There has been only one investigation on the absorption-line spectrum of 0420-388, that of Atwood et al. (1985). Spectra were obtained using the SIT Vidicon detector attached to the echelle spectrograph at the Cassegrain focus of the Cerro Tololo Inter-American Observatory 4 m telescope in October and November 1981. The claimed resolution is 33 km/s (FWHM). Additional spectra, covering the wavelength range 3815-3950 Å were obtained using the Anglo-Australian telescope. When the same quasar has been observed by two or more groups, an intercomparison of the data helps to assess the quality of the data. We have carried out such comparisons for 0805+046 (Varshni, 1985) and 0237-233 (Varshni, 1988c). Such a comparison is, however, not possible in the present case because only one set of data are available. In a recent paper, Cristiani and Shaver (1988) show, in their Figure 1, an intermediate resolution spectrum of 0420-388; however, no wavelengths are listed.
The wavelength discrepancy that one can permit in the identifications depends on the resolution, uncertainty in the wavelength calibration and the width of the line. The claimed resolution in the observations of Atwood et al. (1985) is c/R = 33 km/s for most of the wavelength range for which data are reported. This corresponds to deltalambda=0.5 Å at 4500 Å. We do not know how good the wavelength calibration was; we shall assume that any uncertainty due to this reason can be neglected. The real problem lies with the width of the lines. An examination of an enlargement of the spectrum given in Figure 1(a-e) of Atwood et al. (1985) shows that many of the lines are wider than 1 Å and appear to be blends (inadequate dispersion?). Some of them are much wider than 1 Å, e.g., line numbers 9, 20, 23, 37, 43, etc. Line No. 174 is approximately 10 Å wide. Taking into account these various factors, a guiding value of 1 Å was taken for the tolerance, i.e., the discrepancy between the observed and identified wavelengths. When the line is very wide (blended) clearly a larger value is acceptable.
We should also note here another problem with the available spectrum. The SIT Vidicon has a more or less conventional silicon diode vidicon preceded by an electrostatic image intensifier stage. It is known that the detective quantum efficiency of silicon diode vidicon is severely reduced at low light levels by amplifier and beam noise (Walker, 1987). Thus the observed spectrum shows only strong single lines and blended medium and weak lines. Many single medium strength and most single weak lines have not been detected. Thus the spectrum is very incomplete. This incompleteness has to be borne in mind while considering the number of lines of any element which are identified as against the expected number of lines, and also as regards the completeness of multiplets in the identifications. We would also like to point out here that Figure 1(a-e) of Atwood et al. (1985) shows that there are several lines which are present in the spectrum, but for some reason they have not been recorded in their Table II. In looking for completeness of multiplets we have taken into account such lines; their wavelengths being read from the spectrum. A few of such lines are included in Table II; they do not have a line number.
As regards the relative intensities of lines, for other quasars (Varshni, 1985, 1988c) the available data indicates that the recorded equivalent widths can only be taken as a qualitative guide and this practice was also followed in the present case. Identifications for most of the absorption lines reported by Atwood et al. (1985) are presented in Table II. The average of the absolute value of the difference between the observation and identification wavelength for all the identifications is 0.69 Å which is quite satisfactory. The representation of some of the elements is as follows.
Hydrogen. There is an absorption line at 4862.85 Å. Can this line be identified with Hydrogen beta or not? It is not possible to answer this question with the limited data available. Hydrogen alpha is outside the range observed. There is no absorption line at the position of Hydrogen gamma, but Hydrogen gamma could have been weak and easily missed by the detector. There is a line at 4098.05 Å, which is quite near Hydrogen delta; it is conceivable that Hydrogen delta, if present, might be a component of this 4098.05 Å line. Clearly better data are required to settle this question. From the available data a low concentration of hydrogen cannot be ruled out. The quasar is certainly deficient in hydrogen, but we do not know to what degree (Kaufmann and Theil, 1980).
The other six quasars in Table I also show absorption lines. 0420-388 is, of course, the brightest. It is obvious that high dispersion, high resolution studies of the absorption line spectrum of this quasar (and of the other six) with instrumentation capable of detecting weak lines would be most worthwhile.