the
variable. V 2565 Sgr closely follows and is generally blended with this bright
star. (Both variables are indicated in Figure 1 in the chart for Variable 2;
the charts show South at the top, so that the bright star is below Variable 2.)
These stars are in a crowded region; hence few plates besides some 50 of the
Harvard A series (24-inch Bruce refractor) were useful for magnitude estimates.
Likewise, the third star in the Table is too close to the bright star, CoD
-23deg14427 to be measurable on the Nantucket plates. The period depends upon
Harvard plates taken between 1924 and 1951.

My estimates on IU Sgr (rediscovered by Miss Hu) yield a period of 269 days.
This does not agree with the period of 382 days previously published by Innes
(Union Obs. Cir. No 37, 1917) whose estimate of the period depended upon only
15 observations. These observations are satisfied by the new period, whereas
the older period does not represent the recent observations.

                         [FIGURE 1]
                         [FIGURE 2]

From Miss Hu's estimates of V1938 Sgr on Nantucket plates, I redetermined the
period, confirming the earlier published value (Hoffleit, Astron. Jour.,
Vol. 63, p. 78, 1964).

V963 Sgr was rediscovered by Judy Karpen who estimated its brightness on over
100 Nantucket plates for 1957-71. Innes (loc.cit.) had indicated a period of
258 days based on only 17 plates. Erleksove (Ast. Cir. U.S.S.R. No. 171, p. 23,
1956) published two Julian days, one for a probable, the other for a definite
maximum. The General Catalogue gives this epoch of maximum and a period of
256.5 days. Neither of these periods represents the new observations, whereas
the period of 261 days from the Nantucket observations also satisfies the cited
published earlier observations.

We are grateful to the U.S. National Science Foundation under whose support
the main part of this work, as well as that reported in Information Bulletin
592, was carried out.


DORRIT HOFFLEIT
Maria Mitchell Observatory
Nantucket, Mass., U.S.A.