This paper draws a comparison between the 700-year-old historically reported will-o'-the-wisp phenomenon and the more recent discovery of extremophilic colonization of hostile environments; both have been observed as present in isolated, stressed environmental regions and originating from biological phenomena. However, whereas extremophilic activity can be understood in terms of a survival strategy based upon the synthesis of specific suites of protective biochemicals which are designed to control biogeologically the stressed habitats and to provide protection against the extreme environments, the analytical techniques that have proved so successful for the illumination of these survival strategies of extremophiles and which are now being miniaturized for in-field studies and for extraterrestrial exploration have not been applied to a clarification or evaluation of the phenomenon of will-o'-the-wisp. The reason is simply that the will-o'-the-wispsightings have now disappeared completely. Tantalizingly, all of the most reasonable physico-chemical and biological explanations for the will-o'-the-wisp phenomenon proved to be unsatisfactory in some respect and it is clear that, just as in the case of extremophilic colonization, will-o'-the-wisp would benefit from a modern rigorous analytical study which would produce the data from which the potentially novel biological behaviour could be characterized and which would help a better understanding to be made of our natural world.
Extremophiles, which are defined as organisms that can exist in isolated, stressed environments often where other life cannot, have challenged conventional biological and biochemical explanations for their existence which it is now realized is fundamentally associated with their unique ability to create a suite of chemicals designed to protect them from the effect of high-energy radiation insolation, temperature extremes, desiccation, barometric pressure, high metal toxicity and high salt concentrations [1,2]. The recognition of extremophilic survival strategies and their presence in geological microniches is critically dependent upon the adoption of analytical chemical techniques that can indicate the presence of key biomolecular signatures [3,4], usually demonstrated by the presence of an inorganic mineral host matrix that provides a protective shelter for the colonies under study. The observation of extremophiles in their natural habitats is usually a painstaking and sometimes difficult task and in many instances, their presence is confirmed from the analytical data in conjunction with the visual recognition of morphological indicators. Hence, extremophiles have often as a result been considered as rather secretive entities for normal analytical procedures and a great deal of objective scientific detective work is usually necessary to confirm their presence; the importance of the study of extremophiles lies in the widening of the range of conditions under which we would normally expect to see terrestrial life signatures—for example, the thermophilic or thermotrophic salt-tolerant extremophiles existing at −52°C in Don Juan Pond, Antarctica, to the xerophilic extremophiles found at +130°C near volcanic vents (black smokers) on the ocean bed.
In recent years, the discovery of subsurface water on Mars, surface ice on Mars and Europa, and mineral deposits on Mars which could harbour similar endolithic and halophilic life on which has been identified on Earth such as jarosite and gypsum, have raised the question as to whether extraterrestrial life, extinct or extant, could exist in forms similar to those of the extremophiles that are found on Earth [5,6]. This has prompted the adoption of special missions to our neighbouring planets and their satellites in the Solar System over the next few decades to specifically detect surface and subsurface life signatures, the first of which is ExoMars, the European Space Agency mission to Mars and scheduled for launch in 2018 as part of project Aurora, which will has been designed to address the hazards of a manned human exploration mission to Mars scheduled for 2033 or thereafter [7,8].
At first sight, it may be rather challenging to attempt to draw a parallel between what is known about the phenomenon of will-o'-the-wisp and extremophilic behaviour, but actually to analytical scientists engaged in work on the characterization of the latter, several common features are apparent. Both phenomena have long been unrecognized for what they are, they share an unpredictability of discovery and also cannot be simply defined using one set of characteristic values. In addition, extremophiles still often pass uncharacterized through masking by their host or associated matrices. The definition of an extremophile also engenders several and varied tenets under the blanket statement of an organism that can survive and reproduce in conditions that are stressed or even deemed to be impossible for human occupation. As will be shown later, the will-o'-the-wisp occurrence too is phenomenologically similar to this and especially so when it will be appreciated that the scientific techniques that are now being developed for the successful in-field recognition of extremophilic activity, past or present, are precisely those that could be brought to bear on the mysterious will-o'-the-wisp phenomenon.
2. Observation of will-o'-the-wisp
The will-o'-the-wisp conjures up a description of something rather elusive, mysterious and ephemeral; however, historically, the phenomenon has been described by observers in the literature of many countries and it clearly refers to a real event which was widely reported by travellers in Europe and North America who were crossing treacherous habitats, usually marshland, by night. Many legends have been created around this phenomenon, which has been beautifully described by Milton in Paradise Lost:
‘.…. a wand'ring fire
Compact of unctuous vapour, which the night
Condenses, and the cold environs round
Kindled through agitation to a flame,
Which oft, they say, some evil spirit attends,
Hovering and blazing with delusive light,
Misleads th'amaz'd night-wanderer from his way,
To bogs and mires, and oft through pond or pool,
There swallow'd up and lost, from succour far …’
The sense of mystery of the will-o'-the-wisp as a harbinger of bad news and death in the above poem is reinforced by some of its alternative names in the literature of many countries, including ignis fatuus, corpse-candle, irrlichtern, foxfire, fairy lights, Hob's lantern, feu follet and elf-fire. Was this apparition a figment of the imagination of weary and lonely travellers in desolate places (stressed environments!) or does it have some scientific explanation? The first recorded mention of a will-o'-the-wisp sighting appeared in 1340 when Dafydd ap Gwilym (translated by W. Sykes) noted
‘There was in every hollow a hundred wrymouthed wisps’
and he referred to the phenomenon in Welsh as canwyll corff, literally corpse-candle, so formulating the mysterious association between the wisps and burial places. In the English language, William Fulke first reported the wisp phenomenon, which he termed ignis fatuus, as ‘foolish fire that hurteth not’ in his Book of Meteors published in 1563, closely followed by William Shakespeare's mention of ignis fatuus in Act III of his play Henry IV Part I, published in 1608.
Since the first scientific report of the will-o'-the-wisp as ‘ignis fatuus-vapours shining without heat’ in a text on optics by Isaac Newton in 1704 , which led to some intense scientific debate through the eighteenth and nineteenth centuries, the mystery still remains unexplained.
There is much conjecture and little hard evidence to appreciate and evaluate ; the only certainty is that no modern experimental analysis has been carried out on this phenomenon. A distillation of eye witness accounts by observers and scientists  over the past 300 years yields the following information about the will-o'-the-wisp phenomenon
— small, dancing flames observed in marshland, often in close association with standing water and reeds;
— bluish luminosity, with a yellow centre, which is not sufficient to illuminate the ground;
— the flames persist for up to 15 min in the same place, often located six feet (2 m) above the ground;
— the flames are cool to the touch and are up to six inches (15 cm) tall and one-and-a-half inches (4 cm) in diameter;
— multiple occurrences were reported;
— the flames are susceptible to air currents and could be readily extinguished.
Although the phenomenon has been variously attributed to the spontaneous combustion of marsh gas, or to bioluminescence and chemiluminescence processes, it is clear that on closer inspection none of these actually describe fully the eye witness accounts of the will-o'-the-wisp phenomenon.
3. Preliminary analysis
Let us consider some possibilities
Bioluminescence: as exemplified by the glow-worm and the firefly, which have both been incorrectly attributed to the will-o'-the- wisp phenomenon. It has been pointed out as long ago as 1830  that from as close as two metres, personal observation of these lights indicated that there was no flying or suspended insect involved. Biologically, it has been recognized for some time that only the female glow-worm glows brightly but does not fly and that the male Lampyris noctiluca flies but is not luminescent ; the comments of Derham in the Philosophical Transactions of the Royal Society issue of January 1729, are a 14-page compilation of his detailed observations of ignis fatuus over a period of 50 years and represent perhaps the first serious but unsuccessful attempt to explain the mystery scientifically, in which he rejected the idea of flying insects and preferred the idea of a self-igniting gaseous emanation.
Chemiluminescence: defined as chemical reactions which are accompanied by the emission of light at ambient temperatures but without the emanation of heat. Several observers were fortunate to observe the bluish, dancing lights at close range for up to 2 min or more on each occasion and remarked on the coolness of the flame; a brass-ferruled walking stick held in a flame which burned steadily for 15 min, completely odourless and smoke-free, was cold to the touch afterwards.
Combustion: many accounts refer to the will-o'-the-wisp as a flame, but it is clearly not an ordinary combustion flame. In only one case recorded does the observer report the successful ignition of combustible material from a will-o'-the-wisp. Blesson  noted that bubbles appeared from a ferruginous swamp in Gorbitz during the day and at night blue flames were observed at the surface. It was conjectured that these flames could well have been present during the day but that only at the onset of twilight did they become visible. Close approach of the observer invariably resulted in the disturbance of the air near the flames, and on one occasion he was able to first singe and then ignite a piece of paper in the flame. From several hundred reported sightings, this is the only documented case of a will-o'-the-wisp causing the ignition of material and the assignment of this particular observation to a will-o'-the-wisp occurrence must surely therefore be questionable.
4. Some possible explanations
It is without doubt that the will-o'-the-wisp phenomenon was manifestly real and widely observed throughout Europe and North America, at least through the nineteenth century. It was associated with stagnant water in marshy terrain and was seen on dark, warm nights; its flames seemed to have little or no definition of boundary and were variously described as luminous yellow at the centre with a blue colour at their edges. The flames were not hot enough to produce any perception of warmth, much less resulting in the ignition of combustible materials, with the sole exception of the listed observation  noted above. The phenomenon has been reported as occurring near the ground surface water and up to several feet above it and was of sufficient intensity to illuminate the neighbouring vegetation. It is possible that multiple occurrences could have given rise to the idea of ‘dancing flames’.
Although often misattributed historically to some other luminous phenomena, it can be concluded that the will-o'-the-wisp occurrence is definitely not the same as one of the following natural sightings:
St. Elmo's fire: a coronal electrical discharge from conductors and termini of objects in strong electrostatic fields in the vicinity of an electric storm;
Ball lightning: a metastable luminous globular lightning discharge  generated by forked lightning striking near the ground;
Natural gas: an evolution of ‘firedamp’ or methane in the vicinity of geological faults;
Luminous owls: also known as ‘fox-fire’. The contamination of the plumage of roosting birds with fungal metabolites which glow in the dark, and for which the nocturnal activities of hunting owls has provided several reported sightings , cannot be confused with the essentially static and localized dancing flames of the will-o'-the-wisp.
However, the major problem in attempting to provide a rational explanation of the will-o'-the-wisp phenomenon is its contemporary scarcity. There are no reports at all of modern sightings and it is intriguing to speculate on the reason for this. Given that modern travel has opened up many hitherto inaccessible regions to many more putative observers, coupled with a wide public interest in natural and unexplained phenomena through the influence of the media, this is a very surprising conclusion. The very discovery of extremophilic biology in the least expected of places on the terrestrial surface and subsurface is witness to the increased observational scientific prowess that is being brought to bear on natural phenomena.
More recently, Mills  has tried to revive interest in the will-o'-the-wisp topic with an appeal for modern scientific analysis to be applied to the study of the phenomenon. He has undertaken an elegant analysis of the observed sightings and has tried to correlate these with known natural chemical phenomena such as the self-ignition of impure phosphines and methylated phosphines, perhaps arising from a source of bacterially decomposing protein buried in a wet depositional environment. All apparently reasonable explanations, however, have foundered, and like extremophilic activity initially, will-o'-the-wisp must remain a mysterious biological and biochemical phenomenon. Unlike the case of extremophiles, however, where increasing numbers of scenarios are being discovered and reported every year and their diverse biochemical survival strategies enumerated, the major problem facing chemical analysts who would be interested in contributing to a scientific explanation for what must surely be one of the longest unexplained historical natural mysteries is the total disappearance of the will-o'-the-wisp phenomenon in contemporary times.
Perhaps the most acceptable current explanation under these circumstances which can be proposed for the will-o'-the-wisp lights is the generation of pockets of marsh gas, which comprises approximately two-thirds methane and one-third carbon dioxide, resulting from the bacterial (extremophilic?) anaerobic decomposition of cellulose and proteins in vegetation; methane is lighter than air and burns with a clear blue flame with a distinct yellow luminosity and marsh gas bubbles can be ignited in nature above stagnant pools of rotting vegetable matter, where indeed they are observed to burn with a blue flame, even albeit rather briefly. Tenable though this hypothesis may be superficially, however, a major problem relates to the natural process of spontaneous ignition of the marsh gas.
The spontaneous ignition of hydrocarbons produced from decomposition processes in nature has hitherto been ascribed to phosphine [18,19,20], first reported in 1789 by Lavoiser as phosphoretted hydrogen, whose presence in gaseous emanations is attributed to the putrefaction of phosphatic organic matter. The reduction of phosphate to phosphine under bacterial growth conditions has been challenged thermodynamically ; even more significant, however, is the evidence that the well-known flammability of phosphine (PH3) is believed to be due to traces of a higher hydride of phosphorus such as diphosphine, P2H4, or the spontaneously flammable methylphosphines and it appears that pure phosphine itself is stable in air.
To combat this apparent impasse, several theories to explain the self-ignition of marsh gas from biodegradative processes have been proposed , including an unspecified chemiluminescent oxidation of volatiles and organometallic compounds entrained in the methane.
Whatever the true explanation of the will-o'-the-wisp phenomenon, perhaps the most significant and challenging statement is its apparent disappearance from contemporary observation. There have been very few recorded sightings even in the past century, despite the significant increase in tourism and travel to isolated regions alluded to earlier. Could it be that the reason for the disappearance of this haunting phenomenon can be traced to the removal of the environmental habitats in which it once survived; the draining of large areas of swampland as the habitation of rural areas extended outwards from industrialized centres, which incidentally has already contributed to the decimation of the population of storks and frogs for example in some of the lowland areas of Europe, provides a real possibility of this occurrence?
In terms of a comparison with current knowledge of extremophilic activity, the generation of methane extraterrestrially in the subsurface of Mars has excited much interest; in the absence of volcanic activity, it is possible that bacterial methanogens are responsible and their detection would be dramatic evidence in the search for life experiments being contemplated for future missions to the planet's surface and subsurface. Bacterial methanogens have even been suggested as an explanation of the rather ambiguous and mystifying data retrieved from the early Viking chemical experiments carried out on the Martian surface in the mid-1970s. Whatever the future may hold for the discovery of extraterrestrial methanogenic bacteria, unlike their terrestrial extremophilic counterparts which are now providing us with a valuable resource for testing space mission life-detection concepts, the will-o'-the-wisp phenomenon cannot be evaluated in a similar manner until it is reported again—and that challenge is yet to be met.
Or is there somewhere isolated, where at the dead of night the little blue and yellow flames execute their macabre dance, unobserved by Man and safe from observation by the highly sophisticated in-field portable spectrometers and analytical instrumentation now available for environmental monitoring, which could otherwise provide a unique answer to one of the most historically haunting puzzles in natural science, and which has spawned innumerable theories over the last three hundred years? With the current recognition of the necessity for the development of novel miniaturized instrumentation for life detection on extraterrestrial planetary surfaces and subsurfaces and their evaluation over diverse extremophilic scenarios it is perhaps appropriate that analytical scientists address the unsolved problem of the will-o'-the-wisp occurrence that would itself provide an extreme challenge for potential extraterrestrial exo/astrobiological analytical scientific interpretation in the solution of a centuries-old terrestrial observation—but, unlike examples of known terrestrial extremophiles,whose range of survival and identification is increasing the will-o'-the-wisp is now impossible to find. The instrumentation now being developed is ready and the first 21st Century sightings of will-o'-the-wisp are now eagerly awaited … but what has happened to the will-o'-the-wisp in the interim?
One contribution of 14 to a Theme Issue ‘Raman spectroscopy meets extremophiles on Earth and Mars: studies for successful search of life’.
- © 2014 The Author(s) Published by the Royal Society. All rights reserved.