6. Individual objects
The list of stars was correlated with the catalog of Schwering & Israel (1989a). Most stars are not listed there. For those that are, the number is listed in the remarks column as SMC number . In most cases their photometry is within 20% of the fluxes derived by us using GIPSY.
The list of objects was compared to the lists of optically selected carbon stars in the SMC of Rebeirot et al. (1993) and Morgan & Hatzidimitriou (1995). Two objects are mentioned in the former publication and none in the latter one. They are RAW 644 (S12) and RAW 1543 (S20b). In the latter case there is another red object in the field (S20a). Although the IRAS source is located closer to S20b, plotting both SEDs suggests that the IRAS colours are related to S20a. This is confirmed by our results from ISOCAM imaging where we find a source at 12 micron of comparable flux to the IRAS value at the position of S20a. At the position of S20b only a weak source of order of a few mJy is found in the ISOCAM image.
The counterpart indicated on the finding chart for RAW 644 in Rebeirot et al. is identical to the counterpart identified by us on the NIR images based on the red colours. RAW 644 belongs to the reddest about 20% of stars in their sample. According to the transformation given in RAW the star has a colour of 2.3.
The counterpart indicated on the finding chart for RAW 1543 is identical with the counterpart S20b identified by us based on the red colours. RAW 1543 has colours which are average for the carbon stars in RAW's sample. According to the transformation given in RAW the star has a colour of 1.9. For both S12 and S20b we confirm the carbon star character from the optical spectra.
The object S2 is interpreted as being carbon-rich according to WFMC. Our photometry is in good agreement with that found by WFMC who also monitored the object and found only minor variations in K. Our data also agrees with the observations in Z96 and van Loon et al. (1997). WFMC present a medium resolution 4000 to 7000 Å spectrum which can be compared to our low resolution spectrum. The major difference with respect to their spectrum is the absence of H . H is also much weaker in our spectrum then in theirs (3 continuum), but this is at least partly due to the lower resolution of our spectrum. Contrary to them we find (weak) CN lines in the spectrum. The overall shape of the SED suggests a temperature similar to that of a late K-giant. The spectral type should be close to C3,2. The new data does not provide any further insight into the evolutionary status of the star.
The object S9 was also monitored by WFMC. It is variable on a long time scale but no period could be derived by them. They derive a high luminosity for this object () indicating it is a massive AGB star or a supergiant. Our photometry is somewhat fainter than the faintest dataset in their paper. The K -amplitude is of order 0.6 magnitude (WFMC). Our spectrum confirms it to be a late-type M-star.
The object S16 was also monitored by WFMC who found it to be a luminous long-period variable with a K -amplitude of about 1.1 magnitude and a period of about 800 days. Our photometry is in good agreement with that in WFMC.
Our sample also includes the known oxygen-rich stars HV 11417 (S18, spectral type M5e, P = 1300 days, see Elias et al. 1980, 1985), HV 12956 (S22, enhanced in lithium, P = 518 days; see Smith et al. 1995, WBF, WFMC, Z96, van Loon et al. 1997) and GM 103 (S10), in which Groenewegen et al. (1995) detected the silicate feature in emission.
The spectrum of S14 is peculiar (see Fig. 1). It is unmistakably a carbon star because of the presence of the 5165 Å C2 Swan band and CN band, but these features are very weak. More unusual is the upturn in flux at shorter wavelengths and the emission lines at approximately 5540 Å and 6345 Å This suggests that there may be a binary companion. At the present resolution both lines can not be assigned.
The spectrum of S30 is unusual (see Fig. 1). It shows strong H and is very red suggesting a late-spectral type. The blue region is not well exposed and the C2 bands are not seen. The spectrum is classified as carbon-rich because of the CN bands longward of 9000 Å.
For three objects (S13, S21 and S29) there appears to be no red star in the field. The IRAS coordinates of S21 put this objects close to the edge of the cluster NGC 419, which was clearly visible in our NIR images. It is possible that the IRAS emission is not associated with a single point source. For S13 we have obtained ISOCAM images at different wavelengths indicating an extremely red source close to the IRAS position. Preliminary modelling indicates that it is fainter than 20th magnitude in K, consistent with the non-detection in the ground-based K -images.
According to information in SIMBAD, the radial velocities of S19 and S31 are such that they are foreground objects of spectral type G8/K0 III (radial velocity of +10 km s-1) and K1/K2 III (radial velocity of -28 km s-1), respectively.
© European Southern Observatory (ESO) 1998
Online publication: March 10, 1998