Viking's search for life on Mars


Figure 1. The engineering replica (simulator) of the Viking lander at the Jet Propulsion Laboratory (JPL) at Pasadena, California, was used by scientists to solve problems as they occurred with the actual spacecraft on Mars. They were thus able, for example, to free a locking pin on the Viking 1 soil scoop that had failed to eject. JPL commanded the scoop arm to shake out the pin on the Viking 1 soil scoop that had failed to eject. JPL commanded the scoop arm to shake out the pin after working out a series of movements with the simulator. The 'repair' was effected across more than 340 million kilometers (212 million miles) of space. The sampler scoop, cameras (upper left and center) and meteorology beam (upper right) can clearly be seen in the photograph.

surface of Mars

Figure 2. Surface of Mars from Viking 2 lander at Utopia Planitia. A thin layer of frost can be seen.

Twenty-five seconds after settling onto the red sands of Mars, the Viking 1 lander beamed back its first image to Earth – an unprepossessing, black-and-white snapshot of one of its footpads and some nearby stones. Within a day the first color photo was in, giving a far more impressive view out to the horizon of the red and orange boulder-strewn plain of Chryse Planitia. Weather reports started streaming their way back home: conditions at the Viking 1 site were clear, cold, and uniform; the daytime sky was butterscotch in hue (pink at sunrise and sunset) thanks to a permanent haze of dust.


A few technical glitches, not surprisingly, cropped up during the first few days on the surface. Viking 1's seismometer remained stuck in its protective cage after a pin-pulling device failed to detonate on landing (the same instrument on Viking 2 later deployed properly). The spacecraft's UHF transmitter switched to a low power setting before self-correcting back to its normal 30-watt mode within a day or so. More seriously, the sampler arm, whose task was to deliver soil to the biology experiments, the gas chromatograph-mass spectrometer, and the X-ray fluorescence spectrometer, jammed. Technicians feared it might be an electronic problem. Certainly any delay in fixing it would have impacted on the soil acquisition sequence, scheduled to begin on sol 8. (One sol is a Martian day, which is 40 minutes longer than an Earth day. Sol 8 therefore was the eighth Martian day after the spacecraft landed.) Fortunately, the solution proved satisfyingly simple. Engineers realized the arm might be stuck because a locking pin, which was part of the shroud latching system, hadn't dropped free. Sure enough, extending the boom of the sample arm did the trick, allowing the pin to fall to the ground in front of the craft, where onboard cameras could see it.


Everything was now ready for the main business of the mission to begin. Packed into a space smaller than a microwave oven, Viking's biology payload held the hopes and dreams of a race that had long wondered if it was alone in the universe. With the preliminary testing of the spacecraft done, Viking's biology team sent up commands for the life-detection gear to swing into action. Meanwhile the media on Earth were busily hyping up popular expectations that Viking would "prove", once and for all, whether the Red Planet was alive.


Positive signs

The first soil samples were scooped up, as planned, on sol 8 (28 July 1976). Four samples were dug, the first being placed into the biology instrument distributor assembly, the next two into the GCMS processor, and the fourth into the funnel of the X-ray fluorescence spectrometer. Three days later Viking mission manager Jim Martin was able to report at a news briefing that biology data had begun to roll in.


From the outset, the messages sent back by the Viking probes were extraordinary – and deeply puzzling. Vance Oyama's Gas Exchange (GEX) experiment on Viking 1 was the first to return data, having begun a cycle in "humid mode," in which the nutrient medium was added so that the soil didn't come into contact with the nutrient, but was exposed to water vapor in the atmosphere). A rapid outpouring of oxygen was detected from the sample during the early stages of incubation.


Gil Levin's Labeled Release (LR) experiment reported back next with news of a similar recorded rush of radioactive carbon dioxide when its sample was wetted with radioactive nutrients. Such high levels of activity were surprising, because they were comparable with the results of LR tests conducted on Earth; true, they were at the lower end of the terrestrial measurements but given that Mars is much drier and colder than Earth, the activity was much greater than most scientists had expected. Something remarkable was happening in the Martian soil – although exactly what was far from clear. The rapid outflow of gas in the LR experiment slowed down drastically after only 70 hours. If Earth microbes had been in the sample, the activity would have been less pronounced and more drawn out, continuing for perhaps a week longer.


The world's press seized on these early sensational results as virtual proof of life. But the Viking scientists remained more cautious. Although living organisms could have been responsible for the dramatic changes seen in the GEX and LR experiments, the speed of the reactions was suspiciously like that of chemical processes. Compounds that are highly oxidizing – in other words, very effective at transferring oxygen or, more generally, taking electrons away from other substances, and therefore unusually reactive – might, it was suggested, explain the high rate of climb of the gas emissions. Among the chief suspects were superoxides (chemicals that contained negatively charged oxygen in the form of O2) and hydrogen peroxide, familiar as a bleach. But the oxidizing chemical theory seemed hard to square with the duration of the gas release. As biology team leader Harold Klein remarked early on, though some of the observed activity was probably chemical in origin, there could very well be a life component as well. The three-day period until the gas stopped flowing in the LR experiment was tantalizingly intermediate between what would be expected with chemistry and biology, and the results of later tests, as we'll see, made the puzzle even greater.


First results from Norman Horowitz's Pyrolytic Release (PR) experiment soon followed and proved no less intriguing. They showed beyond doubt that there would been some incorporation of radioactively-labeled carbon dioxide and carbon monoxide into organic molecules – just what would be expected from the metabolism of microbes. Instead of a radioactivity count of 15 per min, which was the value predicted if no gas from the experimental atmosphere were assimilated, the level was 95.9 counts per minute. This was on a par with what would be achieved by the sparse life in soil from the dry Antarctic deserts and showed clearly that organic synthesis was going on.


Taken as a whole, the early Viking data were perplexing. Mars was giving signs of life, but these seem to be mixed up with some sort of exotic chemical reactions.


Long before Viking left the launch pad, mission scientists had agreed on a key criterion: a positive result from any one of the three biology experiments onboard the spacecraft would signify that life had been found. Yet now, with the media spotlight on the team and one of the most momentous calls in human history needing to be made – whether or not we are alone in the universe – the situation no longer seemed so straightforward.


Through a gas darkly

The confusion deepened when a second shot of nutrients was injected into the LR experiment on sol 17 (6 August). A fresh surge of radioactive carbon dioxide would have pointed strongly to a chemical cause of the reaction. Instead what happened was completely unexpected. After an initial brief release of labeled CO2, the amount of gas given off dropped to almost two-thirds the previous levels. Researchers were baffled. Some suggested that the gas had been incorporated into microorganisms, others that the equipment had sprung a leak. A few days after the second nutrient injection, the counts of radioactive carbon dioxide started to edge up and were still rising on sol 22 when the incubation period of the experiment's first cycle ended.


There was a palpable sense of excitement among scientists at this stage that life on Mars was creeping closer. The first batch of experiments in which the soil was sterilized before incubation would, it was hoped, give a clearer idea of what was happening. If Martian microbes were responsible for at least some of the activity seen in the first active round of experiments, then this activity ought to be reduced by killing the bugs in advance.


That's exactly what was seen in the LR and PR control experiments. (A control experiment is one designed to show that the factor being tested is actually responsible for the effect observed.) In both the LR and PR control runs, the soil samples were preheated to sterilizing temperatures for three hours. The rapid release of gas that marked the active runs of the LR experiment plummeted tenfold. while organic synthesis in the PR control run dropped 85%. "Even our chemically oriented skeptics are excited about these results," said Harold Klein. Head of the PR experiment team, Norman Horowitz, who was later to emerge as an arch skeptic made this key point: "If we had these results in a laboratory we would have concluded that we had a weak positive signal – weak but positive." Yet he cautioned: "Since the signal comes from Mars, which is an entirely different world and one we don't understand yet, we have to be very careful in how we interpret these numbers."


At this stage of the Viking investigation, two facts stood out. First, Mars had at least one chemical on its surface that was extraordinarily reactive. No one on the Viking team had any about doubt that. Second, some of the results were very hard to explain without invoking life. Take, for instance, the control runs of the LR and PR experiments, which were particularly interesting because they complimented each other. The LR experiment showed that something present only in unsterilized soil could oxidize a nutrient soup to carbon dioxide. The PR experiment, on the other hand, showed that something present only in unsterilized soil could reduce (take oxygen from) carbon dioxide and carbon monoxide in an atmosphere above the soil. Anyone bent on finding a chemical explanation for this double whammy had to conjure up hypothetical reactions never seen naturally on Earth and bordering on the inexplicable. Highly oxidizing chemicals in the soil, which broke down when heated, could partially explain the LR data and some other results, but a net reducing soil was needed to make sense of the PR measurements. Only an exotic effect, beyond the realm of known chemistry, could be squared with both sets of results. On the other hand, microbes could easily produce both types of reaction – as when a plant photosynthesizes carbon compounds from carbon dioxide in the air by day (a reducing reaction), and produces CO2 when it turns food into energy (an oxidizing reaction).


The first active run of the PR experiment also made a powerful case for a blend of high-energy chemistry and life. The measured activity of the sample was hard to fathom without a chemical component, but the unmistakable synthesis of organic material from inorganic building blocks looked authentically biological. In fact, Klein believed at the time that it offered the most convincing argument for life. Only Horowitz, who later became the Viking project's most vociferous opponent of Martian life theories, played down the findings, which ironically came from his own experiment.


Another key observation was made by the GEX experiment. Not only was oxygen released from the soil in the first active GEX run, so too was nitrogen – an element essential to life as we know it. Before Viking arrived, a major unknown clouding the biological prospects for Mars was the availability of nitrogen, given that there are hardly any nitrogen-containing minerals. The discovery of nitrogen, therefore, was a huge boost to the hopes for finding life. The nitrogen release itself, however, was open to both a chemical and a biological interpretation. If inorganic, it could have been due to water vapor escaping from the soil and carrying dissolved nitrogen with it; alternatively, it could have come from decomposing microbes, killed after they were inundated with water. Other Viking experiments, in which an uptake of nitrogen was observed, were similarly ambiguous: the explanation might have been chemical adsorption (a process whereby gas molecules stick to the surface of a solid) or biological fixation (as in the case of many Earth microbes which extract nitrogen gas from the atmosphere).


Life – but no bodies

By the middle of August 1976, all three biology experiments on the Viking 1 lander had yielded provocative results, and a similar set of findings (with a few twists) would soon come from identical equipment on Viking 2 which was about to touch down. Mars had thrown scientists a huge curve ball with its wildly reactive surface. As mission director Tom Young put it: "We did not properly comprehend how complex the Martian problem was." But for all the puzzles and uncertainties posed by the data, it was hard to avoid the suspicion that Mars looked a little livelier each day.


That was about to change. One simple, shocking piece of news from the Red Planet would abruptly shift mainstream scientific opinion away from the view that Viking might have found life. It came from an instrument whose task was to detect and analyze organic compounds (those containing both carbon and hydrogen) – the GCMS (Gas Chromatograph Mass Spectrometer). Given the uncertain verdict of the biology experiments, the GCMS took on the role of an appeals court. If there were microbes dead or alive in the Martian soil samples then there had to be organic compounds.


Connected to the GCMS was a small oven into which a soil sample could be placed. The sample was heated in steps to various temperatures up to 500°C to vaporize or thermally break down any organic substances. The gas chromatograph then separated out the various chemical components produced by the baking process, before passing them on to the mass spectrometer to have their charge-to-mass ratios determined and thus, indirectly, their molecular makeup. A test version of the instrument, identical to the flight units, was used to analyze a number of different soil and rock samples on Earth before and after the Viking missions to evaluate its performance. The scientists in the GCMS project, led by principal investigator Klaus Biemann of the Massachusetts Institute of Technology, reported that it identified a wide variety of organics in meteorite samples and Antarctic soils in concentrations as low as a few parts per billion. However, this claim applied to gases given off when the sample was heated to 500°C – a temperature not high enough to release gas from some potential organics that might be present in the sample.


On sol 17 (6 August), the Viking 1 GCMS carried out its first analysis of Martian soil. About 300 mass spectra were beamed back to Earth, showing the various compounds the instrument had found in the sample. None were organic. That wasn't too surprising, as the sample had only been heated only to 200°C, which was below the temperature at which organics were expected to break down into small enough fragments to be detected. But what happened next was devastating.


A second run of the GCMS took place on sol 23 (12 August) following reheating of the first sample to a maximum of 500°C. Once again, the GCMS failed to catch even the faintest whiff of organic material. So, too, did a subsequent analysis at the Viking 1 site, and several others at the Viking 2 location. To within the detection limits of the GCMS (a factor to be much debated over the coming years) there were absolutely no carbon-hydrogen-bearing molecules in the Martian surface soil at either site.


This seemed hard to believe. Whether there was life or not on Mars it had been known for many years that some meteorites harbor a slew of organic compounds. Such rocks must have peppered Mars over the eons just as they have the Earth. Two decades after the Vikings landed, it came to light that rocks from the Red Planet have actually found their way to Earth and these chunks of the Martian crust carry a treasure trove of organics. What's more the PR experiment had detected the synthesis of carbon-hydrogen compounds, indicating either the presence of life itself or of chemicals that can easily give rise to substances found in living organisms. Something didn't add up, and a number of Viking scientists questioned what might be behind the negative results of the GCMS at both landing sites.


The team in charge of acquiring samples suggested that the GCMS processor might not actually have received any soil. This was because there was no fool-proof way on the instrument of telling whether a sample was in the test cell. When the GCMS instrument was purged, it was hoped to see the expelled soil on the ground but, in the event, the camera view was obscured by the sampling arm. The fact is, there was never any direct, incontrovertible evidence that full or even partial samples were received. The conclusion that samples had been received was based solely on the detection of carbon dioxide and water vapor – compounds which are common in the Martian atmosphere and would have been picked up even in the absence of a soil sample.


Gil Levin and his assistant on the LR experiment, Patricia Straat, brought up another issue, concerning the sensitivity of the GCMS. They pointed out that their LR apparatus was about a million times more sensitive than the GCMS. Wasn't it possible, they asked, that the GCMS was simply blind to the small amount of organics that would be involved if the LR results were due to microorganisms? Biemann acknowledged that his equipment needed about a billion cells per gram of soil (equivalent to about a million cells per GCMS sample) in order to detect organics. But he assured Levin and Straat that there would be plenty of dead cells from the baking process (paralysis) at 500°C to provide that much raw material. As we'll see later, Biemann's claims of a three-parts-per-billion sensitivity, would eventually come under strong scrutiny – but only long after most scientists had lost its taste for Viking biology.


It took a few months for the bad news from the GCMS instruments on both Vikings to build up and for the objections of dissenters such as Levin to be put to bed. One of the last of the fence-sitters to come to terms with the failure of the GCMS to find organics was Viking Project Scientist Jerry Soffen. At first he argued with Tom Young that the test cells must have been empty because there had to be organic matter of some sort in the soil. But eventually he was persuaded and was heard to mutter one day, as he walked away from where the data was being analyzed, "That's the ball game. No organics on Mars, no life on Mars." Even as the twin Vikings continued their investigations, NASA announced to the world that the spacecraft had drawn a blank in their life quest: the activity observed in the Martian soil could be explained in purely chemical terms.


Divergent paths

Almost everyone on the project, as well as the wider scientific community, jumped on to the GCMS bandwagon and took up the mantra "no organics, no life." Reinforcing the GCMS results were those of the GEX experiment which could clearly be best explained in terms of chemical activity. The biological case, strongly supported by some of the LR and PR measurements, was increasingly dismissed, as too was the fact that the GEX experimental conditions would likely be hostile to any life on Mars, as pointed out much earlier by Wolf Vishniac.


As the months went by, a clear division opened up among the Viking scientists. Harold Klein and most of the other researchers involved in the life-seeking mission concluded that the results of the experiments, taken as a whole, could best be explained by chemical effects alone. The one major dissenter was Gil Levin who argued forcefully against the general opinion that whatever his LR experiment had detected it wasn't microbial life.


Klein urged Levin to keep quiet about his suspicions, and tended to shunt him out of the limelight at press conferences. Mission Director, Jim Martin took a different tack and told him: "Damn it, Gil, why don't you just stand up and say you detected life."


From the outset, Levin was the oddball of the group because of his less academic, more hands-on background. His path to becoming a principal investigator on the Viking mission had been unusual. Prior to his NASA engagement he'd worked as a sanitation engineer for several state health departments, concerned with microbial contamination of drinking and swimming water. He invented a method called radiorespirometry, which involved adding small amounts of radioactive nutrients to the biochemical soups. Organisms couldn't distinguish between normal and radioactive nutrients and any consumption could easily be measured with a radioactivity counter – to an astonishing level of sensitivity (about 10 bacterial cells per sample).


Levin's association with the space agency began in 1958 when he accompanied his wife to a Christmas party and there met the first NASA administrator, Keith Glennan. During a conversation, Levin spoke about his research and Glennan suggested that he send a proposal to Clark Randt, head of the new NASA biology program. Levin's proposal to develop the radiorespirometry experiment for use on Mars was eventually selected, in the face of fierce competition, and renamed the LR experiment.


Even in the early days of Viking, as the mission started to take shape, the potential for disharmony among the project scientists was clear. Levin was an engineer by training and disposition, and upbeat about the prospects for life on Mars. Norman Horowitz, like most of the others on the Viking science team, was a career academic, and of everyone involved in the mission, the least hopeful of finding life. It's hardly surprising that Levin and Horowitz eventually found themselves at loggerheads.


As the Viking landers moved forward with their investigations, Levin and his co-worker Patricia Straat insisted that the LR results were in good accord with a biological interpretation. Central to their claim was the remarkably uniform production of gas from the LR nutrient when it was added to soil samples at both lander sites, and, even more importantly, the biologically-consistent responses from the whole range of heat-treated control samples.1


Exposing a duplicate sample to the one that gave a positive response to 160°C for three hours rendered it inactive. This satisfied a pre-mission criterion for life, argued Levin, because any likely chemical reagents would have survived such heating and given another positive response. What's more, further tests, showed that the active agent in the soil was destroyed as 51°C, reduced by 70% at 46°C, and, most tellingly, eliminated altogether and standing three months in the dark inside the sample container held at 7° to 10°C.


The samples did, however, keep their activity for up to several Martian days in the the chamber held to 10°C before testing. No chemical oxidant, among the many proposed, could duplicate this thermal sensitivity profile.


Despite these arguments, the consensus within the Viking science team, and within NASA as a whole, shifted relentlessly toward a non-biological explanation. Theories about what inorganic compounds and reactions might have caused the Viking results became the order of the day. Most invoked some kind of very strong oxidant, or combination of oxidants, which would react with water to produce oxygen and hydrogen, and with nutrients to generate carbon dioxide. This notion that powerful oxidizers were at work, fed the suspicion that Mars was sterile because such compounds would, it was assumed, be inimical to life as we know it.


Howard Klein freely acknowledged that there was room for doubt: each of the Viking biology experiments operated under conditions that weren't the same as those actually found on Mars. "While we have obtained significant and fascinating data in the Martian experiments," he said, "we may not have hit upon the proper conditions to elicit evidence of Martian metabolism."2 His statement mirrored the realization that the Viking life detection experiments were carried out before we had a proper handle on the Martian environment and therefore had a solid basis on which to interpret the results.


Klein concluded that while some of the results fitted a biological interpretation, most were hard to reconcile with life as we know it. Interestingly, the experiment that he thought made the best case for life was the PR experiment. "An explanation," he said, "for the apparent small synthesis of organic matter in the PR experiment remains obscure." However, the principal investigator of the PR experiment, Norman Horowitz, was never in favor of Martian biology. In fact, from the outset, Horowitz had such strong doubts about finding anything alive on Mars that on several occasions other members of the team wondered aloud why he had remained with the group. 2


In 1978 Klein published his, and effectively NASA's, definitive word on the subject. He weighed the various chemical and biological explanations on offer and concluded that, while some of the data were consistent with a biological interpretation, most were better in tune with a reactive chemistry scenario. 3


Most scientists rallied around Klein's assessment and started theorizing about which strong oxidizer might explain the Viking results. Hydrogen peroxide was an early favorite, but no one could come up with a mechanism for making enough of it to match the responses of the biology experiments, particularly the LR experiment. It also wasn't clear how this compound could be stable enough on the Martian surface to accumulate amid the battering of intense ultraviolet and other destructive forces. If hydrogen peroxide were the mysterious active agent, it would have to be wrapped up somehow in chemical complexes with minerals in the soil.


Although the mysterious oxidant couldn't be pinned down, scientific support for the interpretation that all observations could solely be explained by chemical reactivity remained strong, and Levin and Straat were increasingly sidelined. Meanwhile NASA was so convinced that Mars was dead that the biology program was swiftly wrapped up and its principal investigators left to look for new jobs. Extraterrestrial life research at NASA continued only through the poorly-funded Exobiology Branch until the Astrobiology Institute was set up in 1998. Ironically, the two major events which propelled the formation of the Astrobiology Institute were the possible detection of fossilized life in a Martian meteorite and the surprising discovery that life thrived in a variety of extreme environments on Earth.


Oxidants? what oxidants?

The idea that the Viking results were caused by highly oxidizing compounds born of the interaction between ultraviolet radiation and the Martian soil became the paradigm of choice. This was despite the fact that the theory failed to explain a number of key observations. Most obviously it left unanswered the question of why the PR experiment recording the synthesis of organic material.


Among other things it didn't address a curious outcome of one of the LR experiments on Viking 2. In this experiment, the reactivity of soil samples taken from in the open and under a rock, nicknamed Notched Rock, was compared. It turned out that the soil which had been sheltered by the rock, and therefore in the dark, was almost as reactive as soil that had been lit, and far more reactive than samples that had been heated to sterilizing temperatures – or even to a mere 50°C. Although light was evidently a factor in boosting the reaction rate, just as clearly some of the reaction was still able to take place in darkness. This was hugely significant because it implied that some of the reactivity seen in the PR and LR experiments couldn't be due to an oxidizer that forms from radiation interacting with the Martian soil. The patch sampled under the rock at the Viking 2 site had surely not been illuminated for millions, and perhaps even tens or hundreds of millions of years. Yet the soil gave a positive response to the nutrients offered to it in the LR run. That was a puzzle for anyone favoring a purely chemical hypothesis because it's hard to envision how any highly oxidizing compound – one that can supposedly split apart water into oxygen and hydrogen (as in the GEX experiment) – could come about without the direct influence of strong radiation. The surprising reactivity of the under-rock soil could, however, be explained by the presence of microbes which relied not on sunlight but on a ready-made supply of organics.


The fact is that no chemical model so far devised adequately mimics the Viking results. What's more, no suitable oxidant, including hydrogen peroxide, has been detected at the levels required by any subsequent mission to the Red Planet. The bit of inorganically-produced hydrogen peroxide that is present on Mars occurs at levels less than one percent of those needed to explain the Viking responses.


Most damning of all, the same type of GCMS that failed to detect any organic matter on Mars also reported a sample of Antarctic soil to be sterile, even though a later wet chemistry analysis demonstrated the presence of organic matter and the likely presence of microorganisms in a similar sample. This was the very instrument used to settle the dispute between biological and chemical interpretations of the LR, PR, and GEX data. In 2000, the Viking GCMS came under fresh scrutiny when chemist Stephen Benner, then professor at the University of Florida in Gainesville and now director of the Westheimer Institute of Science and Technology, pointed out that it would have been unable to detect certain organic compounds potentially critical to a life investigation.4 More recent and devastating was the blow to the instrument's credibility dealt by a group of scientists led by Rafael Navarro-Gonzalez of the National Autonomous University of Mexico. The group reported that the sensitivity of the Viking GCMS was several orders of magnitude lower than originally thought.5


At the same time, some of the key arguments leveled against a life interpretation of Viking's findings have been undermined. One of these criticisms is that life needs water and there was no evidence of water at the Viking sites. However, frost was observed at the Viking sites and extensive regions had already been found on Mars by the Mariner 9 orbiter where the surface pressure exceeded the triple-point pressure (the temperature and pressure at which a substance can exist simultaneously in its solid, liquid, and gaseous states) of water thus allowing water to exist in the liquid phase. Furthermore, Viking data indicated that the temperature of the top several millimeters of soil beneath the Viking 2 lander sampling had risen to 0°C (where ice liquefies) and remained there for at least several minutes.6 Also, the advancing science of extremophiles – life in extreme environments – has shown that many organisms can survive being dried out in a dormant state for extraordinarily long periods, perhaps even hundreds of millions of years.


Harold Klein insisted that no known Earth organism had been shown capable of reproducing all the Viking results. However, any Martian microbes would inevitably have adapted over many million of years to the conditions found on the planet today. The critical question then becomes: what are the ultimate limits of life under the environmental conditions to which a soil is exposed? For example, we know that the Martian environment is poor in nutrients. One strategy that terrestrial microbes use to deal with nutrient-poor environments is to become small. Bacteria in a growing state typically have a cell diameter of about one micron (millionth of a meter), but many bacteria can shrink to one-tenth of this size or less when faced with starvation or other extreme stresses. Other survival strategies used by microscopic life are to become highly efficient in metabolizing the few nutrients on tap, or to go dormant and then reproduce quickly when food becomes available.7


Life – but not as we know it

The only spacecraft ever to search for life on another world was built to look for the kind of microbes with which scientists were most familiar in the 1970s. Viking's experiments supplied a specific, narrow range of environments and nutrients that, in pre-flight tests, had proved attractive to common-or-garden Earth bugs. These experiments were also designed in the expectation (known as the Oyama model) that if life did exist on Mars, it would spread over the whole planet at least as densely as bacteria occur on challenging locations such as Antarctica.


A lot has been learned both about the limits of life on Earth and about conditions on Mars in the three decades since Viking. An extraordinary variety of extremophiles has come to light, vastly broadening the scope of potentially habitable places on other worlds, and thanks to a number of advanced robotic probes in recent years we've added terabytes to our library of data on the surface, subsurface, and atmospheric environments of Mars.


Yet, for all this new knowledge, there's still a tendency to think of life from a terrestrial perspective. Nowhere is this more evident than in the persistent arguments put forward to account for the Viking results in chemical rather than biological terms.


A very different picture emerges, however, if we free ourselves from the shackles of geocentrism. Why on Earth – or, rather, on Mars – should organisms on the Red Planet exploit the same biochemistry and evoke the same biological responses as terrestrial microbes? Whether Mars came up with its own life from scratch, or received an early biological gift from Earth via meteorites, adaptation and evolution under alien conditions would surly have given rise to species today that have alien biosignatures.


It was with such thoughts in mind that Joop Houtkooper, of the University of Giessen in Germany, came up with a novel idea of life on the fourth planet. Houtkooper was initially inspired by an essay by the physicist Freeman Dyson called "Warm-Blooded Plants and Freeze-Dried Fish" and first presented his concept of Martian fauna at the Bioastronomy 2004 conference held, appropriately enough, in Iceland. Shortly after, Schulze-Makuch teamed up with Houtkooper to develop this idea further.


Put simply, the two scientists envisioned microbes on Mars that use not plain water inside their cells (as Earth life does) but a mixture of water and hydrogen peroxide.8 The phrase "peroxide blond" is a familiar one. Hydrogen peroxide (H202) is a potent chemical that's just the job if you fancy bleaching your hair, but isn't the sort of stuff you'd want sloshing around inside your body. At the same time, it isn't entirely unknown among creatures on Earth. Some bacteria produce it and even use it in their metabolism, while keeping its reactivity in check with a chemical stabilizer. There's also the remarkable case of the Bombardier beetle, which produces a solution of 25% hydrogen peroxide in water and sprays it on to any unfortunates that it considers a threat.


Given that some terrestrial life forms have evolved to be able to harbor high concentrations of hydrogen peroxide, it's by no means far-fetched to speculate that microbes on Mars might have made it an integral part of their biochemistry. An intracellular cocktail of hydrogen peroxide and water would offer a number of benefits to organisms in the cold, dry Martian environment. The freezing point of a hydrogen peroxide solution can be as low as –56.5°C (depending on the peroxide concentration); below this temperature, the solution becomes firm but doesn't form cell-destroying crystals, as water does. Hydrogen peroxide is hygroscopic, which means that it attracts water vapor from the atmosphere – a valuable trait on a planet where liquid water is scarce. And finally, peroxide offers a rich source of oxygen to power cellular activity.


On Earth, there are good reasons why life doesn't generally embrace the potential benefits of hydrogen peroxide. Earth is two-thirds covered in water and so not the best place for a hydroscopic and reactive compound to prove the merits. If organisms here need a form of antifreeze they tend to use salts, which are highly soluble in water and mimic the chemistry of Earth's oceanic environment. Early terrestrial organisms were exposed to salty ocean water and learned to adapt to salt in high concentrations, and further use it as antifreeze in cold places such as mountainous regions and the Arctic.


On Mars, liquid water was never as plentiful as on Earth. Even during the early, wet phase of Martian history, its seas and lakes were not long-lived in geological terms. Organisms on the fourth planet could have adapted to use the properties on hydrogen peroxide to their advantage, especially as the planet became increasingly dry. Over time, the hygroscopic character of this potent chemical, a handicap on the wet Earth, would become an advantage on our desiccated outer neighbor.


The Viking mystery solved?

The possible existence of powerful oxidizing chemicals, including peroxides, had been the most popular conventional explanation for Viking's surprising observations, and also a reason to assume that life might be impossible on or near the surface of Mars today. What this new theory does is turn the argument on its head by proposing that living Martians actually make use of hydrogen peroxide in running their cellular machinery. If they do, then, remarkably, it would explain all the most puzzling results of the Viking biology experiments.


The peroxide-water hypothesis addresses all the major results from the life detections experiments plus the non-detection of organics by the GCMS. In particular, it offers a good explanation for problematic issues that have haunted the mission since the data were first received.


The Viking GCMS heated the Martian soil samples to temperatures of several hundred degrees Celsius to analyze them for organic compounds. Such heating would kill any organisms. Being a powerful oxidant, hydrogen peroxide, when released from dying cells, would sharply lower the amount of organic matter in its surroundings and produce carbon dioxide instead (which was measured by the GCMS). Combined with the fact that the GCMS was much less sensitive than originally thought, this would explain why the instrument didn't detect any organics on the Martian surface.


The new hypothesis would explain the nature of the mysterious oxidant on Mars, since the hydrogen peroxide in the internal cellular fluid is part of the very biochemistry of the Martian organisms.


The most puzzling result from the GEX experiment was the enormous release of oxygen. This can be interpreted in two ways: as the result of an energy-producing metabolism, or more likely, upon humidification, as due to the decomposition of Martian organisms. After death, the microbes' internal peroxide would disintegrate under Mars' surface conditions into oxygen and water, catalyzed by inorganic oxides in the soil.


The LR experiment, in which samples of Martian soil were exposed to water and a nutrient source including radio-labeled carbon, showed rapid production of radio-labeled carbon dioxide which then leveled off. The initial increase would have been due to metabolism by peroxide-containing organisms, and the later leveling off due to the organisms dying from exposure to the environmental conditions. The possibility of die-off in the LR experiment had been mentioned by Levin earlier,9 but he couldn't pinpoint the reason for it. The peroxide-water hypothesis explains why the experimental conditions would have been fatal: microbes using this mixture would wither "drown" (in scientific terms, suffer hyperhydration), or they would burst, due to water absorption by the hygroscopic peroxide, if suddenly exposed to water.


The possibility that the tests killed the very organisms the Viking team was looking for is also consistent with the results of the PR experiment, in which radio-labeled carbon dioxide was converted to organic compounds by samples of Martian soil. Of the seven tests done, three showed significant production of organic substances and one showed a much higher production prior to death. The variation could be due to patchy distribution of microbes, but perhaps most intriguing is that the sample with the lowest production (Utopia 2) – lower even than the control – was the only one that had been treated with liquid water.


Only on Mars can the peroxide-water hypothesis of life be properly tested. Either we can use data sent back by current and future missions, and look for results consistent with the hypothesis, or we can design a new mission that tackles the issue head-on. Current missions are not particularly helpful in this regard because they're not designed to search for oxidants or life. All they can do is support the case that Mars was, and potentially still is, habitable, based on the availability of water. However, future missions, such as NASA's Mars Science Laboratory and the European Space Agency's ExoMars, may shed more useful light on the subject.



1. Levin, G. V. and Straat, P. A. "Recent results from the Viking labeled release experiment on Mars." Journal of Geophysical Research, 82, 4663–4667 (1977).
2. Ezell, E. C. and Ezell, L. N. 1984, On Mars: Exploration of the Red Planet 1958–1978. NASA History Office NASA SP-4212, Scientific and Technical Information Branch, National Aeronautics and Space Administration, Washington, D.C.
3. Klein, H. P. 1978, "The Viking biological experiments on Mars". Icarus, 34, 666–674.
4. Benner, S. A., Devine, K. G., Matveeva, L. N., Powell, D. H. 2000, "The missing organic molecules on Mars." Proceedings of the National Academy of Science USA, 97, 2425–2430.
5. Navarro-Gozalez, R., et al. "The limitations on organic detection in Mars-like soil by thermal volatilization – gas chromatography-MS and their implications for the Viking results." Proceedings of the National Academy of Science USA, 103: 16089–16094 (2006).
6. Moore, H. J., Hutton, R. E., Scott, F. F., Spitzer, C. R., and Shorthill, R. W. "Surface materials of the Viking landing sites." Journal of Geophysical Research, 82: 4497–4523 (1977).
7. Schulze-Makuch, D., Irwin, I. N., Lipps, J. H., LeMone, D., Dohm, J. M., and Fairén. A. G. "Scenarios for the evolution of life on Mars." Special Edition on Early Mars of Journal of Geophysical Research – Planets 110: E12S23, doi:10.1029/2005JE002430.
8. Houtkooper. J. M. and Schulze-Makuch, D. "A possible biogenic origin for hydrogen peroxide on Mars: the Viking results reinterpreted." Int. J. of Astrobiology 6: 147–152 (2007).
9. Levin, G. V. and Straat, P. V. "Recent results from the Viking labeled release experiment on Mars." Journal of Geophysical Research 82: 4663–4667