The spectrum

The spectrum of RR Cae is reproduced in Fig. 5. While spectra in the blue-yellow region have been published by [Wegner (1975)] and [Bragaglia et al. (1988)], a red spectrum is presented here for the first time. It shows molecular signatures of a late type dwarf and the Stark broadened H profile of the white dwarf embedding a narrow emission component.

 
Figure 5: The red spectra of RR Cae and the M5 V standard star Yale1955. The latter has been downscaled by a factor of 400.

 

In order to attempt a spectral classification of the secondary spectrum we compared our data with templates found in the spectral atlasses of [Jacoby et al. (1984)] and [Torres-Dodgen & Weaver (1993)]. The RR Cae spectrum was initially convoluted with a Gaussian to match the template resolution (4.5 Å). The relative strengths of the TiO and CaH bands were then used to find the best agreement at spectral type M5 V. In Fig. 5 the spectrum of the M5 V standard star Yale1755 is plotted beneath the spectrum of RR Cae in order to enable a comparison. Note that the spectral resolution of the standard star is lower than that of RR Cae. Alternatively the calibration of TiO indices of [Reid et al. (1995)] was used to classify the secondary of RR Cae. Taking the veiling by the hot component into account (see Sect. 6.2) this leads to a spectral class M6 V. It is difficult to assess the uncertainty of this classification, being at least one spectral subtype.

The narrow emission line in the core of H seen in the present case as well as in other pre-cataclysmic binaries is normally thought to arise on the secondary star. Note, however, that the illumination effect which is generally considered to be responsible for the emission is much weaker in RR Cae than in other pre-CVs (see e.g. BE UMa, [Ferguson & James 1994]) due to the very low temperature of the white dwarf ([Bragaglia et al. (1995)]). Of course, the line could also be intrinsic to the secondary; very late main sequence stars often exhibit emission in the Balmer lines. Spectroscopic observations around the orbit would easily permit to distinguish between the alternatives since due to the hight orbital inclination of RR Cae any line which is due to illumination should be strongly variable with phase.