Back Matter

Notes

It may be as well to give the reader some account of the enormous extent of the Celtic folk-tales in existence. I reckon these to extend to 2000, though only about 250 are in print The former number exceeds that known in France, Italy, Germany, and Russia, where collection has been most active, and is only exceeded by the MS. collection of Finnish folk-tales at Helsingfors, said to exceed 12,000. As will be seen, this superiority of the Celts is due to the phenomenal and patriotic activity of one man, the late J. F. Campbell, of Islay whose Popular Tales and MS. collections (partly described by Mr. Alfred Nutt in Folk-Lore, i. 369 - 83) contain references to no less than 1281 tales (many of them, of course, variants and scraps). Celtic folk-tales, while more numerous, are also the oldest of the modern European races; some of them—e.g., "Connla" in the present selection, occurring in the oldest Irish vellums. They include (1) fairy tales properly so-called—i.e., tales or anecdotes about fairies, hobgoblins, &c, told as natural occurrences (2) hero-tales, stories of adventure told of national or mythical heroes; (3) folk-tales proper, describing marvellous adventures of otherwise unknown heroes, in which there is a defined plot and supernatural characters (speaking animals, giants, dwarfs, &c.); and finally (4) drolls, comic anecdotes of feats of stupidity or cunning.

The collection of Celtic folk-tales began in Ireland as early as 1825 with T. Crofton Croker's Fairy Legends and Traditions of the South of Ireland.   This contained some 38 anecdotes of the first class mentioned above, anecdotes showing the belief of the Irish peasantry in the existence of fairies, gnomes, goblins, and the like. The Grimms did Croker the honour of translating part of his book, under the title of Irische Elfenmärchen. Among the novelists and tale-writers of the schools of Miss Edgeworth and Lever folk-tales were occasionally utilised, as by Carleton in his Traits and Stories, by S. Lover in his Legends and Stories, and by G. Griffin in his Tales of a Jury-Room. These all tell their tales in the manner of the stage Irishman. Chapbooks, Royal Fairy Tales and Hibernian Tales, also contained genuine folk-tales, and attracted Thackeray's attention in his Irish Sketch-Book. The Irish Grimm, however, was Patrick Kennedy, a Dublin bookseller, who believed in fairies, and in five years (1866 - 71) printed about 100 folk- and hero-tales and drolls (classes 2, 3, and 4 above) in his Legendary Fictions of the Irish Celts, 1866, Fireside Stories of Ireland, 1870, and Bardic Stories of Ireland, 1871; all three are now unfortunately out of print. He tells his stories neatly and with spirit, and retains much that is volkstümlich in his diction. He derived his materials from the English-speaking peasantry of county Wexford, who changed from Gaelic to English while story-telling was in full vigour, and therefore carried over the stories with the change of language. Lady Wylde has told many folk-tales very effectively in her Ancient Legends of Ireland, 1887. More recently two collectors have published stories gathered from peasants of the West and North who can only speak Gaelic. These are by an American gentleman named Curtin, Myths and Folk-Tales of Ireland, 1890; while Dr. Douglas Hyde has published in Beside the Fireside, 1891, spirited English versions of some of the stories he had published in the original Irish in his Leabhar Sgeulaighteachta, Dublin, 1889. Miss Maclintoch has a large MS. collection, part of which has appeared in various periodicals; and Messrs. Larminie and D. Fitzgerald are known to have much story material in their possession.

But beside these more modern collections there exist in old and middle Irish a large number of hero-tales (class 2) which formed the staple of the old ollahms or bards. Of these tales of "cattle-liftings, elopements, battles, voyages, courtships, caves, lakes, feasts, sieges, and eruptions," a bard of even the fourth class had to know seven fifties, presumably one for each day of the year. Sir William Temple knew of a north-country gentleman of Ireland who was sent to sleep every evening with a fresh tale from his bard. The Book of Leinster, an Irish vellum of the twelfth century, contains a list of 189 of these hero-tales, many of which are extant to this day; E. O'Curry gives the list in the Appendix to his MS. Materials of Irish History. Another list of about 70 is given in the preface to the third volurne of the Ossianic Society's publications. Dr. Joyce published a few of the more celebrated of these in Old Celtic Romances; others appeared in Atlantis (see notes on "Deirdre"), others in Kennedy's Bardic Stories, mentioned above.

Turning to Scotland , we must put aside Chambers' Popular Rhymes of Scotland, 1842, which contains for the most part folk-tales common with those of England rather than those peculiar to the Gaelic-speaking Scots. The first name here in time as in importance is that of J. F. Campbell, of Islay. His four volumes, Popular Tales of the West Highlands (Edinburgh, 1860-2, recently republished by the Islay Association), contain some 120 folk- and hero-tales, told with strict adherence to the language of the narrators, which is given with a literal, a rather too literal, English version. This careful accuracy has given an un-English air to his versions, and has prevented them attaining their due popularity. What Campbell has published represents only a tithe of what he collected. At the end of the fourth volume he gives a list of 791 tales, &c., collected by him or his assistants in the two years 1859-61 and in his MS. collections at Edinburgh are two other lists containing 400 more tales. Only a portion of these are in the Advocates' Library; the rest, if extant, must be in private hands, though they are distinctly of national importance and interest.

Campbell's influence has been effective of recent years in Scotland. The Celtic Magazine (vols. xii. and xiii.), while under the editorship of Mr. MacBain, contained several folk- and hero-tales in Gaelic, and so did the Scotch Celtic Review. These were from the collections of Messrs. Campbell of Tiree, Carmichael, and K Mackenzie. Recently Lord Archibald Campbell has shown laudable interest in the preservation of Gaelic folk- and hero-tales. Under his auspices a whole series of handsome volumes, under the general title of Waifs and Strays of Celtic Tradition, has been recently published, four volumes having already appeared, each accompanied by notes by Mr. Alfred Nutt, which form the most important aid to the study of Celtic Folk-Tales since Campbell himself. Those to the second volume in particular (Tales collected by Rev. D. Maclnnes) fill 100 pages, with condensed information on all aspects of the subject dealt with in the light of the most recent research in the European folk-tales as well as on Celtic literature. Thanks to Mr. Nutt, Scotland is just now to the fore in the collection and study of the British Folk-Tale.

Wales makes a poor show beside Ireland and Scotland Sikes' British Gobllns, and the tale, collected by Prof. Rhys in Y Cymrodor, vols. ii.-vi., are mainly of our first-class fairy anecdotes. Borrow, in his Wild Wales, refers to a collection of fables in a journal called The Greal, while the Cambrian Quarterly Magazine for 1830 and 1831 contained a few fairy anecdotes, including a curious version of the "Brewery of Eggshells" from the Welsh. In the older literature, the Iola MS., published by the Welsh MS. Society, has a few fables and apologues, and the charming Mabinogion, translated by Lady Guest, has tales that can trace back to the twelfth century and are on the border-line between folk-tales and hero-tales.

Cornwall and Man are even worse than Wales. Hunt's Drolls from the West of England has nothing distinctively Celtic, and it is only by a chance Lhuyd chose a folk-tale as his specimen of Cornish in his Archæologia Britannica, 1709 (see Tale of Ivan). The Manx folk-tales published, including the most recent by Mr. Moore, in his Folk-Lore of the Isle of Man, 1891, are mainly fairy anecdotes and legends.

From this survey of the field of Celtic folk-tales it is clear that Ireland and Scotland provide the lion's share. The interesting thing to notice is the remarkable similarity of Scotch and Irish folk-tales. The continuity of language and culture between these two divisions of Gaeldom has clearly brought about this identity of their folk-tales. As will be seen from the following notes, the tales found in Scotland can almost invariably be paralleled by those found in Ireland, and vice versâ. This result is a striking confirmation of the general truth that the folk-lores of different countries resemble one another in proportion to their contiguity and to the continuity of language and culture between them.

Another point of interest in these Celtic folk-tales is the light they throw upon the relation of hero-tales and folk-tales (classes 2 and 3 above). Tales told of Finn or Cuchulain, and therefore coming under the definition of hero-tales, are found elsewhere told of anonymous or unknown heroes. The question is, were the folk-tales the earliest, and were they localised and applied to the heroes, or were the heroic sagas generalised and applied to an unknown τίς? All the evidence, in my opinion, inclines to the former view, which, as applied to Celtic folk-tales, is of very great literary importance; for it is becoming more and more recognised, thanks chiefly to the admirable work of Mr. Alfred Nutt, in his Studies on the Holy Grail, that the outburst of European Romance in the twelfth century was due, in large measure, to an infusion of Celtic hero-tales into the literature of the Romance-speaking nations. Now the remarkable thing is, how these hero-tales have lingered on in oral tradition even to the present day. (See a marked case in "Deirdre.") We may, therefore, hope to see considerable light thrown on the most characteristic spiritual product of the Middle Ages, the literature of Romance and the spirit of chivalry, from the Celtic folk-tales of the present day. Mr. Alfred Nutt has already shown this to be true of a special Section of Romance literature, that connected with the Holy Grail, and it seems probable that further study will extend the field of application of this new method of research.

The Celtic folk-tale again has interest in retaining many traits of primitive conditions among the early inhabitants of these isles which are preserved by no other record. Take, for instance, the calm assumption of polygamy in "Gold Tree and Silver Tree." That represents a state of feeling that is decidedly pre-Christian. The belief in an external soul "Life Index," recently monographed by Mr. Frazer in his "Golden Bough," also finds expression in a couple of the Tales (see notes on "Sea-Maiden" and "Fair, Brown, and Trembling"), and so with many other primitive ideas.

Care, however, has to be taken in using folk-tales as evidence for primitive practice among the nations where they are found. For the tales may have come from another race—that is, for example, probably the case with "Gold Tree and Silver Tree" (see Notes). Celtic tales are of peculiar interest in this connection, as they afford one of the best fields for studying the problem of diffusion, the most pressing of the problems of the folk-tales just at present, at least in my opinion. The Celts are at the furthermost end of Europe. Tales that travelled to them could go no further and must therefore be the last links in the chain.

For all these reasons, then, Celtic folk-tales are of high scientific interest to the folk-lorist, while they yield to none in imaginative and literary qualities. In any other country of Europe some national means of recording them would have long ago been adopted. M. Luzel, e.g., was commissioned by the French Minister of Public Instruction to collect and report on the Breton folk-tales. England, here as elsewhere without any organised means of scientific research in the historical and philological sciences, has to depend on the enthusiasm of a few private individuals for work of national importance. Every Celt of these islands or in the Gaeldom beyond the sea, and every Celt-lover among the English-speaking nations, should regard it as one of the duties of the race to put its traditions on record in the few years that now remain before they will cease for ever to be living in the hearts and memories of the humbler members of the race.

In the following Notes I have done as in my English Fairy Tales, and given first, the sources whence I drew the tales, then parallels at length for the British Isles, with bibliographical references for parallels abroad, and finally, remarks where the tales seemed to need them. In these I have not wearied or worried the reader with conventional tall talk about the Celtic genius and its manifestations in the folk-tale; on that topic one can only repeat Matthew Arnold when at his best, in his Celtic Literature. Nor have I attempted to deal with the more general aspects of the study of the Celtic folk-tale. For these I must refer to Mr. Nutt's series of papers in The Celtic Magazine, vol. xii., or, still better, to the masterly introductions he is contributing to the series of Waifs and Strays of Celtic Tradition, and to Dr. Hyde's Beside the Fireside. In my remarks I have mainly confined myself to discussing the origin and diffusion of the various tales, so far as anything definite could be learnt or conjectured on that subject.

Before proceeding to the Notes, I may "put in," as the lawyers say, a few summaries of the results reached in them. Of the twenty-six tales, twelve (i., ii., v., viii., ix., x., xi., xv., xvi., xvii., xix., xxiv.) have Gaelic originals; three (vii., xiii., xxv.) are from the Welsh; one (xxii.) from the now extinct Cornish; one an adaptation of an English poem founded on a Welsh tradition (xxi., "Gellert") and the remaining nine are what may be termed Anglo-Irish. Regarding their diffusion among the Celts, twelve are both Irish and Scotch (iv., v., vi., ix., x., xiv.-xvii., xix., xx., xxiv); one (xxv.) is common to Irish and Welsh; and one (xxii.) to Irish and Cornish; seven are found only among the Celts in Ireland (i.-iii., xii., xviii., xxii., xxvi); two (viii., xi.) among the Scotch; and three (vii., xiii., xxi.) among the Welsh. Finally, so far as we can ascertain their origin, four (v., xvi., xxi., xxii.) are from the East; five (vi., x., xiv., xx., xxv.) are European drolls; three of the romantic tales seem to have been imported (vii., ix., xix.); while three others are possibly Celtic exportations to the Continent (xv., xvii., xxiv.) though the last may have previously come thence; the remaining eleven are, as far as known, original to Celtic lands. Somewhat the same result would come out, I believe, as the analysis of any representative collection of folk-tales of any European district.

I. CONNLA AND THE FAIRY MAIDEN.

Source.—From the old Irish "Echtra Condla chaim maic Cuind Chetchathaig" of the Leabhar na h-Uidhre ("Book of the Dun Cow"), which must have been written before 1106, when its scribe Maelmori ("Servant of Mary") was murdered. The original is given by Windisch in his Irish Grammar, p. 120, also in the Trans. Kilkenny Archæol. Soc. for 1874. A fragment occurs in a Rawlinson MS., described by Dr. W. Stokes, Tripartite Life, p. xxxvi. I have used the translation of Prof. Zimmer in his Keltische Beitrage, ii. (Zeits. f. deutsches Altertum, Bd. xxxiii. 262-4). Dr. Joyce has a somewhat florid version in his Old Celtic Romances, from which I have borrowed a touch or two. I have neither extenuated nor added aught but the last sentence of the Fairy Maiden's last speech. Part of the original is in metrical form, so that the whole is of the cante-fable species which I believe to be the original form of the folk-tale (Cf. Eng. Fairy Tales, notes, p. 240, and infra, p. 257).

Parallels.—Prof. Zimmer's paper contains three other accounts of the terra repromissionis in the Irish sagas, one of them being the similar adventure of Cormac the nephew of Connla, or Condla Ruad as he should be called. The fairy apple of gold occurs in Cormac Mac Art's visit to the Brug of Manannan (Nutt's Holy Grail, 193).

Remarks.—Conn the hundred-fighter had the head-kingship of Ireland 123-157 A.D. , according to the Annals of the Four Masters, i. 105. On the day of his birth the five great roads from Tara to all parts of Ireland were completed: one of them from Dublin is still used. Connaught is said to have been named after him, but this is scarcely consonant with Joyce's identification with Ptolemy's Nagnatai (Irish Local Names, i. 75). But there can be little doubt of Conn's existence as a powerful ruler in Ireland in the second century. The historic existence of Connla seems also to be authenticated by the reference to him as Conly, the eldest son of Conn, in the Annals of Clonmacnoise. As Conn was succeeded by his third son, Art Enear, Connla was either slain or disappeared during his father's lifetime. Under these circumstances it is not unlikely that our legend grew up within the century after Conn—i.e., during the latter half of the second century.

As regards the present form of it, Prof. Zimmer (l.c. 261-2) places it in the seventh century. It has clearly been touched up by a Christian hand who introduced the reference to the day of judgment and to the waning power of the Druids. But nothing turns upon this interpolation, so that it is likely that even the present form of the legend is pre-Christian—i.e. for Ireland, pre-Patrician, before the fifth century.

The tale of Connla is thus the earliest fairy tale of modern Europe. Besides this interest it contains an early account of one of the most characteristic Celtic conceptions, that of the earthly Paradise, the Isle of Youth, Tir-na n-Og. This has impressed itself on the European imagination; in the Arthuriad it is represented by the Vale of Avalon, and as represented in the various Celtic visions of the future life, it forms one of the main sources of Dante's Divina Commedia. It is possible too, I think, that the Homeric Hesperides and the Fortunate Isles of the ancients had a Celtic origin (as is well known, the early place-names of Europe are predominantly Celtic). I have found, I believe, a reference to the conception in one of the earliest passages in the classics dealing with the Druids. Lucan, in his Pharsalia (i. 450-8), addresses them in these high terms of reverence:

The passage certainly seems to me to imply a different conception from the ordinary classical views of the life after death, the dark and dismal plains of Erebus peopled with ghosts; and the passage I have italicised would chime in well with the conception of a continuance of youth (idem spiritus) in Tir-na n-Og (orbe alio).

One of the most pathetic, beautiful, and typical scenes in Irish legend is the return of Ossian from Tir-na n-Og, and his interview with St. Patrick. The old faith and the new, the old order of things and that which replaced it, meet in two of the most characteristic products of the Irish imagination (for the Patrick of legend is as much a legendary figure as Oisin himself). Ossian had gone away to Tir-na n-Og with the fairy Niamh under very much the same circumstances as Condla Ruad; time flies in the land of eternal youth, and when Ossian returns, after a year as he thinks, more than three centuries had passed, and St. Patrick had just succeeded in introducing the new faith. The contrast of Past and Present has never been more vividly or beautifully represented.

II. GULEESH.

Source.—From Dr. Douglas Hyde's Beside the Fire, 104-28, where it is a translation from the same author's Leabhar Sgeulaighteachta. Dr Hyde got it from one Shamus O'Hart, a gamekeeper of French-park. One is curious to know how far the very beautiful landscapes in the story are due to Dr. Hyde, who confesses to have only taken notes. I have omitted a journey to Rome, paralleled, as Mr. Nutt has pointed out, by the similar one of Michael Scott (Waifs and Strays, i. 46), and not bearing on the main lines of the story. I have also dropped a part of Guleesh's name: in the original he is "Guleesh na guss dhu," Guleesh of the black feet, because he never washed them; nothing turns on this in the present form of the story, but one cannot but suspect it was of importance in the original form.

Parallels.—Dr. Hyde refers to two short stories, "Midnight Ride" (to Rome) and "Stolen Bride," in Lady Wilde's Ancient Legends. But the closest parallel is given by Miss Maclintock's Donegal tale of "Jamie Freel and the Young Lady," reprinted in Mr. Yeats' Irish Folk and Fairy Tales, 52-9. In the Hibernian Tales, "Mann o' Malaghan and the Fairies," as reported by Thackeray in the Irish Sketch-Book, c. xvi., begins like "Guleesh."

III. FIELD OF BOLIAUNS.

Source.—T. Crofton Croker's Fairy Legends of the South of Ireland, ed. Wright, pp. 135-9. In the original the gnome is a Cluricaune, but as a friend of Mr. Batten's has recently heard the tale told of a Lepracaun, I have adopted the better known title.

Remarks.—Lepracaun is from the Irish leith bhrogan, the one-shoemaker (cf. brogue), according to Dr. Hyde. He is generally seen (and to this day, too) working at a single shoe, cf. Croker's story "Little Shoe," l.c. pp. 142-4. According to a writer in the Revue Celtique, i. 256, the true etymology is luchor pan, "little man." Dr. Joyce also gives the same etymology in Irish Names and Places, i. 183 where he mentions several places named after them.

IV. HORNED WOMEN.

Source.—Lady Wilde's Ancient Legends, the first story.

Parallels.—A similar version was given by Mr. D. Fitzgerald in the Revue Celtique, iv. 181, but without the significant and impressive horns. He refers to Cornhill for February 1877, and to Campbell's "Sauntraigh" No. xxii. Pop. Tales, ii. 52-4, in which a "woman of peace" (a fairy) borrows a woman's kettle and returns it with flesh in it, but at last the woman refuses, and is persecuted by the fairy. I fail to see much analogy. A much closer one is in Campbell, ii. p. 63, where fairies are got rid of by shouting "Dunveilg is on fire." The familiar "lady-bird, lady-bird, fly away home, your house is on fire and your children at home," will occur to English minds. Another version in Kennedy's Legendary Fictions, p. 164, "Black Stairs on Fire."

Remarks.—Slievenamon is a famous fairy palace in Tipperary according to Dr. Joyce, l.c. i. 178. It was the hill on which Finn stood when he gave himself as the prize to the Irish maiden who should run up it quickest. Grainne won him with dire consequences, as all the world knows or ought to know (Kennedy, Legend Fict., 222, "How Fion selected a Wife").

V. CONAL YELLOWCLAW.

Source.—Campbell, Pop. Tales of West Highlands, No. v. pp. 105-8, "Conall Cra Bhuidhe." I have softened the third episode, which is somewhat too ghastly in the original. I have translated "Cra Bhuide" Yellowclaw on the strength of Campbell's etymology, l.c. p. 158.

Parallels.—Campbell's vi. and vii. are two variants showing how widespread the story is in Gaelic Scotland. It occurs in Ireland where it has been printed in the chapbook, Hibernian Tales, as the "Black Thief and the Knight of the Glen," the Black Thief being Conall, and the knight corresponding to the King of Lochlan (it is given in Mr. Lang's Red Fairy Book). Here it attracted the notice of Thackeray, who gives a good abstract of it in his Irish Sketch-Book, ch. xvi. He thinks it "worthy of the Arabian Nights, as wild and odd as an Eastern tale." "That fantastical way of bearing testimony to the previous tale by producing an old woman who says the tale is not only true, but who was the very old woman who lived in the giant's castle is almost" (why "almost," Mr. Thackeray?) "a stroke of genius." The incident of the giant's breath occurs in the story of Koisha Kayn, Maclnnes' Tales, i. 241, as well as the Polyphemus one, ibid. 265. One-eyed giants are frequent in Celtic folk-tales (e.g., in The Pursuit of Diarmaid and in the Mabinogi of Owen).

Remarks.—Thackeray's reference to the "Arabian Nights" is especially apt, as the tale of Conall is a framework story like The 1001 Nights, the three stories told by Conall being framed, as it were, in a fourth which is nominally the real story. This method employed by the Indian story-tellers and from them adopted by Boccaccio and thence into all European literatures (Chaucer, Queen Margaret, &c.), is generally thought to be peculiar to the East, and to be ultimately derived from the Jatakas or Birth Stories of the Buddha who tells his adventures in former incarnations. Here we find it in Celtdom, and it occurs also in "The Story-teller at Fault" in this collection, and the story of Koisha Kayn in Macinnes' Argyllshire Tales, a variant of which collected but not published by Campbell has no less than nineteen tales enclosed in a framework. The question is whether the method was adopted independently in Ireland, or was due to foreign influences. Confining ourselves to "Conal Yellowclaw," it seems not unlikely that the whole story is an importation. For the second episode is clearly the story of Polyphemus from the Odyssey which was known in Ireland perhaps as early as the tenth century (see Prof. K. Meyer's edition of Merugud Uilix maic Leirtis, Pref. p. xii). It also crept into the voyages of Sindbad in the Arabian Nights. And as told in the Highlands it bears comparison even with the Homeric version. As Mr. Nutt remarks (Celt. Mag. xii.) the address of the giant to the buck is as effective as that of Polyphemus to his ram. The narrator, James Wilson, was a blind man who would naturally feel the pathos of the address; "it comes from the heart of the narrator;" says Campbell (l.c., 148), "it is the ornament which his mind hangs on the frame of the story."

VI. HUDDEN AND DUDDEN.

Source.—From oral tradition, by the late D. W. Logie, taken down by Mr. Alfred Nutt.

Parallels.—Lover has a tale, "Little Fairly," obviously derived from this folk-tale; and there is another very similar, "Darby Darly." Another version of our tale is given under the title "Donald and his Neighbours," in the chapbook Hiberman Tales, whence it was reprinted by Thackeray in his Irish Sketch-Book, c. xvi. This has the incident of the "accidental matricide," on which see Prof, R. Köhler on Gonzenbach Sicil. Mährchen, ii. 224. No less than four tales of Campbell are of this type (Pop. Tales, ii. 218-31). M. Cosquin, in his "Contes populaires de Lorraine," the storehouse of "storiology," has elaborate excursuses in this class of tales attached to his Nos. x. and xx. Mr. Clouston discusses it also in his Pop. Tales, ii. 229-88. Both these writers are inclined to trace the chief incidents to India. It is to be observed that one of the earliest popular drolls in Europe, Unibos, a Latin poem of the eleventh, and perhaps the tenth, century, has the main outlines of the story, the fraudulent sale of worthless objects and the escape from the sack trick. The same story occurs in Straparola, the European earliest collection of folk-tales in the sixteenth century. On the other hand, the gold sticking to the scales is familiar to us in Ali Baba. (Cf. Cosquin, l.c., i. 225-6, 229).

Remarks.—lt is indeed curious to find, as M. Cosquin points out a cunning fellow tied in a sack getting out by crying, "I won't marry the princess," in countries so far apart as Ireland, Sicily (Gonzenbach, No.71), Afghanistan (Thorburn, Bannu, p. 184), and Jamaica (Folk-Lore Record, iii. 53). It is indeed impossible to think these are disconnected, and for drolls of this kind a good case has been made out for the borrowing hypotheses by M. Cosquin and Mr. Clouston. Who borrowed from whom is another and more difficult question which has to be judged on its merits in each individual case.

This is a type of Celtic folk-tales which are European in spread, have analogies with the East, and can only be said to be Celtic by adoption and by colouring. They form a distinct section of the tales told by the Celts, and must be represented in any characteristic selection. Other examples are xi., xv., xx., and perhaps xxii.

VII. SHEPHERD OF MYDDVAI.

Source.—Preface to the edition of "The Physicians of Myddvai"; their prescription-book, from the Red Book of Hergest, published by the Welsh MS. Society in 1861. The legend is not given in the Red Book, but from oral tradition by Mr. W. Rees, p. xxi. As this is the first of the Welsh tales in this book it may be as well to give the reader such guidance as I can afford him on the intricacies of Welsh pronunciation, especially with regard to the mysterious w's and y's of Welsh orthography. Fur w substitute double o, as in "fool" and for y, the short u in but, and as near approach to Cymric speech will be reached as is possible for the outlander. It may be added that double d equals th, and double l is something like Fl, as Shakespeare knew in calling his Welsh soldier Fluellen (Llewelyn). Thus "Meddygon Myddvai" would be Anglicè "Methugon Muthvai."

Parallels.—Other versions of the legend of the Van Pool are given in Cambro-Briton, ii. 315; W. Sikes, British Goblins, p. 40. Mr. E. Sidney Hartland has discussed these and others in a set of papers contributed to the first volume of The Archæological Review (now incorporated into Folk-Lore), the substance of which is now given in his Science of Fairy Tales, 274-332. (See also the references given in Revue Celtique, iv., 187 and 268). Mr. Hartland gives there an ecumenical collection of parallels to the several incidents that go to make up our story—(1) The bride-capture of the Swan-Maiden, (2) the recognition of the bride, (3) the taboo against causeless blows, (4) doomed to be broken, and (5) disappearance of the Swan-Maiden, with (6) her return as Guardian Spirit to her descendants. In each case Mr. Hartland gives what he considers to be the most primitive form of the incident. With reference to our present tale, he comes to the conclusion, if I understand him aright, that the lake-maiden was once regarded as a local divinity. The physicians of Myddvai were historic personages, renowned for their medical skill for some six centunes, till the race died out with John Jones, fl. 1743. To explain their skill and uncanny knowledge of herbs, the folk traced them to a supernatural ancestress, who taught them their craft in a place still called Pant-y-Meddygon ("Doctors' Dingle"). Their medical knowledge did not require any such remarkable origin, as Mr. Hartland has shown in a paper "On Welsh Folk-Medicine," contributed to Y Cymmrodor, vol. xii. On the other hand, the Swan-Maiden type of story is widespread through the Old World. Mr. Morris' "Land East of the Moon and West of the Sun," in The Earthly Paradise, is taken from the Norse version. Parallels are accumulated by the Grimms, ii. 432; Köhler on Gonzenbach, ii. 20; or Blade, 149; Stokes' Indian Fairy Tales, 243, 276; and Messrs. Jones and Koopf, Magyar Folk-Tales, 362-5. It remains to be proved that one of these versions did not travel to Wales, and become there localised. We shall see other instances of such localisation or specialisation of general legends.

VIII. THE SPRIGHTLY TAILOR.

Source.—Notes and Queries for December 21, 1861, to which it was communicated by "Cuthbert Bede," the author of Verdant Green, who collected it in Cantyre.

Parallels.—Miss Dempster gives the same story in her Sutherland Collection, No. vii. (referred to by Campbell in his Gaelic list, at end of vol. iv.); Mrs. John Faed, I am informed by a friend, knows the Gaelic version, as told by her nurse in her youth. Chambers' "Strange Visitor," Pop. Rhymes of Scotland, 64, of which I gave an Anglicised version in my English Fairy Tales, No. xxxii., is clearly a variant.

Remarks.—The Macdonald of Saddell Castle was a very great man indeed. Once, when dining with the Lord-Lieutenant, an apology was made to him for placing him so far away from the head of the table. "Where the Macdonald sits," was the proud response, "there is the head of the table."

IX. DEIRDRE.

Source.—Celtic Magazine, xiii. pp. 69, seq. I have abridged somewhat, made the sons of Fergus all faithful instead of two traitors, and omitted an incident in the house of the wild men called here "strangers." The original Gaelic was given in the Transactions of the Inverness Gaelic Society for 1887, p. 241, seq., by Mr. Carmichael. I have inserted Deirdre's "Lament" from the Book of Leinster.

Parallels.—This is one of the three most sorrowful Tales of Erin, (the other two, Children of Lir and Children of Tureen, are given in Dr. Joyce's Old Celtic Romances), and is a specimen of the old heroic sagas of elopement, a list of which is given in the Book of Leinster. The "outcast child" is a frequent episode in folk and hero-tales: an instance occurs in my English Fairy Tales, No. xxxv., and Prof. Köhler gives many others in Archiv. f. Slav. Philologie, i. 288. Mr. Nutt adds tenth Century Celtic parallels in Folk-Lore, vol. ii. The wooing of hero by heroine is a characteristic Celtic touch. See "Connla" here, and other examples given by Mr. Nutt in his notes to Maclnnes' Tales. The trees growing from the lovers' graves occurs in the English ballad of Lord Lovel and has been studied in Mélusine.

Remarks.—The "Story of Deirdre" is a remarkable instance of the tenacity of oral tradition among the Celts. It has been preserved in no less than five versions (or six, including Macpherson's "Darthula") ranging from the twelfth to the nineteenth century. The earliest is in the twelfth century, Book of Leinster, to be dated about 1140 (edited in facsimile under the auspices of the Royal Irish Academy, i. 147, seq.). Then comes a fifteenth century version, edited and translated by Dr. Stokes in Windisch's Irische Texte II., ii. 109, seq., "Death of the Sons of Uisnech." Keating in his History of Ireland gave another version in the seventeenth century. The Dublin Gaelic Society published an eighteenth century version in their Transactions for 1808. And lastly we have the version before us, collected only a few years ago, yet agreeing in all essential details with the version of the Book of Leinster. Such a record is unique in the history of oral tradition, outside Ireland, where, however, it is quite a customary experience in the study of the Finn-saga. It is now recognised that Macpherson had, or could have had, ample material for his rechauffé of the Finn or "Fingal" saga. His "Darthula" is a similar cobbling of our present story. I leave to Celtic specialists the task of settling the exact relations of these various texts. I content myself with pointing out the fact that in these latter days of a seemingly prosaic century in these British Isles there has been collected from the lips of the folk a heroic story like this of "Deirdre," full of romantic incidents, told with tender feeling and considerable literary skill. No other country in Europe, except perhaps Russia, could provide a parallel to this living on of Romance among the common folk. Surely it is a bounden duty of those who are in the position to put on record any such utterances of the folk-imagination of the Celts before it is too late.

X. MUNACHAR AND MANACHAR.

Source.—I have combined the Irish version given by Dr. Hyde in his Leabhar Sgeul., and translated by him for Mr. Yeats' Irish Folk and Fairy Tales, and the Scotch version given in Gaelic and English by Campbell, No. viii.

Parallels.—Two English versions are given in my Eng. Fairy Tales, No. iv., "The Old Woman and her Pig," and xxxiv., "The Cat and the Mouse," where see notes for other variants in these isles. M. Cosquin, in his notes to No. xxxiv., of his Contes de Lorraine, t. ii. pp. 35-41, has drawn attention to an astonishing number of parallels scattered through all Europe and the East (cf., too, Crane, Ital. Pop. Tales, notes, 372-5). One of the earliest allusions to the jingle is in Don Quixote, pt. I, c. xvi.: "Y asi como suele decirse el gato al rato, el rato á la cuerda, la cuerda al palo, daba el arriero á Sancho, Sancho á la moza, la moza á él, el ventero á la moza." As I have pointed out, it is used to this day by Bengali women at the end of each folk-tale they recite (L. B. Day, Folk-Tales of Bengal, Pref.).

Remarks.—Two ingenious suggestions have been made as to the origin of this curious jingle, both connecting it with religious ceremonies: (1) Something very similar occurs in Chaldaic at the end of the Jewish Hagada, or domestic ritual for the Passover night. It has, however, been shown that this does not occur in early MSS. or editions, and was only added at the end to amuse the children after the service, and was therefore only a translation or adaptation of a current German form of the jingle; (2) M. Basset, in the Revue des Traditions poplaires, 1890, t. v. p. 549, has suggested that it is a survival of the old Greek custom at the sacrifice of the Bouphonia for the priest to contend that he had not slain the sacred beast, the axe declares that the handle did it, the handle transfers the guilt further, and so on. This is ingenious, but fails to give any reasonable account of the diffusion of the jingle in countries which have had no historic connection with classical Greece.

XI. GOLD TREE AND SILVER TREE.

Source.—Celtic Magazine, xiii. 213-8, Gaelic and English from Mr. Kenneth Macleod.

Parallels.—Mr. Macleod heard another version in which "Gold Tree" (anonymous in this variant) is bewitched to kill her father's horse, dog, and cock. Abroad it is the Grimm's Schneewitchen (No. 53), for the Continental variants of which see Köhler on Gonzenbach, Sicil. Mârchen, Nos. 2-4, Grimm's notes on 53, and Crane, Ital. Pop. Tales, 331. No other version is known in the British Isles.

Remarks.—It is unlikely, I should say impossible, that this tale, with the incident of the dormant heroine, should have arisen independently in the Highlands: it is most likely an importation from abroad. Yet in it occurs a most "primitive" incident, the bigamous household of the hero: this is glossed over in Mr. Macleod's other variant. On the "survival" method of investigation this would possibly be used as evidence for polygamy in the Highlands. Yet if, as is probable, the story came from abroad, this trait may have come with it, and only implies polygamy in the original home of the tale.

Xll. KING O'TOOLE AND HIS GOOSE.

Source.—S. Lover's Stories and Legends of the Irish Peasantry.

Remarks.—This is really a moral apologue on the benefits of keeping your word. Yet it is told with such humour and vigour, that the moral glides insensibly into the heart.

XIII. THE WOOING OF OLWEN.

Source.—The Mabinogi of Kulhwych and Olwen from the translation of Lady Guest, abridged.

Parallels.—Prof. Rhys, Hibbert Lectures, p. 486, considers that our tale is paralleled by Cuchulain's "Wooing of Emer," a translation of which by Prof. K. Meyer appeared in the Archæological Review, vol.i. I fail to see much analogy. On the other hand in his Arthurian Legend, p. 41, he rightly compares the tasks set by Yspythadon to those set to Jason. They are indeed of the familiar type of the Bride Wager (on which see Grimm-Hunt, i. 399). The incident of the three animals, old, older, and oldest, has a remarkable resemblance to the Tettira Jataka (ed. Fausböll, No. 37, transl. Rhys Davids, i. p. 310 seq.) in which the partridge, monkey, and elephant dispute as to their relative age, and the partridge turns out to have voided the seed of the Banyan-tree under which they were sheltered, whereas the elephant only knew it when a mere bush, and the monkey had nibbled the topmost shoots. This apologue got to England at the end of the twelfth century as the sixty-ninth fable, "Wolf, Fox, and Dove," of a rhymed prose collection of "Fox Fables" (Mishle Shu'alim), of an Oxford Jew, Berachyah Nakdan, known in the Records as "Benedict le Puncteur" (see my Fables of Æsop, i. p. 170). Similar incidents occur in "Jack and his Snuff-box" in my English Fairy Tales, and in Dr. Hyde's "Well of D'Yerree-in-Dowan." The skilled companions of Kulhwych are common in European folk-tales (Cf. Cosquin, i. 123-5), and especially among the Celts (see Mr. Nutt's note in Maclnnes' Tales, 445-8), among whom they occur very early, but not so early as Lynceus and the other skilled comrades of the Argonauts.

Remarks.—The hunting of the boar Trwyth can be traced back in Welsh tradition at least as early as the ninth century. For it is referred to in the following passage of Nennius' Historia Britonum ed. Stevenson, p. 60, "Est aliud miraculum in regione quæ dicitur Buelt [Builth, co. Brecon] Est ibi cumulus lapidum et unus lapis super-positus super congestum cum vestigia canis in eo. Quando venatus est porcum Troynt [var. lec. Troit] impressit Cabal, qui erat canis Arthuri militis, vestigium in lapide et Arthur postea congregavit congestum lapidum sub lapide in quo erat vestigium canis sui et vocatur Carn Cabal." Curiously enough there is still a mountain called Carn Cabal in the district of Builth, south of Rhayader Gwy in Breconshire. Still more curiously a friend of Lady Guest's found on this a cairn with a stone two feet long by one foot wide in which there was an indentation 4 in. x 3 in. x 2 in. which could easily have been mistaken for a pawprint of a dog, as may be seen from the engraving given of it (Mabinogion, ed. 1874, p. 269).

The stone and the legend are thus at least one thousand years old. "There stands the stone to tell if I lie." According to Prof. Rhys (Hibbert Lect. 486-97) the whole story is a mythological one, Kulhwych's mother being the dawn, the clover bossoms that grow under Olwen's feet being comparable to the roses that sprung up where Aphrodite had trod, and Yspyddadon being the incarnation of the sacred hawthorn. Mabon, again (l.c. pp. 21, 28-9), is the Apollo Maponus discovered in Latin inscriptions at Ainstable in Cumberland and elsewhere (Hübner, Corp. Insc. Lat. Brit. Nos. 218, 332, 1345). Granting all this, there is nothing to show any mythological significance in the tale, though there may have been in the names of the dramatis personæ. I observe from the proceedings of the recent Eisteddfod that the bardic name of Mr. W. Abraham, M.P., is 'Mabon'. It scarcely follows that Mr. Abraham is in receipt of divine honours nowadays.

XIV. JACK AND HIS COMRADES

Source.—Kennedy's Legendary Fictions of the Irish Celts.

Parallels.—This is the fullest and most dramatic version I know of the Grimm's "Town Musicians of Bremen" (No. 27). I have given an English (American) version in my English Fairy Tales, No. 5, in the notes to which would be found references to other versions known in the British Isles (e.g., Campbell, No. 11) and abroad. Cf. remarks on No. vi.

XV. SHEE AN GANNON AND GRUAGACH GAIRE.

Source.—Curtin, Myths and Folk-Lore of Ireland, p. 114 seq. I have shortened the earlier part of my tale, and introduced into the latter a few touches from Campbell's story of "Fionn's Enchantment," in Revue Celtique, t. i., 193 seq.

Parallels.—The early part is similar to the beginning of "The Sea-Maiden" (No. xvii., which see). The latter part is practically the same as the story of "Fionn's Enchantment," just referred to. It also occurs in MacInnes' Tales, No. iii., "The King of Albainn" (see Mr. Nutt's notes, 454). The head-crowned spikes are Celtic, cf. Mr. Nutt's notes (Maclnnes' Tales, 453).

Remarks.—Here again we meet the question whether the folk-tale precedes the hero-tale about Finn or was derived from it, and again the probability seems that our story has the priority as a folk-tale, and was afterwards applied to the national hero, Finn. This is confirmed by the fact that a thirteenth century French romance, Conte du Graal has much the same incidents, and was probably derived from a similar folk-tale of the Celts. Indeed, Mr. Nutt is inclined to think that the original form of our story (which contains a mysterious healing vessel) is the germ out of which the legend of the Holy Grail was evolved (see his Studies in the Holy Grail, p. 202 seq.).

XVI. THE STORY TELLER AT FAULT.

Source.—Griffin's Tales from a Jury-Room, combined with Campbell, No. xvii. c, "The Slim Swarthy Champion."

Parallels.—Campbell gives another variant, l.c. i. 318. Dr. Hyde has an Irish version of Campbell's tale written down in 1762, from which he gives the incident of the air-ladder (which I have had to euphemise in my version) in his Beside the Fireside, p. 191, and other passages in his Preface. The most remarkable parallel to this incident, however, is afforded by the feats of Indian jugglers reported briefly by Marco Polo, and illustrated with his usual wealth of learning by the late Sir Henry Yule, in his edition, vol. i. p. 308 seq. The accompanying illustration (reduced from Yule) will tell its own tale: it is taken from the Dutch account of the travels of an English sailor, E. Melton, Zeldzaame Reizen, 1702, p. 468. It tells the tale in five acts, all included in one sketch. Another instance quoted by Yule is still more parallel, so to speak. The twenty-third trick performed by some conjurors before the Emperor Jahangueir (Memoirs, p. 102) is thus described: "They produced a chain of 50 cubits in length, and in my presence threw one end of it towards the sky, where it remained as if fastened to something in the air. A dog was then brought forward, and being placed at the lower end of the chain, immediately ran up, and, reaching the other end, immediately disappeared in the air. In the same manner a hog, a panther, a lion, and a tiger were successively sent up the chain." It has been suggested that the conjurors hypnotise the spectators, and make them believe they see these things. This is practically the suggestion of a wise Mohammedan, who is quoted by Yule as saying, "Wallah!  'tis my opinion there has been neither going up nor coming down; 'tis all hocus-pocus," hocus-pocus being presumably the Mohammedan term for hypnotism.

Remarks—Dr. Hyde (l.c. Pref. xxix.) thinks our tale cannot be older than 1362, because of a reference to one O'Connor Sligo which occurs in all its variants; it is, however, omitted in our somewhat abridged version. Mr. Nutt (ap. Campbell, The Fians, Introd. xix.) thinks that this does not prevent a still earlier version having existed. I should have thought that the existence of so distinctly Eastern a trick in the tale, and the fact that it is a framework story (another Eastern characteristic) would imply that it is a rather late importation, with local allusions superadded (cf. notes on "Conal Yellowclaw," No. v).

The passages in verse from pp. 137, 139, and the description of the Beggarman, pp. 136, 140, are instances of a curious characteristic of Gaelic folk-tales called "runs." Collections of conventional epithets are used over and over again to describe the same incident, the beaching of a boat, sea-faring, travelling and the like, and are inserted in different tales. These "runs" are often similar in both the Irish and the Scotch form of the same tale or of the same incident. The volumes of Waifs and Strays contain numerous examples of these "runs," which have been indexed in each volume. These "runs" are another confirmation of my view that the original form of the folk-tale was that of the Cante-fable (see note on "Connla" and on "Childe Rowland' in English Fairy Tales).

XVII. SEA-MAIDEN.

Source.— Campbell, Pop. Tales, No.4. I have omitted the births of the animal comrades and transported the carlin to the middle of the tale. Mr. Batten has considerately idealised the Sea-Maiden in his frontispiece. When she restores the husband to the wife in one of the variants, she brings him out of her mouth! "So the sea-maiden put up his head (Who do you mean? Out of her mouth to be sure. She had swallowed him)."

Parallels.—The early part of the story occurs in No. xv., "Shee an Gannon," and the last part in No. xix., "Fair, Brown, and Trembling" (both from Curtin), Campbell's No. 1. "The Young King" is much like it; also Maclnnes' No. iv., "Herding of Cruachan" and No. viii., "Lod the Farmer's Son." The third of Mr. Britten's Irish folk-tales in the Folk-Lore Journal is a Sea-Maiden story. The story is obviously a favourite one among the Celts. Yet its main incidents occur with frequency in Continental folk-tales. Prof. Köhler has collected a number in his notes on Campbell's Tales in Orient and Occident, Bnd. ii 115-8. The trial of the sword occurs in the saga of Sigurd, yet it is also frequent in Celtic saga and folk-tales (see Mr. Nutt's note, MacInnes' Tales, 473, and add. Curtin, 320). The hideous carlin and her three giant sons is also common form in Celtic. The external soul of the Sea-Maiden carried about in an egg, in a trout, in a hoodie, in a hind, is a remarkable instance of a peculiarly savage conception which has been studied by Major Temple, Wide-awake Stories, 404-5; by Mr. E. Clodd, in the "Philosophy of Punchkin," in Folk-Lore Journal, vol ii., and by Mr. Frazer in his Golden Bough, vol. ii.

Remarks.—As both Prof. Rhys (Hibbert Lect., 464) and Mr. Nutt (Maclnnes' Tales, 477) have pointed out, practically the same story (that of Perseus and Andromeda) is told of the Ultonian hero, Cuchulain, in the Wooing of Emer, a tale which occurs in the Book of Leinster, a MS. of the twelfth century, and was probably copied from one of the eighth. Unfortunately it is not complete, and the Sea-Maiden incident is only to be found in a British Museum MS. of about 1300. In this Cuchulain finds that the daughter of Ruad is to be given as a tribute to the Fomori, who, according to Prof. Rhys, Folk-Lore, ii. 293, have something of the nightmare about their etymology. Cuchulain fights three of them successively, has his wounds bound up by a strip of the maiden's garment, and then departs. Thereafter many boasted of having slain the Fomori, but the maiden believed them not till at last by a stratagem she recognises Cuchulain. I may add to this that in Mr. Curtin's Myths, 330, the threefold trial of the sword is told of Cuchulain. This would seem to trace our story back to the seventh or eighth century and certainly to the thirteenth. If so, it is likely enough that it spread from Ireland through Europe with the Irish missions (for the wide extent of which see map in Mrs. Bryant's Celtic Ireland). The very letters that have spread through all Europe except Russia, are to be traced to the script of these Irish monks: why not certain folk-tales? There is a further question whether the story was originally told of Cuchulain as a hero-tale and then became departicularised as a folk-tale, or was the process vice versâ. Certainly in the form in which it appears in the Tochmarc Emer it is not complete, so that here, as elsewhere, we seem to have an instance of a folk-tale applied to a well-known heroic name, and becoming a hero-tale or saga.

XVIII. LEGEND OF KNOCKMANY.

Source.—W. Carleton, Traits and Stories of the Irish Peasantry.

Parallels.—Kennedy's "Fion MacCuil and the Scotch Giant," Legend. Fict., 203-5.

Remarks.—Though the venerable names of Finn and Cucullin (Cuchulain) are attached to the heroes of this story, this is probably only to give an extrinsic interest to it. The two heroes could not have come together in any early form of their sagas, since Cuchulain's reputed date is of the first, Finn's of the third century A.D. (cf. however, MacDougall's Tales, notes, 272). Besides, the grotesque form of the legend is enough to remove it from the region of the hero-tale. On the other hand, there is a distinct reference to Finn's wisdom-tooth, which presaged the future to him (on this see Revue Celtique, v. 201, Joyce, Old Celt. Rom., 434-5, and MacDougall, l.c. 274). Cucullin's power-finger is another instance of the life-index or external soul, on which see remarks on Sea-Maiden. Mr. Nutt informs me that parodies of the Irish sagas occur as early as the sixteenth century, and the present tale may be regarded as a specimen.

XIX. FAIR, BROWN, AND TREMBLING.

Source.—Curtin, Myths, &c., of Ireland, 78 seq.

Parallels.—The latter half resembles the second part of the Sea-Maiden (No. xvii.), which see. The earlier portion is a Cinderella tale (on which see the late Mr. Ralston's article in Nineteenth Century, Nov. 1879, and Mr. Lang's treatment in his Perrault). Miss Roalfe Cox is about to publish for the Folk-Lore Society a whole volume of variants of the Cinderella group of stories, which are remarkably well represented in these isles, nearly a dozen different versions being known in England, Ireland, and Scotland.

XX. JACK AND HIS MASTER.

Source.—Kennedy, Fireside Stories of Ireland, 74-80, "Shan an Omadhan and his Master."

Parallels.—It occurs also in Campbell, No. xlv., "Mac a Rusgaich." It is a European droll, the wide occurrence of which—"the loss of temper bet" I should call it—is bibliographised by M. Cosquin, l.c. ii. 50 (cf. notes on No. vi.).

XXI. BETH GELLERT.

Source.—I have paraphrased the well-known poem of Hon. W. R. Spencer, "Beth Gêlert, or the Grave of the Greyhound," first printed privately as a broadsheet in 1800 when it was composed ("August 11, 1800, Dolymalynllyn" is the colophon). It was published in Spencer's Poems, 1811, pp. 78-86. These dates, it will be seen, are of importance. Spencer states in a note: "The story of this ballad is traditionary in a village at the foot of Snowdon where Llewellyn the Great had a house. The Greyhound named Gêlert was given him by his father-in-law, King John, in the year 1205, and the place to this day is called Beth-Gêlert, or the grave of Gêlert." As a matter of fact, no trace of the tradition in connection with Bedd Gellert can be found before Spencer's time. It is not mentioned in Leland's Itinerary, ed. Hearne, v. p 37 ("Beth Kellarth"), in Pennant's Tour (1770), ii. 176, or in Bingley's Tour in Wales (1800). Borrow in his Wild Wales, p. 146, gives the legend, but does not profess to derive it from local tradition.

Parallels.—The only parallel in Celtdom is that noticed by Croker in his third volume, the legend of Partholan who killed his wife's greyhound from jealousy: this is found sculptured in stone at Ap Brune, co. Limerick. As is well known, and has been elaborately discussed by Mr. Baring-Gould (Curious Myths of the Middle Ages, p. 134 seq.), and Mr. W. A. Clouston (Popular Tales and Fictions, ii. 166, seq.), the story of the man who rashly slew the dog (ichneumon, weasel, &c.) that had saved his babe from death, is one of those which have spread from East to West. It is indeed, as Mr. Clouston points out, still current in India, the land of its birth. There is little doubt that it is originally Buddhistic: the late Prof. S. Beal gave the earliest known version from the Chinese translation of the Vinaya Pitaka in the Academy of Nov. 4, 1882. The conception of an animal sacrificing itself for the sake of others is peculiarly Buddhistic; the "hare in the moon" is an apotheosis of such a piece of self-sacrifice on the part of Buddha (Sasa Jataka). There are two forms that have reached the West, the first being that of an animal saving men at the cost of its own life. I pointed out an early instance of this, quoted by a Rabbi of the second century, in my Fables of Æsop, i. 105. This concludes with a strangely close parallel to Gellert; "They raised a cairn over his grave, and the place is still called The Dog's Grave." The Culex attributed to Virgil seems to be another variant of this. The second form of the legend is always told as a moral apologue against precipitate action, and originally occurred in The Fables of Bidpai in its hundred and one forms, all founded on Buddhistic originals (cf. Benfey, Pantschatantra, Einleitung, §201). Thence, according to Benfey, it was inserted in the Book of Sindibad, another collection of Oriental Apologues framed on what may be called the Mrs. Potiphar formula. This came to Europe with the Crusades, and is known in its Western versions as the Seven Sages of Rome. The Gellert story occurs in all the Oriental and Occidental versions; e.g., it is the First Master's story in Wynkyn de Worde's (ed. G. L. Gomme, for the Villon Society.) From the Seven Sages it was taken into the particular branch of the Gesta Romanorum current in England and known as the English Gesta, where it occurs as c. xxxii., "Story of Folliculus." We have thus traced it to England whence it passed to Wales, where I have discovered it as the second apologue of "The Fables of Cattwg the Wise," in the Iolo MS. published by the Welsh MS. Society, p. 561, "The man who killed his Greyhound." (These Fables, Mr. Nutt informs me, are a pseudonymous production probably of the sixteenth century.) This concludes the literary route of the Legend of Gellert from India to Wales: Buddhistic Vinaya Pitaka—Fables of Bidpai;—Oriental Sindibad;—Occidental Seven Sages of Rome;—"English" (Latin), Gesta Romanorum;—Welsh, Fables of Cattwg.

Remarks.—We have still to connect the legend with Llewelyn and with Bedd Gelert. But first it may be desirable to point out why it is necessary to assume that the legend is a legend and not a fact. The saving of an infant's life by a dog, and the mistaken slaughter of the dog, are not such an improbable combination as to make it impossible that the same event occurred in many places. But what is impossible, in my opinion, is that such an event should have independently been used in different places as the typical instance of, and warning against, rash action. That the Gellert legend, before it was localised, was used as a moral apologue in Wales is shown by the fact that it occurs among the Fables of Cattwg, which are all of that character. It was also utilised as a proverb: "Yr wy'n edivaru cymmaint a'r Gŵr a laddodd ei Vilgi ("I repent as much as the man who slew his greyhound"). The fable indeed, from this point of view, seems greatly to have attracted the Welsh mind, perhaps as of especial value to a proverbially impetuous temperament. Croker (Fairy Legends of Ireland, vol. iii. p. 165) points out several places where the legend seems to have been localised in place-names—two places, called "Gwal y Vilast" ("Greyhound's Couch"), in Carmarthen and Glamorganshire; "Llech y Asp" ("Dog's Stone"), in Cardigan, and another place named in Welsh "Spring of the Greyhound's Stone." Mr. Baring-Gould mentions that the legend is told of an ordinary tombstone, with a knight and a greyhound in Abergavenny Church; while the Fable of Cattwg is told of a man in Abergarwan. So widespread and well known was the legend that it was in Richard III.'s time adopted as the national crest. In the Warwick Roll, at the Herald's Office, after giving separate crests for England, Scotland, and Ireland, that for Wales is given as figured in the margin, and blazoned "on a coronet in a cradle or, a greyhound argent for Walys" (see J. R. Planché, Twelve Designs for the Costume of Shakespeare's Richard III., 1830, frontispiece). If this Roll is authentic, the popularity of the legend is thrown back into the fifteenth century. It still remains to explain how and when this general legend of rash action was localised and specialised at Bedd Gelert: I believe I have discovered this. There certainly was a local legend about a dog named Gelert at that place; E. Jones, in the first edition of his Musical Relicks of the Welsh Bards, 1784, p. 40, gives the following englyn or epigram:

which he Englishes thus:

No reference was made in the first edition to the Gellert legend, but in the second edition of 1794, p. 75, a note was added telling the legend, "There is a general tradition in North Wales that a wolf had entered the house of Prince Llewellyn. Soon after the Prince returned home, and, going into the nursery, he met his dog Kill-hart, all bloody and wagging his tail at him; Prince Llewellyn, on entering the room found the cradle where his child lay overturned, and the floor flowing with blood; imagining that the greyhound had killed the child, he immediately drew his sword and stabbed it; then, turning up the cradle, found under it the child alive, and the wolf dead. This so grieved the Prince, that he erected a tomb over his faithful dog's grave; where afterwards the parish church was built and goes by that name—Bedd Cilhart, or the grave of Kill-hart, in Carnarvonshire. From this incident is elicited a very common Welsh proverb [that given above which occurs also in 'The Fables of Cattwg;' it will be observed that it is quite indefinite.]" "Prince Llewellyn ab Jorwerth married Joan, [natural] daughter of King John, by Agatha, daughter of Robert Ferrers, Earl of Derby; and the dog was a present to the prince from his father-in-law about the year 1205." It was clearly from this note that the Hon. Mr. Spencer got his account; oral tradition does not indulge in dates Anno Domini. The application of the general legend of "the man who slew his greyhound" to the dog Cylart, was due to the learning of E. Jones, author of the Musical Relicks. I am convinced of this, for by a lucky chance I am enabled to give the real legend about Cylart, which is thus given in Carlisle's Topographical Dictionary of Wales, s.v., "Bedd Celert," published in 1811, the date of publication of Mr. Spencer's Poems. "Its name, according to tradition, implies The Grave of Celert, a Greyhound which belonged to Llywellyn, the last Prince of Wales: and a large Rock is still pointed out as the monument of this celebrated Dog, being on the spot where it was found dead, together with the stag which it had pursued from Carnarvon," which is thirteen miles distant. The cairn was thus a monument of a "record" run of a greyhound: the englyn quoted by Jones is suitable enough for this, while quite inadequate to record the later legendary exploits of Gêlert. Jones found an englyn devoted to an exploit of a dog named Cylart, and chose to interpret it in his second edition, 1794, as the exploit of a greyhound with which all the world (in Wales) were acquainted. Mr. Spencer took the legend from Jones (the reference to the date 1205 proves that), enshrined it in his somewhat banal verses, which were lucky enough to be copied into several reading-books, and thus became known to all English-speaking folk.

It remains only to explain why Jones connected the legend with Llewelyn. Llewelyn had historical connection with Bedd Gellert, which was the seat of an Augustinian abbey, one of the oldest in Wales. An inspeximus of Edward I. given in Dugdale, Monast. Angl., ed. pr. ii. 100a, quotes as the earliest charter of the abbey "Cartam Lewelin, magni." The name of the abbey was "Beth Kellarth"; the name is thus given by Leland, l.c., and as late as 1794 an engraving at the British Museum is entitled "Beth Kelert," while Carlisle gives it as "Beth Celert." The place was thus named after the abbey, not after the cairn or rock. This is confirmed by the fact of which Prof. Rhys had informed me, that the collocation of letters rt is un-Welsh. Under these circumstances it is not impossible, I think, that the earlier legend of the marvellous run of "Cylart" from Carnarvon was due to the etymologising fancy of some English-speaking Welshman who interpreted the name as Killhart, so that the simpler legend would be only a folk-etymology.

But whether Kellarth, Kelert, Cylart, Gêlert or Gellert ever existed and run a hart from Carnarvon to Bedd Gellert or no, there can be little doubt after the preceding that he was not the original hero of the fable of "the man that slew his greyhound," which came to Wales from Buddhistic India through channels which are perfectly traceable. It was Edward Jones who first raised him to that proud position, and William Spencer who securely installed him there, probably for all time. The legend is now firmly established at Bedd Gellert. There is said to be an ancient air, "Bedd Gelert," "as sung by the Ancient Britons"; it is given in a pamphlet published at Carnarvon in the "fifties," entitled Gellert's Grave; or, Llewellyn's Rashness: a Ballad, by the Hon. W. R. Spencer, to which is added that ancient Welsh air, "Bedd Gelert," as sung by the Ancient Britons. The air is from R. Roberts' "Collection of Welsh Airs," but what connection it has with the legend I have been unable to ascertain. This is probably another case of adapting one tradition to another. It is almost impossible to distinguish palœozoic and cainozoic strata in oral tradition. According to Murray's Guide to N. Wales, p. 125, the only authority for the cairn now shown is that of the landlord of the Goat Inn, "who felt compelled by the cravings of tourists to invent a grave." Some old men at Bedd Gellert, Prof. Rhys informs me, are ready to testify that they saw the cairn laid. They might almost have been present at the birth of the legend, which, if my affiliation of it is correct, is not yet quite 100 years old.

XXII. STORY OF IVAN.

Source.—Lluyd, Archæologia Britannia, 1707, the first comparative Celtic grammar and the finest piece of work in comparative philology hitherto done in England, contains this tale as a specimen of Cornish then still spoken in Cornwall. I have used the English version contained in Blackwood's Magazine as long ago as May 1818. I have taken the third counsel from the Irish version, as the original is not suited virginibus puerisque, though harmless enough in itself.

Parallels.—Lover has a tale, The Three Advices. It occurs also in modern Cornwall ap. Hunt, Drolls of West of England, 344, "The Tinner of Chyamor." Borrow, Wild Wales, 41, has a reference which seems to imply that the story had crystallised into a Welsh proverb. Curiously enough, it forms the chief episode of the so-called "Irish Odyssey" ("Merugud Uilix maicc Leirtis"—"Wandering of Ulysses M'Laertes"). It was derived, in all probability, from the Gesta Romanorum, c. 103, where two of the three pieces of advice are "Avoid a byeway," "Beware of a house where the housewife is younger than her husband." It is likely enough that this chapter, like others of the Gesta, came from the East, for it is found in some versions of "The Forty Viziers," and in the Turkish Tales (see Oesterley's parallels and Gesta, ed. Swan and Hooper, note 9).

XXIII. ANDREW COFFEY.

Source.—From the late D. W. Logie, written down by Mr. Alfred Nutt.

Parallels.—Dr. Hyde's "Teig O'Kane and the Corpse," and Kennedy's "Cauth Morrisy," Legend. Fict., 158, are practically the same.

Remarks.—No collection of Celtic Folk-Tales would be representative that did not contain some specimen of the gruesome. The most effective ghoul story in existence is Lover's "Brown Man."

XXIV. BATTLE OF BIRDS.

Source.—Campbell (Pop. Tales, W. Highlands, No. ii.), with touches from the seventh variant and others, including the casket and key finish, from Curtin's "Son of the King of Erin" (Myths, &c., 32 seq.). I have also added a specimen of the humorous end pieces added by Gaelic story-tellers; on these tags see an interesting note in MacDougall's Tales, note on p. 112. I have found some difficulty in dealing with Campbell's excessive use of the second person singular, "If thou thouest him some two or three times, 'tis well," but beyond that it is wearisome. Practically, I have reserved thou for the speech of giants, who may be supposed to be somewhat old-fashioned. I fear, however, I have not been quite consistent, though the you's addressed to the apple-pips are grammatically correct as applied to the pair of lovers.

Parallels.—Besides the eight versions given or abstracted by Campbell and Mr. Curtin's, there is Carleton's "Three Tasks," Dr. Hyde's "Son of Branduf" (MS.); there is the First Tale of Maclnnes (where see Mr. Nutt's elaborate notes, 431-43), two in the Celtic Magazine, vol. xii., "Grey Norris from Warland" (Folk-Lore Journ. i. 316), and Mr. Lang's Morayshire Tale, "Nicht Nought Nothing" (see Eng. Fairy Tales, No. vii.), no less than sixteen variants found among the Celts. It must have occurred early among them. Mr. Nutt found the feather-thatch incident in the Agallamh na Senoraib ("Discourse of Elders"), which is at least as old as the fifteenth century. Yet the story is to be found throughout the Indo-European world, as is shown by Prof. Köhler's elaborate list of parallels attached to Mr. Lang's variant in Revue Celtique, iii. 374; and Mr. Lang, in his Custom and Myth ("A far travelled Tale"), has given a number of parallels from savage sources. And strangest of all, the story is practically the same as the classical myth of Jason and Medea.

Remarks.—Mr. Nutt, in his discussion of the tale (Maclnnes, Tales 441), makes the interesting suggestion that the obstacles to pursuit, the forest, the mountain, and the river, exactly represent the boundary of the old Teutonic Hades, so that the story was originally one of the Descent to Hell. Altogether it seems likely that it is one of the oldest folk-tales in existence, and belonged to the story-store of the original Aryans, whoever they were, was passed by them with their language on to the Hellenes and perhaps to the Indians, was developed in its modern form in Scandinavia (where its best representative "The Master Maid" of Asbjörnsen is still found), was passed by them to the Celts and possibly was transmitted by these latter to other parts of Europe, perhaps by early Irish monks (see notes on "Sea-Maiden"). The spread in the Buddhistic world, and thence to the South Seas and Madagascar, would be secondary from India. I hope to have another occasion for dealing with this most interesting of all folk-tales in the detail it deserves.

XXV. BREWERY OF EGGSHELLS.

Source.—From the Cambrian Quarterly Magazine, 1830, vol. ii. p. 86; it is stated to be literally translated from the Welsh.

Parallels.—Another variant from Glamorganshire is given in Y Cymmrodor, vi. 209. Croker has the story under the title I have given the Welsh one in his Fairy Legends, 41. Mr. Hartland, in his Science of Fairy Tales, 113-6, gives the European parallels.

XXVI. LAD WITH THE GOAT SKIN.

Source.—Kennedy, Legendary Fictions, pp. 23-31. The Adventures of "Gilla na Chreck an Gour'."

Parallels.—"The Lad with the Skin Coverings" is a popular Celtic figure, cf. MacDougall's Third Tale, MacInnes' Second, and a reference in Campbell, iii. 147. According to Mr. Nutt (Holy Grail, 134), he is the original of Parzival. But the adventures in these tales are not the "cure by laughing" incident which forms the centre of our tale, and is Indo-European in extent (cf. references in English Fairy Tales, notes to No. xxvii.). "The smith who made hell too hot for him is Sisyphus," says Mr. Lang (Introd. to Grimm, p. xiii.); in Ireland he is Billy Dawson (Carleton, Three Wishes). In the Finn-Saga, Conan harries hell, as readers of Waverley may remember " 'Claw for claw, and devil take the shortest nails,' as Conan said to the Devil" (cf. Campbell, The Fians, 73, and notes, 283). Red-haired men in Ireland and elsewhere are always rogues (see Mr. Nutt's references, Maclnnes' Tales, 477; to which add the case in "Lough Neagh," Yeats, Irish Folk-Tales, p. 210).