Half a century after the only direct search for life ever attempted on another world, the agency that attempted it has cancelled the mission that could finish the job. Thirty rock cores sit in a Martian crater waiting for a ride that no longer has a budget — while Europe, China and India move into the space the Americans are vacating.
On 20 July 1976, after a fortnight of nerve-shredding delay while controllers hunted for ground flat enough to survive, NASA’s Viking 1 lander set its footpads into the dust of Chryse Planitia. Its twin came down in Utopia Planitia that September. Between them the two spacecraft carried something no machine has carried to another planet before or since: instruments designed not to infer the conditions for life, but to catch life itself in the act of living.
The ambition was extraordinary and, in retrospect, premature. One experiment flooded a scoop of Martian soil with nutrient broth and inert helium and watched for the gaseous exhalations of metabolism; the engineers nicknamed it chicken soup. A second, the labelled release experiment, dosed the soil with nutrients tagged with radioactive carbon-14 and waited for a tagged gas to come back out. A third baked dry soil in tagged carbon dioxide to see whether anything took the bait. Alongside them, a gas chromatograph–mass spectrometer hunted for organic molecules — the carbon skeletons that all terrestrial biology is built from.
The results were maddening. Chicken soup produced gas, but chemistry could explain it. The labelled release experiment produced tagged carbon dioxide, which looked, to some eyes, exactly like respiration. And then the mass spectrometer found essentially no organic molecules at all — nothing but chloromethane and dichloromethane, which the team wrote off as cleaning-fluid contamination from Earth. A world with less organic carbon than the Moon could not plausibly be a world with microbes in its topsoil. Most biologists closed the file.
It took thirty-two years to discover that Viking may have destroyed the evidence it was sent to find. In 2008 NASA’s Phoenix lander found perchlorate salts in Martian soil — powerful oxidisers, the kind of chemistry used in rocket propellant. Laboratory work published in 2010 showed that heating a perchlorate-bearing soil analogue shreds any organic compounds present and leaves behind, of all things, chloromethane and dichloromethane. The contamination signature was not contamination. It was the ash of the very molecules Viking was looking for, incinerated by Viking’s own oven.
That is the lesson the last fifty years of Mars exploration were built on: the question was right, the instrument was wrong, and the planet does not forgive a badly framed enquiry. NASA spent three decades constructing the answer to Viking’s failure — a methodical, unbroken cadence of orbiters and rovers that has kept at least one American spacecraft at Mars continuously since 1997, and that established, beyond serious dispute, that the Red Planet was once warm, wet and habitable. The strategy had a destination. It was called Mars Sample Return.
As Viking’s golden anniversary arrives, that destination has been deleted from the budget.
What Congress actually did
The immediate cause is not scientific but fiscal, and the sequence matters because it is widely misreported as a straightforward assault on science. It was not. It was worse, in a way — it was a targeted excision.
The White House’s fiscal 2026 request proposed cutting NASA to roughly $18.8 billion, a reduction of about 24 per cent, with the agency’s science account facing cuts approaching half. Congress rejected that comprehensively. The Commerce, Justice, Science appropriations package cleared the House of Representatives on 8 January 2026 and the Senate on 15 January, and it gave NASA $24.4 billion, with $7.25 billion for the Science Mission Directorate — a cut of roughly one per cent against the previous year. By the standards of the moment, that was a rout for the science community’s advocates.
One programme did not survive the rescue. In a section headed Mars Future Missions, the accompanying report stated flatly that “the agreement does not support the existing Mars Sample Return (MSR) program.” The campaign to retrieve the rock cores that the Perseverance rover has been drilling and caching in Jezero Crater since 2021 — more than thirty of them — was, in effect, terminated. Nature’s news pages put the position bluntly in January: funding was restored for most of NASA’s space-science missions, but not for bringing home the samples already collected.
What Congress left in its place was $110 million for a Mars Future Missions line intended to preserve the technologies MSR had been developing — radar, spectroscopy, entry, descent and landing systems, and what the legislators called translational precursor technologies — on the reasoning that these capabilities are indispensable not only to future robotic science but to any eventual human expedition. The House had proposed $300 million. The final figure was a third of that, and a rounding error against a mission whose most recent cost estimates ran between $5.8 billion and $7.7 billion.
The programme had been dying for some time. An independent review had concluded that the original architecture could cost as much as $11 billion and might not deliver samples to Earth before 2040. In January 2025 NASA responded by putting two competing architectures on the table — a proven sky-crane landing system at $6.6–7.7 billion, and a commercial option at $5.8–7.1 billion — and promised a decision in the second half of 2026. The decision arrived early, from a different building, and it was neither.
What NASA is doing instead
To understand the pivot, read what replaced it. At an event branded Ignition on 24 March 2026, NASA unveiled the initiatives it says will deliver the President’s National Space Policy. The centre of gravity is unmistakable: return to the Moon inside the presidential term, build a lunar base in three phases, pause the Gateway station in its current form, and reorganise the agency around a great-power competition whose outcome, in the words of Administrator Jared Isaacman, will be measured in months rather than years.
Mars appears in that plan. It does not appear as a scientific question. It appears as a destination and a logistics problem.
The two new Mars vehicles NASA has proposed are a Mars Telecommunications Network — a comms relay — and Space Reactor-1 Freedom, which the agency intends to launch before the end of 2028 as the first nuclear-powered interplanetary spacecraft, demonstrating nuclear electric propulsion in deep space. Associate Administrator Amit Kshatriya described the reactor as finally moving nuclear propulsion out of the laboratory and into deep space, and on its own terms that is a genuinely significant engineering ambition, one the agency has studied for six decades without flying.
When SR-1 Freedom reaches Mars it will release Skyfall, a payload of Ingenuity-class helicopters. What those helicopters will actually investigate has not been established. There is no defined science plan, and no process yet for deciding what one would contain. A telecommunications relay, a propulsion demonstrator and a set of rotorcraft without a research question do not constitute a search for life. They constitute infrastructure for someone else’s arrival.
The Senate’s NASA Authorization Act of 2026 makes the reframing explicit in a way that deserves more attention than it has received. Among the Mars-focused mission concepts the bill directs NASA to study is one that would send human tissue samples to Mars in order to examine what the Martian environment does to them. Biology is still going to Mars. It is our biology, and the question is no longer whether life arose there but whether we can survive there. That is a legitimate question. It is not the same question, and it will not be answered by the same instruments.
The rock in the middle of the argument
The cancellation would be easier to defend if the samples were unremarkable. They are not.
In July 2024, while working through an outcrop called Bright Angel on the flank of Neretva Vallis — a 400-metre-wide channel that once poured water into Jezero Crater — Perseverance came upon an arrowhead-shaped mudstone roughly a metre long, nicknamed Cheyava Falls. Its surface carried millimetre-scale markings the team called leopard spots, and sub-millimetre nodules they called poppy seeds. The rover drilled it and cached a core named Sapphire Canyon.
After a year of scrutiny, a paper led by Joel Hurowitz of Stony Brook University, published in Nature on 10 September 2025, laid out what the spots are made of. Each leopard spot has a structure: a rim of vivianite, a hydrated iron phosphate; a bleached interior from which rusty ferric iron has been leached away; and a core of greigite, an iron sulfide. Those minerals sit in intimate association with organic carbon, in low-temperature sediment, arranged as reaction fronts — the frozen boundaries of a chemical change that swept through the rock.
The reason this matters is redox. Moving electrons between iron, sulfur and organic matter in oxygen-poor sediment is precisely how a very large fraction of Earth’s microbial biosphere earns its living. On Earth, vivianite forms in peat bogs and lake sediments around decaying organic material; greigite forms where microbes cycle sulfur. Find them together, wrapped around organic carbon, at low temperature, and a terrestrial geologist would reach for the word biosignature without much hesitation.
NASA did not reach for it. The agency called Sapphire Canyon a potential biosignature and meant every syllable of the qualifier. Chemistry that could feed life is not chemistry made by life. The same textures can, in principle, be produced abiotically — given sustained heat, or acidity, or organic compounds acting as reagents rather than as corpses. The Hurowitz paper says so itself. Every route to resolving the ambiguity runs through a laboratory on Earth.
Then, in June 2026, the case got heavier. Writing in Science Advances on 24 June, Ashley Murphy of the Planetary Science Institute and colleagues reported that Perseverance’s SHERLOC instrument had detected complex organic carbon across four targets on three separate rocks at Bright Angel — including, for the first time, on a rock the rover had never drilled into. The carbon is mixed both into the original silicate sediment and into carbonate and sulfate minerals that formed later, implying that organics were emplaced at two distinct moments in the rock’s history: when the mud was laid down, and again when fluids moved through it afterwards.
The origin remains open. Paul Byrne of Washington University in St. Louis lists the candidates without prejudice: meteorites and cosmic dust, abiological hydrothermal chemistry, or biology. And there is a wider implication. Curiosity found organics at Gale Crater in 2014, some 3,500 kilometres from Jezero. If organic carbon is that widely distributed, then whatever Mars was doing chemically, it was doing at continental scale.
Perseverance cannot settle any of it. The rover has no instrument capable of resolving the molecular architecture of that carbon or identifying the functional groups hanging off it. The samples are already drilled, already sealed, already sitting in a crater. The scientific work is done. Only the freight remains — and the freight is what was cancelled.
“Deeply concerned”
The reaction from the research community has been unusually plain-spoken, and it has come from the body NASA itself created to advise it.
Victoria Hamilton, a planetary scientist at the Southwest Research Institute in Boulder and chair of NASA’s Mars Exploration Program Analysis Group (MEPAG), told Space.com in January that the steering committee welcomed the overall budget outcome but was “deeply concerned by the cancellation of the MSR program.” Her argument had three parts, and none of them was sentimental.
The first was procedural. Sample return has been the top-ranked planetary science priority in two consecutive Decadal Surveys — the once-a-decade prioritisation exercises the US National Academies conduct at NASA’s own request, and which Congress has historically treated as the sector’s constitution. Discarding that guidance, Hamilton argued, sets a precedent that reaches far beyond Mars.
The second was practical, and it is the argument most likely to land with the people who cancelled the programme. The samples are not only an astrobiology asset. They are a risk-reduction asset for the Moon-to-Mars programme. Knowing what Martian dust actually is — its toxicity, its electrostatics, its perchlorate load, its behaviour against seals and lungs and optics — is worth a great deal of money and possibly some lives to anyone planning to put crews on the surface. Cancelling the samples does not remove that cost. It defers it to a programme where the unknowns are engineered around at far greater expense.
The third was reputational, and it was the sharpest. Hamilton framed the decision as difficult to read as anything other than an admission that the task is “too hard for the United States” — and asked how an agency that concluded robotic sample return was beyond it proposed to succeed at crewed Mars exploration, which is orders of magnitude harder and where lives are at stake. She also warned of the position the US would occupy if it were reduced to watching remarkable discoveries made by scientists of a country with which American researchers are not permitted to collaborate.
MEPAG’s immediate practical demand is narrower and more urgent: that NASA work with the community now to produce a plan that preserves both the samples and the ability to retrieve them later, while allowing Perseverance to continue its science. Sample tubes on the Martian surface are not indefinitely patient, and neither is institutional memory.
Not everyone reads the outcome as terminal. Jack Kiraly, director of government relations at The Planetary Society, which campaigned hard against the deeper cuts, points out that the bill does not cancel the idea of sample return — it cancels one architecture, while instructing NASA to build a common technological heritage for Mars exploration. NASA’s own Mars Exploration Program director, Tiffany Morgan, told MEPAG in April that returning Perseverance’s samples has not been abandoned. The difficulty is that nobody can say how, or when, or with whose money.
The plumbing is failing too
Underneath the argument about missions is a quieter problem that will constrain every option on the table: the machinery that makes Mars legible to Earth is wearing out.
On 3 June 2026, NASA declared MAVEN dead. The Mars Atmosphere and Volatile Evolution orbiter had arrived in September 2014 and worked for eleven years. It went silent in December 2025; fragmentary data on 6 December showed it rotating at about 2.7 revolutions per minute, which it should not have been doing at all. A review board concluded in February that the spin drained the batteries and killed the radio. Project manager Mike Moreau of Goddard Space Flight Center reported that the spacecraft is not recoverable. The root cause is still unexplained.
MAVEN’s science legacy is considerable: it established that the solar wind strips roughly 100 grams of atmosphere from Mars every second — a rate that climbs roughly tenfold during solar storms — and it made the first direct observation of atmospheric sputtering at any planet. Principal investigator Shannon Curry of the University of Colorado Boulder notes that atmospheric escape at Mars is now better understood than at Earth. It is, in other words, the mission that explained how a warm wet world became a freeze-dried one. Hamilton calls it the strongest evidence yet for that transition.
But MAVEN was also a relay. It was one of five satellites in the joint NASA–ESA Mars Relay Network, and its loss leaves four. Tiffany Morgan says the network remains resilient enough to absorb the loss, though rovers now see occasional delays; MAVEN mattered most, she notes, for science data rather than operational traffic. The remaining assets are between ten and twenty-five years old. NASA issued a request for proposals in May for a new Mars Telecommunications Network to provide continuous communications for sample return, surface operations and human missions — but it may not fly before 2030.
Briony Horgan, a planetary scientist at Purdue University, puts the trajectory plainly: the Mars infrastructure is growing more fragile every year. Hamilton adds the sting — surface missions now depend on Mars Odyssey and Mars Reconnaissance Orbiter, and those are themselves exposed to the proposed budget. A programme can be ended by attrition as effectively as by decision, and rather more quietly.
Who is still asking the question
The history of Mars exploration is very nearly a NASA monopoly. The next chapter will not be.
Europe: the drill that goes deep
The European Space Agency’s Rosalind Franklin rover is the only spacecraft now being built with the explicit purpose of doing what Viking tried — detecting life directly — and its history is a case study in institutional persistence. NASA joined the mission, then withdrew in 2012 citing budgets. Roscosmos supplied the launcher and pad, then invaded Ukraine in February 2022, a week before the rover was to ship to Baikonur; project scientist Jorge Vago has described the team as having had the crates packed and the celebration wine stacked. NASA rejoined in 2024. The current administration’s 2026 request proposed leaving again. Congress said no, and in April 2026 NASA approved the Rosalind Franklin Support and Augmentation project and awarded the launch to SpaceX’s Falcon Heavy.
The rover will take a deliberately slow route to Mars to avoid landing in dust-storm season, touching down in 2030 at Oxia Planum — four-billion-year-old terrain thick with phyllosilicate clays, which form only where organics sit in prolonged contact with liquid water, and which some researchers believe marks the shoreline of an ancient ocean.
Its decisive advantage is a drill that reaches two metres down. The standing record for drilling on Mars still belongs to Viking, at fifteen centimetres. Everything shallower than about a metre has been sterilised by radiation for aeons; depth is the difference between reading ash and reading text. The Mars Organic Molecule Analyzer, built under the Max Planck Institute for Solar System Research with NASA contributing components, combines a gas chromatograph, a mass spectrometer, small furnaces and — crucially — an excitation laser that can lift organics off crushed rock without ever triggering the perchlorate reaction that ruined Viking. It can see organics at ten parts per billion by weight.
In July 2026, researchers at Max Planck, Göttingen and the Université Côte d’Azur reported a validation of one of the instrument’s sharpest tricks using replicas of its capillary tubes. The targets are pristane and phytane, two hydrocarbons that are common in terrestrial petroleum because they are the degraded remains of chlorophyll. They are chiral — they come in mirror-image forms — and biology, unlike geology, is notoriously lopsided about which mirror image it uses. MOMA’s coated capillaries make the two forms travel at different speeds, separating them in time. It is a discriminating test of the kind Viking never had.
The team is also braced for the epistemology. Vago’s group devised an ExoMars Biosignature Score to grade any candidate finding across sedimentary morphology, chemistry and geological context, on the explicit principle that no single measurement will ever be enough. He expects to spend at least two years with the data before saying anything, and wants “several independent lines of evidence” before he will use the word.
China: the mission that says the word out loud
China’s Tianwen-3 is the most direct challenge to American primacy in this field in fifty years, and what is striking about it is not its schedule but its stated purpose. Where NASA’s remaining Mars line items are about propulsion and bandwidth, Tianwen-3 names the detection of potential biosignatures as its primary scientific goal. Chief scientist Hou Zengqian, an academician of the Chinese Academy of Sciences, set out the architecture in Nature Astronomy in June 2025.
The plan calls for two Long March 5 launches from Wenchang in the late-2028 window — one carrying a lander and ascent vehicle, the other an orbiter and Earth-return craft — with a rendezvous in Mars orbit and delivery of no less than 500 grams of Martian material to Earth around 2031. There is no rover. Sampling is done three ways: a surface scoop, a drone ranging a few hundred metres, and a drill reaching two metres down, which would be a world first. Candidate landing sites have been narrowed from more than eighty to nineteen, constrained to between 17 and 30 degrees north by engineering, with three finalists due by the end of 2026. A high-security laboratory with ultra-clean and biosafety zones is being built to receive the material.
The programme is moving. Chief designer Liu Jizhong told reporters in March 2026 that key technologies had been demonstrated and that the team was entering flight-model development within the year. In April, at China Space Day, CNSA formally unveiled the mission and offered twenty kilograms of payload capacity to international partners, selecting five cooperative projects. China has committed to opening the returned samples to international researchers after safety verification.
The comparison with MSR has quietly inverted. For years the question was whose samples would arrive first. It is now whether the American–European samples will be collected at all — and the ones sitting in Jezero are, by any scientific reading, the more interesting rocks, chosen after two decades of deliberate reconnaissance.
Where India stands
For Indian readers the question is not merely spectatorial, because India is building a Mars surface programme at precisely the moment the field’s incumbent is retreating from its scientific core.
ISRO’s Mars Orbiter Mission made India the first Asian nation to reach Mars orbit, and the first anywhere to do it at the first attempt, operating from 2014 until contact was lost in 2022. Its successor is a different order of ambition. The Mars Lander Mission — popularly Mangalyaan-2 — was cleared by the Space Commission in February 2025 and awaits Union Cabinet approval. As described by ISRO Chairman V. Narayanan, it is a roughly 4,500-kilogram stack launched on LVM3, comprising a cruise stage and a descent stage, and delivering a lander, a rover and a rotorcraft of about five kilograms to the surface using a sky crane and a supersonic parachute. Launch has been indicated for around 2030. Success would place India in the very short list of nations that have landed and operated on the Martian surface.
It is worth being precise about what that mission is and is not. Its declared objectives — surface composition, mineralogy, geomorphology, atmosphere, interplanetary dust, and the ionospheric measurements of instruments such as ARIS, built at IIST — sit squarely in the habitability and environment tradition. It is not a life-detection mission. That is not a criticism; it is the correct sequence, and it is exactly the logic NASA adopted after Viking taught the field that you cannot ask the ultimate question until you have learned the planet’s grammar. India is at the stage of learning the grammar, and doing so with a maturity of design — sky crane, supersonic parachute, rotorcraft — that would have been unthinkable for an Indian programme a decade ago.
But the American experience carries a warning worth reading carefully in Bengaluru and in Delhi. NASA did the grammar for thirty years, and did it superbly, and then discovered that the political system which funded the preparation would not fund the payoff. A methodical strategy is only as durable as the budget line at its far end. India’s planetary community — at the Physical Research Laboratory’s planetary sciences division, at the Space Applications Centre, at VSSC — is currently designing for the 2030s. The lesson from Jezero is that the hardest part of a long-horizon science programme is not the engineering. It is surviving the fourth election after it starts.
There is a second, subtler point. The man who announced NASA’s reorientation at Ignition, Associate Administrator Amit Kshatriya, is of Indian origin — a reminder that the Indian scientific diaspora is now not merely present in the American space programme but making its strategic decisions. Whether India also acquires the capacity to make such decisions on its own terms, with its own instruments answering its own questions, is precisely what the next decade will test. Tianwen-3’s open call for international payloads and open access to returned samples is a live demonstration of how quickly the etiquette of planetary science can be rewritten by whoever is actually flying.
What happens next
The obituary may be premature, and the next twelve months will decide.
The Senate Commerce Committee passed the NASA Authorization Act of 2026 unanimously on 4 March, co-sponsored by Chairman Ted Cruz and Ranking Member Maria Cantwell. It authorises $24.7 billion for fiscal 2026 and $25.3 billion for 2027. On Mars sample return it does something interesting: it directs the formal termination of the old programme and simultaneously orders the creation of a new one, under a total cost cap of $8 billion, with NASA required to deliver a plan including cost and schedule within 120 days of enactment. It also insists the Mars Telecommunications Orbiter stay administratively separate from that effort. The House Science Committee passed its own version on 4 February. The two bills differ substantially, and neither is law.
Industry has read the room. Whitley Poyser, director of exploration at Lockheed Martin Space, said in April that the company has a plan to return samples for under $3 billion by leaning on proven technologies and cutting mission complexity, and that Lockheed stands behind it. Whether that number survives contact with an actual contract is a separate question; the history of this programme is a history of numbers that did not.
Against that, the fiscal 2027 request released on 3 April proposes cutting NASA science by 47 per cent to $3.4 billion — the same assault Congress rejected a year earlier, returned. It keeps Mars Future Missions at $110 million but repurposes the money away from sample-return technologies and towards a cadence of smaller, cheaper missions, with a solicitation due in October for a 2030 launch capped at $220 million. Four senators — Adam Schiff, Alex Padilla, Mark Kelly and Ben Ray Luján — have written to appropriators asking for at least $400 million for the line in 2027 and warning of severe and irreversible harm without it. Senator Jerry Moran, who chairs the relevant subcommittee and is one of the letter’s recipients, has already signalled he wants a robust and balanced NASA bill.
So the honest summary is this. Mars sample return is not dead in law; it is dead in cash. There is an authorised concept with an $8 billion ceiling, a contractor claiming it can be done for under $3 billion, an appropriations line of $110 million pointed somewhere else, and a rover that has finished its part of the job.
Coda: the question and the instrument
It is tempting to read this as a story about American political dysfunction, and it partly is. But there is an older and more uncomfortable pattern underneath, and Viking is the key to it.
Viking failed not because Mars was barren but because we asked the question before we had built the instrument capable of hearing the answer. We then spent fifty years building the instrument. Perseverance is that instrument’s field half; the world’s isotope laboratories, synchrotrons and electron microscopes are its other half; and the only thing missing is the eighty-kilogram problem of moving thirty small tubes across 225 million kilometres of vacuum. We arrived, at last, at the point where the question is answerable — and stopped asking it.
Ashwin Vasavada, who has worked on Curiosity since 2004, calls the whole history of Mars exploration a roller coaster. The metaphor is generous. A roller coaster returns you to the platform. What has happened here is that the carriage has been detached at the top of the climb and the passengers have been told the view from up here is excellent, and that a nuclear-powered helicopter carrier will be along in 2028.
Sapphire Canyon will keep. Rocks are patient; that is their defining professional quality. The organic carbon in Cheyava Falls has waited something like three and a half billion years for someone to ask what made it, and it can wait a few decades more. The question is whether the people who know how to ask — the instrument builders, the isotope geochemists, the graduate students who would have spent careers on those tubes — will still be there. Programmes do not usually die from a vote. They die when the expertise disperses and nobody notices for ten years.
Jorge Vago, who has spent more than two decades trying to get one rover to Mars through budget cuts, a crashed lander, a pandemic and a war, declines to see any of this as a race. He wishes everybody well, on the grounds that what is being sought matters to humankind rather than to any flag. It is a decent sentiment, and it is also the only one that survives contact with the facts. The rocks are on Mars. Somebody will fetch them. The only remaining question is whose laboratory gets to open the tube — and whether the country that spent fifty years and countless billions figuring out exactly which rock to pick up will be in the room when it is finally cracked.
– Vijaya Yandrapalli
