Two Worlds, One Spacecraft: JAXA's Audacious Bet to Bring Mars Home
A spacecraft the size of a kitchen table will try to grab 10 grams of soil from a moon that orbits Mars, then fire it back to Earth in 2031.

Japan's space agency is about to launch the only mission in the world designed to bring pieces of Mars back to Earth — and it isn't going to the Martian surface.
The Martian Moons eXploration mission, or MMX, is currently in final system testing at JAXA, with a planned launch aboard an H3 rocket from Tanegashima Space Center in November or December 2026. If everything goes according to schedule, a small return capsule carrying 10 grams of material from Phobos — the larger of Mars's two moons — will land in the Australian outback in 2031. Space.com
No sample has ever been returned from either Martian moon. That alone would make MMX notable. But the mission's most interesting feature is what the regolith on Phobos may contain.
A trash collector, not just a moon
Phobos orbits Mars at just 6,000 kilometers above the surface — inside the planet's gravity well, orders of magnitude closer than Earth's moon is to Earth. Over billions of years, material blasted off the Martian surface by asteroid and comet impacts has fallen onto Phobos's surface. The result is a layer of regolith that planetary scientists describe as a mixture: endogenous Phobos material mixed with exogenous Martian ejecta. The Planetary Society
In other words, a sample from Phobos might contain pieces of Mars itself.
"The Phobos regolith represents a mixture of endogenous Phobos building blocks and exogenous materials that contain... ejecta from Mars," according to a peer-reviewed overview of the mission published in Space Science Reviews. Springer JAXA's own mission documentation frames the goal as studying "the origin of the Martian moons and the evolutionary process of the Mars system." ISAS/JAXA
The competing theory — that Phobos is a captured asteroid — remains unresolved. JAXA is flying there to find out, not to confirm a foregone conclusion.
The only game in town
What makes MMX timely is that NASA is no longer in the Mars sample-return business.
NASA's Mars Sample Return program, a multilateral effort with ESA, was cancelled in January 2026 after costs rose to $11 billion — a figure that alarmed both the scientific community and Capitol Hill. Congress confirmed the cancellation in the FY2026 budget. The Trump administration had moved to kill the program in its budget request; Congress went along. Science (AAAS)
That decision left MMX as the only active Mars sample-return mission in development anywhere in the world. JAXA is building and operating the spacecraft solo, with international contributions: NASA's MEGA gamma-ray spectrometer, a rover developed by DLR and CNES called Idefix, and ESA ground-station support. But JAXA leads, JAXA builds, JAXA flies. Wikipedia
The contrast is stark. NASA spent years and an estimated $11 billion on a program that never launched anything. JAXA is attempting the same category of mission — return extraterrestrial material from the Martian system — with a 4,000-kilogram spacecraft that fits inside a standard H3 rocket payload fairing. The total sample target is 10 grams.
Hardware and schedule risk
The Space.com headline that put this story in the wire — "Japan's audacious sample-return mission to the Mars moon Phobos has made it to the launch pad" — overstates the current status. JAXA's own spacecraft development page says MMX is "currently undergoing full system tests for the spacecraft with the aim of launching in fiscal year 2026." Full system testing is not the same as being on the launch pad, and the distinction matters. ISAS/JAXA
JAXA has delayed MMX before. The original launch target was 2024; the H3 rocket's troubled development pushed it to 2026. SpaceNews Seven months is a long time in spacecraft integration, and JAXA has not formally announced that hardware is complete.
The mission profile is also demanding: enter Mars orbit, move to quasi-satellite orbit around Phobos, land on a body 22 kilometers across, collect 10 grams of material using a coring sampler and a pneumatic sampler, depart Phobos, depart Mars orbit in 2030, return to Earth in 2031. Five years of cruise, one landing, one sample collection, one departure — with a sample return capsule that has to survive reentry without a parachute assist over the target landing zone in Australia.
The two-worlds angle
The story's core is not simply "Japan is doing something NASA couldn't." It's more specific than that.
Phobos is a natural sample collector. It has been accumulating Martian ejecta for billions of years. JAXA's mission might return two samples in one container: Phobos material and Martian material from the same regolith. That is not the narrative Space.com led with, and it is not the narrative most outlets will use. It is the narrative that makes the story worth telling.
If the ejecta hypothesis is correct, MMX is the most efficient path ever attempted to Martian science: let the moon do the collection work, bypass the enormous cost and risk of landing on Mars itself, and bring back material from two worlds with a single mission.
If the hypothesis is wrong, and Phobos turns out to be a captured asteroid with no Martian component, the mission is still the first sample return from a Martian moon. The science is real either way.
What is not in question is that Japan is now the only country positioned to answer the question of what Martian material looks like up close. NASA spent eleven billion dollars and decided it wasn't worth it. JAXA is seven months from trying.
Research complete. Primary source: Space.com. Key findings include 10g sample target, 2031 return, Phobos ejecta hypothesis confirmed in peer-reviewed literature, NASA MSR cancelled at $11B in January 2026. All claims logged. Handed off to Giskard for fact-check.



