Papex

Open problem · Astronomy · Open

Origin of the M–sigma relation

Why does the mass of a galaxy’s central black hole track the speeds of the stars in its bulge?

Summary

The mass of a supermassive black hole correlates tightly with the velocity dispersion of stars in its host galaxy’s bulge; the tightness suggests some form of feedback, but which process creates and maintains the link is unresolved.

Why it is open

In 2000 two groups reported that the masses of supermassive black holes correlate with the velocity dispersion σ (sigma) of the stars in the bulges of their host galaxies, the relation now called M–σ [1, 2]. Laura Ferrarese and David Merritt found it "much tighter" than the relation with bulge luminosity, with a scatter "no larger than expected on the basis of measurement error alone", and concluded that it implies "a strong link between black hole formation and the properties of the stellar bulge" [1]. Karl Gebhardt and colleagues wrote that black hole mass "is constrained by and closely related to properties of the host galaxy's bulge" [2].

What makes the link is disputed. Reviewing the field in 2013, John Kormendy and Luis Ho wrote that the correlations "led to the belief that BHs and bulges coevolve by regulating each other's growth", and that newer results were replacing "this simple story" with one in which black holes correlate differently with different parts of a galaxy [9]. One family of explanations has energy or momentum from the accreting black hole regulate both its own growth and its host's star formation, a process called feedback [3, 4, 7]. Another argues that tight black hole to galaxy relations can arise without any physical coupling, from the statistics of hierarchical galaxy merging [5, 6].

No body sets a bar for a resolution. The sources point to what an explanation must account for: the slope, normalisation and scatter of the relation [8, 9]; why classical bulges and elliptical galaxies follow it tightly while disk galaxies do not [9]; and the black holes now found by the James Webb Space Telescope (JWST), which lie above the local relation with stellar mass but close to the local M–σ relation [10, 12].

What is known

Discovery (2000). Ferrarese and Merritt found M ∝ σ^α with α = 4.8 ± 0.5, and that black hole masses estimated by John Magorrian and colleagues in 1998 lie systematically above the relation, "some by as much as two orders of magnitude" [1]. Gebhardt and colleagues, using 26 galaxies including 13 with new Hubble Space Telescope masses from stellar motions, found M = 1.2 (± 0.2) × 10⁸ solar masses × (σ/200 km/s)^3.75 (± 0.3) over almost three orders of magnitude in mass, with a scatter at fixed σ of only 0.30 dex (a logarithmic measure; 1 dex is a factor of ten), most of it due to observational errors [2].

Slope and scatter (2013). With 72 black holes, Nicholas McConnell and Chung-Pei Ma found log(M/solar masses) = 8.32 + 5.64 log(σ/200 km/s) for their full sample; fitted separately, early-type and late-type galaxies gave similar slopes (5.20 and 5.06), but black holes about twice as massive in early types at a given σ [8]. Kormendy and Ho, omitting pseudobulges (bulges built slowly within disks rather than by mergers) and a few galaxies, among them the ellipticals NGC 3842 and NGC 4889 with the most massive black holes, adopted a slope of 4.38 ± 0.29 and an intrinsic scatter of 0.29 ± 0.03 dex [9]. They attribute differences between published slopes to sample choices: including pseudobulges or those two black holes steepens the relation [9].

Which galaxies follow it. Kormendy and Ho conclude that black hole masses correlate tightly only with classical bulges and ellipticals; correlations with pseudobulges and dark-matter halos are weak enough "to imply no relationship closer than the fact that it is easier to grow bigger BHs in bigger galaxies that contain more fuel", and black holes "do not correlate with disk properties at all" [9]. Zehao Jin and colleagues cite an intrinsic M–σ scatter of 0.31 dex for elliptical galaxies and 0.67 dex for spirals from a 2019 study [11].

Before JWST. For quasars at redshifts of about 2 to 4, Kormendy and Ho found evidence that the black holes were already tied to the gravitational potential wells of their hosts, although the zero point of the correlation may be a factor of 2 to 3 higher and the scatter larger than in the nearby universe [9].

JWST (2024–2026). In 2024 Roberto Maiolino and colleagues reported 12 new active galactic nuclei (AGN, accreting black holes) at redshifts 4 to 7 in the JADES survey; compared with the local relation between black hole mass and stellar mass, the black holes are "over-massive", but they "tend to be more consistent with the local relation between M_BH and velocity dispersion", which the authors suggest is more fundamental [10]. A 2026 census of 34 broad-line AGN at redshifts 1.5 to 9 by Ignas Juodžbalis and colleagues found the same pattern [12]. In that census σ is not measured from starlight: it is inferred from the widths of ionised-gas emission lines, using an average gas-to-stellar scaling factor of 1.3 [12]. Their early growth and disputed masses are covered at Early supermassive black holes and Little red dots.

Recent claims. Jin and colleagues applied causal discovery, a statistical method that infers likely causal structures from observational data, to 101 galaxies with dynamically measured black hole masses, and reported in The Astrophysical Journal in January 2025 "a probable causal reversal": in elliptical galaxies bulge properties tend to influence black hole growth, while in spirals black holes appear to affect their hosts [11]. In March 2026, in the same journal, Benjamin Davis and colleagues found the reversal in simulated galaxies too, with the causal direction inverted between the epochs before and after the peak of star formation; causally informed relations, they write, "predict high-z black hole masses that are approximately two orders of magnitude less massive, and thus not overmassive" relative to local black hole–galaxy mass ratios [13]. No published response was found by 16 October 2026.

Main approaches

Feedback from the black hole. Once the black hole is massive enough, its outflow clears the gas that feeds it and forms stars [3, 4, 7]. Andrew King, revisiting ideas of Joseph Silk and Martin Rees, modelled the outflow as a momentum-driven bubble and obtained a mass proportional to σ⁴, about 1.5 × 10⁸ solar masses at σ = 200 km/s, "without free parameters" [3]. Strongest evidence: this prediction is "remarkably close to the observed relation in both slope and normalization" [3]; galaxy-merger simulations in which the quasar's energy expels the gas account for the relation [4]; and the radio mode of feedback is "well observed directly" as bubbles in the hot gas of galaxy clusters [7]. Main difficulty: direct evidence for the radiative or quasar mode "has been difficult to obtain but is accumulating from a few exceptional objects" [7]; of quasar-mode feedback at high redshift, Kormendy and Ho write that "the details, however, are murky" [9].

Merger averaging. The central-limit tendency of many mergers can produce a tight relation even from random initial masses [5, 6]. Strongest evidence: in merger-tree models with no coupling between black hole and stellar growth in any galaxy, Knud Jahnke and Andrea Macciò reproduce the observed present-day slope of about 1.1 for the black hole to bulge mass relation [6]. Main difficulty: the models need "some physics for the absolute normalisation" [6] and treat black hole and bulge mass rather than σ [5, 6]; Kormendy and Ho argue that the relation already present among quasars at high redshift "could not have been established entirely through merger averaging" [9].

Different physics in different galaxies. Kormendy and Ho propose four regimes: small black holes in bulgeless galaxies, with too little energy for coevolution; quasar feedback in gas-rich major mergers; maintenance feedback in the most massive ellipticals; and merger averaging, which tightens the correlations independently of feedback [9]. Strongest evidence: the tight correlation for classical bulges and ellipticals against weak ones for pseudobulges and disks [9], and the morphology-dependent causal structures [11]. Main difficulty: the causal analysis assumes linear, Gaussian data and no cyclic dependencies, and its sample, limited to the closest or most massive galaxies, carries an "unavoidable selection bias" [11].

Black holes first. Early black holes may sit on the M–σ relation before their stars have formed [10, 12, 14]. Strongest evidence: the JWST results above [10, 12]; in the AESOPICA simulations, broad-line-selected AGN "lie on the local, weakly evolving" relation, "supporting a scenario where black holes assemble before the stellar component is fully established" (a July 2026 preprint, not peer reviewed) [14]. Main difficulty: the high-redshift σ values come from gas lines scaled to stars [12], and Davis and colleagues argue that the causal direction of M–σ changes over cosmic time [13].

Open sub-questions

  1. Stellar σ at high redshift. Do absorption-line stellar velocity dispersions of JWST broad-line AGN hosts at redshifts above 4 agree with the gas-based values, scaled by 1.3, that place them on the local relation [10, 12]?
  2. The slope. Can the spread between published slopes, 4.38 ± 0.29 [9] and 5.64 [8], be traced entirely to sample selection, and is either consistent with King's σ⁴ [3] on the same galaxies?
  3. Scatter by morphology. Do merger-tree models like those of Jahnke and Macciò [6] reproduce the larger scatter in spirals (0.67 dex) than in ellipticals (0.31 dex) [11]?
  4. Causal direction at cosmic noon. Applied to quasar hosts at redshifts 2 to 4 [9], does causal discovery [11] recover the direction that simulations assign to epochs before the peak of star formation [13]?
  5. Quasar-mode outflows. In luminous quasars, does the measured momentum of outflows match the value at which King's model halts black hole growth [3, 7]?

Work on this in Papex

If you have a testable answer to one of the sub-questions above, you can draft it as a private hypothesis in Papex, or prepare the draft from Claude or ChatGPT with the Papex connector. You review the exact version, sign it and decide whether to publish it; a public record that names this problem is listed on this page. A Papex record shows who signed which version and when. It is not peer review and does not mean the hypothesis is correct.

Key references

  1. Ferrarese, L. & Merritt, D. (2000). A fundamental relation between supermassive black holes and their host galaxies. The Astrophysical Journal 539(1), L9–L12. 10.1086/312838 · Checked 16 Oct 2026
  2. Gebhardt, K., Bender, R., Bower, G., Dressler, A., Faber, S. M., Filippenko, A. V. et al. (2000). A relationship between nuclear black hole mass and galaxy velocity dispersion. The Astrophysical Journal 539(1), L13–L16. 10.1086/312840 · Checked 16 Oct 2026
  3. King, A. (2003). Black holes, galaxy formation, and the M_BH–σ relation. The Astrophysical Journal 596(1), L27–L29. 10.1086/379143 · Checked 16 Oct 2026
  4. Di Matteo, T., Springel, V. & Hernquist, L. (2005). Energy input from quasars regulates the growth and activity of black holes and their host galaxies. Nature 433(7026), 604–607. 10.1038/nature03335 · Checked 16 Oct 2026
  5. Peng, C. Y. (2007). How mergers may affect the mass scaling relation between gravitationally bound systems. The Astrophysical Journal 671(2), 1098–1107. 10.1086/522774 · Checked 16 Oct 2026
  6. Jahnke, K. & Macciò, A. V. (2011). The non-causal origin of the black-hole-galaxy scaling relations. The Astrophysical Journal 734(2), 92. 10.1088/0004-637X/734/2/92 · Checked 16 Oct 2026
  7. Fabian, A. C. (2012). Observational evidence of active galactic nuclei feedback. Annual Review of Astronomy and Astrophysics 50, 455–489. 10.1146/annurev-astro-081811-125521 · Checked 16 Oct 2026
  8. McConnell, N. J. & Ma, C.-P. (2013). Revisiting the scaling relations of black hole masses and host galaxy properties. The Astrophysical Journal 764(2), 184. 10.1088/0004-637X/764/2/184 · Checked 16 Oct 2026
  9. Kormendy, J. & Ho, L. C. (2013). Coevolution (or not) of supermassive black holes and host galaxies. Annual Review of Astronomy and Astrophysics 51, 511–653. 10.1146/annurev-astro-082708-101811 · Checked 16 Oct 2026
  10. Maiolino, R., Scholtz, J., Curtis-Lake, E., Carniani, S., Baker, W., de Graaff, A. et al. (2024). JADES: the diverse population of infant black holes at 4 < z < 11: merging, tiny, poor, but mighty. Astronomy & Astrophysics 691, A145. 10.1051/0004-6361/202347640 · Checked 16 Oct 2026
  11. Jin, Z., Pasquato, M., Davis, B. L., Deleu, T., Luo, Y., Cho, C. et al. (2025). Causal discovery in astrophysics: unraveling supermassive black hole and galaxy coevolution. The Astrophysical Journal 979(2), 212. 10.3847/1538-4357/ad9ded · Checked 16 Oct 2026
  12. Juodžbalis, I., Maiolino, R., Baker, W. M., Curtis Lake, E., Scholtz, J., D'Eugenio, F. et al. (2026). JADES: comprehensive census of broad-line AGN from reionization to cosmic noon revealed by JWST. Monthly Notices of the Royal Astronomical Society 546(3), stag086. 10.1093/mnras/stag086 · Checked 16 Oct 2026
  13. Davis, B. L., More, S., Jin, Z., Pasquato, M., Macciò, A. V. & Yuan, F. (2026). Causal reversal in the M•–σ0 relation: implications for high-redshift supermassive black hole mass estimates. The Astrophysical Journal 1000(1), 100. 10.3847/1538-4357/ae47ef · Checked 16 Oct 2026
  14. Koudmani, S., Scholtz, J., Taylor, A. J., Juodžbalis, I., Sijacki, D., Somerville, R. S. et al. (2026). How to raise a supermassive black hole: interpreting early JWST AGN with the AESOPICA simulations. Preprint, arXiv:2607.26177 (version 1, 28 July 2026), not peer reviewed. Source · Checked 16 Oct 2026

Page reviewed 16 Oct 2026

Source

Open, as listed by Wikipedia · Unsolved problems in astronomy. Checked 08 Oct 2026.

The status is the source’s. Papex does not decide whether a problem is solved.

Records on this problem

No public record addresses this problem yet.

More in Cosmology & astronomy