2 Mercury-Manganese stars

Adelman (1993) and Adelman et al. (1994) showed that several Mercury-Manganese (HgMn) stars, including the most often cited as variable, were non-variable both by examining published and Four College Automated Photoelectric Telescope filter photometry. A list of HgMn stars (Table 1) was compiled mainly from Aikman (1976), Wolff & Preston (1978), the 5 th edition of the Yale Bright Star Catalog (Hoffleit & Warren 1991), and Renson (1991). Some stars which were not true class members were removed by consulting the literature especially Abt & Morrell (1995) and Garrison & Gray (1994). To classify a star as a HgMn star usually requires a photographic region spectrogram with a resolution greater than that typically used for classification purposes and often with a resolution similar to that needed for an abundance analysis. The paucity of known HgMn stars fainter than about 7 th magnitude is due to this cause. There is also the unresolved problem of whether the HgMn stars have weak magnetic fields.

Figure 1 shows the relation between standard error and mean magnitude for the HgMn stars. Most HgMn stars are constant according to the Hipparcos Catalogue. Only four were found to have periods. Three are in the region defined by the variable normal stars: AR Aur (= HR 1728), a well known eclipsing binary, HR 7911, and 13 And (= HR 8913). The Hipparcos photometry light curves for AR Aur and 13 And well fit their respective photometry, but that for HR 7911 could be improved. According to Hoffleit (1982), HR 7911 is a close binary star and 13 And is a spectroscopic binary. HR 6023 (= $\phi$ Her) with a small amplitude is noted as variable. Its range of magnitudes is 4.219 to 4.226 mag and its standard error is 0.0004 mag. These values strongly suggest that this star is constant. Thus both HR 7911 and 13 And deserve additional study.

If we use as a guide that other possible HgMn star variables should have distances from the constancy line at least as great as the three variable HgMn stars found by the Hipparcos team, then additional candidates are: HR 149, HR 8512, HD 2019, HD 65949, and HD 65950.

Catalano & Leone (1991) found HR 149 (= HD 3322) to be a variable star with a period of 4.6904 days while Aikman (1976) discovered it to be a spectroscopic binary with a period of about 400 days. On the other hand Winzer (1974) noted it was constant to better than 0.01 mag during a 14 day observing run. Its Hipparcos photometry indicates a scatter with an amplitude of about 0.03 mag.

Stickland & Weatherby (1984) derived a binary orbit for HR 8512. Abt & Morgan (1969) classified both HD 65949 and HD 65950 as HgMn stars. Abt & Levy (1972) found the former was a SB1 star while North (1984) used the later as a comparison star for differential photometry.

Figure 2 shows the relation between standard error and amplitude for the HgMn stars. The discrepant position of AR Aurigae was previously noted.