Grand-Design vs Flocculent Spirals

Two flavors of the same machine

Look at a gallery of spiral galaxies and you quickly notice that the word "spiral" hides two very different kinds of object. Some galaxies — M51, the Whirlpool, is the poster child — have two enormous, high-contrast arms you could trace with a finger from the centre out to the edge of the disk, sweeping around for a full turn or more in clean symmetry. Others, like NGC 2841 or NGC 5055, look as though someone took a smooth disk and dabbed it with a sponge: dozens of short, broken, feathery arc segments, none of them connected into a global pattern, scattered across the face of the galaxy.

The first kind is a grand-design spiral; the second is a flocculent spiral (from the Latin flocculus, "tuft of wool"). The distinction was sharpened in the 1980s by Bruce and Debra Elmegreen, who built an "arm class" scale running from 1 (purely flocculent, ragged patches only) to 12 (two long, symmetric, dominant arms). Roughly 10% of disk galaxies sit at the grand-design end, about 30% are clearly flocculent, and the majority are "multi-armed" intermediate cases with several partial arms. The remarkable thing is that these are not two different species of galaxy — they are the same rotating stellar-and-gas disk operating under two different organizing principles.

The winding dilemma that started it all

The whole subject exists because of a paradox. A galactic disk does not rotate like a solid record. It rotates differentially: in the flat part of a typical rotation curve the orbital speed v is roughly constant (~220 km/s for the Milky Way), so the angular speed Ω = v/R falls off as 1/R. Inner material laps outer material.

If a spiral arm were simply a fixed string of stars and gas, differential rotation would wind it tighter and tighter. We can estimate how fast. At the Sun's radius (R ≈ 8 kpc) the orbital period is

T = 2πR / v = 2π(8 kpc) / (220 km/s) ≈ 2.2 × 10⁸ yr

but at R = 4 kpc the period is roughly half that. After only a handful of rotations — a few hundred million years, against a galaxy age of ~10¹⁰ years — a material arm would be coiled into a tight, unrecognizable spring. Yet we see well-defined, open arms everywhere. This is the winding dilemma, and resolving it is exactly what separates the two spiral types.

Grand design: the Lin-Shu density wave

In 1964 C.C. Lin and Frank Shu proposed the quasi-stationary spiral structure (QSSS) hypothesis. A grand-design arm, they argued, is not a fixed group of stars at all. It is a density wave — a pattern of slightly enhanced density (typically only 10–20% above the local mean) that rotates rigidly around the galaxy at a single pattern speed Ω_p, while the individual stars and gas clouds orbit at their own, faster or slower, rates and stream through the pattern.

The standard analogy is a traffic jam on a highway. The slow-moving clot of cars persists for hours and moves down the road at its own pace, even though individual cars are constantly entering it from behind and leaving it ahead. The jam is a wave in the density of cars, not a fixed set of vehicles. A spiral arm is the same: a wave in the density of stars.

Because the pattern rotates rigidly at one Ω_p while the disk rotates differentially, there is a special radius — the corotation radius — where the orbital angular speed equals the pattern speed:

Ω(R_CR) = Ω_p

Inside corotation, stars and gas overtake the arm from behind; outside, the arm overtakes them. Either way, gas slamming into the slowly moving potential well of the arm is shock-compressed on the upstream edge. That compression is what triggers star formation, which is why a grand-design arm is lined on one side by a dark dust lane and on the other by bright blue clusters of newborn O and B stars and red Hα knots. The wave itself is a global m = 2 mode (two-fold rotational symmetry, i.e. two arms) of the disk's gravity.

Where the wave begins and ends: Lindblad resonances

A density wave cannot exist at every radius. It is confined between resonances set by the disk's two characteristic frequencies: the orbital frequency Ω(R) and the epicyclic frequency κ(R), the rate at which a star oscillates radially about its mean orbit. The inner and outer Lindblad resonances (ILR, OLR) occur where

Ω_p = Ω ∓ κ/2        (− gives the ILR, + gives the OLR, for m = 2)

A two-armed density wave propagates only in the band between the inner Lindblad resonance and corotation (and a second branch out to the OLR). At a resonance the wave's energy is absorbed or reflected, which both bounds the spiral pattern and gradually transfers angular momentum outward — the mechanism that lets the disk slowly redistribute mass over cosmic time. This is the same resonance machinery that gates bar-driven spiral modes and ring formation; the pattern speed Ω_p is the single number that fixes where all of these resonances fall.

Swing amplification and the role of bars and companions

Pure Lin-Shu theory describes a standing wave but is quieter on how a strong two-armed mode gets excited and kept alive. The modern answer is swing amplification (Goldreich & Lynden-Bell 1965; Toomre 1981): a leading disturbance, as it shears around into a trailing one, can be amplified by a large factor if the disk is dynamically cool enough — quantified by the Toomre stability parameter Q ≈ 1–2. A weak perturbation can be swing-amplified into a strong arm.

What provides the perturbation? Observationally, two things dominate grand-design galaxies:

• A stellar bar. A rotating bar is a powerful, persistent m = 2 driver. Its gravitational torque launches a two-armed spiral that connects to the bar ends. The great majority of strong grand-design spirals are barred (SB) or weakly barred (SAB).
• A tidal companion. A close passage by another galaxy delivers an m = 2 kick that swing-amplification turns into clean two-armed structure. M51's spectacular grand design was tidally driven by its smaller companion NGC 5195; N-body simulations reproduce the arms with a single recent passage.

Isolated, unbarred, dynamically warm disks have no such persistent driver. They tend to be flocculent or weakly multi-armed instead.

Flocculent: stochastic self-propagating star formation

If grand designs are global waves, what are flocculent arms? The leading model is stochastic self-propagating star formation (SSPSF), introduced by Gerola and Seiden in 1978. The idea is local and bottom-up rather than global and top-down:

1. Massive stars form in a giant molecular cloud and, within a few million years, explode as supernovae and blow stellar winds.
2. These shocks compress neighbouring gas clouds, triggering a new burst of star formation a short distance away.
3. The star-forming activity therefore propagates outward across the disk like the front of a forest fire or a chemical reaction.
4. Differential rotation continuously shears each bright star-forming patch into a short, trailing arc — a "flocculus."

Because the ignition is locally random and the shearing is fast, the result is dozens of disconnected, kiloparsec-scale arc fragments with no global symmetry — precisely the flocculent appearance. There is no long-lived wave and no single pattern speed; the structure is transient, with each patch lasting roughly one shearing time (a few × 10⁷ to 10⁸ years) before fading and being replaced. The arms are made of light (young stars and ionized gas), not of an organized gravitational pattern.

Side-by-side comparison

Property	Grand-design spiral	Flocculent spiral
Arm count / symmetry	Two dominant, symmetric arms (m = 2)	Many short fragments, no global symmetry
Underlying mechanism	Quasi-stationary density wave (Lin-Shu)	Stochastic self-propagating star formation
Arm lifetime	Long-lived (many disk rotations)	Transient (~one shearing time, 10⁷–10⁸ yr)
Pattern speed	Single rigid Ω_p; corotation radius defined	None; arms corotate with local material
Dust-lane / star-formation offset	Sharp; young stars offset from dust lane	No systematic offset
Typical driver	Bar or tidal companion	Isolated, unbarred, self-gravitating gas
Near-IR (old stars) appearance	Smooth two-armed wave clearly visible	Often featureless / weak
Elmegreen arm class	10–12	1–4
Prototype	M51, M81, NGC 1300	NGC 2841, NGC 5055, NGC 4414

Real numbers: pattern speeds, sizes, fractions

Concrete figures anchor the picture. Grand-design pattern speeds Ω_p are typically 15–30 km/s/kpc, placing corotation somewhere in the outer half of the optical disk (for the Milky Way, estimates cluster near R_CR ≈ 8–11 kpc, close to the Sun's orbit). The arm-to-interarm density contrast in the old stellar disk is modest — about 1.1 to 1.5 — even though the optical contrast looks dramatic, because the young blue stars and Hα emission greatly amplify the visual impression.

Quantity	Typical value	Note
Disk rotation speed (flat)	~150–300 km/s	~220 km/s for the Milky Way
Orbital period at 8 kpc	~2.2 × 10⁸ yr	The winding timescale
Grand-design pattern speed Ω_p	15–30 km/s/kpc	Rigid; sets corotation
Old-disk arm/interarm contrast	1.1–1.5	Mass amplitude of the wave
Flocculent arm-segment length	1–5 kpc	Sheared star-forming patch
Toomre Q for amplification	~1–2	Disk must be dynamically cool
Grand-design fraction	~10 %	Multi-armed ~60 %, flocculent ~30 %

Famous examples and how we tell them apart

• M51 (the Whirlpool), ~7.6 Mpc. The canonical grand design: two clean, symmetric arms studded with HII regions, tidally driven by the companion NGC 5195 hanging off the end of the northern arm. The textbook proof that interactions excite m = 2 modes.
• M81, ~3.6 Mpc. A grand-design spiral in a small group; interactions with M82 and NGC 3077 have stirred its disk and its arms are reproduced by tidal N-body models.
• NGC 1300, ~18 Mpc. A spectacular barred grand design (SBbc) where the two arms spring directly from the ends of a long stellar bar — the bar-driven case made visible.
• NGC 2841, ~14 Mpc. The classic flocculent: an unbarred Sb galaxy whose disk is covered in short, choppy, discontinuous arm segments with no master pattern.
• NGC 4414 and NGC 5055 (M63). Further flocculent prototypes; M63 actually shows a smoother two-armed pattern in the near-IR, illustrating the "hidden grand design" effect.

The decisive diagnostic is wavelength. Optical (blue, Hα) light traces patchy young star formation and exaggerates flocculent structure. Near-infrared light at 1.6–3.6 μm traces the old red stellar mass — the actual gravitating disk — and reveals the underlying density wave if one exists. Spitzer 3.6 μm surveys (e.g. S⁴G) repeatedly found smooth two-armed waves hiding inside galaxies that look ragged in optical light. So a galaxy can be "optically flocculent, infrared grand-design."

Common misconceptions and edge cases

• "Spiral arms are made of a fixed set of stars." For grand designs this is exactly wrong — that is the winding dilemma. Stars stream through the wave; the arm is a pattern, not a parcel.
• "Flocculent means fewer stars or a younger galaxy." No. Flocculent disks have full old stellar populations; they simply lack a coherent global density wave. The classification is about the organization of arms, not the amount of mass or age of the disk.
• "Grand design and flocculent are mutually exclusive." Many galaxies are multi-armed intermediates, and the same galaxy can read flocculent in one band and grand-design in another. Elmegreen's 1–12 scale is a continuum, not a binary.
• "More arms means a faster, more disturbed galaxy." Arm number is set by which m-mode is amplified, which depends on the disk's stability (Q) and driver, not on rotation speed. A quietly rotating cool disk can host a strong m = 2; a warm one may show none.
• "The density wave makes the stars." The wave doesn't create stars from nothing; it compresses pre-existing gas, raising it above the threshold for gravitational collapse. Remove the gas (quench the galaxy) and the old stellar wave persists but the bright blue arms vanish.
• "Trailing vs leading is ambiguous." Almost all observed arms are trailing (the outer tip points opposite to the direction of rotation). Leading arms are rare and usually signal a recent retrograde encounter (NGC 4622 is the famous oddity).