drawings of seven Old English shoes, two of which, the circular and the split types, would appear to be rather primitive.

We need not linger over the allusion to horseshoes in Domesday Book. The reference to the blacksmiths of Hereford, who were liable to be called upon to make horseshoes for the king at a fixed rate, is well known to most students.

The circular horseshoe, which has just been mentioned, and which is occasionally dug up in the Fens, is still commonly employed in Northumberland. Through the kindness of the Rev. Hastings M. Neville, of Ford, Cornhill-on-Tweed, I am enabled to give illustrations of this form of shoe (Figs. 84, 85). The shoe, which has been somewhat abraded by wear, is markedly convex on the lower surface, and correspondingly concave where it is fitted to the hoof. In this respect it resembles the broad shoe (Fig. 83 C, D) of the ordinary outline, discovered in making a main drainage trench at Edenbridge, Kent. This latter specimen is now in the possession of Mr C. E. Youens, of Dartford, through whose courtesy a sketch has been obtained. As in the Northumbrian example, the “wale-holes” are very near the margin; but while the iron of the former shoe is carried completely round to give support behind, the Edenbridge specimen does not even possess calkins—that is, portions projecting downward at the “heel.” The Edenbridge shoe appears to be Mediaeval, but it may perhaps be Saxon, or even of earlier date. The specimen should be compared with the Mediaeval examples in the Guildhall Museum.

The circular form of shoe, according to Mr James Weatherston, the blacksmith at Duddo, near Norham, Northumberland, is advantageous to a horse which has a weak “toe” or heel. This shape has been used from time immemorial. Sometimes a leather sole, covering the whole foot, is placed between the hoof and the circular shoe. Again, a detachable iron plate, or “complete shoe,” is occasionally screwed on to the outside surface of the round one, so that, by removing the plate, the horse’s foot can be examined without interfering with the shoe. In this case, the leather sole is omitted. The screws or “cogs” are square-headed, and project to such a degree that the animal walks on them alone.

The purpose of the convexity of the round shoe is to break the shock as the horse’s foot strikes the ground. The efficacy of this shoe is specially noticeable with “foundered” horses—those which have inflamed feet. An animal, thus suffering, tends to tread more on the heel than the toe, and the convexity allows a better grip to be obtained. Indeed, the ordinary form of shoe is sometimes slightly bent for the same reason. The Northumberland practice, with regard to “soles” and round shoes, while now based on expert veterinary principles, seems to represent a primitive plan. It is noteworthy that the round form of shoe is especially prized as a bringer of luck. Obviously, it is the material of the shoe—witch-hated iron—which is there considered important; all folk-memory respecting the virtues of the crescentic shape has perished. Youatt describes a shoe, under the name of the “bar-shoe,” which appears roughly to correspond to the Northumberland type[1149]. The observant person will occasionally see a London dray horse wearing a somewhat similar kind of shoe, the difference being in the hinder portion, which is either straight or slightly re-entering. This peculiar mode of shoeing, in all cases, seems to be due to the advice of the veterinary surgeon. It will be well, for anyone who wishes to pursue the subject, to read the opinions of the authorities already given (pp. 423 n., 424 n.).

A slight digression may here be made to consider a kindred topic. Iron objects of peculiar shape, commonly called hippo-sandals, have been discovered in various places, notably by Pitt-Rivers, in the Romano-British settlements in Cranborne Chase. Some authorities have thought that the hippo-sandal represents a kind of horseshoe, but Pitt-Rivers agrees with Fleming in scouting this theory. The shape, he considers, would be inconvenient for this purpose. Moreover, horseshoes, and—so it is believed—ox-shoes, are represented among the relics of the settlements, so that another type of shoe, he argues, would scarcely be found at the same spot. This objection is, as Mr C. Roach Smith has hinted, not conclusive against the use of hippo-sandals for special occasions. The suggestion was made by Pitt-Rivers that the hippo-sandals were intended for shoeing the poles of sledges, and he figures a form of that vehicle in which the shaft skids along the ground. In a footnote, however, he betrays some uncertainty, and admits that specimens of hippo-sandals which are displayed in the Museum at Mayence must have been fitted to the feet of horses, probably as splints when the hoofs had been accidently broken. The hippo-sandals, in those cases, would doubtless be attached by cords or straps which passed through the iron rings. Mr Roach Smith, writing in 1859, stated that iron pattens, fastened to the hoof by means of leather straps, were still used in Holland. This fact supplies, doubtless, the key to the puzzle. The hippo-sandal shown in the illustration (Fig. 86) was discovered, along with many other relics, at the Roman villa at Darenth, in Kent, and was first figured by Mr George Payne, in Archaeologia Cantiana. About half a dozen specimens are on view in the Guildhall Museum[1150]. In connection with this subject it may be noted that Youatt, in his book on The Horse, describes and illustrates a light kind of open-work sandal for horses with delicate hoofs, made of strap-work and iron clips.

A slight retrospect of the shoeing question will be made when we deal with oxen, but we must now return to the British horsemen of the Early Iron Age. At that period, so engrossing was the craft of the chariot-warrior, that care was often taken to provide the dead chieftain with the means of renewing his pastime elsewhere. Thus it was not unusual to inter a horse, or chariot, or both of these, in the burial mound. Keysler quotes numerous instances of the custom, chiefly with respect to the ancient Scythian and Scandinavian peoples[1151]. Records also implicate Tartars, Franks, Wends, and Finns as agents in like ceremonies. A well-known passage in Virgil seems to show that the practice obtained in classical times. Aeneas, on his descent into the lower regions, views with wonder the empty chariots of the chiefs, and the horses feeding at large on the plain. The heroes retained their old fondness for chariots and shining steeds[1152], and these necessaries had evidently been deposited in the earth at the time of the funeral.

The English records of chariot-burial are fairly numerous. The Rev. E. W. Stillingfleet (c. A.D. 1816), and Canon W. Greenwell (A.D. 1876), excavated several round barrows at a farm called Arras, near Market Weighton, in Yorkshire, and discovered therein remains of horses, chariots, and harness. Associated with these mute memorials, there was found, in one instance, a boar’s tusk, which had been invested with some ceremonial value, since it was perforated with a square hole, and was mounted in a brazen case. Interments of this class belong usually to the Late-Celtic period. A small urn, of unspecified age, dug up near Eastbourne in 1778, contained about a dozen horse’s teeth. Mr J. Romilly Allen compiled a considerable list of instances of chariot-burial. Many others are given in the Guide to the Early Iron Age (British Museum), as well as in the writings of Messrs L. Jewitt and J. R. Mortimer. So recently as 1906, a chariot-burial was discovered at Hunmanby, in Yorkshire, and was described by Canon Green well in Archaeologia[1153]. The “trappings” found in connection with the other remains comprised bridle-bits, buckles, head-ornaments, and similar articles.

One of the Wold barrows, which was opened by Canon Greenwell, contained a whole chariot and the bones of two horses, placed alongside a human skeleton. In another mound the wheels alone had been buried[1154]. A third grave yielded wheels and an iron bit[1155]. Similar discoveries were made at Nanterre, in France; horses were found entombed with portions of their trappings, together with tires of wheels, and various bronze and iron objects, evidently betokening a Transitional period[1156]. In all the foregoing cases we must suppose that the horse was sacrificed at its master’s funeral—the coincidence of natural or of violent death must have been exceedingly uncommon save in warfare. The custom of chariot-burial persisted for centuries into the Christian era, and an assemblage of relics, kindred to those mentioned, is common in Scandinavian graves of the Viking Age (A.D. 700-1000), wherein unburnt bodies were interred[1157]. Again, an Anglo-Saxon grave at Reading was shown to contain the skeleton of a horse, human bones, and a sword remarkably rich in its ornament[1158]. At a still later date, A.D. 1389, when Bertrand Duguesclin was buried at St Denis, several horses, which had been previously blessed by the Bishop of Auxerre, were sacrificed, or, as one account says, compounded for by the owners.

Were we ignorant of the foregoing facts, certain modern practices of an analogous nature would have to be dismissed as inexplicable. Once acquainted with the ancient instances, however, the student can account for the atavism which here and there betrays itself. We cannot, of course, in the absence of overlapping evidence, be certain that the burial of the horse along with its master is a custom which has never died out. There may have been a continuous bond of tradition, or again, folk-memory may have lain almost dormant for centuries, to be unconsciously revived at a later time. A few instances will now be rapidly surveyed.

A surgeon, one Mr Thomas Sheffield, dying in 1798, at Downton, in Wiltshire, left instructions that he should be interred in his garden, and that, when his favourite horse should die, it was to be laid by his side. Mr Sheffield was buried as he desired, but in 1807 his body was removed to the village churchyard[1159]. We are left to infer that the horse was placed in its master’s grave, as was undoubtedly done in the case, quoted by Southey, in which a man of Salisbury, “in derision of religion,” commanded that his horse should be slaughtered and buried with him[1160]. Again, so recently as 1866, when Queen Victoria’s huntsman died, his favourite horse was shot, and its ears were placed in his coffin and buried in his grave in Sunninghill churchyard, Berkshire[1161]. Parenthetically, we notice that, in Patagonia, the horse of a deceased person is still killed at the grave[1162]. Such incidents as these do not seem to be far removed in time from the days of barrow burial. Not quite so apposite is the case of Wellington’s horse, Copenhagen, which was buried (A.D. 1836) with full military honours at Strathfieldsaye (Hampshire), and which was commemorated by a tombstone bearing an appropriate inscription and epitaph. In thus honouring his charger, however, the Duke had a prototype in the Emperor Augustus, who, as Pliny relates, erected a tomb to his horse, on which occasion Germanicus Caesar wrote a poem[1163]. Turn the facts which way we will, they seem to tell of an ingrained instinct which unexpectedly reveals itself to the surprise of the majority of folk—surprise, nevertheless, which speedily becomes tinged with sympathy.

Not so distinctly a reversion, but still probably a custom derived from primitive observances, was the Mediaeval ceremony, when a great person was buried, of leading his horse before the body and presenting the animal to the ecclesiastical authorities as an obituary due[1164]. Such legacies were very common, so that a single example will suffice. At the obsequies of Henry V., three war-steeds were led to the altar, and were there formally bequeathed to the Church[1165]. It will be fresh in the memory of all, how, at the funeral of King Edward VII., that monarch’s favourite horse was led by a groom behind the body of his late master.

In considering how far these lingering customs may represent real survivals, it will be of some assistance to collect examples showing to what extent the horse cult was observed in the ceremonial routine of the ancient Celts and Teutons. In the first place, we are struck by the respect which was paid to white horses in particular. Tacitus, in a familiar passage, asserts that the German tribes kept milk-white horses in consecrated woods and groves[1166]. From these horses, which were never degraded by being put to any kind of labour, warnings and auguries were received by the priestly caste. Grimm tells us that there existed, at Drontheim, temples in which sacred horses were kept and fed[1167]. Other peoples have betrayed a similar affection for the white horse. Such animals, Virgil relates, are not usually put to work, since they are beloved of the gods; it is criminal to kill or wound them, except for sacrifice. Herodotus describes how the sacred white horses of the Persians were drowned when Cyrus was endeavouring to cross the river Gyndes[1168]. The same writer states that, in his day, Russia teemed with white horses[1169]. White was pre-eminently the noble colour. In the Apocalypse, a white horse is symbolical of victory and triumph[1170]. This idea is also common among classical writers[1171]. The figure of a white horse appeared on the Standard of the Saxons, and later, in the arms of Saxony and the House of Brunswick. In our day, a white horse constitutes the Kentish emblem, and is popular as a tavern sign. The celebrated “White Horse” carved on the Chalk downs near Uffington, Berkshire, and its fellow, incised on Bratton Hill, near Westbury, Wiltshire, though usually believed to commemorate victories over the Danes, are more probably to be referred to the Late Bronze, or Early Iron Age. In each case, the neighbouring country abounds with prehistoric remains—earthworks, barrows, and trackways[1172]. Certain details of the carvings, such as the bird-like head of the Uffington Horse, and the crescentic tail of the original, but now destroyed, “Horse” of Bratton, have been compared with corresponding features on early British coins[1173]. These coins were probably debased representations of the gold stater of Philip II. of Macedon. In modern times, other intaglios have been cut on our hillsides; these, while reviving the practice, have introduced breeds of horses unknown to the Britons. This seems a fitting place to observe that we have some indication of horse figures in Late-Celtic ornament. On a Late-Celtic bucket (c. first century B.C.) unearthed near Marlborough, in 1807, and enclosing burnt human bones, curious representations of the horse were carved[1174]. Belonging to about the same period are the queer horse-like figures depicted on a bronze-mounted wooden bucket, coming from the Late-Celtic cemetery of Aylesford, Kent[1175]. And, to conclude this section of our subject, we will note that the Anglo-Saxon tumulus in Taplow churchyard, Bucks. (cf. p. 81 supra), yielded portions of a bucket decorated with horseshoe symbols[1176]. We find representations of supposed horses appearing later on church fonts; the celebrated eleventh-century font of Burnsall, in Wharfedale, will serve as an example[1177].

Our discussion of the white horse has carried us far afield, and may have momentarily masked the general question. Not white horses alone were used in sacrifice and divination. The sacrifice of any horse was a most solemn event, attended with much ceremony, alike among Persians and Indians, among Teutons, Finns and Slavs[1178]. In auguries, too, the animal bore an honoured part. The Greeks, Strabo informs us, deemed the neighing of a horse an omen of good[1179]. In Germany, divinations by means of the horse lasted till the seventh century, for, when St Gall died,

unbroken horses were charged with the burden of his coffin, and to their decision was entrusted the choice of a burial-place[1180]. In Denmark, horse-sacrifices lingered until the early part of the eleventh century; a specific instance is given by Keysler, on the authority of the historian Dithmar, who was the Bishop of Mersburg, or Merseburg, and who died A.D. 1028. Dithmar relates that the Danes were wont to celebrate the Feast of the Epiphany by sacrificing ninety human victims, together with an equal number of dogs and cocks, in order to appease the infernal deities[1181]. The custom indicates a not infrequent kind of early compromise. Kemble states that, although bulls are known to have been used for divination in England, he knows of no allusion to augury by means of horses[1182]. A few faint traces, however, suggestive of the horse cult, may be detected. There is, for example, that curious story, told by Bede, how the priest Coifi rode on a stallion when he went to destroy the images in the heathen temple at Godmundingham (now Goodmanham) in Yorkshire[1183] (cf. p. 32 supra). As Bede’s narrative runs its length, we learn that a high priest among the pagan Saxons might lawfully ride only on a mare[1184], and one is inclined to speculate whether any of the idols took the form of this animal. We know that the stallion was the most honoured among horses[1185], and it is expressly stated that, when Coifi borrowed the king’s stallion, he did so in contempt of his former superstitions. The change of steed, at any rate, coincided with an onslaught upon established custom, and we shall see later that the priestly rule about riding mares only was abandoned. Another vestige of the horse cult was the belief, common among Teutonic peoples, that the last wisp of corn in the harvest field was inhabited by the sacred horse. For this reason, a horse, representing the corn-god, was customarily slaughtered, and eaten with special rites by the reapers at the harvest supper. Professor Frazer describes some quaint harvest customs, prevalent in Hertfordshire and Shropshire, which furnish examples of the corn-spirit, appearing in the shape of a horse or mare. And, again, in his recent work, Totemism and Exogamy, he records the Red Indian practice of sacrificing costly horses to appease the “medicine” or corn-spirit[1186].

Underlying such observances as those which have been described, there is an idea which gives a clue to a much-discussed problem. Folk of our generation are continually asking why the flesh of such a clean-feeding animal as the horse—a true vegetarian—should be despised as food. The question is not indeed altogether of recent date, for it was propounded in A.D. 1720 by Keysler, who reviews the subject at some length[1187]. He contends that the stringent prohibition must not be credited to the influence of the Mosaic Law; first, because no flesh, in itself, was deemed unclean for the saints, and secondly, because other articles of the ceremonial law had already at various times been abandoned with impunity[1188]. The rejection of horseflesh for food, Keysler concludes, was due to the Christian teachers, who found our pagan ancestors employing the animal in sacrifices and auguries, and eating its flesh in the subsequent repasts; hence, as a mark of disapprobation, this kind of food was forbidden to converts. The results, it is urged by the old antiquary, have been deplorable, more especially, because there is no law of Christ which prescribes this rule of conduct (Christi certe lex nulla exstat, quae eum agendi modum praescribat)[1189].

These propositions, in the main, seem undeniable. We have seen that the Palaeolithic cave-man ate horseflesh freely, and that the Britons of the Round Barrow Period were probably addicted to a like custom. There is little doubt, again, that throughout Roman Britain horseflesh was a common article of food. This is attested by the frequency of the occurrence of broken bones of the horse in the “Brit-Welsh” caves of the Iron Age[1190]. Corroboration of Keysler’s theory is afforded by historical facts. Pope Gregory III. (ruled A.D. 731-741), in a letter to St Boniface, the Apostle of Germany, forbade the eating of the flesh of wild horses as an unclean and execrable act[1191]. Yet at a somewhat earlier date, Gregory II., when consulted on the same perplexing subject, had sent a temporizing answer, shielding himself behind the famous passage in the Epistle to the Corinthians respecting meat offered to idols[1192]. So long as the new faith held its converts insecurely, and wherever Christianity was merely nominal, the frontier line of authority alternatively advanced and receded. Nearly half a century after the death of Gregory III., at the Council of Celchyth (A.D. 787), the consumption of horseflesh was noted as a stain on the character of the British Christians; their fellow-believers in the East were not guilty of such a sin (quod nullus Christianorum in orientalibus facit)[1193]. Yet the monks of St Gall not only ate horse-flesh, but returned thanks for it in the metrical grace, written by the monk Ekkehard III. (died c. A.D. 1036): “Sit feralis equi caro dulcis sub cruce Christi.” Elsewhere, too, the habit seemed incurable. The Norwegians, apparently in paying devotional honour to Odin, still ate the forbidden food during the eleventh and twelfth centuries[1194]. The growth of superstition tended to strengthen the Christian ban against horseflesh. This food was the reputed diet of giants and witches[1195]; its preparation was associated with sacrifices; it was eaten with hallowed salt. The sacrifices, in turn, were connected with popular assemblies or folk-moots[1196]. Now witches and trolls were supposed to live under mounds. Inside these mounds they held their dances, and played on pipes made of horse bones[1197]. The hillocks were, as a rule, actually barrows, the burial-places of bygone peoples, and the folk who had once raised them probably not only ate horseflesh ceremonially, but regarded it as welcome fare in times of dearth and scarcity. Successors of the mound builders continued to partake of horseflesh, and coupled the act with the worship of Odin. The ecclesiastical decrees were thus primarily directed against the pagan practice, but, because of superstition, the ban remained when its original necessity had passed away.

L’Abbé Valentin Dufour, who in the year 1868 translated and edited Keysler’s valuable chapter on the eating of horseflesh, adds a few facts which bring the story down to modern times. He tells us that the sale of horseflesh was forbidden in Paris in A.D. 1739, no reason being assigned for the prohibition. When, however, in A.D. 1784, a similar promulgation was issued, the ostensible motive was to prevent disease—there were certain maladies “que l’usage de pareilles chairs ne pouvait manquer d’occasionner[1198].” Since considerable importance was also attached to the assumed novelty of eating horseflesh, Dufour is at some pains to show that slaughter-houses (boucheries, écorcheries) existed, and that the forbidden flesh was vended, during the early part of the fifteenth century[1199]. Statutes continued to be passed against the use of horseflesh in France, until, in the early nineteenth century (1814, 1816, 1817), the commodity was allowed to be sold by certain persons who had secured the special privilege[1200]. Scarcity of food was doubtless a factor in bringing about a relaxation. By some writers it is supposed that the revulsion of feeling dates from the siege of Copenhagen (A.D. 1807), when the Danes ate horseflesh from necessity, and that the habit gradually spread all over Europe[1201]. This may be true in the general sense, but, archaeologically considered, one may doubt whether the practice had ever been really quite extinct.

The old pre-Christian veneration of the horse probably touches the groundwork of much of the folk-lore about the animal. Professor A. de Gubernatis, in his work on Zoological Mythology, deals fully with horse legends as exemplified in the Vedic, Greek, and Latin literatures, and particularly with the horse as the favourite animal of the solar hero[1202]. It is common, in ancient art, to find symbols of sun-worship associated either with the horse or the chariot, or with both. All that can be done in this place is to supply the reference. One old story may, nevertheless, be noted: that which tells how the Emperor Caligula spoke of raising his horse to the consulship. The usual explanation attributes the remark to a passing caprice, but another interpretation is conceivable. May it not be that Caligula intended the observation as a compliment to British and Gallic opinions concerning the sanctity of selected horses, opinions with which he must have been well acquainted?

We retrace our steps a little. Evidence seems to show that when the early Palaeolithic cave-men hunted the horse, they were accustomed to carry into their shelters only the fleshy parts of the carcass, together with the head of the animal, and—for ornamental purposes—the valuable tail. The long bones, which were crushed to obtain the marrow, do not appear, as a rule, to have been taken into the caves[1203]. Light may be cast on the anomalous separation of flesh and bones by a study of ancient Egyptian custom as described by Herodotus. This writer states that imprecations were heaped on the head of the sacrificial victim, so that any impending evil might fall thereon; the Egyptians, in consequence, would never eat the head of any animal[1204]. Strict taboo, as imposed among common folk, is not inconsistent with ceremonial eating by privileged individuals, and numerous instances might be given in support of this antinomy of custom. Merely as a speculation, it might be suggested that the head of the victim was at one time a delicate morsel reserved for the chieftain. In the caves of the Neolithic and Bronze Ages, the skull does not seem to be of common occurrence, but in the early historic period, as shown by folklore, we catch echoes of its legendary repute. Tacitus relates that the ancient German tribes hung the heads of animals on trees as offerings to Odin. In Teutonic fairy tales, the horse’s head works miracles, especially when played upon as an instrument[1205]. It was thrown by witches into the Midsummer fire[1206]—a notable collocation of details. Russian magic teaches that ambrosia comes out of a horse’s head, and enables its possessor to do deeds of prowess. By virtue of this ambrosia one hero discomfited ninety-nine hostile monsters[1207]. In parts of Germany, horses’ heads were buried in stables; in Holland, they were hung over pigstyes; in Mecklenburg, they were placed under a sick man’s pillow[1208]. Again, in Lower Saxony, horses’ heads, projecting outwards, were carved on the gables of buildings, ostensibly for ornament, but in reality, it is probable, to prevent mischief to the horses kept within. A similar practice was observed by the builders of the older houses in Rhaetia[1209]. In modern Norway, the handles of bowls, and the ends of the wooden lever by which the primitive mangles are worked, are often formed of carved horse-heads. Numerous examples may be seen in the Horniman Museum, London. Specimens are said to have been met with in English houses also. Some authorities considered that the figures represented a Celtic legacy, but Grimm claims that the custom of carving these images, like that of horse-worship generally, belongs “equally to Celts, Teutons, and Slavs[1210].” The domain might be much extended. Even in our own day (1865), such carvings as those described have been recorded from Jutland. They were once common, it is stated, in Sussex, and Miss M. Braitmaier has figured a series of modern gable ornaments from different parts of Germany[1211]. When someone asked the meaning of the horses’ heads on the Jutish gables, the natives answered, “Oh, they are Hengist and Horsa[1212].” (Note that the name Hengist = a stallion, and Horsa = a mare.) Whether or not Hengist and Horsa were historical personages may be left in abeyance, the fact remains that they were sometimes represented by horses’ heads carried in front of the army as tutelary deities.

In certain parts of England, notably in Kent, there still survives the custom of a group of men going round at Christmas carrying a horse’s head, crudely carved in wood, and known as the “hoodening horse.” Sometimes, it would appear, a skull long buried in the soil, and afterwards dug up by chance, formed the “wooser,” “wooset,” or “husset[1213].” Mr P. Maylam, who has carefully collated the records of analogous customs from both England and Germany, considers that the word “hoodening” is not, as popularly supposed, derived either from the Norse word Odin or the Low German form Woden. He prefers to connect it with those old performances in which the hobby horse and characters representing Robin Hood and Maid Marian were prominent. A writer in the Athenaeum ridicules this idea, and prosaically refers the name to the hood or sack which concealed the supposed body of the horse—really the body of the hoodener, or performer[1214]. Yes, but why should the horse, hooded or otherwise, enter into the ceremonies at all? It is easy to deride the early school, of which Grimm is a representative, as old-fashioned and full of extravagances. But we have to face a series of converging customs, which were not begotten of a complex society like that of modern or even Mediaeval England. Only by an appeal to some primitive form of the horse cult can an ultimate solution be really obtained.

Standing in close relationship to the “hoodening horse” custom, is a somewhat weird Welsh practice, which is now nearly extinct, but which some enthusiasts have lately attempted to revive. A horse’s skull is dressed up and carried about by a performer who is enveloped in a cloak. He makes the jaws of the skull snap to the accompaniment of Welsh rhymes. Houses are visited, and largesse is demanded. The performance, known colloquially as Mari Lwyd, is traced by some to pre-Reformation usage. But doubtless, it goes back, like the “hoodening horse,” of which, perhaps, it is a mere variant, to pagan times[1215].

Virgil relates a curious tradition which bears on our subject. As the Carthaginians were digging near a venerable wood, they dug up a horse’s skull—a “courser’s head,” as the phrase runs in Dryden’s translation, and this discovery was accounted such a prosperous omen that a temple was raised to Juno on that spot[1216]. Professor Conington, garnering his knowledge from several classical writers, gives us the additional information that the head of an ox was first lighted upon, and that this was thought to portend servitude, but after further excavation, the horse’s head appeared—an earnest of plenty, combined with success in war[1217]. From Vishnu mythology comes a contradictory item, for, in that system, the mouth of hell is conceived as a huge horse head[1218].

Before quitting this department of folk-lore, we may scan the wider field of skull superstitions in general. A fox’s head, nailed to a Scotch stable door, was supposed to keep off the dreaded witch. I noticed an instance of this custom at Rottingdean, near Brighton, in 1908, but cannot be sure that any significance was attached to the fox’s head[1219]. Why, again, does the gamekeeper suspend rows of weasels, stoats, cats, magpies, and jays on his gibbet? Certainly he does this, in the first place, to prove his zealous stewardship, and perhaps with some dimly conscious belief that similar “evil-doers” will take warning. But from observation of other curious practices, scarcely to be discussed here, one suspects that the origin was ceremonial. We must also remember cases like that described by Mr Baring-Gould, who once saw, hanging on a magnificent elm at Westmeston, under Ditchling Beacon, in Sussex, the carcasses of two horses and three calves. The reason offered for this custom was that the suspension of the bodies was lucky for cattle. Keeping, however, to a consideration of heads, we notice that Sir G. L. Gomme records a peculiar instance from Hornchurch, in Essex, where the lessee of the tithes used to pay, as a Christmas tribute, a boar’s head. This payment could not depend upon the intrinsic value of the toll, nor could the destruction of a single boar be counted meritorious in itself. The tribute was obviously symbolical. Camden relates that a stag was formerly paid as part of the rent of Church lands situated in Essex. He adds that, when he was a boy, namely, in the third quarter of the sixteenth century, the priests of St Paul’s Cathedral were accustomed to meet the stag as it was brought up the steps of the sacred building. The animal’s head was then carried on a spear round the cathedral, which echoed meanwhile to the sound of horns. Of this curious ceremony, the young antiquary was an eye-witness[1220].

From a review of these facts, we may deduce that the head of a slaughtered animal bore an imputed sanctity. This was essentially the case when the animal had been offered in sacrifice, and it is to pagan and prehistoric ritual that we must look for an interpretation of the facts. The species of animal esteemed most sacred would vary with the time and the place—here, the horse, there, the ox. Later days brought other competitors for the position of honour. Only by keeping well in mind the widespread belief in the efficacy of skulls, are we enabled to understand another series of records, which we now proceed to summarize.

More than half a century ago, when the chancel of St Botolph’s church, Boston, was being rebuilt, a quantity of horses’ bones and the jaw-bones of sheep were found under the floor[1221]. Again, we have seen that, on the site of the present St Paul’s Cathedral, a deposit of bones of oxen and other animals was discovered indicating a pagan site[1222] (cf. p. 83 supra). Secular buildings have also yielded horse remains. In 1895, when Colonel Stanley Scott was taking up the ground floor of a house in North Devon, he discovered, laid in order and well preserved, the skulls of eight horses and ten bullocks[1223]. In Wharfedale, again, under the floor of a house, probably from two to three hundred years old, the workmen took up the skulls of seven horses and a cow[1224]. With these discoveries one naturally associates the Dutch and German customs already mentioned (p. 440 supra). And, of course, the primitive idea must be connected with that which underlies foundation sacrifices, although complication arises from the unique merit attached to skulls. The foundation sacrifice is widely prevalent, but the burial of skulls is a more specialized custom.

It will be noticed that, of the last two examples, one is recent, and the other comparatively recent, therefore any folk-memory associated with the deposition of the skulls was probably defective. According to popular belief in Ireland, the skulls which are nowadays placed under buildings are intended to “cause an echo.” Just as a public building has, or has not, a horses skull buried beneath it, so will it be good or bad for the purpose of hearing. A certain field which possessed a good echo was commonly believed to have a horse interred in it; the tradition was sound, but it is not known whether the horse was buried for that purpose[1225]. Why, it may be pertinently asked, does a field need a good echo?

On broad grounds, it is sufficiently obvious that the sacrificial idea preceded the economic, yet there must have been a period of overlapping. For not only was some variation of the custom observed, as we shall see, by Mediaeval church builders, but the practice was kept up until our own days. Noticeably has this been the case in the Scottish Presbyterian Church[1226]. When the old Bristol Street meeting-house, in Edinburgh, was being demolished a century ago, eight horse skulls were found concealed in the sounding-board of the pulpit[1227]. Less than half a century back, the same class of object was put under an organ in a parish church in the province of Munster to increase the effect of the music[1228].

The modern theory of the acoustic purpose of the skulls fades as we trace the custom to more remote times. A small chamber in the belfry of Elsdon church, Northumberland, appeared to have been built specially to contain three horse skulls, which had lain piled against each other for hundreds of years[1229]. The masons of old time doubtless imagined that the skulls would make the tones of the bells more resonant, but, “lulled in the countless chambers of the brain” there must have been almost-forgotten memories of these traditional talismans. These sacred and oracular heads, there can be little question, were built into heathen temples before the dawn of history, and the habit was passed on from one generation to another[1230]. Does this theory seem far-fetched? Consider the conditions at Elsdon. Here is a district teeming with earthworks and other British and Roman remains. The population is scanty, the moorland wild and pathless; there was, until recently, little inter-communication among the scattered folk. Hereditary custom held firm sway. Such was the preference for burial in Elsdon churchyard, that corpses were carried many miles over the moors for interment. Yet pagan customs were rife. Well-worship was carried on here until our own times, and not many decades have passed since cattle were driven through the Midsummer bonfires to ward off disease. How much stronger was superstition when the village church of Elsdon was first built! There must have been dark, undisturbed depths of paganism in the lives of the countryfolk. We really know little of the true beliefs of the Mediaeval peasant, as recorded by himself. Even our information about the faiths held by the official classes, though somewhat exiguous, reveals a basis of gross superstition. The gap between the twentieth century and the sixteenth is almost immeasurable as compared with that between the sixteenth century and the Neolithic period.

Let us halt, to draw a comparison from Brittany. Who, in the absence of direct evidence, would have imagined that, in our own generation, a people, nominally Christian, could have been found to set out dishes of cream for the dead on All Souls’ Eve, to employ grave-earth for the cure of fevers, to pour out milk on tombs as a libation, to anoint menhirs with oil and honey, to scatter the ashes of the festival fires over the fields to ensure a fat harvest? Yet all these customs have been practised by the Bretons in recent times. Well-worship, the blessing of oxen at Carnac, the ghastly reverence paid to images personifying Death, and all such rites, we pass by, as being everywhere somewhat persistent. The parallel which I wish to draw is between the Mediaeval Englishman and the more modern Breton. Could we turn back and thoroughly understand the pages of history, I am convinced that even the seventeenth century peasant of the English Cornwall, for example, would be found quite as superstitious as the nineteenth century peasant of the French Cornwall. What is true of Cornwall, holds good for the Highlands of Scotland, for Ireland and Wales, and, in a lesser degree, for the whole of rural England.

To return: the acoustic idea had its birth so far back as Roman times at least, though at that period it was associated with the use of sounding jars. Probably horse skulls were still buried sacrificially, but the purpose was being forgotten. The belief in the efficacy of horse skulls as reverberators seems to have been derived from the employment of these jars, at a rather later time when the sacramental idea concerning skulls was obsolete. About the jars themselves there has been a vast controversy, which, even at the risk of being discursive, we must briefly notice.

To take a Roman example first: along the seats of the Coliseum there was a peculiar arrangement of horizontal pots, which Sir E. Beckett (Lord Grimthorpe) believed were intended to augment the sound. As a result of experiment, this authority found that the vessels acted much in the same way as would a series of short, wide tubes, if presented to a hemispherical bell when this was struck[1231]. Vitruvius mentions brazen vessels, perhaps comparable to the gong or kettle-drum, as being in use in Roman theatres. Some writers have thought that the purpose was to make the voices of the actors more distinct, others consider that the vessels were accessories in the imitation of thunder.

Coming to Mediaeval times, we find that the church of the Celestins at Metz was furnished (A.D. 1439) with jars, expressly to improve the chanting, but it is affirmed that experience showed them to be useless. Mr Gordon M. Hills, in a valuable paper on this subject, says that the jars were “a great disfigurement to the building, the marvel of all beholders, and the jest of fools[1232].” There are other Continental records of acoustic jars from Strasburg, Angers, Paris, and other places. L’Abbé Cochet discovered numerous specimens in the churches of Upper Normandy, together with “cornets” of baked earth in the church of St Blaise, at Arles. Illustrations of some of these are given in Cochet’s paper in the Gentleman’s Magazine[1233], and the statement is made that similar “cornets” are found in the interior walls and vaults of many churches in Sweden, Denmark, and Russia[1234]. Didron, after referring to specimens from the two first-named countries, and to those discovered at Arles and Metz, decides against the acoustic hypothesis: “Ce mode [d’acoustique] me semblait aussi puéril qu’inefficace[1235].” And, seeing that the men of the Middle Ages made bells and organs so commonly, why, he asks, are not the sounding “poteries” of more frequent occurrence[1236]?

In England, notable finds of jars are on record, though the number of churches concerned is but a trivial percentage of the whole. At St Clement’s, Sandwich, the jars were built into the walls of the chancel, overlooking the altar[1237]. At Barkway, Hertfordshire, they were likewise embedded in the chancel wall, but on the floor level[1238]. They are also found in the thickness of the wall, a few inches below the floor level, as at Fountains Abbey, where they had been placed at the base of the old choir screen. The Fountains vases lay on their sides, and both in and around them there was an abundance of charcoal. The charcoal, it is conjectured, may have had no more mysterious origin than a fire which occurred at the Dissolution[1239]. Jars, supposed to be of Romano-British make, were found on the top of the chancel wall at East Harling, Norfolk; in each case the mouth of the jar faced the interior of the chancel. For a long time a coating of lath and plaster had concealed these curiosities[1240], and one is led to wonder whether other jars may not, even now, lie hidden elsewhere. The gables of Newington church, Kent, yielded three jars. Other records come from Fairwell (Staffs.), Denford (Northants.), St Peter’s Mancroft and St Peter-per-Mountergate in Norwich, Upton, near Newark, and from Youghal, in Ireland[1241]. But the greatest collection of all was uncovered at the village church of Leeds, near Maidstone, in 1878. Altogether, about fifty earthenware pots were revealed. They were found on the top of each wall of the nave, below the wall plate. The walls and oaken roof belonged to the fifteenth century. The best judges at first declared that the vessels were of Romano-British manufacture, and dated a thousand years earlier than the fabric. This would seem to indicate that a series of urns had been discovered in the neighbourhood, and pressed into service by the Mediaeval masons. Later expert opinion, however, declares that the jars, though possessing some Celtic characteristics, are of Mediaeval date. The bodies of the jars were cylindrical, and about 8 or 9 inches in diameter, while the mouths narrowed to 3 or 4 inches. The height averaged 10-12 inches. The bottom of each jar was convex and perforated. Mr Hills calls attention to some perplexing general considerations. The jars are of any form and every form, they are old and new, they are placed, as if at hazard, from the floor to the roof. He therefore concludes that the intentions were several, although he does not himself suggest any other purpose to supplement the acoustic theory[1242]. In such a matter as this, difference of purpose, variety in underlying belief, changeable custom according to locality, confused folk-memory and tradition, need cause the antiquary no surprise. The prime motive having vanished, the custom is bereft of its full meaning, and the course of development runs along divergent rather than parallel lines. Two other discoveries which seem to favour the acoustic theory may be given—those at Ashburton, in Devon (1838), and Luppitt, also in Devon (1880). The Ashburton jars (Fig. 88 A), though convex at the base, and exhibiting chevron ornament, are assigned not to the Late-Celtic period, but to the close of the fourteenth or the beginning of the fifteenth century. The jars from Luppitt (Fig. 88 B) are comparable to those found at Leeds. They were apparently made especially for insertion in a wall, as they are flattened a little in one portion. They probably belong to the fifteenth century[1243].

This question of “acoustic jars” has been dilated upon because it seems to involve an indirect derivative of the skull superstition, and one is induced to outline the story, however roughly and tentatively. We start with a period when the horse cult is rife, and when solemnity is the note of the priest and soothsayer. At a later date, a horse, or among some peoples, an