News & Views | Published:


The dark cradles of stars

Nature volume 409, pages 140141 (11 January 2001) | Download Citation


Stars are born in dark clouds of molecular gas, which remain shrouded in mystery. Astronomers have found a new way to peer inside these star factories.

One of the success stories of twentieth-century astronomy was the development of a physical understanding of the nature, evolution and death of stars — replacing thousands of years of mystical speculation with firm knowledge. But one piece of the puzzle is still missing: how stars are born. Hidden in the dark interiors of molecular clouds lie the birthplaces of stars, inaccessible at optical wavelengths to the probing eyes of astronomers. Technological breakthroughs in detectors operating at infrared and sub-millimetre wavelengths have improved our picture of stellar genesis in the past two decades, but it is still too early to declare that we understand how stars, including our own Sun, were formed. On page 159 of this issue, João Alves and colleagues1 bring our knowledge of pre-stellar processes an important step forward.

It is widely agreed that most stars in the Galaxy are born in giant molecular clouds. Hundreds to several thousand stars are produced in these star factories, which convert tenuous gases into objects dense and hot enough to ignite nuclear processes. In the course of a few million years, roughly 10% of a molecular cloud may be turned into stars. But what conditions determine whether a particular clump of gas and dust will become gravitationally unstable and form a star? Such information is critical for calculations of the early stages of star formation.

Alves and co-workers imaged a type of dark cloud known as a Bok globule, at optical and infrared wavelengths, by using telescopes of the European Southern Observatory, including the world's biggest, the Very Large Telescope in the Chilean Atacama desert. Although most stars are born in giant molecular clouds, these enormous stellar nurseries are a mixture of complex environments that are not readily disentangled. Bok globules, on the other hand, are relatively simple, tiny, dense blobs of gas and dust, which can completely blot out the visible light of background stars (see Fig. 1 on page 159). When the eighteenth-century astronomer Sir William Herschel first encountered a Bok globule in his telescope, he exclaimed: “Mein Gott, da ist ein Loch in Himmel” (“My God, there is a hole in the skies”)2. Bok globules are thought to be the original dense cores of larger molecular clouds, which have been broken up by powerful radiation from nearby massive stars3. The destruction of the rest of the molecular cloud allows us a clear view of the surviving cloud core.

Figure 1: A star-forming Bok globule (GDC1).
Figure 1

GDC1 is similar to the B68 globule studied by Alves et al.1, but it has already passed through the collapse phase and formed a Sun-like star in its interior. Although the newborn star is still embedded within thick layers of gas and dust, it reveals its existence by spewing a two-sided supersonic jet into the surroundings. Such jets rid the forming star of rapidly rotating material that would otherwise hinder its formation. Jets also help the newborn star to blow away the remnant cloud material8. Image: BO REIPURTH/ESO

There have been many previous attempts to probe the structure and physical conditions of pre-stellar cores of gas and dust by using modern radio telescopes operating at millimetre wavelengths, which are sensitive to emissions from complex molecules4,5. These and other studies have shown that the main ingredient of such cores is molecular hydrogen (H2) mixed with small but important traces of heavier molecules, along with a sprinkling of dust particles. We also know that temperatures are very low, of the order of 10 degrees above absolute zero. The gas densities of 104–10 5 cm−3 appear high relative to the interstellar environment, but are still at least twenty orders of magnitude smaller than that of their ultimate destination inside stars. Because hydrogen cannot be observed directly at millimetre wavelengths, other trace gases are used to infer gas densities. Unfortunately, this introduces uncertainties in the physical parameters of cloud cores derived from millimetre observations.

In their study, Alves et al.1 have exploited the fact that at infrared wavelengths (a few micrometres) a Bok globule becomes more transparent, even though it is virtually opaque at optical wavelengths. By combining optical and infrared images they measure the degree to which dust particles in the Bok globule B68 block out the starlight of background stars seen through the cloud. They chose to study B68 because it is an almost spherical, isolated globule that is viewed against the rich stellar backdrop near the Galactic centre. So although it is small (0.4 light years in diameter), the light of thousands of stars is attenuated while passing through the dusty globule.

In the same way that the light from the setting Sun becomes redder as it passes through more of the Earth's atmosphere, starlight is reddened as it passes through B68. Alves et al.1 construct a detailed map of how the reddening of light from nearly 4,000 stars changes across the globule, from which they determine the profile of the dust density through the cloud with unprecedented resolution and signal-to-noise ratio. By assuming that the dust is mixed throughout the gas in the cloud, and that the gas-to-dust ratio is 100:1, Alves et al. get a good picture of the internal structure of this gas globule. Remarkably, they find a close correspondence between the observed dust profile and theoretical predictions for the structure of a certain type of gas sphere that is stable but close to collapse6. With a total mass of only twice that of the Sun, B68 appears to be a prime candidate for a pre-stellar cloudlet.

The formation and evolution of stars represent a slow but inexorable effort by gravity to compress interstellar gas into the increasingly high densities encountered in stellar interiors, and ultimately into the extreme densities of white dwarfs, and even into neutron stars or black holes. B68 seems to be hovering near the brink of gravitational contraction and will most likely collapse to form a star (Fig. 1), unless stabilized by other forces. Despite its low temperature of 16 K, the primary stabilizing force is thermal pressure. In addition, it is likely that the globule possesses a weak magnetic field, which could help stabilize it further7. Nonetheless, B68 appears to be only marginally stable, and so could easily be destabilized by physical processes, such as further cooling or a drop in the internal magnetic pressure, or by being hit by a travelling blast wave from a supernova explosion. So Alves and co-workers predict that the exquisite dark silhouette of B68 may well, sometime in the future, be converted into yet another little shining star in the Milky Way.


  1. 1.

    , & Nature 409, 159– 161 (2001).

  2. 2.

    Pub. Astron. Soc. Pacif. 89, 597–611 (1977).

  3. 3.

    Astron. Astrophys. 117, 183–198 (1983).

  4. 4.

    & Astrophys. J. (suppl.) 71, 89–108 (1989).

  5. 5.

    & Astrophys. J. 518, L37–L40 ( 1999).

  6. 6.

    An Introduction to the Study of Stellar Structure (Univ. Chicago Press, 1939).

  7. 7.

    Pub. Astron. Soc. Jpn 31, 697–712 (1979).

  8. 8.

    et al. Astron. J. 112, 1141– 1168 (1996).

Download references

Author information

Corresponding author

Correspondence to Bo Reipurth.

About this article

Publication history




  1. Search for Bo Reipurth in:


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing