One room of the palace, a stately hall about 80 ft. in length by 26 ft. in breadth, traversed by a row of square-sectioned pillars, has been named by its discoverer "the Hall of the Double Axes," from the frequent occurrence of this symbol therein. Not only does the sacred axe occur as a more or less crude engraving on the stone blocks composing certain pillars in the palace, but little models of it were found associated with an altar, and, in the Dictaean cave, some miles distant, several bronze specimens of the axe were discovered in circumstances which show that they were votive offerings. Sometimes the sacred symbol was set up on a socketed pedestal (Fig. 9). Moreover, in more recent excavations a curious "tomb" was found (Figs. 10 and 11) which was double-axe shaped in plan and was evidently the repository of a giant emblem (Fig. 12. See plate, p. 42).

Long before Dr. Evans' excavations in Crete the great German archaeologist Schliemann, during his researches at Mycenae on the mainland, unearthed from one of the graves an ox-head of gold plate, with a double axe between the upright horns. The double axe was also the sign of the Zeus worshipped at Labraunda in Caria, a country to the north-east of Crete, on the mainland of Asia Minor, where the implement was known as the labrys.

The cult of the bull was also much in evidence in the palace remains. Schliemann, in excavating the site of Tiryns in 1884, came across an extraordinary wall-painting depicting a man holding one horn of a great bull whilst he leaps over its back, the animal meanwhile charging at full speed. Several examples of such scenes have since been discovered, painted upon walls, engraved on gems, or stamped on seal-impressions. Amongst the debris of one of the rooms in the palace at Knossos was found a painting of a scene in which two girls are engaged in dodging the charge of a bull, whilst a boy, who has evidently just left hold of its horns, turns a somersault over its back.

Near the main north entrance to the palace was brought to light a large plaster relief of a bull's head, no doubt originally forming part of the complete beast. This relief was a masterpiece of Minoan art. It was of life-size and beautifully coloured, and particular attention had been given to the modelling and colour of the eye, the fierce stare of which, in conjunction with the open mouth, conveys a fine effect of frenzied excitement.

These are only a few examples, amongst many, which go to demonstrate that the sport of "bull-leaping"—or ταυροκαθαψια [Greek: taurokathapsia], as it was called by the Greeks—was beloved of the Minoans and was probably practised in the precincts of the palace.

In the light of these discoveries Dr. Evans concludes that the palace of Knossos was the Labyrinth, or House of the Labrys, which gave rise to the classic legend, the idea of the Minotaur originating in the practice of training captives to participate in the dangerous sport of bull-leaping. (Tauros = bull, hence Minotaur = Bull of Minos.) We will refer further to the etymology of "labyrinth" in a later chapter. The palace was certainly of sufficient complexity to render it difficult for the uninitiated to find their way about it, but the plan of its remains exhibits no resemblance to a designed labyrinth of the conventional type. There is, however, a suggestion of the latter in the meander pattern painted on one of the walls, to which reference has been made above. The notion as to the Labyrinth having been a prison from which escape was impossible may also have some connection with two deep pits beneath the palace, whose function was possibly that of dungeons for prisoners.

In considering the origin of the legend, we must remember that a period of several centuries elapsed between the destruction of the Knossian buildings and the first written account of the Labyrinth, and must take into account the probability that the people who in later ages became the dominant race in Crete would be likely to make ample use of their imagination in formulating an explanation of the vast and complicated ruins of the burnt city, with their mysterious frescoes and enigmatic symbols.

It may also be borne in mind that the excavations in Crete have by no means reached a final stage, and that, although no architectural remains of a plan conforming to the usual conception of a formal labyrinth are yet forthcoming, there is a possibility that something of the kind may yet turn up, though indeed the chance seems very remote. Even as this book is going to press appears an article in The Times by Sir Arthur Evans announcing yet further enthralling discoveries; he finds abundant signs of a great earthquake, causing ruin over the whole Knossian area, about 1600 B.C., also evidence—including portable altars and huge ox-skulls—indicating an expiatory sacrifice recalling Homer's words, "in bulls doth the Earthshaker delight"; and finally, on a floor-level about thirty feet down, the opening of an artificial cave with three rough steps leading down to what was apparently the lair of some great beast. "But here, perhaps," says Sir Arthur, "it is better for imagination to draw rein."

CHAPTER VII
THE ETRUSCAN OR ITALIAN LABYRINTH

In addition to the Egyptian and Cretan labyrinths, a few other structures are referred to as being in the same category, but not until a fairly late period.

Pliny (died A.D. 79) mentions one built by Smilis of Aegina, after the Egyptian model, on the island of Lemnos, and says that it was renowned for the beauty of its 150 columns and that remains of it existed in his time. He also mentions one at Samos, said to have been built by Theodorus, and says that "all of these buildings are covered with arched roofs of polished stone." No other details concerning these edifices have come down to us, but Pliny quotes from Varro (116–27 B.C.) a detailed description of a very extraordinary tomb at Clusium (the modern Chiusi), said to be that of the great Etruscan general Lars Porsena. This is the only Etruscan tomb described by the ancient writers, and is mentioned by Pliny solely because it was alleged to contain a subterranean labyrinth. It must have been a most elaborate, not to say extravagant, monument. Even Pliny feels some qualms about accepting responsibility for the description, and therefore makes it clear that he is simply quoting from information received.

"It is but right that I should mention it," he says, "in order to show that the vanity displayed by foreign princes, great though it is, has been surpassed. But in view of the exceedingly fabulous nature of the story I shall use the words given by M. Varro himself in his account of it: 'Porsena was buried below the city of Clusium in the place where he had built a square monument of dressed stones. Each side was three hundred feet in length and fifty in height, and beneath the base there was an inextricable labyrinth, into which, if anybody entered without a clue of thread, he could never discover his way out. Above this square building there stand five pyramids, one at each corner and one in the centre, seventy-five feet broad at the base and one hundred and fifty feet high. These pyramids so taper in shape that upon the top of all of them together there is supported a brazen globe, and upon that again a petasus3 from which bells are suspended by chains. These make a tinkling sound when blown about by the wind, as was done in bygone times at Dodona. Upon this globe there are four more pyramids, each a hundred feet in height, and above them is a platform on which are five more pyramids.' The height of the latter, Varro is ashamed to add, but, according to the Etruscan stories, it was equal to that of the rest of the building. What utter madness is this, to attempt to seek glory at a great cost which can never be of use to anyone; not to mention the drain upon the resources of the country. And all to the end that the artist may receive the greater share of the praise!"

There have been many discussions as to the possibility of a monument of this nature having existed, and various reconstructions have been attempted, notably one (Fig. 13), based on Varro's account, by a celebrated French scholar of a century ago, M. Quatremère de Quincy.

One enthusiast, a certain Father Angelo Cortenovis, even wrote a treatise to prove that the whole contrivance was nothing more nor less than a huge electrical machine!

Most writers on the subject have been inclined to look upon Varro's description as at best a gross exaggeration, but Professor Müller gave it as his opinion that the labyrinth described did actually exist, and that the upper part, though no doubt highly embellished in the description, was not the mere offspring of fancy. He thought it quite probable that there was a square basement of regular masonry supporting five pyramids as recounted by Varro, but that the latter described the upper part from hearsay. He drew attention to the fact that a tomb somewhat of this nature is still in existence on the Appian Way at Albano, the pyramids being represented in this case by cones. It is commonly called the Tomb of the Horatii and Curiatii.

In the early part of last century a British traveller, G. Dennis, made a study of the antiquities of Etruria and gave particular attention to the remarkable rock-cut labyrinths of which that province furnishes several examples.

He pointed out that the possession of a labyrinth was the distinguishing feature of Porsena's tomb which alone caused Pliny to mention it. The expression "Sub Clusio" used by Pliny, he says, led subsequent writers to infer that the subterranean passages beneath Chiusi were intended, but such an arrangement would be at variance with the general sepulchral practice of the Etruscans, and the tomb of Porsena must be looked for outside the city walls. Dennis then goes on to describe the great cemetery that had recently been discovered in the hill called Poggio Gajella, about three miles to the north-east of Chiusi. His sketch of the principal "storey" of this labyrinthine excavation is shown in Fig. 14.

Here again we may note that the design of the passages, although perhaps puzzling to a stranger, especially with imperfect illumination, in no respect approaches the traditional "labyrinth" pattern. That the conventional form was not unknown to the Etruscans, however, is shown by the occurrence of a design of this type on a vase found at Tragliatella which we shall mention later.

It is, of course, possible that the tomb of Porsena was erected on the hill above this labyrinth, but we have not much evidence on the point. If the tomb possessed a labyrinth, no doubt the latter would have been something of this type. Dennis also mentions various other labyrinths of this nature in Etruria—for example, one near Volterra, "a long passage cut in the rock, six feet wide but only three high, so that you must travel on all-fours. From time to time the passage widens into chambers, yet not high enough to permit you to stand upright, or it meets the passages of similar character opening in various directions and extending into the heart of the hill, how far no one can say. In short, this is a perfect labyrinth in which, without a clue, one might very soon be lost."

He also mentions one at San Pietro, Toscanella, "in the cliffs below the Madonna dell' Olivo, about half a mile from the town. Here a long, sewer-like passage leads into a spacious chamber of irregular form, with two massive columns supporting its ceiling and a rude pilaster on the wall behind. But the peculiarity of the tomb lies in a cuniculus or passage cut in the rock, just large enough for a man to creep through on all-fours, which, entering the wall on one side after a long gyration and sundry branchings, now blocked with earth, opens in the opposite wall of the tomb."

These Etruscan labyrinths were all of a sepulchral character, and one is naturally reminded of the catacombs of Rome, Paris, and Naples, to which, however, the term "labyrinth" is not customarily applied. Strabo uses the word in reference to a catacomb near Nauplia, which he calls the Labyrinth of the Cyclops. In Pliny's time the word would appear to have been used to denote a winding path following a more or less formal design of intricate pattern, but not necessarily connected with sepulchral purposes.

When speaking of the Labyrinth of Crete he says, "We must not compare this to what we see traced upon our mosaic pavements or to the mazes formed in the fields for the entertainment of children, and thus suppose it to be a narrow path along which we may walk for many miles together, but we must picture to ourselves an edifice with many doors and galleries which mislead the visitor...." This passage shows that the term "labyrinth" had come to have a fairly broad significance. It had long been used in a metaphorical sense, even as we find Plato, over four centuries earlier, employing it to describe an elaborate argument. We also find it applied by extension to other objects, such as traps for fish, to judge by a certain passage in the works of Theocritus.

The only buildings to which the ancient writers applied the term, however, were those to which we have already referred.

Of the two phrases which we have italicised in the above quotation from Pliny, the second is of interest in connection with a matter we shall deal with later on, whilst the former brings us to the subject of our next chapter.

CHAPTER VIII
THE LABYRINTH IN ANCIENT ART

There has been considerable speculation as to how the typical labyrinth form first came into existence. It became stereotyped long before the Christian era and retained its character for many centuries.

The coins of Knossos furnish us with abundant examples of it, and, from the fact that in certain of the earlier specimens the corresponding figure is a simple repeated meander, it has been supposed that the typical labyrinth design arose by elaboration of the meander. The resemblance between this form and the very widespread and primitive sign known as the fylfot or svastika has also attracted some attention. It is a somewhat long step, however, from a loose combination of meanders like that shown in, say, Fig. 20, to the compact conventional labyrinth of Fig. 30. The adoption of the former design may possibly have been inspired by the fresco on one of the walls of the Minoan palace, to which we have made reference in Chapter VI (Fig. 8), portions of which may have been visible among the ruins for several generations. There does not appear to be any evidence that the complex meander pattern of the fresco was an allusion on the part of the Minoans to an actual constructional labyrinth; it may quite well have been a purely ornamental conception, without any symbolical significance. Meander designs were used by the Minoans at a much earlier date than this, one example, though of simpler character, having been found in the older palace, and others, either snake-like or of a squarish nature, on ivory seals unearthed at other Minoan sites (Zakro and Hagia Triada). Similar designs exist on certain Egyptian "button-seals" of an approximately contemporary period—from the VIth Dynasty onwards—and Sir Arthur Evans has expressed the opinion that these will possibly prove to constitute the source of the Labyrinth in Art. Figs. 15, 16, 17, 18 show specimens of early Egyptian seals and plaques of this character in the British Museum. Professor Flinders Petrie very kindly drew the writer's attention to a steatite plaque in his collection at University College, London, which is somewhat similar to one of those mentioned above, but of rather more elaborate design (Fig. 19). The labyrinthine pattern on this is surmounted by a representation, in the peculiar "linear" fashion often adopted by early Egyptian artists, of two seated human figures facing one another, the knees being drawn up. Professor Petrie acquired the plaque at Memphis. He considers that it dates from a period round about 3000 B.C., and points out that if the broken lines be completed there would appear to be five false turns to be avoided before reaching the centre.

In discussing the designs on these seals and plaques, Sir A. Evans alludes to a possible connection with two of the hieroglyphs of the period, which are of the nature of simple square meanders of a kind extensively employed in ancient ornament. One of them (mer) is the sign used for indicating irrigated land. The other (aha) is a simplified form of a more elaborate sign representing the plan of a palace court, a figure to which one of the Minoan signs bears a close resemblance.

The Knossian coins shown in Figs. 20 to 31 are from the British Museum collection and are reproduced by the courtesy of the Keeper of the Coins and Medals Department, who supplied the writer with plaster casts for the purpose. They date, of course, from times greatly posterior to those of the Minoan civilisation, from times when the culture of Greece had long replaced that of the Mycenaeans, or whatever similar race it was that succeeded the Minoans (see Appendix IV, i.).

Figs. 20, 21, and 22 show silver coins dating from about 500 to 430 B.C. They portray on one side the Minotaur and on the other a symmetrical meander pattern which, it needs very little imagination to see, has reference to the labyrinth in which the monster was alleged to dwell.

Fig. 23 shows a silver coin of a rather later date, representing on its obverse a female head which is thought to be that of Demeter or Persephone, and on the reverse a meander-labyrinth containing a star at its centre.

Fig. 24 shows a similar obverse, but on the reverse we see a bull's head surrounded by a simple meander frame.

Fig. 25, the obverse of which is likewise adorned with a female head, gives on the reverse the design of a square labyrinth of the conventional type that thereafter predominates.

Fig. 26 shows a bronze coin having on one side the head of Apollo and on the other a labyrinth with a star.

The four coins last described date from a period between 430 and 350 B.C. The next (Fig. 27) is rather later in date and shows on its obverse the head of Hera and on the reverse a square labyrinth together with an arrow-head and thunderbolts and the Greek characters ΚΝΩΣΙΩΝ [Greek: KNÔSIÔN].

The bronze coin of about 220 B.C., shown in Fig. 28, bears on its obverse the figure of Europa seated on a bull, with two dolphins below, and on the reverse a square labyrinth, the Knossian superscription being again evident.

The remaining three figures represent silver coins of the two succeeding centuries, but not later than 67 B.C.

Fig. 29 exhibits on one side the head of Pallas, and on the reverse a little square labyrinth placed beside an owl standing upon a prostrate amphora.

In Fig. 30 the obverse is occupied by the head of Apollo, the reverse by a labyrinth of circular shape, but conforming to the conventional plan.

The head on the coin shown in Fig. 31 may be intended for that of Minos or Zeus. On the reverse is a square labyrinth.

Labyrinthine designs are also found on certain Lydian, Phrygian, and Ionian coins.

It will be noticed that when once the labyrinth pattern has been definitely conventionalised it remains very constant in principle, whether its general conformation be rectangular or circular. Starting from the exterior, the "path" runs inwards a short distance, turns so as to run parallel with the outer wall until nearly a full circuit has been completed, then doubles back on itself and runs round in the opposite direction, doubles upon itself again, and so on until it finally comes to a stop in a blind end, having traversed all of the space within the outer walls without covering any part twice and without forming any branches or loops.

Obviously there is no "puzzle" about this kind of labyrinth; one has simply to follow the one path, either to penetrate to the inner goal or to escape thence to the exterior.

A labyrinth of precisely this type was discovered traced on the surface of a crimson-painted pillar in the peristyle of the building known as the House of Lucretius, in the excavated portion of Pompeii (Fig. 32). It was evidently scratched with a nail or stylus by some idler of 2000 years ago (Pompeii was overwhelmed by Vesuvius in A.D. 79) and is accompanied by the words "LABYRINTHUS. HIC HABITAT MINOTAURUS," possibly in waggish reference to the owner or occupier of the premises.

Another house has, in consequence of its mosaic and pictorial references to the Cretan Labyrinth, received the name of the Casa del Labirinto or House of the Labyrinth. One mosaic discovered therein depicts Theseus and the Minotaur struggling on the ground, watched by a group of affrighted maidens.

The Romans excelled in the art of designing and executing mosaic pavements, abundant remains of which have been preserved. These were of various kinds. There was the pavimentum sectile, composed of pieces of marble of various sizes, shapes, and colours arranged in uniform sets, so as to form when put together an ornamental pattern; the pavimentum tessellatum, in which the pieces of marble, though variously coloured, were all of the same size and shape, generally small squares; the pavimentum vermiculatum, composed of very small pieces of coloured marble of irregular shape so arranged as to portray objects in their natural shapes and colours; and finally the pavimentum scalpturatum, in which the design was engraved or inlaid. Opus alexandrinum is a variant of sectile.

Several Roman pavements embodying labyrinthine devices, and in some cases commemorating the victory of Theseus over the Minotaur, or other exploits of the hero, have come to light from time to time, not only on the continent of Europe but also in England; they are usually executed in opus alexandrinum.

Fig. 33 shows in outline a beautiful specimen, 18 ft. long and 15 ft. broad, discovered at Salzburg, in Austria. It bears the device of a labyrinth, with, at the centre, a representation of Theseus about to give the fatal blow to the Minotaur.

On the left side we see Theseus and Ariadne joining hands over the altar. In the upper panel Theseus appears to be putting Ariadne ashore, and to the right we see the disconsolate maiden deserted by her lover, presumably on the Isle of Naxos.

A labyrinth of the type shown also occurs on a Roman mosaic which was unearthed in the churchyard at Caerleon-on-Usk. It was in a poor state of preservation, but sufficient remains to show that the labyrinth, of a design similar to that of the Salzburg specimen, is surrounded by scrolls proceeding from two vases (Fig. 34).

A very fine specimen of this type of labyrinth was discovered in 1904 beneath a ploughed field at Harpham, in the East Riding of Yorkshire. Another, of which details are not to hand, is said to have been found in Northamptonshire.

In 1790 a pavement, about eighteen feet by twelve, was unearthed at Aix, near Marseilles. It portrayed the combat between Theseus and the Minotaur, within a framed square, the remainder of the mosaic consisting of a complicated interlaced meander representing the labyrinth.

In Fig. 35 is reproduced from A. de Caumont's "Abécédaire d'Archéologie" a rough sketch of the Roman baths at Verdes (Loir-et-Cher), showing a pavement with a labyrinth mosaic.

A pavement found in 1830 at Cormerod, in the Canton of Friburg, Switzerland, is shown in Fig. 36. A few years afterwards another was brought to light in the neighbouring Canton of Vaud, from beneath the ruins of the ancient town of Orbe.

A splendid mosaic labyrinth of Roman times was found some forty or fifty years ago on a family tomb in the ancient necropolis of Susa, Tunis (Hadrumetum). It was afterwards destroyed by looters, but a careful drawing of it was fortunately made on its first discovery (Fig. 37). The whole mosaic measured about seventeen feet by ten, and contained a very finely executed labyrinth of four paths, like the Harpham and Caerleon examples mentioned above, the central space being occupied by the Minotaur, who is shown in an attitude of defeat. Sailing towards the labyrinth was a boat containing figures which presumably represented Theseus and his companions. The design was accompanied by the words "HIC INCLUSUS VITAM PERDIT."

Another well-preserved mosaic of this character was discovered in 1884 at Brindisi, and placed in the municipal museum of that town. It measures 17 ft. by 10 ft. 6 in., and shows within a square labyrinth Theseus in the act of clubbing the Minotaur, who has fallen on his knees. Around the labyrinth are various perches with birds thereon, perhaps in allusion to the automatic birds reputed to have been made by Daedalus (cf. Fig. 36).

We shall examine other mosaic and pavement labyrinths when we come to consider the question of the use of this symbol by the Church.

Apart from the designs on Knossian coins, Greek art does not appear to have left us any definite representations of the labyrinths, although with the Romans, who acquired the idea at a later date, it was a favourite motif.

We cannot, however, ignore the suggestion that has been made that certain structures discovered in the ruins of Tiryns and Epidaurus, two cities in that part of ancient Greece known as the Argolid, are architectural labyrinths, used for ritual purposes. The foundations of the tholos, or rotunda, of the sanctuary of Aesculapius at Epidaurus, which was excavated by P. Kabbadias, Director of the Greek Archaeological Society, in the 'eighties, do certainly suggest something of the kind. They consist of concentric circular walls, the three innermost being connected by a radial wall, separated by narrow spaces which intercommunicate by an opening or doorway in each wall, forming in plan a figure somewhat in the style of the "Pigs in Clover" toy mentioned in a later chapter. When the peculiar nature of the upper part of the building is considered, however, it seems very reasonable to suppose that these walls, with their passages, were designed only with a view to the requirements of the superstructure which they had to support.

As for the slightly similar concentric foundations unearthed by German excavators at Tiryns in 1912, the analogy is too imperfect to afford reliable grounds for the statement in question.

Greek ceramic art, on the other hand, furnishes us with very many allusions to the Theseus-Minotaur myth, and also with a profusion of frets and meanders, which are thought in some cases to be symbolical of the labyrinth.

Consider, for instance, the "kylix" or bowl in the British Museum which is shown in Fig. 38. (A similar bowl is preserved in Harrow School Museum.) On it are represented most of the exploits of the hero up to his Knossian adventure. All who are familiar with the legend will recognise at a glance Periphetes the Club-bearer, Sinis the Pine-bender, the Wild Sow of Krommyon, Kerkyon the Wrestler, Procrustes of the Standard Bed, and other gentlefolk that Theseus successively encountered and appropriately dealt with on his initial journey to Athens. In the centre of the bowl is shown the adventure of the Labyrinth, the hero being seen in the act of despatching the monster at the very door of his lair. The meander on the door-post has been thought to symbolise the Labyrinth, but there is more reason to suppose that it is purely decorative.

The Minotaur exploit is also shown on the smaller bowl shown in Fig. 39.

In the previous chapter we have already referred to an Etruscan vase found at Tragliatella. This was very roughly decorated with incised figures, representing amongst other things a circular labyrinth of the traditional type and some horsemen who are thought to be engaged either in the attack on Troy or in the game known as the Lusus Trojae or Game of Troy. That there can be no doubt about the artist's identification of the labyrinth in some way with the celebrated city in question is clear from the word Truia scratched within it (Fig. 133).

Representations of the labyrinth were sometimes engraved on ancient gems, a fine specimen of which is figured in P. A. Maffei's "Gemme Antiche" (Fig. 40), published in 1707. The Minotaur in this case is shown as a centaur. A similar representation appears on a sixteenth-century bronze plaquette of Italian workmanship exhibited in the Plaquette Room at the British Museum (Fig. 41. See plate, p. 60).

Before leaving the subject of the Labyrinth in ancient art we must take notice of a reference in an ancient manuscript which tends to show that the symbol figured on the robes of Roman Emperors. This manuscript was discovered by A. F. Ozanam in the Laurentian Library at Florence. It is entitled "Graphia Aurea Urbis Romae" and contains, under the heading "De diarodino imperatoris," the following passage:

Habeat et in diarodino laberinthum fabrefactum ex auro et margaritis, in quo sit Minotaurus, digitum ad os tenens ex smaragdo factus, quia sicut non valet quis laberinthum scrutare, ita non debet consilium dominatoris propalare.

"Let there be represented on it (the Emperor's robe) a labyrinth of gold and pearls, in which is the Minotaur, made of emerald, holding his finger to his mouth, thus signifying that, just as none may know the secret of the labyrinth, so none may reveal the monarch's counsels."

It has been pointed out by Mr. A. B. Cook that in the Fitzwilliam Museum at Cambridge is a painting by Bartolommeo Veneto (1502–1530) representing an unknown man who wears on his breast a labyrinth resembling that described above.

CHAPTER IX
CHURCH LABYRINTHS

The consideration of labyrinths worked in Roman mosaic pavements leads us on to a very interesting development of the subject which deserves a chapter to itself, namely, the Labyrinth in the Church.

Probably the oldest known example of this nature is that in the ancient basilica of Reparatus at Orléansville (Algeria), an edifice which is believed to date from the fourth century A.D. In the pavement near the north-west entrance of the church is the design shown in outline in Fig. 42. It measures about 8 ft. in diameter and shows great resemblance to the Roman pavement found at Harpham and the tomb-mosaic at Susa. At the centre is a jeu-de-lettres on the words SANCTA ECLESIA, which may be read in any direction, except diagonally, commencing at the centre. But for the employment of these words the labyrinth in question might well have been conceived to be a Roman relic utilised by the builders of the church to ornament their pavement. Such pavement-labyrinths, however, with or without central figures or other embellishments, and of various dimensions and composition, are found in many of the old churches of France and Italy.

They seem to have been constructed chiefly during the twelfth century, and although several of them have been destroyed many fine examples still remain. Some are formed on the walls instead of the pavements, and in such cases are of smaller dimensions.

On the whole, too, those in the Italian churches are much smaller than the French specimens. On the wall of Lucca Cathedral (Fig. 43) is one of a diameter of only 1 ft. 7½ in. It formerly enclosed at the centre a representation of Theseus and the Minotaur, but owing to the friction of many generations of tracing fingers this has become effaced. Opposite the "entrance" is the inscription:

A similar small labyrinth, with a central Theseus-Minotaur design, is to be found on the wall of the church of San Michele Maggiore at Pavia (Fig. 44). It is thought to be of tenth-century construction. This is one of the few cases where the Minotaur is represented with a human head and a beast's body—as a sort of Centaur, in fact. It is accompanied by the words "TESEUS INTRAVIT MONSTRUMQUE BIFORME NECAVIT."

Of about the same period was the example in the church of San Savino at Piacenza. It is described by P. M. Campi in his "Ecclesiastical History of Piacenza" (1651), under the year A.D. 903. The signs of the Zodiac were placed in juxtaposition to it. The accompanying legend in this case consisted of four hexameters, to the effect that the labyrinth represented the world we live in, broad at the entrance, but narrow at the exit, so that he who is ensnared by the joys of this world, and weighed down by his vices, can regain the doctrine of life only with difficulty.

In the Cathedral of Cremona, which, like Pavia and Piacenza, is on the banks of the River Po, is a mutilated mosaic of early date—possibly eighth or ninth century—showing part of an interlaced pattern which was evidently intended to refer to the Cretan Labyrinth, as it was placed close to two figures in fighting attitudes and armed with swords and shields, the right-hand figure having the head of a beast and the label "CENTAVRVS." (There was apparently little distinction between a Minotaur and a Centaur in the minds of some mediaeval artists.)

A rather larger specimen, 5 ft. in diameter, may be seen in the church of Sta. Maria-in-Aquiro, Rome. It is composed of bands of porphyry and yellow and green marble, surrounding a central plate of porphyry, and is similar in design to that at Lucca. Another church in the same city, Sta. Maria-di-Trastavera, has a labyrinth composed of variously coloured marbles worked in the floor. It is 11 ft. across and was probably constructed about 1190 A.D. (Fig. 45). It is now somewhat mutilated, but was originally a most beautiful example. The fact that the inner paths consist of a series of concentric rings rather suggests that it has at some time been repaired without regard to the original design; unless we accept the hypothesis of M. Durand that they bore a symbolic reference to the various degrees of beatitude by which the soul approaches heaven, as figured by Dante. Fig. 46 shows another old Italian specimen. It is nearly 11½ ft. in diameter and is to be found in the church of San Vitale, Ravenna.