5 Comparison with other authors; spectroscopic variability

The most extensive work on this star was performed by Brosch et al. (1978) who identified the following elements: H, He I, O I, [O I], Mg II, Si II, Fe II and [Fe II]. They remark that few [Fe II] lines are present and that except for [O I] no other forbiden lines are seen. Another point is that they were unable to see absorption lines. Although part of this may be attributed to their low resolving power, the absence of Ca II in absorption is striking. Furthermore Kuhi in 1976 quoted a number of higher Balmer lines in absorption (Brosch et al. 1978), which in Brosch appear in emission. Since on our spectra we are limited to $\lambda \gt 4095$ Å, we see only H$\delta$, H$\gamma$, H$\beta$ and H$\alpha$ in emission, so that our spectra agree with the description by Brosch et al. (1978). A further point of agreement with their description is that we see no other forbidden emissions except those of [O I] and [Fe II]. The difference is that we see many lines of the latter species, although none has an equivalent larger than 0.80.

We may also compare with the verbal description by Swings & Struve (1943). They found double emissions in the Balmer series, with $R\gg V$ and deep absorption cores, characteristic of shells. These features extend at least up to n=13. They observe also strong permitted lines of Fe II, all of which are somewhat diffuse but show no duplicity; the strongest Fe II lines have P Cyg type profiles. Mg II is weakly in emission, Si II 4128-30 absent and Ca II shows a relatively weak interstellar absorption line. Several moderately strong emission lines of [Fe II] are also present.

From the descriptions, we see that changes in the spectrum of the star have taken place. Due to the lack of coverage of the star, we are however unable to establish when these changes took place.