All that lives and breathes on Earth owes its existence to the sun, that great power plant in the sky. But to what does the sun owe its existence?

The process by which the sun and its fellow stars form, transforming a cold cloud of dust and gas to a nuclear fusion reactor that burns for billions of years, is reasonably well understood, thanks to ever more refined theoretical work supported by sophisticated computer simulations and detailed observations of star-forming regions in space.

But not all stages of star formation have been directly observed. In particular, a fleeting phase known as the first hydrostatic core, or simply the first core, has so far escaped detection, even though models of how low-mass stars such as the sun form have long suggested its existence. But in recent months two groups have identified objects that may be nascent stars in the first core stage, both of them approximately 800 light-years away in the constellation Perseus. If either could be confirmed it would fill in an evidentiary gap, albeit a narrow one, in the star-formation process.

The first core of low-mass stars (those stars that will evolve to no more than a few times the mass of the sun), which has been predicted to exist by theorists writing as far back the 1960s, develops when a thick cloud of gas and dust begins to gravitationally fall in on itself, releasing heat as material is compressed at its heart. When the central region of the cloud becomes sufficiently dense and opaque the heat trapped within prevents further collapse, producing a relatively stable first core with temperatures of only about 100 kelvins (–175 degrees Celsius), large enough in size to engulf the planets of the inner solar system if it were in the sun's place. But although the infall slows with the formation of the first core, it does not stop completely, and the core's density and temperature continue to inch up over time.

At approximately 2,000 kelvins (1,700 degrees C), the core's predominant gas, hydrogen, breaks apart from molecules into individual atoms, and the core begins a more dramatic collapse. "All the energy that went into keeping it hot goes into breaking up the molecular hydrogen," explains Melissa Enoch, a postdoctoral astronomer at the University of California, Berkeley, who co-authored a study in the October 10 issue of The Astrophysical Journal Letters identifying a possible first core of a future star. "So it no longer has the ability to resist its collapse." What emerges from the collapse is much smaller, denser and hotter—a protostar.

That a first core has not been conclusively observed for any star by astronomers and their telescopes is no surprise: it lasts just tens of thousands of years, if that, and glows ever so faintly. "It's a very short-lived stage, and that's one of the reasons we hadn't observed it, and it radiates very tenuously in the infrared," says Héctor Arce, a Yale University astronomer who co-authored a paper in the June 1 issue of The Astrophysical Journal describing another object that fits the bill for a nascent star in the first core stage. "We see starless cores, cores that do not yet have protostars, and we see very young protostars, but we do not see this epoch in between."

Both groups looked for radiation from regions where there should be none—dense clouds that had previously been considered starless. In deep observations made by the Spitzer Space Telescope, Enoch's group identified infrared emission from a region known as Per-Bolo 58 at long infrared wavelengths of 70 microns (a micron is one millionth of a meter) and very little at shorter wavelengths, as would be expected from a cool first core.

Arce and his colleagues used a ground-based array of submillimeter telescopes to locate an outflow of molecular gas from an area called L1448 IRS2E. Some simulations have predicted such outflows should emanate from a first core as the rotating disk of material surrounding the core interacts with the object's magnetic field. The researchers did not detect the candidate object at 70 microns; they had access only to a broad-ranging Spitzer survey of the area and not the kind of deep observations of Per-Bolo 58 attained by Enoch's group in 2008. (The space telescope ran out of coolant in 2009, rendering its detectors useless at long infrared wavelengths.) But in survey images of the source no infrared was detected at shorter wavelengths, either, indicating that the candidate first core is indeed colder than the known young protostars nearby, all of which appeared in the Spitzer data.

But neither first core candidate is rock-solid, and the astronomer who predicted the existence of the first core more than 40 years ago is not yet convinced that his prediction has been borne out. "It's a bit premature to be gratified, because these are not definitive discoveries," says Richard Larson, a Yale astronomer whose numerical simulations of star formation in 1969 predicted the existence of the short-lived first core. "In both cases they are suggestive but not conclusive."

The Arce group's L1448 IRS2E has not been detected in the infrared, and Enoch and her colleagues' Per-Bolo 58 shows a tenuous signal of relatively warm emission that, according to the simplest models of a cool first core, should not be there. Enoch notes that the emission could originate from an outflow carving out a cavity in the material surrounding the core, making it less opaque and allowing more radiation to escape. "The other possibility is it could just be a really, really faint protostar, so we still need to do further observations to rule that out," she says. The European Space Agency's space-borne Herschel observatory, which launched in 2009, and the Atacama Large Millimeter/submillimeter Array (ALMA), now under construction in Chile, may help identify and follow up on possible first cores in the coming years.

Whether the two groups have seen first cores or simply very young, extremely faint protostars, Larson notes, the astronomers appear to have identified stars at a very early point in their histories—perhaps earlier than anyone has yet seen. "In either case, they're looking at a very early stage," Larson says. "It's interesting that it's now possible to zoom in on these very early stages."