On June 1, 2026, NASA published a sentence never before written about an object from beyond our solar system: "For the first time on an interstellar visitor, Webb directly detected methane gas" [source: NASA, 2026]. The visitor was comet 3I/ATLAS, and by then it had already rounded the Sun and was leaving, on a path that does not bring it back.
Only three interstellar objects — bodies not born around our Sun and not held by it — have ever been confirmed crossing the solar system, and 3I/ATLAS is the third, after 1I/ʻOumuamua in 2017 and 2I/Borisov in 2019 [source: ESA, 2026]. Such an object arrives unannounced, passes once, and leaves. Every measurement had to be taken while it was still here.
This article is about what the instruments recorded: which quantities were measured, which were only bracketed between bounds, and which statements are interpretation rather than readout.
A visitor that could not stay
The first observation reached the Minor Planet Center on July 1, 2025, from the ATLAS (Asteroid Terrestrial-impact Last Alert System) telescope at Rio Hurtado, Chile, and the circular carried the name that stuck: 3I/ATLAS = C/2025 N1 (ATLAS) [source: Minor Planet Center, 2025]. The object was then about 670 million km from the Sun, inside Jupiter's orbit, approaching from the direction of Sagittarius [source: NASA, 2026].
"Comet" came from activity, not appearance. NASA's record states the object "was active, which means it has an icy nucleus and a coma … This is why astronomers categorized it as a comet and not an asteroid" [source: NASA, 2026] — a coma being the envelope of gas and dust that forms when ice on a nucleus turns to vapor.
"Interstellar," the "3I" prefix marking the third of its kind, came from the orbit. JPL's solution gives an eccentricity near 6.14, and anything above 1 is an open hyperbola that leaves the Sun for good, along with an inclination of 175.1°, meaning the comet crossed the solar system almost directly against the direction the planets travel [source: NASA/JPL, 2026].
Two kinds of rows are missing on purpose from the three-object comparison alongside this article. Published sizes measure different things — a length for the elongated 1I, a diameter for 2I, a bracketed nucleus for 3I [source: NASA, 2025] — and the published speeds for 1I and 2I carry no epoch at all.
The numbers that were pinned down, and the one that was not
JPL puts perihelion — closest approach to the Sun — at 1.356 au, about 203 million km, passed on October 29, 2025 at 11:52:48 TDB [source: NASA/JPL, 2026], and ESA's FAQ gives the same pair [source: ESA, 2026]. NASA's FAQ states October 30, 2025 and about 1.4 astronomical units [source: NASA, 2026]; the difference is notation, not disagreement.
Speed on such a path is not one number: it changes continuously with distance from the Sun, so every figure needs a moment attached to it. NASA gives about 246,000 km/h at perihelion and notes the comet leaves at the speed it entered with [source: NASA, 2026] — the figure ESA rounds off and calls "the highest ever recorded for a Solar System visitor" [source: ESA, 2026].
Closest approach to Earth came on December 19, 2025, at 1.8 au, roughly 270 million km [source: NASA, 2026], and both agencies stated plainly that the comet posed no danger to Earth or to any other planet [source: ESA, 2026].
The one quantity that stayed unmeasured is the nucleus. From Hubble observations, NASA reports, "as of Aug. 20, 2025, they saw that the diameter of its nucleus was not less than 1,400 feet (440 meters) and not greater than 3.5 miles (5.6 kilometers)" [source: NASA, 2026]. The upper bound is about 12.7 times the lower one, because the activity that identified the object as a comet also wraps the nucleus in gas and dust and hides it. That bracket is the result, not a vague way of stating a size.
Outgassing also nudges a trajectory. NASA states that for 3I/ATLAS "these perturbations were indeed small and compatible with this process" [source: NASA, 2026], and JPL models the non-gravitational acceleration with a carbon-dioxide-driven law [source: NASA/JPL, 2026]. That the molecule in the model matches what spectroscopy later found dominating the coma is a consistency, not evidence that carbon dioxide steered the orbit.
What Webb actually measured
A comet dominated by carbon dioxide
The first composition measurement came before perihelion. On August 6, 2025, with the comet 3.32 au from the Sun, the James Webb Space Telescope's NIRSpec instrument found a gas coma dominated by carbon dioxide rather than water: (1.70 ± 0.01) × 10²⁷ molecules per second for carbon dioxide against (2.23 ± 0.08) × 10²⁶ for water, with carbon monoxide at (3.7 ± 0.2) × 10²⁶ and a tentative detection of carbonyl sulfide [source: The Astrophysical Journal Letters, 2025]. That puts the carbon-dioxide-to-water ratio at 7.6 ± 0.3, far above what is typical of comets formed in our own solar system.
The first direct detection of methane
The headline result came later and at longer wavelengths. On December 15–16 and December 27, 2025, with the comet outbound at 2.20 and 2.54 au, Webb's MIRI Medium-Resolution Spectrometer recorded spectra from 5 to 28 µm — "the first spectroscopic characterization of an interstellar object at mid-infrared wavelengths" [source: The Astrophysical Journal Letters, 2026].
Both epochs show fluorescence features from water, carbon dioxide and methane, together with a line of atomic nickel; the band identifications are set out in the measurement steps alongside this article [source: The Astrophysical Journal Letters, 2026]. Both the paper and NASA keep the qualifier "direct" [source: NASA, 2026].
The nickel line deserves context rather than a headline. The gas-phase atomic nickel implies a nickel-to-oxygen ratio near 0.002% against roughly 0.34% in the Sun, accounting for only "a fraction of a percent" of the nickel the comet carries, and gaseous nickel has been found before in the comae of distant comets, including 2I/Borisov [source: The Astrophysical Journal Letters, 2026].
Why did methane appear only on the outbound leg? NASA keeps the answer a suggestion: "Its delayed appearance … suggests it was buried below the comet's top surface layer and protected from sublimation until heat from the comet's close pass to the Sun reached deeper parts of the icy subsurface" [source: NASA, 2026]. The paper reads it the same way [source: The Astrophysical Journal Letters, 2026].
Why the methane ratio doubled while the methane itself declined
Between the two MIRI epochs the methane-to-water mixing ratio rose from 11.0% ± 0.5% to 21.6% ± 1.3%, and the carbon-dioxide-to-water ratio from 2.30 ± 0.03 to 5.16 ± 0.13 [source: The Astrophysical Journal Letters, 2026]. Published methane-to-water values for solar-system comets span roughly 0.1% to 10%, with the hypervolatile-rich C/2016 R2 standing apart at 181% ± 25%.
The ratio did not double because the comet produced more methane. Across those 12 days every measured production rate fell, and the published rates work out to declines of about 72% for water, 45% for methane and 38% for carbon dioxide — water "dropping more steeply than other species," as the paper puts it [source: The Astrophysical Journal Letters, 2026]. Water falling fastest lifts every ratio measured against it: "Water, which is less volatile than methane or carbon dioxide, is quicker to 'shut off' its gas production," as NASA explains [source: NASA, 2026].
Where the gas sits, not only what it is
Water vapor appeared spread far beyond the nucleus while carbon dioxide and methane stayed concentrated close to it [source: NASA, 2026], because much of the water comes off icy grains carried out into the coma rather than from the nucleus surface alone [source: The Astrophysical Journal Letters, 2026]. Seeing that difference at all takes an instrument that returns a spectrum for every point in its field of view instead of one average for the whole coma.
NASA states the upshot carefully: both findings "point to a very different formation environment and chemistry than the vast majority of comets that formed within our solar system" [source: NASA, 2026]. That verb marks an interpretation. The mixing ratios are the measurement; the formation environment is the inference drawn from them.
One comet, a dozen instruments
No single telescope could have produced that record. The pass happened once, across a few months, so the campaign was assembled out of whatever instruments could see the comet in the time available [source: NASA, 2026].
Geometry decided who had the best seat. On October 3, 2025 the comet passed 29 million km from Mars, about a ninth of its closest distance to Earth, and ESA reports that ExoMars Trace Gas Orbiter data from early October served to "improve the comet's predicted location by a factor of ten" [source: ESA, 2026].
Solar observatories and spacecraft in transit covered what ground telescopes could not: Parker Solar Probe imaged the comet with its WISPR camera across perihelion at about 10 images a day, SOHO's LASCO coronagraph tracked it through late October, and Psyche and Europa Clipper observed it in passing from 53 million and 164 million km [source: NASA, 2025].
One result came at a wavelength no interstellar object had been seen in. ESA reports that in late November and early December 2025 the X-ray telescopes XRISM and XMM-Newton "observed the comet, revealing a diffuse X-ray glow around the comet nucleus," making 3I/ATLAS "the first interstellar comet to have been observed in X-ray light" [source: ESA, 2026]. This one rests on ESA's public documentation; no peer-reviewed paper on it was located for this article.
An independent instrument landing near the same answer is the most useful kind of confirmation. NASA's SPHEREx, which observes in 102 wavelength bands, tracked a dramatic brightening about two months after perihelion and identified methanol, cyanide, methane, carbon dioxide, carbon monoxide and water ice [source: NASA, 2026]. From its December 7–15, 2025 data the MIRI paper cites a methane-to-water ratio near 14%, an upper limit because methanol feeds the same signal — between MIRI's two values, a partial cross-check rather than a competing claim [source: The Astrophysical Journal Letters, 2026].
TESS re-observed the comet from January 15 to 22, 2026, with a gap to the 18th while the spacecraft was in safe mode, by which point it had faded to about magnitude 11.5, roughly 100 times fainter than the naked-eye limit [source: NASA, 2026]. Its activity and rotation were left for further analysis.
Where it came from: a search that came back empty
The obvious next question — which star did it leave? — has an answer, and the answer is that nobody can name one. A team led by Y. Guo back-traced the orbits of 3I/ATLAS and about 30 million Gaia stars and found 25 encounters with a median distance under one parsec, none of which qualified as a home: "because the encounter speeds between 3I/ATLAS and each encounter exceed 20 km/s, none is a plausible host under common ejection mechanisms" [source: The Astronomical Journal (accepted), 2025]. The strongest gravitational scatterer they identified — a wide M-dwarf binary that passed 0.242 parsecs away 1.64 million years ago at 28.39 km/s — is a perturber, not a parent.
What the kinematics do support is broader and less satisfying: "we find that a thin-disk origin is strongly favored" [source: The Astronomical Journal (accepted), 2025]. The comet's motion matches the population of stars in the flat component of our galaxy rather than any individual system.
Age estimates come from models, not from a clock on the comet. Applying a population model of interstellar objects, Hopkins and colleagues derived "an age of over 7.6 Gyr" and judged it "very unlikely that 3I shares an origin with either of the previous two" visitors [source: The Astrophysical Journal Letters, 2025], while the MIRI paper notes a composition consistent with dynamical ages spanning 3 to 11 billion years [source: The Astrophysical Journal Letters, 2026]. Billions of years, with a range wide enough to hold most of the galaxy's history, is the honest summary; a single number would not be.
What 3I/ATLAS left behind
For an object nobody could plan for, the record is unusually clean: an orbit from 782 observations over a 280-day arc, hyperbolic beyond doubt at an eccentricity near 6.14 [source: NASA/JPL, 2026]; a perihelion distance and time fixed to the second; a nucleus honestly bracketed between 440 m and 5.6 km; a coma measured twice in the mid-infrared, yielding the first direct detection of methane on an interstellar object; and a systematic search for a home star that returned no candidate.
The observing window has since closed. Astrometric observations resumed on October 31, 2025 after the comet passed behind the Sun, and NASA's guidance was that it would stay within reach of small telescopes through spring 2026 [source: NASA, 2026]; as of this writing, in September 2026, that period has passed. Through all of it both agencies' public documentation classified the object as a comet on the basis of what was observed — NASA that "3I/ATLAS's characteristics, color, speed, and direction are all consistent with what we expect from a comet" [source: NASA, 2026], and ESA that "its shape and behaviour indicate that it is a comet" [source: ESA, 2026].
Two threads stay open. The activity and rotation analysis of the TESS data was left for later work [source: NASA, 2026], and the MIRI figures quoted here come from the accepted version of the paper, whose final typeset version may differ in its last digits. As for how common a chemistry like this is among bodies formed around other stars, answering that takes a fourth interstellar visitor, and then a fifth. Until one turns up, what 3I/ATLAS left behind is the set of numbers taken during its one pass.