From: eric@flesch.org (Eric Flesch) Subject: Re: quasars and elliptical galaxies Date: 1998/01/25 Message-ID: <34d19938.12453562@news.nn.iconz.co.nz> Content-Transfer-Encoding: 7bit References: Content-Type: text/plain; charset=us-ascii Organization: Internet Company of New Zealand Mime-Version: 1.0 Newsgroups: sci.physics,sci.astro On 24 Jan 98 21:42:10 GMT, Stupendous Man wrote: > Eric Flesch wrote: >> ... A more germane point is that >> a quasar has *never* been observed to be resident inside an elliptical >> galaxy. Go ahead, prove me wrong. Quote a paper. > > Okay, I'll quote three papers. > > Direct evidence for 12 quasars which occur in elliptical galaxies: > 1. "Hubble Space Telescope Images of a Sample of 20 Nearby > Luminous Quasars," Bahcall et al., ApJ, 479, 642, 1997. > From the Abstract: > "Observations with the Wide-Field Camera of the Hubble > Space Telescope (HST) are presented for a representative > sample of 20 intrinsically luminous quasars with redshifts > smaller than 0.30. These observations show that luminous > quasars occur in diverse environments that include > ellipticals as bright as the brightest cluster galaxies (two), > apparently normal ellipticals (10), Well spotted, Stup, however you're better at reading abstracts than the papers themselves. I have this paper before me and this is what it says: Bahcall does claim the ellipticals and includes comprehensive HST photos for all the QSOs. Some "host" galaxies look like pretty good spirals. The trouble with the ellipticals is they are quite hard to see -- it looks like diffuse emission from the QSO itself. If you start out with the assumption that the QSOs are at their cosmological-redshift distances, then you could deduce the ellipticals. If not, these are just QSO fuzz. Bahcall discusses this problem (the actual-vs-imagined detection of the host elliptical galaxy) on pages 35-38 in the section titled "10.2 Host Luminosities". He speaks of a previous paper by McLeod and Rieke (1995) in which a dependency was found between luminosities of the QSO and "host" elliptical galaxies. This paper has been used as a basis for Bahcall's confidence in his deduced "elliptical" hosts and yet he qualifies the points made by that paper: "McLeod and Rieke (1995) have suggested that there is a linear relation between the quasar absolute magnitude and the minimum host galaxy magnitude. They interpret this linear relation as indicating that a more luminous host galaxy is required to fuel a more luminous quasar. The linear relation that they find between M(host) and M(QSO) is essentially identical to our minimum detection limit for smooth ellipticals..." Observe Bahcall is saying that McLeod and Rieke have claimed the presence of galaxies *at Bahcall's resolution limit*, thus emboldening Bahcall should he choose to declare fuzz to be galaxies. Bahcall continues: "Galaxies that are smooth ellipticals are the most difficult to detect. For the 8 quasars discussed in paper II, smooth elliptical hosts are, on average, visible on our images down to 3.5+/-0.5 mag fainter than the quasar. The limiting brightnesses were determined by visually inspecting simulated galaxies placed in the actual HST quasar observations and are therefore somewhat subjective." Note what Bahcall says he has done. He couldn't decide if the fuzz around the QSO was an elliptical galaxy, so he whipped up a simulated picture and then compared the real shot with the simulation. If they look enough alike, Bahcall goes ahead and annoints the real thing as an "elliptical". He admits this is "somewhat subjective". He's right. As a matter of fact, it's *goddamn* subjective !! Bahcall continues: "As pointed out by McLeod, the relation described by McLeod and Rieke cannot be an artifact produced by detection limits if all their detections are real detections. An artificial correlation would be introduced only if true non-detections were interpreted as marginal detections. " Re-read that. What is says is "If we're right, we're right. And if we're wrong, we're wrong". Nasty. This is the bottom line criterion for the establishment of elliptical host galaxies for QSOs. Bahcall is preaching to the converted here. Those who say he is right will proclaim the ellipticals. Those who seek more rigorous standards will not. I am not swayed by this "science". However, I admit that three of the quasars, PHL909, HE1029-140, and 3C273 look like they are sited inside ellipticals. So for these three I say a "definite maybe", and so well done Stupendous Man, you have proved your point. But the definite establishment of elliptical galaxies as QSO hosts is *no way* proved for those who do not start out with the premise that the QSOs are located at their cosmologically-set redshifts. As an aside, Bahcall proceeds to take a swipe at McLeod & Rieke, whose work he has partly premised his own on, by observing "There is not convincing evidence for a significant dependence of host luminosity upon the luminosity of the quasar". So first Bahcall uses McLeod & Rieke's luminosity relationship as a partial justification of his own work, and then he disavows their luminosity relationship! Nasty. > 2. "The unusual near-infrared morphology of the radio-loud > quasar 4C+09.17", Armus et al., MNRAS 289, 621 (1997) This paper only speculates. "Diffuse emission" is observed around the QSO. In the conclusion Armus writes, "If this emission is starlight at the redshift of 4C+09.17, an elliptical host galaxy with a luminosity larger than about 40L is implied." The last mentioned paper by Stockton & Ridgeway does not seek to establish elliptical hosts for the 3C QSOs. Eric Flesch