G. Large chamber.

H. Slot for stabbing a.s.sailants.

K. Ventilating shaft.]

A typical _souterrain refuge_ is that of the Chateau de Fayrolle, not far from Riberac on the Dordogne.

It was accidentally discovered when the proprietor was levelling for terraces and gardens. A glance at the plan will save a description.

A refuge at S. Gauderic has been explored. The region is one of lacustrine deposits called the Sandstone of Carca.s.sonne; it is friable, argilaceous marl. The opening into the hypogee is in the middle of a field, and there are no indications around of the deposition of the material extracted in the formation of the retreat, so as to betray its presence. The visitor descends by a dozen steps into a long corridor, sinuous, and inclining downwards, about 1 foot 8 inches wide, and 4 feet 6 inches high. The pa.s.sage exhibits rebates in several places, into which door-frames had been fitted, as well as square holes into which the beams were run that fastened the doors. It leads past several side-chambers into which the defenders might retire, so as to burst forth suddenly and unexpectedly on the foe, smite him and extinguish any torch he bore. The corridor leads to a rectangular hall 22 feet long and 7 feet high, vaulted and ventilated by three circular airholes, 6 inches in diameter. There are numerous silos in the floor, and fragments of coa.r.s.e grey pottery turned on the wheel have been found there. [Footnote: _Revue de l'Art Chretienne_, Paris, 1868, p. 498 _et seq_.]

M. L. Druyn, in his _La Guyenne Militaire_, Bordeaux, 1865, gives the following account of a refuge he explored. "Ascending the valley that separates the castle of Roquefort from the church of Luga.s.son, after having pa.s.sed the village of Fauroux, one reaches, on the left side of the road, a splendid quarry of hard stone, but a few paces further on, upon the same side, the stone becomes soft. Here on the right, in a little coppice beside the road, is found a place of refuge of which I give the plan as accurately as it was possible for me to take it where one had to crawl on hands and knees, and sometimes wriggle forward lying on one's stomach, over earth that was damp and rubble fallen from above, and in corridors completely filled by one human body.

"The entrance is at A on a level with the soil outside against the rock, but this cannot have been the original place of admission. It is a round hole and very narrow. The real entrance was at K, where one can distinguish a circular opening like the orifice of a silo, but which is now in the open and is choked with stones; or else at the end of the gallery H B. The chamber Y containing silos for preservation of grain must have been the furthest extremity. It is 6 feet 3 inches high, and the floor is higher above the mouth of the silos than elsewhere. The cavern is hewn out of the rock. All the chambers are circular. They are vaulted for the most part in the form of low cupolas. The domes of some are so low that one cannot stand upright in them. The corridors are still lower than the chambers, and one can only get along them by creeping. The extremities of the corridors and the entrances to the chambers had doors originally. One can see the notches for the reception of the closing beams. I saw no trace of hinges. The pa.s.sages are all arched over in semicircle."

[Ill.u.s.tration: Cluseau de Fauroux.]

Lacoste, speaking of the Saracen invasion and devastation of Quercy, says that "in Lower Quercy, where caverns are not common as they are in Upper Quercy, the inhabitants dug _souterrains_ with a labour that only love of life could prompt. Three of vast extent have been discovered at Fontanes, Mondoumerc, and Olmie. That of Mondoumerc is cut in the tufa, and is about 20 feet deep. It consists of an infinity of cells, or small chambers, united by a corridor. But the vastest and most remarkable for its extent and the labour devoted on it, is that of Olmie. The chambers are scooped out of a very hard sandstone. In some of them are little wells or reservoirs that were filled with water as a precaution against thirst, if refugees were obliged to remain long in this asylum. The pa.s.sages, with their turns, const.i.tute a veritable labyrinth whence it would be hard to find one's way out without the a.s.sistance of a guide."

The entrance to these hiding-places was either under a ledger stone in a church, or through a cellar, or half-way down a well, or in a thicket.

It must be remembered that it was the duty of every feudal seigneur to provide for the safety of his va.s.sels, and the security of their goods.

Consequently a great number of such _souterrains_ are under castles or in the grounds of a feudal lord. The rock on which his towers stood was often drilled through and through with galleries, chambers, and store places, for this purpose. On the alarm being given of the approach of an army marching through the land, of a raid by a marauding neighbour, or the hovering of a band of brigands over the spot, within a few hours all this underground world was filled with ploughs, looms, bedding, garments, household stuff of every description, and rang with the bleating of sheep, the lowing of oxen, the neighing of horses, and the whimpering of women and children. At Vendome, the rock on which stands the castle is riddled with pa.s.sages and halls, access to which is obtained not from the castle, but from the town. At Lavardin by Montoire it is the same. At Paulin in Tarn is a n.o.ble castle standing on a rock 300 feet high, and in this rock are storerooms, halls, a kitchen, a winding staircase. At Montvalon- Tauriac, in the same department, under the castle are refuges and granaries. At Murat in Cantal is the castle of Anterroche, and the rocks about it are traversed with galleries leading to chambers containing silos. At Salles-la-Source in Aveyron, in a cleft of the plateau, is the castle of the Count of Armagnac, and here also there is the same provision. At S. Sulpice in Tarn are the remains of a castle built in 1247, with its chapel over crypts and galleries carved out of the living stone. At Contigne, in Maine-et-Loire, is the manor of Gatines, underneath which are _souterrains_ that extend for a mile, with store-chambers and chapels, hewn out of the tufa. I might mention a hundred more. But all these pertain to a period before the feudal system had sunk into one of oppression, and when the va.s.sals had confidence in their seigneur. In process of time the conditions altered, and then they contrived their own private hiding-places from their lords and masters.

The stories everywhere prevalent where there are castles, that there are under them pa.s.sages connecting them with a church, a river, or another castle, are probably due to the fact of there having been these subterranean retreats intended for the use of the va.s.sals. But when these latter ceased to look to their lords to protect them, and cast about instead to shelter themselves from their lords, the original purport of these _souterrains_ was forgotten and misinterpreted.

One has but to look through the brief notices of towns and villages in Joanne's Departmental Geographies to see what a number of these refuges are already known to exist in France. And he records, be it remembered, only the most interesting. There are thousands more that have either not yet been discovered or remain unexplored. Some are revealed by accident; a peasant is ploughing, when his oxen are suddenly engulfed, and he finds that they have broken through the roof of one of these hiding-places. A gentleman is building his chateau, when in sinking his foundations he finds the rock like a petrified sponge--but not like a sponge in this, that the galleries are artificial. A _paysan_ lets himself down his well to clean it out, as the water is foul. He finds that in the side of the shaft is the opening of a pa.s.sage; he enters, follows it, and finds a labyrinth of galleries.

As an instance of the abundance of the _souterrains_ in France, I will take the department of Vienne and give in a note below a list of the communes where they are known to be, from _De Longuemar, Geographie du dep. de la Vienne_, Poitiers, 1882, and also from several editions of Joanne's Geography. [Footnote: Natural grottoes that may have served as refuges are not included. Availles, Bellefonds, Bethines, Beruges, Bonnes, Bussieres, Chateau Gamier, Champniers, Curzay, Civeaux, Gouex, Ingrandes, S. Julien Lars, Jazneuil, Leugny- sur-Creuse, Loudun, Lautiers, Lusignan, Marnay, Maire-le-Gautier, S.

Martin-Lars, S. Martin-la-Riviere, Maslou Montmorillon, Mazerolles, Mondion, Maulay, Montreuil-Bonnin, Naintre, Princai, Romagne, S. Remy- sur-Creuse, Saulge, Nouvaille, Persac, S. Savin, Sossais, Thure, Usson, Varennes, Le Vigean, Veniers, Velleches, Verrieres, Venneuil-sur-Biard.

Several of these are under churches, others under castles. At some of these places are three or more distinct _souterrains._]

Victor Hugo, in his _Quatrevingt Treise_, speaking of the war in La Vendee, says: "It is difficult to picture to oneself what these Breton forests really were. They were towns. Nothing could be more secret, more silent, and more savage. There were wells round and small, masked by coverings of stones or by branches. The interiors at first vertical, then carried horizontally, spread out underground like tunnels, and ended in dark chambers." These excavations, he states, had been there from time immemorial. He continues: "One of the wildest glades of the wood at Misdon, perforated by galleries and cells, out of which came and went a mysterious society, was called 'The Great City.'

The gloomy Breton forests were servants and accomplices of rebellion.

The subsoil of every forest was a sort of sponge, pierced and traversed in all directions by a secret highway of mines, cells and galleries.

Each of these blind cells could shelter five or six men. Usually the cover, made of moss and branches, was so artistically fashioned that, although impossible on the outside to distinguish it from the surrounding turf, it was very easy to open and close from the inside.

In several of these forests and woods there were not only subterranean villages grouped about the burrow of the chief, but also actual hamlets of low huts hidden under the trees. These underground belligerents were kept perfectly informed of what was going on. Nothing could be more rapid, nothing more mysterious, than their means of communication.

Sometimes they raised the cover of their hiding-places and listened to hear if there was fighting in the distance." He mentions the ability of the ambushed men to spring up, as it were, under the feet of the armies sent against them. And to show the numbers of the concealed forces, he continues: "There are in existence lists which enable one to understand the powerful organisation of that vast peasant rebellion. In Isle-et- Villaine, in the forest of Pertre, not a human trace was to be found, yet there were collected 6000 men under Focard. In the forest of Meullac, in Morhiban, not a soul was to be seen, yet it held 8000 men.

These deceptive copses were filled with fighters, lurking in an underground labyrinth."

On March 26, 1807, Napoleon demanded a fresh conscription of 80,000 men. This was the third levy that had been called for since the Prussian War began. The three conscriptions supplied no less than 240,000 men in seven months, and the call for the third produced consternation throughout France. The number of young men who reached the age of eighteen annually in half a year, more than the entire annual generation, had been swept off to lay their bones in the East of Europe. Great numbers of young fellows fled to the woods, caves, and secret refuges, and concealed themselves; and the gendarmes were employed in hunting them out, but not often with success unless aided by a traitor. Again in 1812, when Napoleon meditated an invasion of Russia, fresh calls were made on the male population. Every male capable of bearing arms was forced to a.s.sume them, and again, as in 1807, the young men disappeared as rabbits underground. It is quite possible that the peasants, who have found these refuges so convenient in the past, should know more about them and where they are situated than they pretend, thinking that at some future time, another revolution or another German invasion, the knowledge may prove serviceable.

And now let us turn to Picardy, perhaps the one of the ancient provinces of France most undermined. On the night of February 13, 1834, after heavy rains, a portion of the wall of the apse of the parish church of Gapennes, half-way between Aussy-le-Chateau and S. Ricquier, collapsed, and in the morning the inhabitants of the commune were stupefied to see the desolation of the holy place. Not only was a large breach gaping in the sanctuary, but all the walls of the chancel were fissured, and the pavement of the nave was upheaved in places and in others rent.

At first it was supposed that this was the result of an earthquake, but after a while the true cause was discovered. The church had been erected over a vast network of subterranean pa.s.sages and chambers, and the roofs of some of these had given way. This led to an exploration, and the plan of this subterranean refuge--for such it had been--was traced as far as possible.

But Gapennes is not the only place where such retreats exist throughout the province. Something like a hundred have been found, and more are every now and then coming to light. Indeed, it may safely be said that there is scarcely a village between Arras and Amiens and between Roye and the sea, betwixt the courses of the Somme and Authie, that was not provided with these underground refuges. The character of all is very much the same. They consist of pa.s.sages communicating with square or circular chambers that served as stores. They have been described at length by M. Bouthers in _Memoires de la Societe d'Archeologie du departement de la Somme_, Amiens, 1834, t. i.

To what date, or period rather, do they belong?

Some doubtless are of extreme antiquity, but the majority are comparatively modern. It is a significant fact that the entrance to perhaps the majority is in the sacristy of the parish church, and in that at Gapennes care was taken not to undermine the tower of the church. M. de Carpentin, who explored and reported on the excavation at Gapennes, remarks on the care taken to so distribute the chalk brought up from these pa.s.sages and vaults that no heaps were anywhere visible.

"The motive that can have induced the undertaking of such an extensive work can only have been that necessity drove the inhabitants to create for themselves a refuge in time of war." In it he found two pieces of common pottery, a lock and a hinge of iron, some straw and leather soles of women's shoes. He adds: "At the entrance of several of the chambers the stone is worked to receive doors, and here portions of decayed wood were found. And many of the chambers had their walls blackened by smoke as of lamps."

At Naours in Somme, the underground galleries have been explored thoroughly; there are several circular chambers for stores, and corn has been found in them, also fourteen gold coins of Charles VI or Louis XIV. In all there are 201 galleries and 300 chambers and the labyrinth extends to the distance of 6000 feet. At Santerre, which possesses three of these refuges, that portion of its territory was called _Territorium Sanctae Libertatis_.

The north-east of France, Picardy and Artois, were always exposed to attack from pirates by sea, Northmen and Saxon, and from invaders over the border. But none of these can have exceeded in barbarity that of 1635 to 1641, when Spanish armies--the first under John de Werth and Piccolomini, 40,000 in number, and made up of Germans, Hungarians, Croats as well as Spaniards--poured over the provinces committing the most frightful atrocities. And precisely to this period some of the refuges may be referred.

A MS. account of this invasion, by a priest of Hiermont, named Claude G.o.dde, leaves this in no manner of doubt. He says: "The Spaniards committed great outrages in Picardy, as they did later in 1658. These wars compelled the inhabitants of Hiermont in 1647 to construct the quarry which we now see. This quarry or cavern, which is a great masterpiece, was first undertaken by five or six of the inhabitants "-- he gave their names. "They first of all dug out the entrance in 1647, but owing to its having given way several times, had to be repaired, and was not completed till 1648. The other inhabitants, seeing its great utility, wanted also to have their chambers, but they were not admitted unless they contributed to the cost of the undertaking, and to this they willingly agreed. This quarry was of great service to the inhabitants in the Wars of Louis XIV. against England, Holland, and the Empire during the years 1708, 1709, 1710 and 1711, which were the days of Marlborough. It was accordingly made by the inhabitants of Hiermont, to hide themselves, their cattle, their grain and their furniture, to preserve them from pillage by the soldiers, whether of the enemy or French. Each family had its own chamber."

In a proces of 1638, one of those interrogated, a nun named Martha Tondu, stated that at Reneval and the neighbouring villages "the peasants are on the look out, and if alarmed, retire and conceal their cattle in ditches and quarries, without abandoning their houses or neglecting their agricultural work."

Some, accordingly, of these subterranean refuges are of comparatively late date; but this does not apply to all. At every period of danger, instinctively the peasants would take advantage of the nature of the chalk to form in it suitable hiding-places, and although some of the finds in these labyrinths are of recent date, others go back to the Gallo-Roman period. In the Arras and Cambrai Chronicle of Balderic (1051), we are told that in the fifth century in those parts a persecution of the Christians occurred, on the invasion of the barbarians, and that the priests celebrated the Divine Mysteries in secret hiding-places. "Many," he adds, "were suffocated in caves and in subterranean pa.s.sages."

There is, in fact, evidence both from archaeology and from history that these refuges were taken advantage of, and doubtless extended from a remote antiquity down to the eighteenth century.

It was not against the foreign foe only that the peasants excavated their underground retreats. Froissart paints the chivalry of his time in the brightest colours, and only here and there by a few touches lets us see what dark shadows set them off. Who paid for the gay accoutrements of the knights? Who were the real victims of the incessant wars? From whom came the ransom of King John and of the n.o.bles taken at Crecy and Poitiers? From the peasant. The prisoners allowed to return on parole came to their territories to collect the sums demanded for their release, and the peasant had to find them. He had his cattle, his plough and tumbril. They were taken from him; no more corn was left him than enough to sow his field. He knew how he would be exploited, and he hid his precious grain that was to make bread for his wife and children. The seigneur endeavoured to extort from him the secret as to where it was concealed. He exposed the man's bare feet before the fire; he loaded him with chains. But the peasant bore fire and iron rather than reveal the hiding-place. Here is Michelet's account of the seigneur in the first half of the fifteenth century. "The seigneur only revisited his lands at the head of his soldiery to extort money by violence. He came down on them as a storm of hail. All hid at his approach. Throughout his lands alarm resounded --it was a _sauve-qui-peut_. The seigneur is no longer a true seigneur; he is a rude captain, a barbarian, hardly even a Christian.

_ecorcheur_ is the true name for such, ruining what was already ruined, s.n.a.t.c.hing the shirt off the back of him who had one; if he had but his skin, of that he was flayed. It would be a mistake to suppose that it was only the captains of the _ecorcheurs_--the b.a.s.t.a.r.ds, the seigneurs without a seigneurie, who showed themselves so ferocious.

The grandees, the princes in these hideous wars, had acquired a strange taste for blood. What can one say when one sees Jean de Ligny, of the house of Luxembourg, exercise his nephew, the Count of Saint-Pol, a child of fifteen, in ma.s.sacring those who fled? They treated their kinsfolk in the same manner as their enemies. For safety--better be a foe than a relation. The Count d'Harcourt kept his father prisoner all his life. The Countess of Foix poisoned her sister; the Sire de Gial his wife. The Duke of Brittany made his brother die of starvation, and that publicly; pa.s.sers-by heard with a shudder the lamentable voice pleading piteously for a little bread. One evening, the 10th of January, the Count Adolphus of Gueldres dragged his old father out of bed, drew him on foot, unshod, through the snow for five leagues to cast him finally into a moat. It was the same in all the great families of the period--in those of the Low Countries, in those of Bar, Verdun, Armagnac, &c. The English had gone, but France was exterminating herself. The terrible miseries of the time find expression, feeble as yet, in the 'Complaint of the poor Commoner; and of the poor Labourers.' It comprises a mixture of lamentations and threats; the starving wretches warn the Church, the King, the Burgesses, the Merchants, the Seigneurs above all, that 'fire is drawing nigh to their hostels.' They appeal to the king for help. But what could Charles VII.

do? How impose respect and obedience on so many daring men? Where could he find the means to repress these flayers of the country, these terrible little kings of castles? They were his own captains. It was with their aid that he made war against the English." [Footnote: _Mist, de France,_ v. p. 184 _et seq._]

Thus, the subterranean refuges that had served at one time as hiding- places against Saracens, Normans, English, became places of retreat for the wretched people against their own masters. They no longer carried their goods into the _souterrains_ under the castles, but into refuges contrived by themselves in the depths of forests, known only to themselves; hidden, above all, from their seigneurs.

The peasantry might have said then, what was said long after by Voltaire: "Il faut etre dans ce monde enclume ou marteau; j'etais ne enclume." Voltaire, however, speedily became a hammer, and after 1789 the Tiers etat also became a hammer, and the n.o.blesse the anvil.

In Iceland there were underground retreats, as we learn from the same Saga that tells us of those in Aquitaine. Orvar Odd found a king's daughter concealed in one. So, also, a very large one in Ireland is spoken of in the Landnama Bok. In England we have, both in Ess.e.x and in Kent, subterranean pa.s.sages and chambers very similar to those described in Picardy and in Aquitaine. These also are excavated in the chalk. They are the so-called Dene Holes, of which there are many in Darenth Wood and near Chislehurst, and they have given occasion to a lively controversy. Some have supposed them to be retreats of the Druids, some that they were places of refuge during the invasions of the Saxons first, and then of the Danes, and others again contend that they were merely quarries for the excavation of chalk to burn into lime.

Here is an account of the Dene Hole at Chislehurst by Mr. W. J.

Nichols. [Footnote: Nichols (W. J.), "The Chislehurst Caves,"

_Journal of the Archaeological a.s.sociation_, Dec. 1903.] "At the foot of the hill is a gap, which is the present entrance to the caves.

A guide meets us here, who, unlocking a door, and switching on the electric light, introduces the visitor to a gallery or tunnel, about 150 feet long, 10 feet to 12 feet high, and with a width of 12 feet to 15 feet, narrowing to about 7 feet at the roof. This, and the galleries so far explored, have been cut through the chalk bed, at a depth of about 6 feet below the Thanet sand which covers it. At the end of the gallery, extending both right and left, are pa.s.sages of like character.

These again open into others so numerous that the visitor is fairly bewildered, and loses all idea of the direction in which he is travelling. The effect of the coloured electric lamps on the old chalk walling is remarkably beautiful. Proceeding on our way we get beyond the range of the electric lamps. Here candles or hand-lamps are lighted; and we pa.s.s, in Cimmerian gloom, through a succession of galleries of various dimensions, some of which, being only 4 feet wide and 5 feet high, are possibly of earlier construction than those already described. There is one gallery of the last-mentioned height and width 63 feet long, with several sharp turns which formerly terminated in a chamber about 12 feet high and 10 feet wide, and a like length, and near it is a seat cut into an angle of the walling. At no great distance from this chamber and near a Dene-hole shaft is a short gallery, at the end of which is a shaft originally level with the flooring, but now bricked round and further protected by an iron cover.

On removing the cover and lowering a lamp, a well of excellent workmanship is discovered. Owing to the quant.i.ty of material thrown down from time to time by explorers, its present depth is no more than 43 feet. Further progress is made, and presently we notice a streak of daylight some distance ahead; here we find that we have reached the foot of a shaft 85 feet deep, which, though now partly covered in, had its mouth in what is at the present time the garden of a modern villa."

There are numerous other Dene Holes or Danes' Pits at East Tilbury, Crayford, and Little Thurrock. As to the theory that they were places of Druidical worship, we may dismiss it as not deserving serious consideration.

At East Tilbury the entrance to the Danes' pit is from above, by narrow pa.s.sages that widen and communicate with several apartments, all of regular forms. One of these pits consists of a shaft descending to chambers arranged like a sixfoiled flower. The shaft is 3 feet in diameter and 85 feet deep. This may be likened to one at Doue-la- Fontaine (Maine et Loire), where a descent is made under a private house into an area from which radiate on all sides chambers, some of which contain tombs.

That these Dene Holes were used as hiding-places when the sails of the Danish Vikings appeared on the horizon is probable enough, but originally they were chalk quarries--some very ancient--for British coins have been found in them. The existence of old lime-kilns near the Chislehurst caves places their origin beyond a doubt. Chalk was largely exported in early times from the Thames to Zealand, whence it was pa.s.sed through the Low Countries and used in dressing the fields.

Altars to Nethalennia, the patroness of the chalk quarries, have been found in the sand on the coast of Zealand; some bear votive inscriptions from dealers in British chalk, and Pliny, writing of the finer quality of chalk (_argentaria_) employed by silversmiths, obtained from pits sunk like wells, with narrow mouths, to the depth of a hundred feet, whence they branch out like the adits of mines, adds, "Hoc maxime Britannia ut.i.tur." [Footnote: Roach Smith, _Collectanea Antiqua_, vi. p. 243, "British Archaeological a.s.soc. Journal," N.S., ix.-x. (1903 and 1904).]

In Cornwall, moreover, there are what are locally called _fogous_.

These are either excavated in the rock with pa.s.sages leading to the sea or to houses, or else they are built of stone slabs standing erect, parallel and covered with other slabs leading to chambers similarly constructed, and all buried under turf or sand. Of the former description there is a very interesting example at Porthcothan in S.

Ervan; of the latter the most remarkable is at Trelowaren. The former may have been excavated by smugglers. An interesting account of the excavation of two caves at Archerfield, in Haddingtonshire, is given in the Proceedings of the Society of Antiquaries of Scotland for 1909.

Both caves are natural, but one had been walled up in front, with a doorway and window and with oven; both had paved hearths in the centre, and there was evidence that they had been tenanted some time after the Roman occupation of Britain, as among the fragments of pottery found was some Samian ware. It would appear that both had been inhabited simultaneously, but not consecutively, for a lengthy period, and no doubt can exist that they were mere rock refuges. In a note to the article we read: "On the coast of Island Magee (Ireland) there is a cave, south of the Gobbins, which has been frequently used as a place of refuge. So late as 1798 it was inhabited by outlaws, who constructed a kind of fortification at the entrance, the remains of which still exist." [Footnote: Cree (J. R.), "Excavation of Two Caves," in "Proceedings of the Soc. of Arch. of Scotland," Edin., 1909, vol.

xliii.]

A cave in the Isle of Egg, one of the Hebrides, has a very narrow entrance, through which one can creep only upon hands and knees, but it rises steeply within and soon becomes lofty, and runs into the bowels of the rock for 225 feet. The stony, pebbly bottom of this cavern was for long strewn with the bones of men, women and children, the relics of the ancient inhabitants of, the island, two hundred in number, of whose destruction the following account is given. "The Macdonalds, of the Isle of Egg, a people dependent on Clanra.n.a.ld, had done some injury to the Lord of Macleod. The tradition of the isle says that it was by a personal attack on the chieftain, in which his back was broken; but that of the two other isles bears that the injury was offered by two or three of the Macleods, who, landing upon Egg and behaving insolently towards the islanders, were bound hand and foot, and turned adrift in a boat, which the winds safely conducted to Skye. To avenge the offence given, Macleod sailed with such a body of men as rendered resistance hopeless. The natives, fearing his vengeance, concealed themselves in the cavern; and, after strict search, the Macleods went on board their galleys after doing what mischief they could, concluding the inhabitants had left the isle. But next morning they espied from their vessels a man upon the island, and immediately landing again, they traced his retreat by means of a light snow on the ground to the cavern. Macleod then summoned the subterranean garrison, and demanded that the inhabitants who had offended him should be delivered up. This was peremptorily refused. The chieftain thereupon caused his people to divert the course of a rill of water, which, falling over the mouth of the cave, would have prevented his purposed vengeance. He then kindled at the entrance of the cavern a large fire, and maintained it until all within were destroyed by suffocation." [Footnote: Lockhart's "Life of Sir Walter Scott," Edin., 1844, p.285.]

A no less horrible deed was committed during the campaign of Ess.e.x against the Irish rebels in 1575. This shall be given in the words of Froude. [Footnote: "Hist. of England," 1870, x. p. 527 _et seq._]