Distant galaxies in Webb images suggest we need to rethink star and galaxy evolution in the early universe.

Webb's first deep field
Webb’s first image, revealed on Monday, July 11th, shows thousands of galaxies in a tiny sliver of sky. Many galaxies are part of a massive cluster named SMACS 0723, but the image reveals far more distant ones, too. The heft of the galaxies and the dark matter around them lenses background galaxies, distorting their shapes but also magnifying them and making them easier to see.

The very first results from the James Webb Space Telescope seem to indicate that massive, luminous galaxies had already formed within the first 250 million years after the Big Bang. If confirmed, this would seriously challenge current cosmological thinking. For now, however, that’s still a big “if.”

Shortly after NASA published Webb’s first batch of scientific data, the astronomical preprint server arXiv was flooded with papers claiming the detection of galaxies that are so remote that their light took some 13.5 billion years to reach us. Many of these appear to be more massive than the standard cosmological model that describes the universe’s composition and evolution.

“It worries me slightly that we find these monsters in the first few images,” says cosmologist Richard Ellis (University College London).


JWST deep-field image showing many galaxies
This image is part of a larger mosaic taken with the Near Infrared Camera (NIRCam) on the James Webb Space Telescope. It’s from a patch of sky near the handle of the Big Dipper. This is one of the first images obtained by the Cosmic Evolution Early Release Science Survey (CEERS) collaboration. Distant galaxies will show up in many JWST images, but CEERS is one of the surveys specifically set up to help find them.
NASA / STScI / CEERS / TACC / S. Finkelstein / M. Bagley / Z. Levay

Young, massive stars in newborn galaxies emit vast amounts of energetic ultraviolet radiation. As this light moves through expanding space for billions of years, the wavelengths stretch (redshift) all the way into the infrared – radiation that Webb’s instruments are sensitive to.

It takes careful spectroscopic measurements – either by Webb’s spectrometers or by the ground-based ALMA observatory that operates at even longer wavelengths – to precisely determine the redshifts, which tells you how far out into space — and thus how far back in time — you’re looking. But there’s a quick (albeit less reliable) workaround that gives a rough idea.

Neutral hydrogen atoms in intergalactic space absorb ultraviolet radiation at wavelengths shorter than 91.2 nanometers. For remote objects, this threshold also redshifts to longer wavelengths, into the infrared for the most distant galaxies. Since Webb’s near-infrared camera NIRCam takes measurements through a large number of filters, each covering a different wavelength band, a galaxy may be visible in some channels but not in others. The wavelength band in which the galaxy disappears roughly indicates its redshift, and the corresponding look-back time.

Distant galaxy seen in different wavelength filters
These postage-stamp images taken by JWST’s near-infrared camera show the galaxy CEERS-93316 in six different filters: F115W, F150W, F200W, F277W, F356W, and F444W.
Donnan et al. 2022

Just six days after Webb’s first science data became available, on July 19th, two independent teams of astronomers presented their analysis based on this technique. Both groups, one led by Rohan Naidu (Center for Astrophysics, Harvard & Smithsonian) and the other by Marco Castellano (Rome Observatory, Italy), found two relatively bright galaxy candidates at redshifts of about 11 and 13, residing in a universe on the order of 400 and 325 million years old, respectively.

In the days that followed, another two independent teams, led by Callum Donnan (University of Edinburgh) and by Yuichi Harikane (University of Tokyo), announced the tantalizing find of an unexpectedly massive galaxy at a redshift of 17. That corresponds to looking back to just 225 million years after the Big Bang.

In yet another study, Haojing Yan (University of Missouri) and his colleagues even claimed that some of their candidate galaxies might reach a redshift of 20 (180 million years after the Big Bang).

“It’s understandable that young teams are racing ahead” to put out their results, says Ellis. According to extragalactic astronomer Mariska Kriek (Leiden Observatory), some of these groups may have written large parts of their paper in advance, so they only had to fill in a couple of numbers and other details. “They have picked the low-hanging fruit,” she says. “For some people, it’s just very important to be first. And of course everyone is very curious about what’s in the data.”

Before the community accepts these claims, the reported redshifts have to be confirmed spectroscopically. Mark McCaughrean, the senior science adviser of the European Space Agency (a major partner on Webb) commented on Twitter: “I’m sure some of them will be [confirmed], but I’m equally sure they won’t all be. […] It does all feel a little like a sugar rush at the moment.”

Ellis agrees: “It’s one thing to put a paper on arXiv,” he says, “but it’s quite something else to turn it into a lasting article in a peer-reviewed journal.”

So far, astronomers have found distant galaxy candidates in four areas of the sky. Some scoured the neighborhood of SMACS 0723-73, the galaxy cluster in the southern constellation Volans (the Flying Fish) showcased in the first Webb image to be released. Others pored over two ongoing surveys, the Grism Lens-Amplified Survey from Space (GLASS) and the Cosmic Evolution Early Release Science (CEERS), in Sculptor and Boötes, respectively. In addition, three candidates were uncovered in another early-release image, the one of Stephan’s Quintet, a compact group of galaxies in Pegasus.

It’s hard to keep track of all the new findings, partly because each team uses its own numbering scheme. For instance, the galaxy candidate at a redshift of 17 is variously known as ID93316, CEERS-1749, and CR2-z17-1.

This galaxy is also emblematic of some of the problems with detecting distant galaxies in this way. In fact, it has earned the nickname “Schrödinger’s Galaxy” because of its undecided nature — it turns out, it might actually be a much closer galaxy that’s so dusty that it appears to disappear at longer wavelengths in the same way that more distant galaxies do. A team led by Jorge Zavala (National Astronomical Observatory of Japan), make the case that this galaxy is at a redshift of 5, corresponding to a lookback time of a “mere” 12.6 billion years.

“But it’s fun,” says Kriek. The fast pace of Webb science is keeping everyone on their toes, and there’s a lot of work to do to confirm the most distant galaxies are really so far away. “Every day is a little adventure,” adds Ellis.


In many of the papers that have been posted so far, the authors state that their results, if confirmed, may challenge the standard model of cosmology. According to this model, known as Lambda Cold Dark Matter (ΛCDM), the universe’s evolution is governed by dark energy (denoted by the Greek letter lambda, Λ) and the equally mysterious cold dark matter (CDM), which makes up almost 85% of all matter.

According to ΛCDM, the very first galaxies could well show up at just 200 million years after the Big Bang, but they’d be puny and faint, resembling small dwarf galaxies. Instead, some of the remote candidate galaxies in the Webb data appear to contain about 1% of the mass of our Milky Way galaxy, which is already quite a lot at that early epoch.

Ivo Labbé (Swinburne University of Technology, Australia) and his colleagues even found one candidate at a redshift of 10 (500 million years after the Big Bang) that is already comparable in mass to our home galaxy. According to a recent study by Michael Boylan-Kolchin (University of Texas, Austin), ΛCDM predicts at most one such massive galaxy in a survey area that is 1,000 times larger.

But theorist David Spergel (Princeton University) is not yet alarmed. “I think that we need to be cautious about these statements,” he says. As Spergel explains, estimates for the mass of a remote galaxy are based on its observed luminosity at various wavelengths (which, incidentally, might be affected by ongoing instrument calibration). But the estimates also assume that the relative numbers of low-mass and high-mass stars are the same as in the Milky Way. However, higher pressures and temperatures in the early universe might have suppressed the formation of low-mass stars back then.

“At low redshifts, most of the mass is in low-mass stars,” Spergel says. “This may not be true at high redshifts. I suspect that we are learning that high-mass star formation was very efficient” in the early universe. Again, the final verdict has to await detailed spectroscopic follow-up observations. According to Kriek, astronomers will be vying for observing time on Webb to sort things out.

One thing’s for sure, though: in its first weeks of operation, the new space telescope has already surpassed most astronomers’ expectations. “It feels like opening a box of toys for the first time,” says Ellis. “It’s just fabulous.”

Editorial note: This article was updated with information from the Zavala et al. study, posted on the arXiv on August 3rd.