Molecular clouds and protostellar evolution: Jeans collapse, the Hayashi track, and the H-R diagram pre-main-sequence

Intuition Beginner

Stars are born in the coldest, densest gas in the galaxy: clouds of molecular hydrogen at roughly 10 Kelvin, colder than liquid air. Inside such a cloud two forces fight. Gravity pulls the gas inward, trying to gather it into a star; the gas's own pressure (plus magnetic fields and turbulence) pushes outward, trying to disperse it. When a patch becomes cold enough and massive enough, gravity wins and the region collapses under its own weight. A protostar forms at the centre, a hot ball of gas still swallowing material from the surrounding envelope.

As the protostar shrinks it heats up, and its changing temperature and brightness trace a path across the Hertzsprung-Russell diagram, a plot of temperature against luminosity. First it slides almost straight down at roughly constant surface temperature: the Hayashi track, when the star is fully convective and churns like boiling water. Then it drifts left and slightly down: the Henyey track, as a radiative core develops and energy starts flowing outward as radiation. When the centre reaches about 10 million Kelvin, hydrogen fusion ignites. The star has arrived on the main sequence.

Why care about any of this? Star formation sets the tempo of every galaxy, builds the planetary systems around new stars, and supplies the raw material for every later generation. Understanding how a cold gas cloud becomes a burning star is the first step in explaining where stars, planets, and the elements come from.

Visual Beginner

The diagram sketches the journey from diffuse molecular cloud to main-sequence star. A cold dense core crosses the Jeans threshold and collapses isothermally; the protostar heats through the Class 0, I, II, III sequence; and its track across the H-R diagram descends the near-vertical Hayashi arm before turning along the shallower Henyey arm to land on the main sequence at a position set by the star's mass.

Below the main-sequence arrival sits the brown-dwarf regime: objects too light to ignite hydrogen, which cool forever after a brief deuterium-burning phase.

Worked example Beginner

The Orion Nebula cluster, about 1344 light-years from Earth, is the nearest region where massive stars are forming. The Orion Molecular Cloud behind the glowing nebula has given birth to roughly 700 young stars, each less than one million years old, along with about 100 protostars still wrapped in their envelopes and about 50 brown dwarfs, objects too light to fuse hydrogen.

Step 1. The four brightest stars at the cluster's heart, the Trapezium, are hot O-type stars whose ultraviolet light strips electrons from the surrounding gas and makes the nebula glow. The most massive of them, Theta-1 Orionis C, is about 40 times the Sun's mass and ionises a bubble dozens of light-years across.

Step 2. Using the Hubble Space Telescope and the James Webb Space Telescope, astronomers have imaged about 200 protoplanetary disks (proplyds) around the young cluster stars. Some are being scorched away by the Trapezium's radiation; others, deeper in the cloud, still hold the dust and gas from which planets will form.

Step 3. Counting the 50 brown dwarfs among the 850 young objects gives a brown-dwarf fraction near 6 percent. The boundary at about 0.08 solar masses separates these failed stars from true hydrogen-burning stars. The lighter objects cool forever; the heavier ones ignite and shine for billions of years.

What this tells us: the Orion Nebula shows every stage of star birth side by side, cold cloud cores, embedded protostars, illuminated disks, and a full range of stellar masses from 40 solar masses down to brown dwarfs.

Check your understanding Beginner

Formal definition Intermediate+

The formation of an individual star is governed by the competition between self-gravity and pressure support in a cold molecular cloud, followed by an accretion phase during which the central object contracts and heats until hydrogen ignition. Following Stahler and Palla ^{[Stahler-Palla 2004]}, the relevant quantities are the Jeans length and Jeans mass for the initial collapse, the Class 0/I/II/III sequence for the embedded protostellar phases, and the Hayashi and Henyey tracks for the approach to the main sequence.

Definition (Jeans length and Jeans mass). For a uniform isothermal gas of density $\rho$ and sound speed $c_s$, the Jeans length and Jeans mass are

$${\lambda }_J=c_s\sqrt{\frac{\pi }{G\rho }},M_J=\frac{{\pi }^{5/2}}{6}\frac{c_s^3}{\sqrt{G^3\rho }}.$$

Regions with size larger than ${\lambda }_J$ or mass larger than $M_J$ are gravitationally unstable and collapse. Numerically, with $T=10$ K, $\mu =2.3$, and number density $n=100$ cm${}^{-3}$, the sound speed is $c_s\approx 0.19$ km/s and $M_J$ is of order $10^5{M}_{\odot }$ (giant molecular clouds); at the dense-core density $n=10^4$ cm${}^{-3}$, $M_J$ falls to roughly $1M_{\odot }$ (individual stars).

Definition (Class 0/I/II/III protostellar sequence). Following Adams, Lada, and Shu ^{[AdamsLadaShu1987 ApJ 312:788]} and the Class 0 extension of Andre, Ward-Thompson, and Barsony ^{[Andre1993 ApJ 406:122]}, young stellar objects are classified by the slope of the spectral energy distribution across the near-infrared, equivalently by the ratio of envelope to disk to stellar mass:

Class	Age (yr)	Signature	Dominant component
0	$\sim 10^4$	submillimeter only; no optical or near-IR	envelope $>$ star
I	$\sim 10^5$	bipolar outflow, rising IR SED	star + envelope + disk
II	$\sim 10^6$	classical T Tauri / Herbig Ae/Be; infrared excess	star + disk
III	$\sim 10^7$	weak-lined T Tauri; little IR excess	star, disk dissipating

Definition (Hayashi and Henyey tracks). The Hayashi track ^{[Hayashi1961 PASJ 13:450]} is the vertically descending pre-main-sequence path in the H-R diagram traced by a fully convective star contracting at roughly constant effective temperature $T_{\text{eff}}\approx 3000$ to $4500$ K. The Henyey track ^{[Henyey1955 PASP 67:154]} is the subsequent leftward drift in the H-R diagram taken when a radiative core develops and the star moves to higher $T_{\text{eff}}$ at nearly constant luminosity, ending on the main sequence at the ZAMS position set by its mass.

Counterexamples to common slips Intermediate+

• "Any cloud will collapse if it is dense enough." No. Magnetic pressure and turbulent support can hold a cloud well above its Jeans mass. The magnetic critical mass $M_{\Phi }\approx \Phi /(2\pi \sqrt{G})$ (the Mouschovias-Spitzer criterion) sets a separate threshold; clouds with mass-to-flux ratio below critical remain magnetically subcritical and do not collapse unless ambipolar drift decouples the neutrals.

  Feedback from existing stars (photoionisation, supernovae, winds) further disrupts clouds and can unbind material that would otherwise have collapsed.

• "More massive stars form faster." True at the low-mass end, where accretion time $t_{\text{acc}}\sim M/\dot{M}$ at fixed $\dot{M}$ grows with mass. The trend reverses at the high-mass end: once the protostar exceeds roughly $8M_{\odot }$, radiation pressure on dust halts spherical accretion. Massive stars therefore form through disk accretion and bipolar outflows (the McKee-Tan turbulent-core model), not by simply scaling up the low-mass recipe.

• "The Hayashi track is a sequence of different stars." No. It is the path in the H-R diagram of a single star evolving as it contracts. Different masses trace parallel Hayashi tracks; each star descends its own.

• "Brown dwarfs are just small stars." No. By the standard definition a star fuses hydrogen; brown dwarfs do not. The deuterium-burning threshold at $0.013M_{\odot }$ (13 Jupiter masses) sets the planet-star boundary, but the hydrogen-burning threshold at $0.08M_{\odot }$ sets the star boundary.

• "All protostellar disks go on to form planets." No. Disk dissipation times are 3 to 10 million years, observed through the infrared-excess lifetime of T Tauri stars. Planet formation must beat that clock; many disks are photoevaporated or stripped by close encounters before planetesimal growth completes.

Key theorem with derivation Intermediate+

Theorem (Jeans 1902: gravitational instability of a self-gravitating isothermal gas). For a uniform isothermal gas of density ${\rho }_0$ and sound speed $c_s$, linear perturbations ${\rho }_1\propto \exp (i(\mathbf{k}\cdot \mathbf{x}-\omega t))$ obey the dispersion relation

$${\omega }^2=c_s^2{k}^2-4\pi G{\rho }_0.$$

Modes with $k<k_J=\sqrt{4\pi G{\rho }_0}/c_s$ have ${\omega }^2<0$ and grow exponentially, signalling gravitational collapse.

Proof. Begin with the equations of inviscid self-gravitating hydrodynamics, namely the continuity equation, the Euler equation, and Poisson's equation for the gravitational potential,

$$\frac{\partial \rho }{\partial t}+\nabla \cdot (\rho \mathbf{v})=0,$$ $$\rho \left(\frac{\partial \mathbf{v}}{\partial t}+(\mathbf{v}\cdot \nabla )\mathbf{v}\right)=-\nabla p-\rho \nabla \Phi ,$$ $${\nabla }^2\Phi =4\pi G\rho ,$$

closed by the isothermal equation of state $p=c_s^2\rho$. Linearise about a static uniform background $({\rho }_0,\mathbf{v}=0,{\Phi }_0)$ by writing $\rho ={\rho }_0+{\rho }_1$, $\mathbf{v}={\mathbf{v}}_1$, $\Phi ={\Phi }_0+{\Phi }_1$ with all perturbed quantities small. To first order the equations become

$$\frac{\partial {\rho }_1}{\partial t}+{\rho }_0\nabla \cdot {\mathbf{v}}_1=0,$$ $$\frac{\partial {\mathbf{v}}_1}{\partial t}=-\frac{c_s^2}{{\rho }_0}\nabla {\rho }_1-\nabla {\Phi }_1,$$ $${\nabla }^2{\Phi }_1=4\pi G{\rho }_1.$$

(The background gradient of ${\Phi }_0$ is dropped by the standard Jeans swindle: a strictly uniform static density is inconsistent with Poisson's equation, but the local analysis of perturbations is correct for scales much smaller than the self-gravitating scale of the cloud.)

Take the time derivative of the continuity equation and substitute the Euler equation,

$$\frac{{\partial }^2{\rho }_1}{\partial t^2}=-{\rho }_0\nabla \cdot \frac{\partial {\mathbf{v}}_1}{\partial t}=-{\rho }_0\nabla \cdot \left(-\frac{c_s^2}{{\rho }_0}\nabla {\rho }_1-\nabla {\Phi }_1\right)=c_s^2{\nabla }^2{\rho }_1+{\rho }_0{\nabla }^2{\Phi }_1.$$

Using Poisson's equation to replace ${\nabla }^2{\Phi }_1=4\pi G{\rho }_1$,

$$\frac{{\partial }^2{\rho }_1}{\partial t^2}=c_s^2{\nabla }^2{\rho }_1+4\pi G{\rho }_0{\rho }_1.$$

Insert the plane-wave ansatz ${\rho }_1\propto \exp (i(\mathbf{k}\cdot \mathbf{x}-\omega t))$, so that ${\partial }_t^2\to -{\omega }^2$ and ${\nabla }^2\to -k^2$. The result is

$$-{\omega }^2=-c_s^2{k}^2+4\pi G{\rho }_0,$$

equivalently ${\omega }^2=c_s^2{k}^2-4\pi G{\rho }_0$. Setting $\omega =0$ gives the critical wavenumber $k_J=\sqrt{4\pi G{\rho }_0}/c_s$ and wavelength ${\lambda }_J=2\pi /k_J=c_s\sqrt{\pi /(G{\rho }_0)}$. For $k<k_J$ the perturbation grows exponentially with growth rate $\gamma =\sqrt{4\pi G{\rho }_0-c_s^2{k}^2}$.

The Jeans mass follows by taking the mass enclosed in a sphere of diameter ${\lambda }_J$,

$$M_J=\frac{4\pi }{3}{\left(\frac{{\lambda }_J}{2}\right)}^3{\rho }_0=\frac{{\pi }^{5/2}}{6}\frac{c_s^3}{\sqrt{G^3{\rho }_0}},$$

as required. With $c_s^2=k_{B}T/(\mu m_{\mathrm{H}})$ this scales as $M_J\propto T^{3/2}{\rho }_0^{-1/2}$: colder and denser gas has a lower Jeans mass, which is why star formation is confined to the cold molecular phase 28.07.01. $\square$

Bridge. The Jeans criterion builds toward the inside-out collapse solution of Shu ^{[Shu1977 ApJ 214:488]} derived in the Full proof set below, in which a rarefaction wave propagates outward at the sound speed and material inside it free-falls onto the central protostar, and appears again in 28.02.01 as the condition separating pressure-supported molecular gas from main-sequence stars on the H-R diagram. The foundational reason gravitational instability is governed by the ratio of the sound-crossing time to the free-fall time is exactly the competition between pressure response and free fall that the dispersion relation encodes, and this is exactly the bridge to the Class 0/I/II/III sequence: isothermal collapse continues until the rising opacity traps the contraction heat, the gas becomes adiabatic, a first hydrostatic core forms, and the object begins its descent down the Hayashi track toward the main sequence.

Exercises Intermediate+

Advanced results Master

Theorem 1 (Jeans 1902: gravitational instability). Jeans's linearisation of the continuity, Euler, and Poisson equations for a self-gravitating isothermal gas ^{[Jeans1902 Phil. Trans. R. Soc. A 199]} yields the dispersion relation ${\omega }^2=c_s^2{k}^2-4\pi G{\rho }_0$ derived above, with the critical wavenumber $k_J$ below which pressure cannot oppose collapse. The Jeans length ${\lambda }_J$ and Jeans mass $M_J$ remain the canonical thresholds for gravitational instability in modern star-formation theory, extended but not superseded by magnetic and turbulent modifications.

Theorem 2 (Larson 1969: the isothermal sphere and Larson's relations). Larson's numerical integration of the self-gravitating collapse equations ^{[Larson1969 MNRAS 145:271]} showed that a roughly uniform cloud collapses from the inside out, with the interior approaching the singular isothermal sphere $\rho (r)=c_s^2/(2\pi G{r}^2)$ before a central hydrostatic core forms. The same paper established three empirical scaling relations among molecular cloud properties (size-line-width, size-density, and mass-size), now known as Larson's laws, which underlie the modern picture of supersonic turbulent support.

Theorem 3 (Hayashi 1961: the Hayashi track and forbidden region). Hayashi proved that a fully convective star of given mass and composition admits hydrostatic-equilibrium solutions only above a minimum effective temperature T_\text{eff, \min}, the Hayashi line ^{[Hayashi1961 PASJ 13:450]}. The region to the cool side of this line is forbidden: no hydrostatic star can sit there. A contracting protostar that is fully convective must therefore move down the Hayashi line at nearly constant $T_{\text{eff}}$, with $L$ falling as $R^2$, until a radiative core develops.

Theorem 4 (Henyey 1955: the radiative-core Henyey track). Once the central temperature rises enough that radiative transport replaces convection in the core, the pre-main-sequence star leaves the Hayashi line and moves leftward in the H-R diagram along the Henyey track ^{[Henyey1955 PASP 67:154]} at nearly constant luminosity. The star reaches the zero-age main sequence when the core temperature crosses the hydrogen-ignition threshold near $10^7$ K.

Theorem 5 (Shu 1977: inside-out collapse). Shu's self-similar solution ^{[Shu1977 ApJ 214:488]} begins from the singular isothermal sphere and shows that a rarefaction wave propagates outward from the centre at exactly the sound speed $c_s$. Inside the wave front, material is in free fall onto a central point mass; outside, the gas remains static. The accretion rate is constant at $\dot{M}=0.975c_s^3/G$, the canonical result for isolated low-mass star formation.

Theorem 6 (Adams-Lada-Shu 1987, Andre 1993: the Class 0/I/II/III sequence). Adams, Lada, and Shu organised the infrared spectral-energy-distribution slopes of embedded sources into the Class I, II, III sequence ^{[AdamsLadaShu1987 ApJ 312:788]}. Andre, Ward-Thompson, and Barsony added the submillimeter-bright, optically invisible Class 0 phase ^{[Andre1993 ApJ 406:122]}, in which the envelope mass exceeds the central mass. The four classes track a single star's evolution through $\sim 10^4$ to $\sim 10^7$ yr, from first hydrostatic core to disk dissipation.

Theorem 7 (McKee and Tan 2003: the turbulent core model). Massive star formation cannot proceed by the spherical accretion of low-mass theory, because radiation pressure on dust reverses the flow for protostars above $8M_{\odot }$. McKee and Tan ^{[McKeeTan2003 ApJ 585:850]} proposed that massive stars form from turbulent, magnetised cores in which accretion is channelled through a disk while radiation escapes along the polar axes, driving bipolar outflows. The model reproduces the observed massive-star formation rate and explains the upper-initial-mass-function cutoff near $150M_{\odot }$.

Theorem 8 (ALMA Partnership 2015: protoplanetary gaps in HL Tau). The ALMA long-baseline image of HL Tau at 1-3 mm wavelength ^{[ALMA_HLTau2015 ApJ 808:L3]} resolved the Class II protoplanetary disk into a series of concentric bright rings and dark gaps. The gap radii are consistent with the clearing action of nascent planets of Neptune to Jupiter mass on timescales of $10^4$ to $10^5$ yr, providing the first direct images of ongoing planet formation inside a protoplanetary disk and confirming that Class II disks are the sites of planetary system assembly.

Synthesis. The pre-main-sequence framework is the foundational reason that every star in the galaxy can be placed on a single evolutionary track from cold molecular gas to hydrogen ignition, and the central insight is that the Jeans criterion, the Hayashi convective track, and the Class 0/I/II/III sequence partition that journey into four physically distinct regimes. The Coulomb-barrier ignition threshold of 28.02.05 is exactly the wall the protostar must climb through to arrive on the main sequence, and putting these together identifies the pre-main-sequence tracks with the observed populations of embedded submillimeter sources, T Tauri stars, and Herbig Ae/Be stars across the H-R diagram. The bridge is between the isothermal gas dynamics of the Jeans collapse and the radiative-transfer regime of the Hayashi forbidden region, and the pattern generalises to the high-mass turbulent-core model of McKee and Tan, where radiation pressure halts spherical accretion above roughly eight solar masses and the formation channel switches to disk-fed infall with bipolar escape. The sequence appears again in 13.08.02 in its primordial Pop III form, where zero-metallicity gas collapses at $T\sim 200$ K (hydrogen molecular cooling) and the Jeans mass is orders of magnitude larger, producing the first massive stars.

Full proof set Master

Proposition 1 (Jeans-mass scaling). With $c_s^2=k_{B}T/(\mu m_{\mathrm{H}})$ the isothermal sound speed, the Jeans mass scales as $M_J\propto T^{3/2}{\rho }_0^{-1/2}$.

Proof. Starting from the formula

$$M_J=\frac{{\pi }^{5/2}}{6}\frac{c_s^3}{\sqrt{G^3{\rho }_0}},$$

substitute $c_s^2=k_{B}T/(\mu m_{\mathrm{H}})$ so that $c_s^3=(k_{B}T/(\mu m_{\mathrm{H}}){)}^{3/2}$,

$$M_J=\frac{{\pi }^{5/2}}{6G^{3/2}}\frac{1}{\sqrt{{\rho }_0}}{\left(\frac{k_B}{\mu m_{\mathrm{H}}}\right)}^{3/2}{T}^{3/2}.$$

All prefactors are dimensionful constants, leaving $M_J\propto T^{3/2}{\rho }_0^{-1/2}$ as claimed. Two consequences follow. First, isothermal collapse is self-similar: as ${\rho }_0$ rises during contraction, $M_J$ falls, so parcels of gas that were stable at the parent cloud density become unstable at the compressed density, producing hierarchical fragmentation. Second, heating the gas (by stellar feedback, cosmic rays, or shocks) raises $M_J$, suppressing further star formation, the basis of feedback regulation of galactic star-formation rates. $\square$

Proposition 2 (free-fall time of a uniform sphere). A pressure-free uniform sphere of initial density ${\rho }_0$ collapses to a point in time

$$t_{\text{ff}}=\sqrt{\frac{3\pi }{32G{\rho }_0}}.$$

Proof. Consider a thin spherical shell at initial radius $r_0$ in a uniform sphere of density ${\rho }_0$. By Newton's shell theorem the shell feels only the mass interior to it, $M(r)=(4\pi /3)r_0^3{\rho }_0$, which is constant during the collapse. The equation of motion is

$$\frac{d^{2}r}{d{t}^2}=-\frac{GM}{r^2}.$$

Multiply both sides by $dr/dt$ and integrate once,

$$\frac{1}{2}{\left(\frac{dr}{dt}\right)}^2=GM\left(\frac{1}{r}-\frac{1}{r_0}\right),$$

using the boundary condition $dr/dt=0$ at $r=r_0$. Taking the negative root (the shell moves inward),

$$\frac{dr}{dt}=-\sqrt{2GM}\sqrt{\frac{1}{r}-\frac{1}{r_0}}.$$

Substitute $r=r_0{\cos }^2\theta$, so $dr=-2r_0\cos \theta \sin \theta d\theta$ and $1/r-1/r_0={\tan }^2\theta /r_0$. Then

$$\frac{dr}{dt}=\frac{dr/d\theta }{dt/d\theta }=-2r_0\cos \theta \sin \theta \frac{d\theta }{dt},$$

and equating,

$$2r_0\cos \theta \sin \theta \frac{d\theta }{dt}=\sqrt{\frac{2GM}{r_0}}\tan \theta .$$

Cancelling $\sin \theta$ and rearranging,

$$dt=\frac{2r_0{\cos }^2\theta }{\sqrt{2GM/r_0}}d\theta =\sqrt{\frac{2r_0^3}{GM}}{\cos }^2\theta d\theta .$$

Integrate from $\theta =0$ (the initial state $r=r_0$) to $\theta =\pi /2$ (the final state $r=0$), using ${\int }_0^{\pi /2}{\cos }^2\theta d\theta =\pi /4$,

$$t_{\text{ff}}=\sqrt{\frac{2r_0^3}{GM}}\frac{\pi }{4}=\frac{\pi }{4}\sqrt{\frac{2r_0^3}{G(4\pi /3)r_0^3{\rho }_0}}=\sqrt{\frac{3\pi }{32G{\rho }_0}},$$

independent of $r_0$ as expected for a uniform density field. $\square$

Numerically, $t_{\text{ff}}\approx 1.0\times 10^7{n}^{-1/2}$ yr with $n$ in cm${}^{-3}$: a giant molecular cloud at $n=100$ cm${}^{-3}$ has $t_{\text{ff}}\approx 10^6$ yr, while a dense core at $n=10^4$ cm${}^{-3}$ has $t_{\text{ff}}\approx 10^5$ yr. The free-fall time sets the natural clock for the Class 0 and Class I phases.

Connections Master

• The interstellar medium and star formation survey 28.07.01. This unit deepens the molecular-cloud content of the survey by deriving the Jeans criterion from the linearised self-gravitating fluid equations, computing the Class 0/I/II/III sequence, and tracing the Hayashi and Henyey tracks across the H-R diagram. Each cold molecular phase in the survey's ISM table is the upstream reservoir from which the dense cores treated here fragment and collapse.

• Stellar nucleosynthesis: the B²FH process network 28.02.05. The protostellar evolution described here is the prequel to stellar nucleosynthesis. Once a protostar descends the Hayashi track and reaches the main sequence with a core temperature near $10^7$ K, the proton-proton chain ignites and the B²FH burning-stage sequence of 28.02.05 takes over. The ignition threshold at $0.08M_{\odot }$ is precisely the brown-dwarf boundary derived in the Formal definition above; below it the B²FH sequence never starts.

• Stars and stellar evolution survey 28.02.01. The H-R diagram introduced in 28.02.01 is the canvas on which the Hayashi and Henyey tracks are drawn. The full stellar lifecycle of 28.02.01 begins where the pre-main-sequence tracks of this unit terminate: arrival on the zero-age main sequence. The brown-dwarf regime treated here occupies the lower-left quadrant of the H-R diagram and cools indefinitely rather than ascending the red-giant branch.

• Cosmology — FLRW, inflation, nucleosynthesis, CMB, and structure 13.08.02. Primordial (Population III) star formation bridges cosmology and stellar physics. After recombination, zero-metallicity gas cooled by trace molecular hydrogen collapsed at temperatures near 200 K with Jeans masses of order $10^3{M}_{\odot }$, far above the present-day value. The first stars were therefore massive and short-lived, reionising the universe and seeding the metals that enabled the present-day molecular cooling channels in this unit. The Jeans criterion is the same; only the cooling physics and metallicity differ.

• Exoplanet detection methods and habitability 28.05.01. The Class II disk phase of Adams-Lada-Shu and the HL Tau disk gaps imaged by ALMA are the birth sites of the planetary systems whose detection and habitability are treated in 28.05.01. The 3-10 Myr disk-dissipation timescale constrains planet-formation models, and the protoplanetary disk masses and architectures inferred from submillimeter imaging directly shape the exoplanet occurrence rates measured by Kepler.

Historical & philosophical context Master

James Jeans established the gravitational-instability criterion in The Stability of a Spherical Nebula ^{[Jeans1902 Phil. Trans. R. Soc. A 199]}, linearising the continuity, Euler, and Poisson equations about a uniform self-gravitating gas and obtaining the critical length scale ${\lambda }_J=c_s\sqrt{\pi /(G\rho )}$ below which pressure dominates and above which gravity runs away. The criterion has been refined but never replaced: magnetic (Mouschovias and Spitzer 1976) and turbulent (McKee and Tan ^{[McKeeTan2003 ApJ 585:850]}) modifications add support terms but preserve the basic competition between sound-crossing and free-fall times.

Bart Bok and Edith Reilly ^{[BokReilly1947 ApJ 105:255]} identified the small, dark, round molecular clouds now called Bok globules as the likely sites of individual star formation, a conjecture confirmed by infrared and submillimeter surveys half a century later. Chushiro Hayashi then proved the existence of the forbidden region to the cool side of the convective Hayashi line ^{[Hayashi1961 PASJ 13:450]}, showing that pre-main-sequence stars are constrained to a nearly vertical track in the H-R diagram and resolving the long-standing puzzle of how young low-mass stars could be so luminous while still cool.

The numerical collapse calculations of Richard Larson ^{[Larson1969 MNRAS 145:271]} established the inside-out character of gravitational collapse and produced the empirical scaling relations that bear his name. Frank Shu then provided the analytic self-similar solution ^{[Shu1977 ApJ 214:488]} in which a rarefaction wave propagates outward at the sound speed, giving the canonical constant accretion rate $\dot{M}=0.975c_s^3/G$ for isolated low-mass star formation. The protostellar classification sequence was codified by Adams, Lada, and Shu ^{[AdamsLadaShu1987 ApJ 312:788]} and extended to the earliest Class 0 phase by Andre, Ward-Thompson, and Barsony's submillimeter detections ^{[Andre1993 ApJ 406:122]}. The ALMA long-baseline campaign image of HL Tau ^{[ALMA_HLTau2015 ApJ 808:L3]} closed the loop between protostellar disks and planetary systems by resolving the gap structure produced by forming planets inside a Class II disk.

Bibliography Master

@article{Jeans1902,
  author  = {Jeans, James H.},
  title   = {The Stability of a Spherical Nebula},
  journal = {Philosophical Transactions of the Royal Society of London, Series A},
  volume  = {199},
  pages   = {1--53},
  year    = {1902},
  doi     = {10.1098/rsta.1902.0012},
}

@article{BokReilly1947,
  author  = {Bok, Bart J. and Reilly, Edith F.},
  title   = {Small Dark Nebulae},
  journal = {Astrophysical Journal},
  volume  = {105},
  pages   = {255--257},
  year    = {1947},
  doi     = {10.1086/144901},
}

@article{Henyey1955,
  author  = {Henyey, Louis G. and Lelevier, Robert and Levee, Ronald D.},
  title   = {The Early Phases of Stellar Evolution},
  journal = {Publications of the Astronomical Society of the Pacific},
  volume  = {67},
  pages   = {154--164},
  year    = {1955},
  doi     = {10.1086/126791},
}

@article{Hayashi1961,
  author  = {Hayashi, Chushiro},
  title   = {Stellar Evolution in Early Phases of Gravitational Contraction},
  journal = {Publications of the Astronomical Society of Japan},
  volume  = {13},
  pages   = {450--452},
  year    = {1961},
}

@article{Larson1969,
  author  = {Larson, Richard B.},
  title   = {Numerical Calculations of the Dynamics of a Collapsing Proto-Star},
  journal = {Monthly Notices of the Royal Astronomical Society},
  volume  = {145},
  pages   = {271--295},
  year    = {1969},
  doi     = {10.1093/mnras/145.4.271},
}

@article{Shu1977,
  author  = {Shu, Frank H.},
  title   = {Self-Similar Collapse of Isothermal Spheres and Star Formation},
  journal = {Astrophysical Journal},
  volume  = {214},
  pages   = {488--497},
  year    = {1977},
  doi     = {10.1086/155274},
}

@article{AdamsLadaShu1987,
  author  = {Adams, Fred C. and Lada, Charles J. and Shu, Frank H.},
  title   = {Spectral Evolution of Young Stellar Objects},
  journal = {Astrophysical Journal},
  volume  = {312},
  pages   = {788--806},
  year    = {1987},
  doi     = {10.1086/164924},
}

@article{Andre1993,
  author  = {Andre, Philippe and Ward-Thompson, Derek and Barsony, Mary},
  title   = {Submillimeter Continuum Observations of Rho Ophiuchi A: The Candidate Protostar VLA 1623 and Early Stelllar Evolution},
  journal = {Astrophysical Journal},
  volume  = {406},
  pages   = {122--141},
  year    = {1993},
  doi     = {10.1086/172425},
}

@article{McKeeTan2003,
  author  = {McKee, Christopher F. and Tan, Jonathan C.},
  title   = {The Formation of Massive Stars from Turbulent Cores},
  journal = {Astrophysical Journal},
  volume  = {585},
  pages   = {850--871},
  year    = {2003},
  doi     = {10.1086/346149},
}

@article{ALMA_HLTau2015,
  author  = {{ALMA Partnership} and Brogan, C. L. and P{\'e}rez, L. M. and Hunter, T. R. and Dent, W. R. F. and others},
  title   = {The 2014 ALMA Long Baseline Campaign: First Results from High Angular Resolution Observations toward HL Tau},
  journal = {Astrophysical Journal Letters},
  volume  = {808},
  pages   = {L3},
  year    = {2015},
  doi     = {10.1088/2041-8205/808/1/L3},
}

@book{StahlerPalla2004,
  author    = {Stahler, Steven W. and Palla, Francesco},
  title     = {The Formation of Stars},
  publisher = {Wiley-VCH},
  year      = {2004},
}

@book{WardThompsonWhitworth2011,
  author    = {Ward-Thompson, Derek and Whitworth, Anthony P.},
  title     = {An Introduction to Star Formation},
  publisher = {Cambridge University Press},
  year      = {2011},
}