SpaceTRCSpaceTRC/Research platform
TürkçeLet’s collaborate
SpaceTRC / Cosmos

The Milky Way: Structure and History of Our Galaxy

The Milky Way's disc, bar and halo, the Sgr A* black hole, dark matter, merger history, Gaia maps and its uncertain future with Andromeda.

At a glance

We see the Milky Way, a barred spiral galaxy, from the inside. Some quantities, such as the mass of the central black hole, are therefore known very precisely, while others, such as the total mass, remain surprisingly uncertain. Each value below is given with its source and stated uncertainty.

Key quantities for the Milky Way, with stated uncertainties
Quantity Value Uncertainty / note Source
Diameter of the stellar disc more than 100,000 light-years (over about 30 kpc) The disc has no sharp edge, so this is an approximate figure NASA [1]
Stellar mass 6.08 × 1010 M☉ ± 1.14 × 1010 M☉ (Kroupa IMF) Licquia & Newman 2015 [3]
Total mass including dark matter (M200) 1.17 × 1012 M☉ +0.21 / −0.15 × 1012 (68%). Other methods give values several times lower; see below Callingham et al. 2019 [4]
Number of stars of order 1011 (around 100 billion) Order of magnitude only. ESA describes Gaia’s one-billion-star census as about 1% of the Galaxy’s stars ESA [2]
Sun’s distance from the Galactic Centre (R0) 8,277 pc (about 27,000 light-years) ± 9 pc (statistical), about ± 30 pc (systematic) GRAVITY Collaboration 2022 [8]
Circular speed at the Sun’s radius 229.0 km/s ± 0.2 km/s (formal), with systematic errors of about 2–5%. Masers give 236 ± 7 km/s Eilers et al. 2019 [6]; Reid et al. 2019 [7]
Galactic year (one orbit of the Sun) about 200–240 million years NASA quotes about 240 Myr. Reid et al. give the Sun’s angular speed as 30.32 ± 0.27 km/s/kpc, which works out to 203 ± 2 Myr NASA [1]; Reid et al. 2019 [7]
Mass of Sagittarius A* 4.297 × 106 M☉ ± 0.012 × 106 (statistical), about ± 0.04 × 106 (systematic) GRAVITY Collaboration 2022 [8]

Sources checked October 2026. Papers: Licquia & Newman (ApJ, 2015), Callingham et al. (MNRAS, 2019), Eilers et al. (ApJ, 2019), Reid et al. (ApJ, 2019), GRAVITY Collaboration (A&A, 2022). The 203 Myr galactic year is calculated here from the angular speed (2π/Ω☉). M☉ means one solar mass, and 1 kpc is about 3,260 light-years.

Sources: [1] [2] [3] [4] [6] [7] [8]

Structure: disc, bar, halo and spiral arms

Most of the Milky Way’s stars lie in a flattened disc. Star counts from the Sloan Digital Sky Survey (SDSS) show that the disc has two components with different thicknesses. Jurić and colleagues found a thinner component (scale height about 300 pc, scale length 2.6 kpc) and a thicker one (about 900 pc and 3.6 kpc) that makes up about 12% of the local star density. These are conventionally called the thin disc and the thick disc. The same survey found the stellar halo to be flattened (axis ratio about 0.64), with a local density only about 0.5% of the thin disc’s.

The Milky Way seen edge-on, to scaleA cross-section with the same scale horizontally and vertically: the thin disc with a scale height of about 300 pc, the thick disc with about 900 pc, the box- or peanut-shaped bulge at the centre and the flattened stellar halo with an axis ratio of about 0.64. The Sun lies in the mid-plane, 8.3 kpc from the centre. Sun centre to Sun: 8.3 kpc stellar disc diameter: more than 30 kpc (over 100,000 light-years) stellar halo flattened: axis ratio ≈ 0.64 local density ≈ 0.5% of the thin disc’s bulge thickened inner part of the bar thick disc scale height ≈ 900 pc ≈ 12% of local stars thin disc scale height ≈ 300 pc gas and young stars 5 kpc (same scale across and up)
Figure 1. The Milky Way edge-on. Horizontal and vertical scales are the same; the disc bands show ±1 scale height. The bulge outline and the edge of the halo are schematic. Data: sources [1], [8] and [9] on this page.

At the centre lies a box- or peanut-shaped bulge. Red clump giants show that this bulge is the thickened inner part of a longer, flatter bar. Wegg, Gerhard & Portail (2015) measured a bar half-length of about 5.0 ± 0.2 kpc, at an angle of roughly 28–33° to our line of sight. Licquia & Newman put the mass of the bulge and bar together at 0.91 ± 0.07 × 1010 M☉, about one-sixth of all the stellar mass. The current star formation rate is 1.65 ± 0.19 M☉ per year.

Mapping the spiral arms from inside the dusty disc is hard. Using radio interferometry, Reid and colleagues measured trigonometric parallaxes and proper motions for about 200 molecular masers around very young high-mass stars. They concluded that these measurements “strongly suggest that the Milky Way is a four-arm spiral”. The Sun lies near a smaller arm segment called the Local or Orion Arm, between the Sagittarius and Perseus arms.

Schematic face-on view of the Milky WayThe Milky Way seen from the north, rotating clockwise: Sgr A* and the bulge at the centre, the bar with a half-length of about 5 kpc at about 28–33 degrees to the Sun–centre line, four main spiral arms and the Local (Orion) Arm near the Sun. The Sun lies 8.3 kpc from the centre, between the Sagittarius and Perseus arms. Sun 8.3 kpc from the centre ≈ 27,000 light-years the Sun’s orbit and direction of rotation Sagittarius Arm Perseus Arm Local (Orion) Arm Scutum–Centaurus Arm Norma–Outer Arm bar half-length ≈ 5 kpc, at ≈ 28–33° to the Sun–centre line Sgr A* central black hole disc: more than 30 kpc across, with no sharp edge 5 kpc
Figure 2. Schematic view of the Milky Way from the north Galactic pole. The four arms are drawn as logarithmic spirals of constant pitch; their shapes and widths are schematic. Data: sources [1], [7] and [8] on this page; bar dimensions from Wegg et al. (2015), linked in the text.

Sources: [3] [7] [9]

The Sun’s orbit and stellar populations

The GRAVITY instrument measures the Sun’s distance from the centre geometrically, from stellar orbits around Sagittarius A*: R0 = 8,277 ± 9 pc, with a systematic uncertainty of about 30 pc. The maser data give a consistent 8.15 ± 0.15 kpc and place the Sun only 5.5 ± 5.8 pc above the mid-plane, practically in the plane.

The Sun moves at roughly 230–250 km/s, depending on whether one quotes the circular speed at its radius or adds its own peculiar motion (Reid et al. give 247 ± 4 km/s for the total). One orbit therefore takes about 200–240 million years.

The different components hold different stellar populations. The thin disc contains gas, ongoing star formation and relatively young, metal-rich stars. The thick disc and the halo are dominated by older, more metal-poor stars. The stellar halo is light, about 1.4 ± 0.4 × 109 M☉ including the Sagittarius dwarf according to Deason, Belokurov & Sanders, or about 2% of the stellar mass, yet it preserves much of the Galaxy’s history.

Sources: [1] [7] [8]

Sagittarius A*: the central black hole

Since the early 1990s, teams led by Reinhard Genzel and Andrea Ghez have independently tracked stars around the radio source Sagittarius A*. Both found that the stars move around an invisible object of about four million solar masses, packed into a region no larger than our Solar System. The Nobel Committee said a supermassive black hole is “the only currently known explanation”. Genzel and Ghez shared half of the 2020 Nobel Prize in Physics; the other half went to Roger Penrose.

Today the GRAVITY interferometer at ESO’s Very Large Telescope gives the most precise mass. It combines the orbits of the stars S2, S29, S38 and S55 to give 4.297 ± 0.012 × 106 M☉. The orbits fit a single point mass, and S2 shows Schwarzschild precession at 7σ significance. Any extended mass inside S2’s orbit is limited to about 3,000 M☉, roughly 0.1% of the black hole’s mass.

On 12 May 2022, the Event Horizon Telescope (EHT) collaboration released the first image of Sgr A*: a dark shadow surrounded by a bright ring. The image was built from observations made in April 2017 with eight radio observatories linked into a single Earth-sized telescope. Sgr A* was harder to image than M87*, because gas orbits it in minutes rather than days to weeks.

Sources: [8] [10] [11]

Dark matter halo, rotation curve and total mass

If visible matter were all there is, orbital speeds would fall outside the bright disc; in the Milky Way they stay high. Eilers and colleagues used APOGEE spectra together with Gaia and infrared photometry to measure the circular velocity curve from 5 to 25 kpc. They found it declines only gently, by about −1.7 ± 0.1 km/s per kpc. Fitting this curve, they inferred a dark halo of about 7.25 ± 0.26 × 1011 M☉ within the virial radius. That is more than ten times the mass in stars.

The total mass is still debated. Methods based on the motions of satellite galaxies give larger values: Callingham et al. find M200 ≈ 1.17 × 1012 M☉. In contrast, Jiao and colleagues analysed Gaia DR3 data and reported a Keplerian decline between about 19 and 26.5 kpc, where the speed drops by about 30 km/s. Their total mass is only 2.06 (+0.24/−0.13) × 1011 M☉, with an upper limit of 5.4 × 1011 M☉. Published estimates therefore differ by a factor of five or more. The result depends on the tracers used, how far they reach, and assumptions about the halo’s shape and equilibrium.

Sources: [4] [5] [6]

Interactive lab: the Milky Way’s rotation curve

The lab opens with a Milky Way-like model: see how the computed curve follows the white dots that trace the measured curve. Then choose the “Visible matter only” preset: the speed at the Sun’s position falls far below the measured 229 km/s, and the curve drops clearly in the outer disc. Change the halo speed and core radius to see how the total mass within 25 kpc, and the unseen share of it, respond.

Satellites and the merger history

The brightest companions of the Milky Way are the Large and Small Magellanic Clouds, visible to the naked eye from the southern hemisphere. From the way the LMC bends the Orphan stellar stream, Erkal et al. (2019) estimate its total mass at 1.38 (+0.27/−0.24) × 1011 M☉. The Sagittarius dwarf galaxy was found by Ibata, Gilmore & Irwin in 1994, hidden behind the bulge on the far side of the Galactic Centre. It is being torn apart by tides and its stars are being added to the halo.

In 2018 Helmi and colleagues combined Gaia’s six-dimensional data with APOGEE chemistry. They showed that the inner stellar halo is dominated by debris from one galaxy, which they named Gaia–Enceladus. At infall it was slightly more massive than the Small Magellanic Cloud. It merged about ten billion years ago with a mass ratio of about 4:1 and heated the disc that became the thick disc. In the same year, Belokurov et al. independently identified the same strongly radial population, now often called the “Sausage”. They traced it to a progenitor of more than 1010 M☉ that arrived 8–11 billion years ago.

Sources: [12]

Fermi bubbles and the future meeting with Andromeda

In 2010 Su, Slatyer and Finkbeiner found two giant gamma-ray lobes in Fermi Large Area Telescope data, extending about 50° above and below the Galactic Centre and about 40° wide in longitude. Their spectrum is hard and uniform, roughly dN/dE ∝ E−2. The authors argued that the Fermi bubbles most likely come from a large injection of energy at the centre, such as past accretion onto Sgr A* or a nuclear starburst within the last ~10 million years. The question is still open.

The Andromeda galaxy (M31) is approaching us, and a merger in about five billion years was long treated as near-certain. In a study published in Nature Astronomy in 2025, Sawala and colleagues ran about 100,000 simulations using Hubble and Gaia data. They included the LMC and M33. They found a probability of close to 50% that the two galaxies do not merge within the next ten billion years. According to the University of Helsinki, a direct collision within 4–5 billion years has a probability of only about 2%. M33 slightly raises the chance of a merger and the LMC lowers it. The largest remaining uncertainty is Andromeda’s sideways (transverse) velocity.

Sources: [13] [14]

How we map it, and how to go further

ESA’s Gaia mission made science observations from July 2014 to 15 January 2025, and ESA retired the spacecraft on 27 March 2025. Gaia DR3 was released on 13 June 2022 and contains 1,811,709,771 sources, about 33.8 million of them with radial velocities, from 34 months of data. Gaia DR4 is scheduled for 2 December 2026 and is based on 66 months of observations. The final DR5 is expected no earlier than the end of 2030.

Radio surveys of the 21-cm line of neutral hydrogen trace gas across the whole disc. The all-sky HI4PI survey reaches 16.2′ resolution. Infrared surveys such as 2MASS, Spitzer and WISE see through the dust in the plane and the bulge. SDSS’s APOGEE took near-infrared spectra (1.51–1.70 µm) of about 657,000 stars from Apache Point and Las Campanas; its abundances help separate accreted stars from those born in the Galaxy.

All of these data are public. The Gaia ESA Archive supports ADQL queries over the full catalogues. SDSS DR17 provides the APOGEE spectra and abundances. NASA’s IRSA hosts 2MASS, WISE and Spitzer data, and HEASARC is NASA’s archive for X-ray and gamma-ray missions, the bands used to study the Galactic Centre and the Fermi bubbles.

Sources: [2] [6] [12]

Known vs open questions

Established

  • The Milky Way is a barred spiral with a thin disc, a thick disc, a box/peanut bulge that is part of the bar, and a diffuse stellar halo.
  • The Sun is about 8.2–8.3 kpc from the centre, measured by GRAVITY to better than 0.5% (statistical).
  • Sgr A* is a black hole of about 4.3 million solar masses. Stellar orbits and the 2022 EHT image confirm this.
  • The inner halo is dominated by debris from a single major merger about 10 billion years ago (Gaia–Enceladus/Sausage).
  • Dark matter dominates the Galaxy’s mass at large radii.

Open

  • The total mass: estimates range from about 2 × 1011 to more than 1012 M☉, and the shape of the outer rotation curve is disputed.
  • The number and exact layout of the spiral arms, especially on the far side of the Galaxy.
  • What powered the Fermi bubbles: past black-hole activity or a nuclear starburst.
  • Whether the Milky Way and Andromeda will merge at all. The answer depends on Andromeda’s transverse velocity and on the LMC.

Sources and further reading

  1. NASA Science — Galaxies — NASA
  2. ESA — Gaia fact sheet — European Space Agency
  3. Licquia, T. C. & Newman, J. A. — Improved Estimates of the Milky Way’s Stellar Mass and Star Formation Rate from Hierarchical Bayesian Meta-Analysis (ApJ, 2015) — arXiv
  4. Callingham, T. et al. — The mass of the Milky Way from satellite dynamics (MNRAS, 2019) — arXiv
  5. Jiao, Y. et al. — Detection of the Keplerian decline in the Milky Way rotation curve (A&A, 2023) — arXiv
  6. Eilers, A.-C. et al. — The Circular Velocity Curve of the Milky Way from 5 to 25 kpc (ApJ, 2019) — arXiv
  7. Reid, M. J. et al. — Trigonometric Parallaxes of High-Mass Star Forming Regions: Our View of the Milky Way (ApJ, 2019) — arXiv
  8. GRAVITY Collaboration — The mass distribution in the Galactic Centre from interferometric astrometry of multiple stellar orbits (A&A, 2022) — arXiv
  9. Jurić, M. et al. — The Milky Way Tomography with SDSS: I. Stellar Number Density Distribution (ApJ, 2008) — arXiv
  10. Nobel Prize — The Nobel Prize in Physics 2020: press release — Royal Swedish Academy of Sciences
  11. Event Horizon Telescope — Astronomers reveal first image of the black hole at the heart of our galaxy (12 May 2022) — EHT Collaboration
  12. Helmi, A. et al. — The merger that led to the formation of the Milky Way’s inner stellar halo and thick disk (Nature, 2018) — Nature
  13. Su, M., Slatyer, T. R. & Finkbeiner, D. P. — Giant Gamma-ray Bubbles from Fermi-LAT: AGN Activity or Bipolar Galactic Wind? (ApJ, 2010) — arXiv
  14. Sawala, T. et al. — Apocalypse When? No Certainty of a Milky Way–Andromeda Collision (Nature Astronomy, 2025) — arXiv
SPACETRC / KNOWLEDGE

Follow your curiosity.

SPACETRC / GET STARTED

What would you like to do?

This short guide is available in six languages. Full scientific articles are in Turkish and English.

Open a workspace

Search publicationsExamine dataSpace eventsPlan a team projectPrepare a publication

Take selected reading with you

Open Turkish and English dossiers and introductory lessons in the reading area. Offline preparation and installation options are available there.

Open the offline library ↗

The language preference stays in this browser; no account or location permission is required.