Early Papers

For the reader who likes to browse original papers, this page contains the articles in which quasar absorption lines were proposed (1965,1966 with Ed Salpter, before the absorption line were observed), in which Monte Carlo simulations were first used (1968) to establish the validity of the line identifications for multiple absorption redshifts in a single QSO spectrum, in which the hypothesis of Galactic Halos as the origin of metallic absorption line systems was suggested (1969, with Lyman Spitzer), in which quasar absorption lines were first used to constrain the temperature of the CMB at larger redshifts (1973, with Paul Joss and Roger Lynds), in which the hypothesis was advanced (1975) that absorption line systems can be resolved into sub-clouds with structure on a scale of order 100 km/s, and in which the use of the HST for quasar and galactic absorption line studies was initially described (1980). The collection also includes the suggestion that 21 cm absorption lines may be detectable in the spectra of distant sources (1969).

On The Interaction of Radiation from Distant Sources with the Intervening Medium

Authors:John N. Bahcall and E. E. Salpeter

Journal: The Astrophysical Journal, 142, No. 4, 1677--1681 (November 1965).

postscript file     pdf file

Abstract: We discuss several ways that a distant radiation source (with a large redshift assumed due to the cosmological expansion) can provide information over a wide range of distances about the intervening medium. As we shall show [cf. Gunn and Peterson (1965)] neutral hydrogen (or other atoms) at various distances between the source and us will give rise to an ``absorption trough'' in the continuous spectrum of a distant source. If the neutral hydrogen is instead concentrated in clusters of galaxies, this trough is replaced by a number of sharp absorption lines. Besides discussing (i) absorption troughs and (ii) absorption lines from clusters, we also consider (iii) photon scattering by free electrons in the intervening medium and (iv) spreading of a radio beam due to inhomogeneities in the ionized gas that is traversed. Present observations furnish some stringent upper limits, and we suggest other feasible cosmological tests.

Absorption Lines in the Spectra of Distant Sources

Authors:John N. Bahcall and E. E. Salpeter

Journal: The Astrophysical Journal, 144, No. 2, 847--852 (May 1966).

postscript file     pdf file

Abstract: We discuss the properties of absorption lines that might appear in the continuum spectra of distant quasi-stellar sources (henceforth abbreviated ``QSS'') and outline what can be learned from a search for such lines. We consider first absorption by intergalactic gas in clusters of galaxies that are in the line of sight. We extend our previous discussion (Bahcall and Salpeter 1965)¹ of absorption lines caused by neutral hydrogen to resonant absorption lines in other elements and use a variety of cosmologies to estimate some of the uncertainties. If the intergalactic gas in clusters has relative abundances comparable to those inside our Galaxy, densities above 10^-5 cm^-3, and temperatures below 5 × 10⁵° K, a number of different ion species could cause pronounced absorption lines; we present a table of the spectral lines expected to be prominent in absorption for various temperatures. The detection of more than one absorption feature in the same spectrum corresponding to the same redshift for the absorber would help to eliminate the possibility that an apparent absorption feature was really a valley between two emission peaks. In particularly favorable cases identification of the absorber with an optically detectable cluster might be possible. A number of QSS's are known with sufficiently large redshifts so that an interposed cluster might contribute several absorption features redshifted into the visible. The characteristics (such as wavelengths, depth, and width) of absorption lines formed in clusters of galaxies could provide otherwise unobtainable information about distant clusters (such as the temperatures, chemical composition, and velocity dispersion of the absorbing gas). We show in what follows that the identification of absorption lines with a common redshift appreciably different from that of the emitting QSS's would provide strong evidence for the hypothesis that the redshifts of QSS's are of cosmological origin. We also discuss absorption features which could be formed by gas clouds in the neighborhood of a QSS and which might lead (under special circumstances) to apparent variations in the observed wavelengths of some QSS emission lines. A program is under way at the California Institute of Technology (J. N. Bahcall, B. A. Peterson, and M. Schmidt) to search for absorption lines in spectra (taken by Schmidt) of QSS's.

¹Eq. (6) of our paper should contain the factor (/100 Mc/s)^-4 instead of (/100 Mc/s)⁴. Observations of the ultraviolet spectra of 3C 9 (down to 3100 Å by Wampler [1965]) and of 0106+01 (Burbidge 1966) enable one to conclude that the cosmological number density of molecular hydrogen is less than 10^-10 cm^-3 if it is uniformly distributed.

A Systematic Method for Identifying Absorption Lines as Applied to PKS 0237-23

Authors:John N. Bahcall

Journal: The Astrophysical Journal, 153, 679--688 (August 1968).

postscript file     pdf file

Abstract: A systematic method for identifying the absorption lines in quasi-stellar sources with rich absorption spectra is described, and the lines that are expected on theoretical grounds to have the highest probability of appearing in absorption in quasi-stellar sources are listed. The computational scheme that was developed for recognizing and analyzing candidate redshifts is then described, this computational scheme is supplemented by a set of rules used to determine if a candidate redshift is acceptable. The computational scheme and formal rules were applied to a set of nonsense spectra, i.e., random-number spectra, with the same main characteristics as the absorption spectrum of PKS 0237-23 determined by Greenstein and Sargent. The average number of acceptable redshifts per nonsense spectrum was 0.7. Five acceptable redshifts were found in the real spectrum: z = 2.2015, 1.671, 1.656, 1.513, and 1.364. No acceptable redshift was found at z = 1.95. A list of additional lines that should be detectable if the identifications proposed here are correct is given for slightly longer and shorter wavelengths than covered in the observations of Greenstein and Sargent. Some characteristics of the identifications are also described.

Absorption Lines Produced by Galactic Halos

Authors:John N. Bahcall and Lyman Spitzer, Jr.

Journal: The Astrophysical Journal, 156, L63-L65 (May 1969).

postscript file     pdf file

Abstract: We propose that most of the absorption lines observed in quasi-stellar sources with multiple absorption redshifts are caused by gas in extended halos of normal galaxies.

On the Temperature of the Microwave Background Radiation at a Large Redshift

Authors:John N. Bahcall, Paul C. Joss, and Roger Lynds

Journal: The Astrophysical Journal, 182, L95 (June 1973).

postscript file     pdf file

Abstract: It is shown that the temperature of the microwave background radiation at a redshift of the order of 2.5 is certainly less than 200° K and probably less than 45° K. Further detailed studies of the absorption spectra of large-redshift quasars can improve these limits.

Splittings in Quasar Absorption Lines

Authors:John N. Bahcall

Journal: The Astrophysical Journal, 200, L1-L3 (August 15, 1975).

postscript file     pdf file

Abstract: It is shown that the apparent constancy of the splitting between quasar absorption lines observed by Boksenberg and Sargent could be understood if the typical velocity in the absorbing material were somewhat less than the observing resolution (estimated to be 100 km s^-1). If this interpretation were correct, then there should be many more redshifts that are split with velocity separations that are less than 100 km s^-1 than have been so far observed at velocity separations of the order of 150 km s^-1, The above conclusions are independent of whether or not the absorption lines originate close to, or very far away from, the quasar. Some results are presented for a specific model in which the splittings are supposed due to absorption in halos or clouds surrounding galaxies or small groups of galaxies and some tests of this model are suggested. The model parameters are consistent with recent radio observations of H I clouds or halos that extend far from their parent galaxies.

Absorption Lines in the Spectra of Distant Objects

Authors:John N. Bahcall

Journal: Scientific Research with the Space Telescope, IAU Colloquium Number 54, ed. M. S. Longair and J. W. Warner (1980), pp. 215-240.

postscript file     pdf file

Abstract: This article discusses critical observations of quasar absorption lines that are made possible by the Hubble Space Telescope. The article reviews the state of quasar absorption line research circa 1980, including a prediction of the absorption lines that are most likely to be observed with HST and two tests for the origins of the absorption systems. According to the interpretation given here, many or most of the absorption lines are caused by intervening material, although references are given to contemporary claims that the absorption lines are intrinsic to the quasars. An iconoclastic proposal is made, namely, that most of the Ly-alpha lines are caused by gas in the outer reaches of galaxies (halos 200 kpc) or of clusters of galaxies (halos of 30 Mpc). Several important observing projects are suggested.

On the Possibility of Detecting Redshifted 21-cm Absorption Lines in the Spectra of Quasi-Stellar Sources

Authors:John N. Bahcall and Ronald D. Ekers

Journal: ApJ, 157, 1055 (September 1969).

pdf file

Abstract: The expected strengths of redshifted 21-cm absorption lines are estimated by using the properties of the observed optical absorption lines in quasi-stellar sources. It is shown that the absence of OI and NI absorption lines in the observed optical absorption spectra of quasi-stellar radio sources indicates that 21-cm absorption lines wider than 100 kHz are likely to be weak unless the heavy-element abundance in the absorbing material is low. Lines narrower than 100 kHz may be strong; their optical counterparts would have escaped detection. A general expression is given for the spin temperature of neutral hydrogen when (following Field) Ly α excitation and de-excitation, 21-cm absorption and emission, and particle collisions are all included. The results are expressed simply in terms of the strength and distance of the radio source and the separation between absorber and emitter. It is shown that the spin temperature of neutral hydrogen is large near a quasi-stellar radio source and is determined either by the ambient 21-cm flux or the ambient Ly α flux for separations between absorber and emitter of less than about 10⁵ lt-yr. The expected absorption (and emission) strengths of redshifted 21-cm lines are compared with observational capabilities at radio observatories. Some interesting candidates for study are listed.