Bok Globule

A black drop on a glowing background

If you point a telescope at a bright nebula — IC 2944 in Centaurus, the Carina Nebula, the Tarantula in the LMC — and look carefully at the surface, you sometimes see what looks like an inkblot on the page: a small, sharp-edged dark patch superimposed on the glowing background. The patch is not a hole in the gas; it is a clump of cold dust and molecular gas in the foreground, opaque enough to block the bright emission behind it. These small, dense, isolated, dark clouds are Bok globules, and each one is a probable stellar nursery on the verge of producing a single low-mass star.

Bok globules were named after Bart Bok, the Dutch-American astronomer who together with Edith Reilly proposed in 1947 that the small dark patches he had been cataloguing in front of HII regions were not coincidences of perspective but physical condensations destined to collapse into stars. The proposal was speculative for decades because the embedded young stars were invisible at optical wavelengths. Only with infrared and millimetre surveys from the 1980s onwards — IRAS, Spitzer, Herschel, and most recently ALMA — could observers look inside, and Bok was vindicated: most well-studied globules harbour Class 0/I embedded protostars or are about to form one.

Anatomy of a globule

A typical Bok globule has the following parameters:

Property	Typical value	Comment
Diameter	0.05 – 0.3 pc	Some isolated giants reach 1 pc
Mass	1 – 50 M_⊙	Most are 2 – 10 M_⊙
Central density	10⁴ – 10⁶ cm⁻³	n_H₂; few × 10⁵ at centre of B68
Central temperature	7 – 12 K	Cosmic-ray heating vs CO/dust cooling
Visual extinction A_V	> 10 mag	Often > 30 at centre
Velocity dispersion	0.15 – 0.3 km/s	Subsonic — thermal-dominated
Outer pressure	~ 10⁵ K cm⁻³	From surrounding ISM

The morphology depends on environment. Isolated globules — far from any massive star — are nearly spherical and follow the Bonnor–Ebert isothermal-sphere equilibrium with remarkable precision. Globules embedded in or near an HII region are reshaped by the ionising radiation: the irradiated face is sharpened into a cometary "nose" by the photoevaporation pressure, with a tail of evaporated gas trailing away from the ionising star. The classic "cometary" or "elephant-trunk" globules of NGC 281, IC 1396 and the Carina Nebula are this type.

Barnard 68: the textbook globule

The most studied isolated globule is Barnard 68 (B68), about 125 pc away in Ophiuchus. It is a near-perfect spherical condensation of about 2 solar masses spread over roughly 0.1 pc. In a beautiful 2001 paper, João Alves and collaborators used near-infrared photometry of more than a thousand background stars to measure the radial reddening across the face of B68. Each star's intrinsic colour was estimated from its photometric type; the difference between observed and intrinsic colour gave the dust column at that line of sight. The resulting column-density profile traced the globule's internal density structure with extraordinary precision.

The profile fits a Bonnor–Ebert solution — the equilibrium between thermal pressure and self-gravity for a pressure-confined isothermal sphere — out to the critical dimensionless radius ξ_max ≈ 6.9, where the configuration becomes gravitationally unstable. B68's measured ξ ≈ 7, slightly past the critical value, suggests it is right on the verge of collapse. Millimetre observations confirm a small inward velocity of about 0.05 km/s in the inner regions: collapse has begun but is still subsonic and gentle, not yet a runaway free-fall. B68 has become the canonical "pre-stellar core" — a globule before the protostar.

Barnard 335: the canonical Class 0 source

B335 is a smaller globule 100 pc away in Aquila, of order 0.05 pc across, also nearly spherical. Unlike B68, B335 contains a deeply embedded young stellar object, IRAS 19347+0727, classified as Class 0 — the earliest, most embedded stage of protostellar evolution. The protostar is invisible at wavelengths shorter than ~10 μm but blazes in the far infrared and submillimetre. Around it spreads a flattened envelope of infalling gas, bipolar outflows extending several thousand AU, and the seeds of a forming disk. ALMA observations have mapped the inflow at radii from a few thousand down to a few hundred AU. B335 is the prototype for what a globule looks like during, rather than before, collapse.

Where do globules come from?

Three formation pathways are debated.

Detached dense cores. Most isolated Bok globules are probably dense cores that have separated from their parent giant molecular cloud through tidal or feedback effects, or because the parent cloud was dispersed around them. The internal physics — density, temperature, line widths — match those of cores still embedded in larger clouds.

Photoevaporation residue. The cometary globules seen in HII regions are the dense remnants of clumps that survived the radiation field of nearby O stars. Pre-existing density inhomogeneities in the parent cloud were able to shield themselves while less dense material was ionised and stripped away. The photoevaporation-driven shock can compress the surviving clump, raising its density enough to trigger collapse — the so-called "radiation-driven implosion" mechanism for star formation, mathematically explored by Bertoldi (1989), Lefloch and Lazareff (1994), and others.

Independent condensation. A small fraction may have condensed independently of any larger structure, from converging atomic flows that crossed the H₂-formation threshold locally. These are difficult to distinguish observationally from detached cores.

The path from globule to protostar

A subcritical Bonnor–Ebert sphere is stable; a supercritical one is not. As the central density rises (perhaps through slow ambipolar diffusion of magnetic flux out of the core, or through external pressure increases), the configuration eventually crosses the threshold and contracts. The contraction is initially quasi-static, growing on a 10⁵–10⁶ year timescale set by ambipolar drift. Once the central region becomes optically thick to its own cooling radiation, a first hydrostatic core forms — a few AU across, roughly 100 K. As more mass piles on and the central temperature rises to about 2,000 K, H₂ dissociation triggers the so-called second collapse, ending with the formation of the second hydrostatic core — the protostar — a few thousandths of an AU across.

From there the system evolves through the standard Class 0 → I → II → III sequence over roughly 10⁶ years. The Class 0 phase is invisible in optical light but characterised by strong submillimetre dust emission and powerful bipolar outflows; B335 is the prototype. As more of the envelope is accreted onto the central star, the system becomes detectable in the infrared (Class I) and eventually in the optical as a classical T Tauri star (Class II). The remaining globular envelope — initially a few solar masses — is mostly accreted onto the star, with the rest dispersed by stellar winds and outflows. The total time from pre-stellar globule to optical pre-main-sequence star is of order a million years.

Famous globules and where to find them

• Barnard 68 (B68). Ophiuchus, 125 pc, 0.1 pc, 2 M_⊙, pre-stellar. Textbook example.
• Barnard 335 (B335). Aquila, 100 pc, 0.05 pc, 1 M_⊙ core, Class 0 source. Outflow prototype.
• CB 26. Taurus, edge-on disk + jet + envelope, well-studied with mm interferometry.
• L1551 IRS 5. Taurus, embedded Class I binary, prominent jets.
• Thackeray's Globules. In IC 2944 (Centaurus); the famous "cometary" globules photographed by Hubble in 2002.
• Pillars of Creation (M16). Tip cores in the dense pillars of the Eagle Nebula — exceedingly dense globules being photoevaporated; many host EGGs (evaporating gaseous globules) at their tips.
• NGC 1999 / V380 Ori globule. A nearby globule once thought to contain a "hole" — actually a hole punched by a young stellar outflow.

How are globules studied?

Bok globules require multi-wavelength observation precisely because they are opaque in optical light.

• Optical imaging — to see the silhouette against the background and map the outer extinction structure.
• Near-infrared photometry — to count and redden background stars and reconstruct the column-density profile (the Alves method on B68).
• Submillimetre dust continuum — Herschel, JCMT/SCUBA-2, APEX/LABOCA: maps the cold dust emission, peaking near 200 μm for 12 K dust, and traces total column density.
• Molecular lines — IRAM 30 m, ALMA, NOEMA. C¹⁸O, N₂H⁺, NH₃, and DCO⁺ map the density structure and velocity field. Low J transitions are excited even at 10 K.
• Polarimetry — submillimetre polarisation traces magnetic field structure inside the globule.
• Infrared imaging — Spitzer, JWST: detects deeply embedded protostars at 24–70 μm even when optically obscured by 30 magnitudes.

Why globules are scientifically useful

• Isolated laboratories. A single, simple, near-spherical, gravitationally isolated cloud is the cleanest astrophysical environment in which to test theories of dense-core physics. B68 is the calibrator for column-density profiles, kinetic temperature, and chemistry models in cold molecular gas.
• Single-star ground truth. Most observed star formation occurs in clusters where stellar interactions and feedback are entangled. A Bok globule produces one or two stars in isolation, so the contributions of physics other than gravity, thermal pressure, magnetic support and the initial mass-to-flux ratio can be untangled.
• Cosmic chemistry. The cold, dense interior is the chemistry laboratory for grain-surface reactions, freeze-out of CO onto grain mantles, and deuteration enhancement at low temperatures.
• Binary formation. Class 0 binaries discovered in globules (B335, CB 26, IRAS 16293) constrain when in the collapse sequence companions appear.

Common pitfalls

• Calling every dark patch a globule. Many dark patches are simply unrelated foreground dust lanes, fragments of larger clouds, or projection effects in confused fields. The defining features are isolation, compactness, and high column density.
• Assuming all globules are pre-stellar. Many already contain a Class 0/I protostar; the embedded source is invisible in optical light, so the globule still looks dark.
• Treating cometary globules like spherical ones. The physics is different: photoevaporation pressure and radiation-driven implosion drive a much faster collapse history than slow isothermal contraction.
• Underestimating dust opacity at the centre. A_V > 30 mag is not unusual; the central density and temperature must be probed at long wavelengths.
• Confusing B68's ξ ≈ 7 with proof of collapse. The configuration is unstable, but the actual collapse can still proceed slowly because magnetic support and ambipolar diffusion limit the inflow rate.