COMMISSION 27 OF THE I. A. U.
         INFORMATION BULLETIN ON VARIABLE STARS
                     Number 1639

                                  Konkoly Observatory
                                  Budapest
                                  1979 July 12




             TX SCUTI - A MYSTERY SOLVED



    TX Sct was discovered by Hubble (1920) who estimated that
it varies from 12.5 to 14.5 pg. On the basis of only 9 plates he
assumed it to be a long period variable. In the summer of 1978
at the Maria Mitchell Observatory I assigned this star to  an
undergraduate research participant, Wendy Whiting (1978), to
identify and, if possible, determine a period. However, we were
unable to find any star close to the given position which varied
appreciably.
An inquiry to Dr. K. Cudworth at Yerkes Observatory led to
his resolving the mystery. He succeeded in locating the plates on
which Hubble had marked the variable. Cudworth measured its
position and found that it disagreed with the approximate position
Hubble had estimated from a BD chart. The correct position is

	(1950) 18h25m27.5s          -11d17.1'
	(1900)    22 40                 18.8

Whereas Hubble had given
	(1900) 18 23 25             -11 15.1.
The spectral class given in the GCVS is M6:. This stems
from Cameron and Nassau (1956) who were uncertain as to the
identification of the variable TX Sct. Dr. Bidelman kindly suplied 
a marked print and references, also indicating that the IR
source at 18h24m09s -11d13.8' (1900) identified as the M7 type star
BD -11d4642 (Nassau and Blanco 1954) might be the variable. The
stars Bidelman indicated, however, do not correspond with the
one Cudworth found marked on Hubble`s plate.
I have now examined Hubble's TX Sct (Fig.1) relative to
surrounding stars on 22 available Harvard plates of the MC series
(16-inch Metcalf telescope) taken between 1915 and 1974, and on
132 RB and RH plates (3-inch Ross-Fecker lenses) taken between
1935 and 1951. These observations are reasonably represented
(Fig.2) by
             Max = JD 2425830 + 24.344d n.

                      [FIGURE 1]

  Figure 1. Finder chart for TX Sct, approximate 10' x 10'.

                      [FIGURE 2]

 Figure 2.  Observations of TX Sct (on arbitrary scale) fitted to a
 period of 24.344d. Small dots for the less reliable observations.
 On some of the plates the image was partially blended with the
 star S prec by about 0.4'. Abscissa markers at intervals of 0.2
 period.

Figure 2 also includes Hubble's maximum discovery plate and non-
maximum observations on Barnard's Atlas plates 24 July and 1 August 
1905. For its period the light curve resembles a W Virginis
star. On the other hand, the variable appears red relative to its
neighbors on the Palomar sky survey, and might therefore be a
semi-regular variable of exceptionally short period. A spectral
type is needed for definitive classification.
    I am grateful to Drs. Bidelman and Cudworth for their very
helpful correspondence.


                          DORRIT HOFFLEIT
                          Yale University Observatory
                          New Haven, Conn., U.S.A.


References:
Cameron, D. and Nassau, J.J. 1956, Ap.J., 124, 346 [BIBCODE 1956ApJ...124..346C ]
Hubble, E.P. 1920, P.A.S.P., 32, 161 [BIBCODE 1920PASP...32..154. ]
Nassau, J.J. and Blanco, V.M. 1954, Ap.J., 120, 118 [BIBCODE 1954ApJ...120..118N ]
Whiting, W.A. 1978, Jour. A.A.V.S.O., 7, 70 [BIBCODE 1978JAVSO...7...70W ]