The search for life on Mars

Fifty years of looking, and the question of what would count as an answer

A The search for life on Mars has been conducted for half a century and has produced, so far, no organism, no fossil and no unambiguous chemical signature. This is frequently reported as failure, which mistakes the nature of the enterprise. The searches have progressively established what is not there and what conditions once were, and the second of these has been transformed: a planet that was assumed in the 1960s to be a cold desert throughout its history is now known to have held rivers, lakes and possibly a sea for a period comparable to the interval in which life appeared on Earth.

B The first attempt was also the most direct, and its result has never been entirely settled. Two landers in 1976 carried experiments that added nutrients to Martian soil and watched for the products of metabolism. One returned a positive signal that persisted and behaved, in several respects, as a biological response should; another instrument on the same craft found no organic molecules whatever in the same soil, which was taken to be decisive, since organisms without organic chemistry are not possible. The reconciliation offered at the time was chemical: an oxidising compound in the soil could produce the reaction without life. It was a reasonable inference, and it was made without any direct evidence for the compound in question.

C That evidence arrived thirty years later, and it complicated rather than settled matters. A later lander identified perchlorates in Martian soil — salts that are strongly oxidising and that, when heated, destroy organic molecules and release chlorinated compounds. The 1976 instrument had worked by heating soil. It is now considered likely that organic material was present and was incinerated by the analysis intended to detect it, which means the most-quoted negative result of the whole search may have been an artefact of its own method.

D Methane has supplied the longest-running dispute. A rover on the surface has repeatedly measured small quantities in the air around it, varying with the seasons, while an orbiter designed specifically to detect the gas from above has consistently failed to find it at all. Both instruments are well calibrated and neither team disputes the other's data. The favoured reconciliation is that the gas seeps from the ground at night, accumulates near the surface where the rover sits, and is destroyed after sunrise before it can mix into the atmosphere the orbiter samples. Methane can be produced by organisms and also by reactions between water and rock, so even a confirmed detection would not settle the question it is used to ask.

E The most contentious claim was made about a rock. A meteorite from Mars, recovered in Antarctica, was announced in 1996 to contain structures resembling fossilised bacteria along with minerals and organic compounds of the kind organisms produce. The claim was made carefully, presented as suggestive rather than conclusive, and reported as though it were a discovery. Each strand was subsequently shown to have a non-biological explanation, and the structures proved smaller than any known free-living cell. The episode is now taught less as a scientific error than as a lesson in what happens to a hedged claim when it enters public circulation.

F The current approach abandons the attempt to detect life directly and concentrates on preserving material. A rover is collecting and sealing samples from an ancient river delta, chosen because such a setting concentrates fine sediment in which biological traces are most likely to survive, and leaving them for a future mission to retrieve. The reasoning is sound: the instruments capable of settling the question fill laboratories, cannot be miniaturised, and improve every year, whereas a sealed sample does not deteriorate. The difficulty is that the retrieval mission has proved so expensive that its future is uncertain, and samples in a cache on Mars answer nothing at all.

G Beneath all of this lies a problem the missions cannot solve, which is that there is no agreed definition of what would constitute proof. Every candidate signature — organic molecules, isotopic ratios, structures resembling cells, gases out of chemical equilibrium — has a plausible non-biological origin, and the terrestrial record shows how difficult the judgement is even when the samples are in hand: the oldest claimed fossils on Earth have been argued about for thirty years by people examining rocks they can hold. A definitive answer from another planet, obtained by remote instruments or from a few grams of returned sediment, may be more than the evidence can ever support. What the search can do, and has done, is narrow the range of places where the question remains open, which is a slower and less satisfying result than the one it is funded to produce.