🔴 At 4:41:03 a.m. JST on Oct. 20, 2026 (19:41 UTC on Oct. 19), Japan's H3 flight No. 10 left Tanegashima on the exact second it was scheduled to. About 1 hour 11 minutes later, the Martian Moons eXploration (MMX) spacecraft was on its way to Mars, with plans to bring soil from the moon Phobos back to Earth in 2031. Earlier in 2026, the American and European plan to bring back rocks from Mars itself came to a halt. For NASA, MMX's 10 grams count for more than they did a year ago.

Right on the second

According to JAXA, the rocket flew as planned and released MMX about 1 hour 11 minutes 13 seconds after liftoff. On Oct. 21 the agency said the spacecraft had deployed its solar arrays, settled into three-axis attitude control and released its medium-gain antenna, and that it was healthy.

H3 now stands at 8 successes in 10 flights, with three straight successes since flights resumed in June 2026 after the December 2025 failure. This was the first H3 to fly with the long fairing, and the first H3 launch to send a payload out of Earth orbit. MMX was originally planned for 2024, but the failure of the first H3 in 2023 cost it two years. Makoto Arita, JAXA's H3 project manager, said after the launch that he was relieved the 10th flight, the first in double digits, could be marked by MMX's success.

We covered the two competing theories about where Phobos came from, and why MMX cannot sample the way Hayabusa2 did, in our pre-launch story.

NASA's share of 10 grams

MMX is supposed to gather 10 grams of material or more across two landings on Phobos. NASA is spending 87 million dollars over the mission's 14-year life, and in return it gets at least 15% of the sample. Its share will go to the Johnson Space Center in Houston, where it will be studied and stored for the long term. NASA also has two instruments on board: the MEGANE gamma-ray and neutron spectrometer from the Johns Hopkins Applied Physics Laboratory, and a pneumatic sampler built by Honeybee Robotics.

Fifteen percent of 10 grams is only about 1.5 grams. Even so, it is, for now, the only material from the Mars system that NASA has a contractual share of.

Sample tubes with no ride home

Since 2021, NASA's Perseverance rover has been drilling Martian rock and sealing the cores in tubes. In September 2025, NASA said one of them, a core called Sapphire Canyon taken in July 2024, contains potential biosignatures: features that could point to past life but have to be studied further before anyone can say so. As a backup, the rover has already set 10 tubes down on the floor of Jezero Crater. Picking those up and flying them to Earth was the job of Mars Sample Return (MSR), a joint program of NASA and the European Space Agency (ESA).

In April 2024, NASA said that under the existing design the program could cost up to 11 billion dollars, with samples back in 2040. In January 2025, it laid out two redesigned options at 5.8 to 7.7 billion dollars, with samples back as early as 2035. Even so, the U.S. spending law for fiscal 2026, enacted in January 2026, stated in its agreement that it does not support the existing MSR program. Instead, it gave 110 million dollars to a Mars Future Missions line for technology work.

NASA's fiscal 2027 request is again 110 million dollars. In April, four senators asked for at least 400 million dollars, arguing that the smaller amount cannot sustain work on precision landing and launching off the Martian surface. NASA's budget proposal does not say when or how it might try again to bring Mars samples back. As of October 2026, no fiscal 2027 budget has been enacted, and NASA is running on a stopgap measure through Dec. 11 that keeps it at the previous year's level.

Europe is pulling back too. In March 2026, ESA member states asked to cancel the Earth Return Orbiter, the spacecraft that was supposed to carry the samples home, and ESA began talks with prime contractor Airbus on reusing its technology. The agency has floated turning the orbiter into a mission to study the Martian atmosphere, and has said Europe cannot afford a full sample return on its own.

Two kinds of sample

China's Tianwen-3 is planned to launch around 2028 and bring 500 grams or more of Martian surface material back around 2031. It would come home at about the same time as MMX, but one scoops ground from Mars itself and the other from a small moon circling it. If both succeed, two very different kinds of Martian sample could reach Earth around 2031.

The hard part is getting back

MMX should be circling Mars by August 2027, and its study of Phobos starts from there. It will drop IDEFIX, a 25-kilogram rover built by France and Germany, in early 2029, and the main spacecraft will then land twice to collect samples. It is scheduled to leave the Mars system in November 2030, and its capsule should come down in Australia in July 2031.

Yasuhiro Kawakatsu, MMX project manager at JAXA, said what sets this apart from the asteroids Ryugu and Itokawa is going deep into a gravity well and climbing back out. Leaving an asteroid takes very little effort; coming home from a moon that circles close to Mars does not. If MMX succeeds, it will return the first sample ever taken from a moon of Mars.

The trip will be recorded by two cameras built by NHK and JAXA, an 8K facing forward and a 4K facing sideways, each shooting a JPEG every 10 seconds. The plan is to capture the Phobos landing, the surface, the climb back up and the capsule release on the way home.

If your own country's plan stalled and its scientists had to wait on samples carried by someone else's spacecraft, how would people there take it?

References