We present spectroscopic observations which reveal Lanning 90 to be a previously uncataloged cataclysmic variable. Low-resolution spectra of the object display strong and broad emission lines of hydrogen and neutral helium superimposed on a continuum with roughly constant F-lambda. In addition to the Balmer and He I emission lines, the spectrum shows high excitation lines of He II, C II, C III, and N III. After a brief discussion of its general spectral character, we argue that Lanning 90 is probably a nova-like variable. We note that the photometric and spectroscopic variability exhibited by the object is similar to the variability shown by the AM Herculis systems.
Key words: variable stars: cataclysmic variables--spectrophotometry
Lanning (1973) has surveyed a large fraction of the galactic plane searching for ultraviolet-bright (UV) sources. As pointed out by Margon and Downes (1981), the survey appears to be a efficient tool for the identification of a variety of interesting emission-line stars. For example, the classification of Lanning 10 as a cataclysmic variable (Margon and Downes 1981; Horne, Lanning, and Gomer 1982 ), as well as the recovery of the dwarf nova WZ Sagittae, has demonstrated the usefulness of the survey for identifying cataclysmics.
Recently, Lanning (in preparation) has compiled a supplement to his earlier list of UV objects. In this paper we report spectroscopic observations of a 17th magnitude star from this new list: Lanning 90. These observations reveal the object to be a previously unknown cataclysmic variable. A finding chart and coordinates are presented in Figure 1. This position does not coincide with any listed variable in the General Catalogue of Variable Stars or its supplements (Kukarkin et al. 1969).
The spectroscopic observations were obtained at the Cassegrain focus of the Shane 3-meter telescope using the Robinson-Wampler Image Dissector Scanner (Robinson and Wampler 1972). Lanning 90 was identified as a cataclysmic variable froma low-resolution (20 A FWHM) spectrum taken on 1982 September 13. In order to study the general spectral character in greater detail and to search for possible variability, a series of additional spectra were obtained the next night.
The observations were obtained with a 300 lines/mm survey grating which, in conjunction with the ``red'' image-tube chain, give ~4400 A of useful spectral coverage. Unfortunately, this image-tube chain is not sufficiently sensitive below ~3800 A to include the Balmer jump in our observations. The resulting spectral measurements have a wavelength coverage of 3800 A-8200 A. The data have been divided by the Spectrum of a quartz lamp in order to remove channel-to-channel irregularities and have been reduced to absolute fluxes by comparison with the standard star Feige 110 which has been calibrated by Stone (1977). The observations were made using 2"x4" slit. The seeing was quite good on both nights of observation (~1") and the sky was photometric. Consequently we feel that, even with the relatively small slit, our absolute fluxes should be good to ~10%.
Figure 2 shows the sum of all of our low-resolution observations from both nights of observations. The spectrum, displaying broad emission lines of hydrogen and helium, is obviously that of a cataclysmic variable. After making a correction for the instrumental profile, we find that the emission lines have a FWHM of ~1200 km/s. A list of all the prominent emission lines, along with their equivalent widths, is presented in Table I.
The line strengths presented in Table I are uncharacteristic of the spectra of most dwarf novae and old novae. For example, the high-excitation He II 4686, N III, C III 4640, 4650, and C II 4267 emissions are rarely seen with any appreciable strength in the spectra of dwarf novae. On the other hand, the spectra of old novae, which usually show these high-excitation lines, typically have emission lines with smaller equivalent widths, and continua which are somewhat more blue than the spectrum presented in Figure 2 (Robinson 1976); Webbink 1982). There are, however, a few exceptions. For example, the old novae DQ Herculis and BT Monocerotis have relatively strong emission lines superimposed on Paschen continua with roughly constant F-lambda. Consequently, we cannot rule out the possibility that Lanning 90 is an old nova solely on the basis of its spectrum.
Although the generalizations described above are not without exception, the spectrum of Lanning 90 appears to be more characteristic of a nova-like variable than of either a dwarf nova or an old nova. Furthermore, as we point out below, the photometric and spectroscopic variability exhibited by the object seems to be characteristic of the AM Herculis subclass of the nova-like systems. The possibility that Lanning 90 is an AM Herculis object may at first appear unlikely due to the relatively week strength of the He II 4686 line. However, there is at least one AM Herculis star, CW 1103+254 (Shore et al. 1962; Stockman et al. 1982), whose spectrum (including the He II strength) is reasonably similar to our spectrum of Lanning 90.
A comparison of the two charts presented in Figure 1 demonstrates that Lanning 90 is highly variable. The two charts were obtained from plates taken as part of the Palomar Sky Survey. Both plates were the blue or ``O'' plates. The plate used in creating Figure 1a was taken in July 1952 and was later rejected because of a plate flaw, while the plate used in producing Figure 1b was taken in June 1954 and was accepted as the final ``O'' plate for the region. It is difficult to make an accurate determination of the magnitude from a visual inspection of the two plates. Nevertheless, we estimate that Lanning 90 has a visual magnitude of 16-17 in Figure 1b and that in Figure 1a it is approximately 1-2 magnitudes fainter. Although Figure 1 establishes that Lanning 90 is variable, it does not provide any useful information on the time scale or nature of the variability. Fortunately, our spectroscopic observations described below are able to shed more light on these points.
Lanning 90 exhibited significant photometric and spectroscopic variability which was noted during both nights of observation. The four spectra presented in figure 3 represent consecutive 16-minute integrations from the September 14 observations. The spectral variability is clearly evident. A good example is provided by the strength of the C II 4267 line. There is no trace of this line in the first spectrum, but by the next integration it is present with an equivalent width of ~7 A. Such rapid and dramatic spectral variability, while relatively rare among the majority of cataclysmic variables, is not uncommon among the AM Herculis stars (e.g., see Liebert et al. 1978; Chiappetti, Tanzi and Treves 1980; Liebert et al. 1982).
In order to best illustrate the photometric variability, we have further divided our September 14 observations into eight consecutive 8-minute integrations. We have estimated the B and V magnitude of each spectrum by integrating the flux in the appropriate filter bandpass given in Matthews and Sandage (1963). Figure 4 shows the V magnitude, the (B-V) color index, and the equivalent widths of H-alpha and H-beta derived from the eight spectra. Although the series of observations were obtained using a relatively small 2"x4" slit, the excellent seeing we enjoyed during our run makes it likely that we were capturing a large, and, more importantly, a constant fraction of the total light. Consequently, our relative photometry (i.e., brightness variations from one 8-minute scan to the next) should be reliable.
We are aware that at sufficiently large zenith angles, the combined effects of atmospheric dispersion and a television guiding system which is primarily sensitive in the red, can lead to color dependent (differential) slit losses (e.g., see Filippenko 1982). Fortunately, differential slit losses are unlikely to pose a serious problem in our spectroscopy for the following reasons.
We realize that the spectroscopic evidence presented here is far from sufficient to argue definitively that Lanning 90 is an AM Herculis object, particularly in view of the relatively week He II 4686 emission. Nevertheless, we fee that the evidence is tempting enough to encourage future observers to look for linear and circular polarization, which is the defining characteristic of the AM Herculis systems. In any case, the high-excitation lines in the spectrum along with the dramatic photometric and spectroscopic variability which we have observed promises to make Lanning 90 and interesting object for future study.
Chiappetti, L., Tanzi, E. G., and Treves, A. 1980, Space Sci. Rev. 27, 3.
Filippenko, A. V. 1982, Pub. A.S.P. 94, 715.
Horne, K., Lanning, H. H., and Gomer, R. H. 1982, Ap. J. 252, 681.
Kukarkin, B. V. et al. 1969, General Catalogue of Variable Stars (Moscow: Sternberg State Astr. Inst.).
Lanning, H. H. 1973, Pub. A.S.P. 85, 70.
Liebert, J., Stockman, H. S., Angel, J. R. P., Woolf, N. J., Hege, K., and Margon, B. 1978, Ap. J. 225, 201.
Liebert, J., Stockman, H. S., Williams, R. E., Tapia, S., Green, R. F., Rautenkranz, D., Ferguson, D. H., and Szkody, P. 1982, Ap. J. 256, 594.
Margon, B., and Downes, R. A. 1981, A.J. 86, 747.
Matthews, T. A., and Sandage, A. R. 1963, Ap. J. 138, 30.
Middleditch, J., and Cordova, F. A. 1982, Ap. J. 255, 585.
Priedhorsky, W. C., and Krzeminski, W. 1978, Ap. J. 219, 597.
Robinson, E. L. 1976, Ann. Rev. Astr. and Ap. 14, 119.
Robinson, L. B., and Wampler, E. J. 1972, Pub. A.S.P. 84, 161.
Shore, S. N., Foltz, C. B., Wasilewski, A. J., Byard, P. L., and Wagner, R. M. 1982, Pub. A.S.P. 94, 682.
Stockman, H., Foltz, C., Tapia, S., Schmidt, G., and Grandi, S. 1982, I.A.U. Circ. No. 3686.
Stone, R. P. S. 1977, Ap. J. 218, 767.
Szkody, P. 1978, Pub. A.S.P. 90, 61.
Warner, B. 1976 in Structure and Evolution of Close Binary Systems P. Eggleton, ed. (Dordrecht: Reidel), p. 85.
Webbink, R. F. 1982, Invited review, presented at the workshop, ``Pulsations in Classical and Cataclysmic Variables,'' held at the Joint Institute for Laboratory Astrophysics, Boulder, Colorado.
Williams, G., and Hiltner, W. A. 1982, Ap. J. 252, 277.
Williams, R. E., and Ferguson, D. H. 1982, Ap. J. 257, 672.
UV Survey Intro Page