杏吧原创

From the deep

Question: The following is from a gazetteer, published in 1790, about Harlech
in Wales. 鈥淎 remarkable phenomenon was seen near this town in the year 1694 and
it continued about the space of eight months. It was a livid vapour, or fiery
exhalation which seemed to arise from the sea on the borders of Caernarvonshire.
It made its first appearance on the side of a bay, a little after sunset, and
from thence spread itself in the most gradual manner, until it had set all the
houses in the neighbourhood on fire.

鈥淣ot only the ricks of hay, corn, and all other forts of grain were
destroyed, but also the vegetables in the gardens, for it had so noxious a smell
that everything perished where it diffused its baleful influence. Its effects
were severely felt by the cattle, to whom it communicated a contagious
distemper, but which many of them died.

鈥淚t made its appearance regularly every night, always rising at the same
place; nor did it stop its course either by rain or storms. It was sometimes
visible by day, but it was very remarkable that it never did any damage except
in the night. The flames were no way violent, but its continuance at last
consumed everything that opposed it. The methods used to extinguish the flames
were to blow bellows, horns and such like instruments, but that was seldom
attended with the desired effect.鈥�

Can anyone explain what this phenomenon might have been?

Answer: The article concerning houses on fire near Harlech is one of several
based upon an original observation by M. Lister published in the
Philosophical Transactions in 1694 (vol 18, p 49). The original article
differs somewhat from later versions but the account presented by your reader is
very similar.

The phenomenon occurred over the winter of 1693-1694 and describes the
burning of hay ricks and buildings in the vicinity of Tydhin Sion Wyn, a stone
house which still stands on the hillside above Morfa Harlech. Observers noted
that 鈥� 鈥檛was a blew weak flame, easily extinguished . . . that it did not the
least harm to any of the men who interposed to save the Hay鈥�. This is a clue to
the cause of the fire鈥攁 will-o鈥�-the-wisp resulting from the spontaneous
oxidation of diphosphane gas produced by decaying organic matter. Diphosphane is
a hydride which becomes a vapour at between 20 and 30掳 C and can
spontaneously combust in air at low concentrations.

In this case, the oxidation itself is unlikely to cause combustion but it
will probably chemiluminesce in a similar manner to phosphorus vapour undergoing
oxidation. However, there is no doubt that combustion did take place and this
could be due to methane being ignited by disphosphane oxidation. There are
reports of such ignitions. Many years ago the Italian scientist Quirico
Filopanti stationed himself by a pile of rotting hemp for several evenings and
was rewarded with a bright flame with which he could light a torch. The low
coastal plain below Tydhin Sion Wyn would probably be waterlogged and contain
much combustible methane, but I suspect that the resulting fires were much
closer to the source of the gas than stated, unless they were carried by
wind.

Much as I would like to believe this explanation there is a curious twist to
the tale. In the early 1900s there was much excitement along the coast road to
the south of Harlech where strange aerial lights were regularly observed at
night. Later work showed that the phenomena were aligned with the large
north-south Mochras Fault. There is some evidence to suggest that the fault is
still alive. If you trace the line of the fault north of Harlech it runs
coastward of Tydhin Sion Wyn.

And there is a modern parallel too. Four years ago a similar dramatic series
of spontaneous fires erupted in the town of Moirans-en-Montagne in the Jura in
France. 杏吧原创s, exorcists and even occultists were enlisted to explain the
phenomenon. The release of inflammable gases created by small earthquakes
beneath the town was considered a possibility, though this was not accepted
universally.

Allan Pentecost

London

Answer: This coast is one of the most tectonically active in onshore Britain
and lies parallel to the Mochras Fault, which separates the Cambrian rocks of
the Harlech Dome from the Tertiary rocks of Tremadog Bay Basin.

We don鈥檛 know the exact cause of the flames and exhalations, but together
with these accounts and those of similar phenomena nearby, most notably some
written in the 16th century, we can limit the possibilities. None of the
accounts was written by eyewitnesses and the cattle poisoning was said to have
occurred before the fires in the original account, but still may be
connected.

The possibilities are triboluminescence (light caused by friction between
rocks along the Mochras Fault), sonoluminescence (produced by ultra-high
frequency sounds in some organic liquids) and chemiluminescence. The latter is
responsible for the will-o鈥�-the-wisp phenomenon in marshes where decaying
organic matter can produce methane (CH₄) and phosphane (PH₃).
If diphosphane (P₂H₄), which is often present in phosphane as
an impurity, reacts spontaneously with air it could ignite the methane to create
a weak blue flame.

Sediments accumulating offshore in Tremadog Bay Basin may have supplied
clouds of phosphane and methane gas, periodically over eight months, poisoning
grass and setting fire to hay stacks and houses. Most