Spiral Galaxies and Density Wave Theory

Spiral arms are the most photographed structures in the extragalactic sky, yet for a century they hid a paradox in plain sight. A galactic disk rotates differentially: with a nearly flat rotation curve the angular velocity $\Omega(R)=v_c/R$ falls with radius, so inner material laps outer material every couple of hundred million years. If an arm were a fixed chain of stars, that shear would wind it into a tight useless coil within a few galactic turns — yet grand-design spirals show open, coherent arms after tens of rotations. The arms cannot be made of the same matter forever. The resolution, due to Bertil Lindblad and made quantitative by C. C. Lin and Frank Shu in 1964, is that a spiral arm is a wave, not a caravan. It is a quasi-stationary density wave: a spiral region of deepened gravitational potential and enhanced surface density that rotates rigidly at a single pattern speed $\Omega_p$ , independent of radius. Stars and gas stream through it at their own local $\Omega(R)$ , crowding and slowing where the potential is deepest, then speeding away — as cars bunch behind a slow truck. The arm is a persistent traffic jam of matter, not a fixed set of cars. The astrophysical payoff is the gas response. Inside corotation the disk overtakes the pattern supersonically; the interstellar gas shocks against the arm potential, is compressed, and collapses into stars — which is why arms glow in short-lived O and B stars and HII regions, with a dust lane marking the shock on the concave inner edge. This article assembles the winding dilemma, the epicyclic frequency, the Lindblad resonances, the dispersion relation, and the Toomre criterion into the working theory of why disks make spirals.

Key takeaways

A differentially rotating disk ( $\Omega\propto 1/R$ for a flat curve) would wind any material spiral into a vanishing pitch angle within a few rotation periods — the winding dilemma.
Lin–Shu density-wave theory makes the arm a quasi-stationary pattern rotating rigidly at $\Omega_p$ while stars and gas orbit through it, so the arm is a crest in a density wave, not a fixed collection of stars.
The epicyclic frequency $\kappa$ governs radial oscillations; for a flat curve $\kappa=\sqrt2\,\Omega$ , making $\Omega-\kappa/2$ nearly constant — why a single $\Omega_p$ can sustain a two-armed spiral across a wide annulus.
Corotation sits where $\Omega=\Omega_p$ ; the inner and outer Lindblad resonances at $\Omega_p=\Omega\mp\kappa/2$ bound the zone where a steady $m=2$ wave lives.
Gas overtaking the arm inside corotation shocks supersonically, so young blue stars, HII regions, and a dust lane appear on the arm's concave inner edge.
The Lin–Shu dispersion relation and the Toomre parameter $Q=\sigma_R\kappa/(3.36\,G\Sigma_0)$ set which wavelengths propagate and when a disk is stable; real disks sit at $Q\sim1$ – $2$ .
Key idea: spiral arms are not fixed collections of stars but crests of a density wave rotating at a constant pattern speed $\Omega_p$ ; stars and gas stream through the pattern, are compressed as they cross the arm potential, and light it up with newly formed stars.

Where this lives today

Density-wave theory is now tested with a precision Lin and Shu could not have imagined. The ESA Gaia mission has measured phase-space coordinates for over a billion Milky Way stars, revealing the disk's arms directly and exposing a corrugated "phase spiral" (the Gaia snail) that records a recent perturbation — evidence our own arms are dynamic and at least partly transient. In external galaxies, resolved PHANGS maps (ALMA, MUSE, JWST/HST) trace the predicted sequence across an arm — dust lane and molecular gas on the inner edge, then HII regions, then aging clusters downstream — a direct check of the shock-triggered star-formation picture and its age gradient. High-resolution simulations largely support a transient, recurrent, self-regulating view: arms that wind up, dissolve, and regenerate rather than rotate rigidly forever. The central open problem is the one the theory was built to answer — are grand-design arms genuine long-lived density waves with a single pattern speed, or a superposition of shorter-lived radius-dependent modes? Measuring $\Omega_p(R)$ from stellar-age and gas-offset gradients across many galaxies is the current frontier.

The physical picture

Picture a slow truck on a busy road. It crawls at a steady speed — the pattern, moving at $\Omega_p$ — while faster cars pile up behind, inch through, then accelerate away. The jam persists even though no single car stays inside it. That is a density wave: the cars are stars and gas clouds orbiting at their local $\Omega(R)$ , the jam is the spiral arm. Because $\Omega(R)>\Omega_p$ inside corotation, material catches the arm from behind, lingers in its deepened potential, and moves on. The subtlety that makes the wave hold together is orbital. In a frame turning at $\Omega_p$ a star traces a small epicyclic loop of frequency $\kappa$ about a slowly drifting guiding centre. Lin and Shu's insight was that if these loops precess at the right rate, their elongated axes nest and phase so the crowding of orbits traces a standing spiral held together by self-gravity. The near-constancy of $\Omega-\kappa/2$ across a flat-curve disk lets one pattern speed serve a broad range of radii — the deep reason spirals are common rather than rare.

Figure

A trailing two-armed density wave: the pattern rotates rigidly at Ωₚ while disk material at radius R orbits at the faster Ω(R) = v_c/R. Inside corotation (dashed circle, where Ω = Ωₚ) stars and gas overtake the arm from behind, shock against its potential well, and light the inner concave edge with a dust lane and young blue stars

The gas is the theory's crucial witness. Stars stream past the arm with a mild density enhancement; gas is compressible and cannot slip through, so it shocks. Because the disk overtakes the pattern from behind inside corotation, the shock sits offset to the inner, concave edge of the potential minimum — where dust lanes, the youngest OB associations, and HII regions appear. The second figure turns the resonance bookkeeping into a diagram of which radii can host a steady wave.

Figure

Lindblad diagram for a flat rotation curve v_c = 220 km/s: Ω, Ω − κ/2 and Ω + κ/2 versus radius with κ = √2 Ω. A pattern speed Ωₚ = 20 km/s/kpc (horizontal line) meets Ω − κ/2 at the inner Lindblad resonance near 3.2 kpc, Ω at corotation 11 kpc, and Ω + κ/2 at the outer Lindblad resonance near 18.8 kpc — the annulus where a steady two-armed wave can live

The core equations

Everything begins with how fast disk material circulates. For a nearly flat rotation curve the circular speed $v_c$ is roughly constant across the disk, so the angular velocity falls off inversely with radius — the source of the shear that motivates the whole theory.

\Omega(R)=\frac{v_c(R)}{R} \qquad (1)

where: $\Omega(R)$ — circular angular velocity at radius $R$ ; $v_c(R)$ — circular (rotation) speed; $R$ — galactocentric radius. With $v_c\approx$ const, $\Omega\propto 1/R$ : the inner disk turns faster and laps the outer. A star nudged radially from a circular orbit oscillates about its guiding centre at the epicyclic frequency $\kappa$ . Fixed by the local rotation curve, it is the natural frequency that resonates with the pattern.

\kappa^2(R)=R\frac{\mathrm{d}\Omega^2}{\mathrm{d}R}+4\Omega^2=\frac{2\Omega}{R}\frac{\mathrm{d}(R^2\Omega)}{\mathrm{d}R} \qquad (2)

where: $\kappa$ — epicyclic (radial oscillation) frequency of a nearly circular orbit; $\Omega$ — circular angular velocity; $R^2\Omega=L_z$ — specific angular momentum. The two forms are identical; the second makes the flat-curve limit $\kappa=\sqrt2\,\Omega$ immediate. The whole hypothesis rests on one assertion: the arm pattern is rigid, rotating at a single pattern speed while the material's $\Omega$ varies with $R$ . Corotation is the one radius where the two coincide.

\Omega_p=\text{const},\qquad \Omega(R_{\rm CR})=\Omega_p \qquad (3)

where: $\Omega_p$ — pattern angular speed, independent of $R$ ; $R_{\rm CR}$ — corotation radius; inside it material overtakes the arm, outside it the arm overtakes material. Grand-design spirals are usually $m=2$ . A star feels the $m$ -armed pattern as periodic forcing at frequency $m|\Omega-\Omega_p|$ . When that matches $\kappa$ , the orbit resonates — the Lindblad resonances that bound the wave zone.

\Omega_p=\Omega(R)-\frac{\kappa(R)}{m}\ \ (\text{ILR}),\qquad \Omega_p=\Omega(R)+\frac{\kappa(R)}{m}\ \ (\text{OLR}) \qquad (4)

where: $m$ — number of arms (the classic two-armed case uses $\kappa/2$ ); $\Omega,\kappa$ — circular and epicyclic frequencies. ILR and OLR flank corotation; a coherent $m=2$ wave lives between them. For a tightly wound ( $|kR|\gg1$ ) wave, linearizing the fluid and Poisson equations of a thin disk yields the dispersion relation — the balance of rotation, self-gravity, and pressure deciding which waves propagate.

(\omega-m\Omega)^2=\kappa^2-2\pi G\Sigma_0|k|+k^2c_s^2 \qquad (5)

where: $\omega=m\Omega_p$ — wave frequency; $k$ — radial wavenumber; $\Sigma_0$ — surface mass density; $c_s$ — gas sound speed (for stars, $\sigma_R$ with a reduction factor); $G$ — gravitational constant. Self-gravity destabilizes; rotation ( $\kappa^2$ ) and pressure ( $k^2c_s^2$ ) restore stability at long and short wavelengths. Setting $m=0$ and demanding the disk never collapse defines the Toomre parameter — the single number for whether a differentially rotating disk is locally stable.

Q\equiv\frac{\sigma_R\,\kappa}{3.36\,G\,\Sigma_0}\qquad(\text{stars}),\qquad Q=\frac{c_s\,\kappa}{\pi\,G\,\Sigma_0}\qquad(\text{gas}) \qquad (6)

where: $Q>1$ locally stable, $Q<1$ unstable to axisymmetric collapse; $\sigma_R$ — radial stellar velocity dispersion; the constant is $3.36$ for a collisionless stellar disk and $\pi$ for a gas disk. Real disks hover at $Q\sim1$ – $2$ : marginally stable, amplifying spirals without fragmenting. Finally the geometry: how tightly the arm winds is fixed by the ratio of azimuthal to radial wavenumber — small pitch angles are tightly wound Sa-type arms, large ones open Sc-type arms.

\tan i=\left|\frac{m}{kR}\right| \qquad (7)

where: $i$ — pitch angle between the arm and the local tangent circle; $m$ — number of arms; $kR$ — radial wavenumber times radius (large for tightly wound arms). Grand-design spirals have $i\approx10^\circ$ – $25^\circ$ .

Derivation AThe winding dilemma made quantitative

The goal is to prove that a spiral made of a fixed set of stars cannot survive: in a flat-rotation-curve disk the differential shear winds any material pattern to a vanishing pitch angle within a handful of rotation periods, forcing us to abandon the chain-of-stars picture for a wave whose crest rotates at a single pattern speed. We take two rings aligned along a radial spoke and track how it shears.

Step 1. Assign each ring its angular velocity from equation (1): a ring at $R_1$ turns at $\Omega(R_1)=v_c/R_1$ and one at $R_2>R_1$ at the slower $\Omega(R_2)=v_c/R_2$ . Why: a flat curve is observed for real disks and maximizes the difficulty — inner material genuinely outruns outer.

Step 2. The azimuthal separation that opens between the two rings grows linearly in time,

\Delta\theta(t)=[\Omega(R_1)-\Omega(R_2)]\,t=v_c\left(\frac{1}{R_1}-\frac{1}{R_2}\right)t.

Why: a radial spoke is trailed into a spiral as the inner end laps the outer, and the winding accumulates without bound.

Step 3. Scale the winding to a real galaxy using one rotation period at the solar circle,

P=\frac{2\pi R_0}{v_c}\approx\frac{2\pi(8.2\ \mathrm{kpc})}{220\ \mathrm{km\,s^{-1}}}\approx2.3\times10^{8}\ \mathrm{yr}.

Why: the Galaxy is roughly 13 Gyr old, so it has completed about 50 rotations — dozens of winding times, not a few.

Step 4. Track the pitch angle: since $\cot i\propto\Delta\theta$ grows without bound, $i\to0$ . After a few rotations a material arm winds far below the observed $10^\circ$ – $25^\circ$ , an unobservably tight coil. Why: theory predicts near-circular filaments, observation shows open arms — the material hypothesis is ruled out.

Step 5. The arm therefore cannot corotate with its matter. The only escape is a pattern whose shape rotates rigidly at one speed $\Omega_p$ while stars and gas flow through it — a density wave. Why: rigid pattern rotation sidesteps the shear, because the arm is not made of any fixed matter. *This gives the Milky Way rotation period $2.29\times10^8$ yr at $R_0=8.2$ kpc, $v_c=220$ km/s.*

Go deeper · Derivation B · 1 minEpicyclic frequency of a flat-rotation-curve disk

Derivation BEpicyclic frequency of a flat-rotation-curve disk

The aim is to evaluate the epicyclic frequency for the flat rotation curve of real disks, obtaining the clean result $\kappa=\sqrt2\,\Omega$ , and then to show that $\Omega-\kappa/2$ varies only weakly with radius — the single fact that explains why a two-armed density wave can be sustained by one pattern speed across a wide annulus rather than at a single special radius.

Step 1. Insert the flat curve into the angular momentum. With $\Omega=v_c/R$ the specific angular momentum is $R^2\Omega=v_c R$ ,

\Omega=\frac{v_c}{R},\qquad R^2\Omega=v_c R.

Why: writing $\kappa^2$ through $R^2\Omega$ turns the derivative into the slope of a straight line.

Step 2. Differentiate: because $v_c R$ is linear in $R$ , its derivative is just the constant $v_c$ ,

\frac{\mathrm{d}(R^2\Omega)}{\mathrm{d}R}=v_c.

Why: the flat curve removes all radial structure from the angular-momentum gradient.

Step 3. Assemble equation (2):

\kappa^2=\frac{2\Omega}{R}\,v_c=\frac{2v_c^2}{R^2}=2\Omega^2\ \Rightarrow\ \kappa=\sqrt2\,\Omega.

Why: the flat-curve anchor between the textbook limits — Keplerian $\kappa=\Omega$ and solid-body $\kappa=2\Omega$ — fixing the resonance spacing for real disks.

Step 4. Form the two-armed pattern combination,

\Omega-\frac{\kappa}{2}=\left(1-\frac{1}{\sqrt2}\right)\Omega\approx0.293\,\Omega.

Why: both scale as $1/R$ , but $\Omega-\kappa/2$ falls gently, so a single $\Omega_p$ can nearly match it over a broad range of radii — the reason a coherent two-armed spiral spans the disk between its Lindblad resonances. *This recovers the standard limits: flat curve $\kappa=\sqrt2\,\Omega$ , Keplerian $\kappa=\Omega$ , solid-body $\kappa=2\Omega$ .*

Go deeper · Derivation C · 1 minFrom the dispersion relation to the Toomre criterion

Derivation CFrom the dispersion relation to the Toomre criterion

The purpose is to extract the local stability threshold of a rotating disk from its axisymmetric dispersion relation, deriving the Toomre parameter $Q$ as the condition that the wave frequency stay real for every wavelength. We set $m=0$ , require $\omega^2\ge0$ for all $k$ , minimize over $k$ , and read off the constant separating a placid disk from a fragmenting one.

Step 1. Impose stability. For the axisymmetric relation $\omega^2=\kappa^2-2\pi G\Sigma_0|k|+k^2c_s^2$ , real oscillation requires

\omega^2(k)\ge0\quad\text{for all }k.

Why: a negative $\omega^2$ means imaginary $\omega$ — exponential growth, i.e. collapse at that wavelength.

Step 2. Find the most unstable wavenumber by minimizing $\omega^2(k)$ ,

\frac{\mathrm{d}\omega^2}{\mathrm{d}k}=-2\pi G\Sigma_0+2kc_s^2=0\ \Rightarrow\ k_*=\frac{\pi G\Sigma_0}{c_s^2}.

Why: self-gravity ( $\propto|k|$ ) and pressure ( $\propto k^2$ ) compete, so one wavenumber is closest to instability; survive there and the disk survives everywhere.

Step 3. Evaluate the minimum,

\omega^2(k_*)=\kappa^2-\frac{\pi^2G^2\Sigma_0^2}{c_s^2},

and demand it be non-negative. Why: this converts "stable for all $k$ " into one inequality between rotation, pressure, and self-gravity.

Step 4. Solve to define $Q$ ,

\kappa^2\ge\frac{\pi^2G^2\Sigma_0^2}{c_s^2}\ \Longleftrightarrow\ Q\equiv\frac{c_s\kappa}{\pi G\Sigma_0}\ge1.

Why: $Q$ collects the three effects into one dimensionless ratio; for a collisionless stellar disk the same argument replaces $\pi$ by $3.36$ and $c_s$ by $\sigma_R$ .

Step 5. Interpret the value. Real disks sit at $Q\sim1$ – $2$ : stable against axisymmetric collapse yet close enough to threshold that swing-amplified spirals grow without shattering into clumps. Why: marginal stability is the sweet spot the density-wave and swing-amplification pictures both need. *This recovers the Toomre $Q$ with the standard constants: $\pi$ for a gas disk and $3.36$ for a collisionless stellar disk.*

Limiting cases

Case	Result	Meaning
Corotation	$\Omega(R_{\rm CR})=\Omega_p$	Pattern and material move together; no net streaming. Divides gas overtaking the arm (inside) from gas lagging it (outside).
Inner Lindblad resonance	$\Omega_p=\Omega-\dfrac{\kappa}{2}$	Inner edge of the two-armed wave zone; strong wave absorption, often present in barred and nuclear regions.
Outer Lindblad resonance	$\Omega_p=\Omega+\dfrac{\kappa}{2}$	Outer edge of the coherent region; beyond it a trailing two-armed pattern cannot be maintained.
Flat rotation curve	$\kappa=\sqrt2\,\Omega,\ \Omega-\dfrac{\kappa}{2}\approx0.293\,\Omega$	Nearly radius-independent precession lets one $\Omega_p$ serve a wide annulus — why spirals are common.
Tightly wound (WKB)	$\vert kR\vert\gg1$	Local dispersion relation valid; small pitch angle; the regime where density-wave theory is cleanest.
Marginal stability	$Q=\dfrac{\sigma_R\kappa}{3.36\,G\Sigma_0}\simeq1$	Disk poised between stability and collapse; observed disks sit at $Q\sim1$ – $2$ , favouring spiral amplification.

Common trapSpiral arms are permanent streams of the same stars

This is the winding dilemma trap, exactly backwards. If arms were fixed chains of matter, differential rotation ( $\Omega\propto1/R$ ) would coil them to a vanishing pitch angle within a few of the disk's $\sim2\times10^8$ -yr rotation periods, yet open arms persist over dozens of rotations — so arms cannot be material. They are crests of a density wave whose shape rotates rigidly at $\Omega_p$ while stars and gas stream through. A subtler version: assuming a single $\Omega_p$ works everywhere. Because a flat curve makes $\Omega(R)=\Omega_p$ at only one radius, a steady two-armed wave is confined between the inner and outer Lindblad resonances; outside that band the pattern cannot be sustained, so real spirals fade beyond their OLR and modern work invokes transient, recurrent, and bar- or tidally-driven arms.

Worked examples

1 · Resonance radii of a model Galaxy (fully worked)

A disk has a flat rotation curve $v_c=220\ \mathrm{km\,s^{-1}}$ and hosts a two-armed spiral with pattern speed $\Omega_p=20\ \mathrm{km\,s^{-1}\,kpc^{-1}}$ . Locate its corotation radius and its inner and outer Lindblad resonances, and say which arm zone the Sun ( $R_0=8.2$ kpc) sits in.

Step 1. Corotation is where the material angular velocity equals the pattern: $\Omega(R_{\rm CR})=v_c/R_{\rm CR}=\Omega_p$ , so $R_{\rm CR}=v_c/\Omega_p=220/20=\mathbf{11.0\ kpc}$ .

Step 2. For a flat curve $\kappa=\sqrt2\,\Omega$ , so $\Omega-\kappa/2=(1-1/\sqrt2)\Omega=0.293\,\Omega$ . The inner Lindblad resonance sets $0.293\,\Omega=\Omega_p$ , i.e. $\Omega_{\rm ILR}=\Omega_p/0.293=68.3\ \mathrm{km\,s^{-1}\,kpc^{-1}}$ , giving $R_{\rm ILR}=v_c/\Omega_{\rm ILR}=220/68.3=\mathbf{3.2\ kpc}$ .

Step 3. The outer Lindblad resonance uses $\Omega+\kappa/2=(1+1/\sqrt2)\Omega=1.707\,\Omega=\Omega_p$ , so $\Omega_{\rm OLR}=20/1.707=11.7\ \mathrm{km\,s^{-1}\,kpc^{-1}}$ and $R_{\rm OLR}=220/11.7=\mathbf{18.8\ kpc}$ .

Step 4. The steady wave lives between the ILR at 3.2 kpc and the OLR at 18.8 kpc, with corotation at 11 kpc. The Sun at 8.2 kpc lies inside corotation, so local material ( $\Omega=220/8.2=26.8\ \mathrm{km\,s^{-1}\,kpc^{-1}}>\Omega_p$ ) overtakes the arms from behind — consistent with gas shocking on the inner edge. ILR 3.2 kpc, CR 11 kpc, OLR 18.8 kpc; the Sun is inside corotation.

2 · Toomre stability and the fragmentation scale at the solar circle

Estimate the Toomre $Q$ of the stellar disk near the Sun, taking $\sigma_R\approx30\ \mathrm{km\,s^{-1}}$ , $\Sigma_0\approx50\ M_\odot\,\mathrm{pc^{-2}}$ , and $\kappa\approx37\ \mathrm{km\,s^{-1}\,kpc^{-1}}$ , and find the most unstable wavelength $\lambda_{\rm crit}=4\pi^2G\Sigma_0/\kappa^2$ .

Step 1. Use $G=4.301\times10^{-3}\ \mathrm{pc}\,(\mathrm{km\,s^{-1}})^2\,M_\odot^{-1}$ and convert $\kappa=37\ \mathrm{km\,s^{-1}\,kpc^{-1}}=0.037\ \mathrm{km\,s^{-1}\,pc^{-1}}$ . Then $Q=\dfrac{\sigma_R\kappa}{3.36\,G\Sigma_0}=\dfrac{(30)(0.037)}{3.36\,(4.301\times10^{-3})(50)}=\dfrac{1.11}{0.7226}=\mathbf{1.54}$ .

Step 2. With $Q\approx1.5$ the local stellar disk is stable to axisymmetric collapse but sits comfortably in the $1$ – $2$ band where non-axisymmetric spiral perturbations amplify — exactly the marginal condition the theory needs.

Step 3. The critical wavelength: $\lambda_{\rm crit}=\dfrac{4\pi^2G\Sigma_0}{\kappa^2}=\dfrac{39.48\,(4.301\times10^{-3})(50)}{(0.037)^2}=\dfrac{8.49}{1.369\times10^{-3}}\approx\mathbf{6.2\ kpc}$ .

Step 4. This several-kpc scale is the longest wavelength self-gravity can destabilize against rotation; shorter ones are stabilized by random motion. Comparable to the disk scale length, it is why the response is global spiral structure rather than fine clumping. Q ≈ 1.5 (marginally stable), λ_crit ≈ 6 kpc.

Physics of the portrait

Density-wave theory is where celestial mechanics, fluid dynamics, and observational morphology meet. It began with Bertil Lindblad's struggle to reconcile stellar orbits with the regularity of spiral arms, crystallized when Lin and Shu recast the arm as a self-consistent wave, and was grounded, then tempered, by Alar Toomre.

Origins · 1 mindensity wave theory and how this was discovered

density wave theory

Lindblad 1925 · Lin & Shu 1964 · Toomre 1964 · spiral structure and stability of galactic disks

The theory of spiral structure is a rare case where the name honours an idea rather than one person, because it took four contributors and forty years. Bertil Lindblad, in Stockholm from the mid-1920s, was the first to take the dynamics seriously: he recognized differential rotation from the Oort–Lindblad analysis of stellar streaming, identified the winding problem, and spent decades exploring how orbital resonances and the combination $\Omega-\kappa/2$ might hold a pattern together — the resonances now bear his name. But his kinematic picture never quite became a self-consistent wave. That step came in 1964 from Chia-Chiao Lin and Frank Shu at MIT, whose quasi-stationary spiral structure hypothesis made the arm a density wave with a fixed pattern speed $\Omega_p$ , backed by the dispersion relation (sharpened by Alar Kalnajs and others). In the same year Alar Toomre published the stability criterion $Q$ , the other pillar. Toomre later became the theory's most productive critic: his 1972 models with Juri Toomre showed tidal encounters excite transient arms, and his swing-amplification work argued that many spirals are recurrent and short-lived. The modern synthesis keeps the density-wave framework as organizing language while acknowledging that real arms range from long-lived grand-design waves to fleeting flocculent responses.

History

1925

— Bertil Lindblad begins his dynamical theory of galactic rotation and spiral structure

1927

— Oort and Lindblad establish the differential rotation of the Galaxy through the Oort constants

1964

— Lin and Shu propose the quasi-stationary density-wave hypothesis and its dispersion relation

1964

— Toomre derives the local axisymmetric stability parameter Q for a rotating disk

1972

— Toomre and Toomre model tidally driven transient spiral arms in interacting galaxies

Going further

Texts. Carroll & Ostlie, An Introduction to Modern Astrophysics, 2nd ed., Ch. 24–25 (spiral structure and the Milky Way) · Binney & Tremaine, Galactic Dynamics, 2nd ed., Ch. 3 (orbits, epicycles) and Ch. 6 (disk dynamics and spiral structure) · Mo, van den Bosch & White, Galaxy Formation and Evolution, Ch. 11 · Lin & Shu, ApJ 140, 646 (1964); Lin, Yuan & Shu, ApJ 155, 721 (1969) · Toomre, ApJ 139, 1217 (1964). Key numerical anchors used throughout: the solar-circle rotation period $2\pi R_0/v_c\approx2.3\times10^8$ yr at $R_0=8.2$ kpc, $v_c=220$ km/s; $\kappa=\sqrt2\,\Omega$ and $\Omega-\kappa/2\approx0.293\,\Omega$ for a flat curve; resonance radii for $\Omega_p=20\ \mathrm{km\,s^{-1}\,kpc^{-1}}$ — ILR 3.2 kpc, CR 11 kpc, OLR 18.8 kpc; and the axisymmetric stability threshold $Q=c_s\kappa/(\pi G\Sigma_0)\ge1$ (constant $3.36$ for stars), with $Q\approx1.5$ and $\lambda_{\rm crit}\approx6$ kpc at the solar circle.

Prerequisites. Structure of the Milky Way Galaxy · Galactic Rotation Curves · Hubble Sequence and Galaxy Morphology · Interstellar Medium of Galaxies. Leads to. Stellar Populations I, II, and III · Neutral Hydrogen 21-cm Line and Galactic Kinematics · Galactic Bulge, Bar, and Central Regions · Evidence for Dark Matter in Galaxies.

Ask the tutor: Want to work through why $\Omega-\kappa/2$ stays nearly flat, or trace a gas streamline shocking through an arm?

Why doesn't the density wave itself wind up like a material arm would?

Because the wave is a shape, not a chunk of matter. The pattern is defined by where orbits crowd together, and Lin and Shu showed that crowding can rotate rigidly at a single $\Omega_p$ even though the stars and gas forming the crest are constantly replaced as they orbit through. The winding argument only bites on features made of fixed matter; a pattern speed independent of radius is immune to shear, held together by self-gravity plus the near-constancy of $\Omega-\kappa/2$ .

What is the difference between the pattern speed and the rotation speed?

The rotation speed refers to the material: stars and gas at radius $R$ circle at $\Omega(R)=v_c/R$ , large in the inner disk and small outside. The pattern speed $\Omega_p$ refers to the arm shape and is one number for the whole spiral. At most radii the two differ, so material streams through the pattern; only at corotation do they coincide. Inside corotation material overtakes the arm from behind; outside, the arm outruns the material. This relative motion drives the gas shocks and the offset star formation.

Why are the Lindblad resonances important?

They mark where a star's radial (epicyclic) oscillation resonates with the periodic push from the arm. In the frame turning at $\Omega_p$ a star meets the $m$ -armed pattern at frequency $m|\Omega-\Omega_p|$ ; when that equals $\kappa$ , energy and angular momentum are exchanged strongly between wave and orbit. For a two-armed spiral this is $\Omega_p=\Omega\mp\kappa/2$ , the inner and outer Lindblad resonances, which bound the region a coherent wave can occupy — it is absorbed or reflected there. Corotation sits between them, so the density-wave zone is the annulus ILR–CR–OLR.

How does the theory explain the young blue stars that trace the arms?

Through the gas response, not the stars directly. Inside corotation the gas orbits faster than the pattern and slams supersonically into the arm's potential minimum, driving a shock offset to the concave inner edge. The shock compresses gas over the threshold for collapse, so star formation concentrates at that front. The most massive O and B stars and their HII regions live only a few million years — far less than the arm-crossing time — so they light the arm brilliantly near their birthplace, while longer-lived stars drift downstream, producing the observed colour and age gradient across the arm.

Do all spiral arms fit the quasi-stationary density-wave picture?

No, and this is where the modern field departs from the strict 1964 hypothesis. Grand-design spirals — clean two-armed systems like M51, often with a companion or bar — are the best candidates for long-lived density waves with a well-defined $\Omega_p$ . But many disks are flocculent, with patchy arms better described by stochastic self-propagating star formation sheared by rotation. Simulations show a large class of spirals are transient and recurrent — growing by swing amplification, winding up, dissolving, and regenerating within a rotation or two — so a single global pattern speed fits poorly. Density-wave theory remains the organizing language and is essentially correct for grand-design systems, but it is one mechanism among several.

What sets whether a galaxy's arms are tightly or loosely wound?

The pitch angle $i$ , with $\tan i=|m/kR|$ , is fixed by the radial wavenumber $k$ the disk can support, which depends on the mass distribution. A more centrally concentrated potential — bigger bulge, larger $\kappa$ — supports shorter-wavelength, tightly wound waves, the small pitch angles of Sa spirals; less concentrated, gas-rich, lower- $\kappa$ disks support longer waves and the open arms of Sc spirals. This ties the Hubble sequence to the underlying dynamics rather than leaving it purely descriptive.