Partly Based on Observations with the Eso 2.2 M Telescope at La Silla/chile (mpi Time)

1 The discovery of the supersoft X-ray source RX J0527.8–6954 during the ROSAT first light observation (Trümper et al. 1991) of the Large Magellanic Cloud (LMC) in June 1990 has directed some attention to the optical variable HV 2554 because its location is within the X-ray error circle of RX J0527.8–6954 (Trümper et al. 1991, Greiner et al. 1991). Later ROSAT observations improved the X-ray position resulting in a larger offset to HV 2554 (Cowley et al. 1993, Greiner et al. 1996a,b). However, it also became clear that there are no exact coordinates available (no SIMBAD entry) for HV 2554. To our knowledge the only finding chart available for HV 2554 is the Large Magellanic Cloud atlas by Hodge & Wright (1967) (see Fig. 1), but unfortunately the scale is too poor and the variable itself invisible. While only a summary of the variability of HV 2554 is published in form of table entries in Shapley & Mohr (1940) (based on the investigation of only 12 plates) and Shapley & McKibben Nail (1955), the detailed notes of the Gaposchkins (C.H. Payne-Gaposchkin and S. Gaposchkin) on the brightness estimates of HV 2554 on 380 plates (taken between 1896 and 1954) of mainly the A series are unpublished. Given these facts we went back to the original plates and re-identified HV 2554. From the unpublished individual brightness estimates (recently archived by D.L. Welch and electronically available on http://www.physics.mcmaster.ca/HCO/) we selected four plates: two with HV 2554 being brightest and two plates with it being in a faint state. A comparison of the brightest/faintest plate pairs quickly revealed a clearly variable object with an amplitude consistent with the value of ∆m∼1.6 mag (Shapley & McKibben Nail 1955). Our independent relative brightness estimates on nearly 30 further plates are in good agreement to those of the Gaposchkins and thus confirm the correctness of our re-identification. The astrometry on plates showing HV 2554 in the bright state is over-plotted on a CCD frame taken in March 1995 (small circle on Fig. 2) and demonstrates that its position is within the 5 ′′ X-ray error circle of RX J0527.8–6954. These findings provided the motivation to determine the pattern of optical variability of HV 2554 over the last six years during which RX J0527.8–6954 was found to gradually decline in X-ray intensity (Greiner et al. 1996a,b). For this purpose, we investigated about 140 blue plates out of …