ii. 7, when he says, anile sane et plenum superstitionis fati nomen

ipsum. On the other hand, it rises to the grim dignity of a name for the dark, inexorable power which the whole universe is conceived to obey, a power before which the great and resplendent Zeus of the Aryan race is a mere puppet.

Perhaps I have dwelt only too long on the policy of 'give and take' which ought to obtain between mythology and glottology. Unfortunately, one can add without fear of contradiction, that, even when that policy is carried out to the utmost, both sciences will still have difficulties more than enough. In the case of mythology these difficulties spring chiefly from two distinct sources, from the blending of history with myth, and from the mixing of one race with another. Let us now consider the latter: the difficulties from this source are many and great, but every fresh acquisition of knowledge tending to make our ideas of ethnology more accurate, gives us a better leverage for placing the myths of mixed peoples in their proper places as regards the races composing those peoples. Still, we have far fewer propositions to lay down than questions to ask: thus to go no further afield than the well-known stories attaching to the name of Heracles, how many of them are Aryan, how many Semitic, and how many Aryan and Semitic at one and the same time? That is the sort of question which besets the student of Celtic mythology at every step; for the Celtic nations of the present day are the mixed descendants of Aryan invaders and the native populations which those Aryan invaders found in possession. So the question thrusts itself on the student, to which of these races a particular myth, rite, or custom is to be regarded as originally belonging. Take, for instance, Brân's colossal figure, to which attention has already been called, pp. 552-3 above. Brân was too large to enter a house or go on board a ship: is he to be regarded as the outcome of Celtic imagination, or of that of a people that preceded the Celts in Celtic lands? The comparison with the Gaulish Tricephal would seem to point in the direction of the southern seaboard of the Baltic (p. 553): what then?

The same kind of question arises in reference to the Irish hero Cúchulainn: take, for instance, the stock description of Cúchulainn in a rage. Thus when angered he underwent strange distortions: the calves of his legs came round to where his shins should have been; his mouth enlarged itself so that it showed his liver and lungs swinging in his throat; one of his eyes became as small as a needle's, or else it sank back into his head further than a crane could have reached, while the other protruded itself to a corresponding length; every hair on his body became as sharp as a thorn, and held on its point a drop of blood or a spark of fire. It would be dangerous then to stop him from fighting, and even when he had fought enough, he required for his cooling to be plunged into three baths of cold water; the first into which he went would instantly boil over, the second would be too hot for anybody else to bear, and the third only would be of congenial warmth. I do not ask whether that strange picture betrays a touch of the solar brush, but I should be very glad to know whether it can be regarded as an Aryan creation or not.

It is much the same with matters other than mythological: take, for instance, the bedlamite custom of the couvade [246], which is presented to us in Irish literature in the singular form of a cess, 'suffering or indisposition,' simultaneously attacking the braves of ancient Ulster. We are briefly informed in the Book of the Dun Cow, fo. 60a, that the women and boys of Ulster were free from it. So was any Ultonian, we are told, who happened to be outside the boundaries of his country, and so were Cúchulainn and his father, even when in Ulster. Any one who was rash enough to attack an Ultonian warrior during this his period of helplessness could not, it is further stated, expect to live afterwards either prosperously or long. The question for us, however, is this: was the couvade introduced by the Aryan invaders of Ireland, or are we rather to trace it to an earlier race? I should be, I must confess, inclined to the latter view, especially as the couvade was known among the Iberians of old, and among the ancient Corsicans [247]. It may, of course, have been both Aryan and Iberian, but it will all the same serve as a specimen of the sort of question which one has to try to answer.

Another instance, the race origin of which one would like to ascertain, offers itself in the curious belief, that, when a child is born, it is one of the ancestors of the family come back to live again. Traces of this occur in Irish literature, namely, in one of the stories about Cúchulainn. There we read to the following effect:--The Ultonians took counsel on account of Cúchulainn, because their wives and girls loved him greatly; for Cúchulainn had no consort at that time. This was their counsel, namely, that they should seek for Cúchulainn a consort pleasing to him to woo. For it was evident to them that a man who has the consort of his companionship with him would be so much the less likely to attempt the ruin of their girls and to receive the affection of their wives. Then, moreover, they were anxious and afraid lest the death of Cúchulainn should take place early, so they were desirous for that reason to give him a wife in order that he might leave an heir; for they knew that it was from himself that his rebirth (athgein) would be. That is what one reads in the eleventh-century copy of the ancient manuscript of the Book of the Dun Cow, fo. 121b; and this atavistic belief, which was touched upon in connexion with the transformations discussed in the last chapter, I need scarcely say, is well known elsewhere to the anthropologist, as one will find on consulting the opening pages of Dr. Tylor's second volume on Primitive Culture. He there mentions the idea as familiar to American Indians, to various African peoples, to the Maoris and the aborigines of Australia, to Cheremiss Tartars and Lapps. Among such nations the words of Don Diègue to his victorious son, the Cid, could hardly fail to be construed in a sort of literal sense when he exclaims:--

............ ton illustre audace Fait bien revivre en toi les héros de ma race.

Let us return to Cúchulainn, and note the statement, that he and his father, Sualdaim, were exempt from the couvade, which marks them out as not of the same race as the Ultonians, that is to say, as the Fír Ulaid, or 'True Ultonians'--presumably ancient inhabitants of Ulster. Furthermore, we have an indication whence his family had come, for Cúchulainn's first name was Setanta Beg, 'the Little Setantian,' which points to the coast of what is now Lancashire, as already indicated at p. 385 above. Another thing which marks Cúchulainn as of a different racial origin from the other Ultonians is the belief of the latter, that his rebirth must be from himself. The meaning of this remarkable statement is that there were two social systems face to face in Ulster at the time represented by the Cúchulainn story, and that one of them recognized fatherhood, while the other did not. Thus for Cúchulainn's rebirth to be from himself, he must be the father of a child from whom should descend a man who would be a rebirth or avatar of Cúchulainn. The other system implied was one which reckoned descent by birth alone [248]; and the Cúchulainn story gives one the impression that it contemplated this system as the predominant one, while the Cúchulainn family, with its reckoning of fatherhood, comes in as an exception. At all events, that is how I now understand a passage, the full significance of which had till recently escaped me.

Allusion has already been made to the story of Cúchulainn being himself a rebirth, namely, of Lug, and the story deserves still further consideration in its bearing on the question of race, to which the reader's attention has been called. It is needless, however, to say that there are extant fragments of more stories than one as to Cúchulainn's origin. Sometimes, as in the Book of Leinster, fo. 119a, he is called gein Loga, or Lug's offspring, and in the epic tale of the Táin Bó Cuailnge, Lug as his father comes from the Síd or Faery to take Cúchulainn's place in the field, when the latter was worn out with sleeplessness and toil. Lug sings over him éli Loga, or 'Lug's enchantment,' and Cúchulainn gets the requisite rest and sleep [249]: this we read in the Book of the Dun Cow, fo. 78a. In another version of the story, Cúchulainn is an incarnation of Lug: the narrative relates how a foster-son was accepted by Dechtere, sister to Conchobar MacNessa, king of Ulster. But her foster-son died young, to the great grief of Dechtere; and her lamentations for him on the day of his funeral having made her thirsty, she inadvertently swallowed with her drink a diminutive creature which sprang into her mouth. That night she had a dream, in which a man informed her that she was pregnant, that it was he who was in her womb, that he had been her foster-son, and that he was Lug; also that when his birth should take place, the name was to be Setanta. After an incident which I can only regard as a clumsy attempt to combine the more primitive legend with the story which makes him son of Sualdaim, she gives birth to the boy, and he is duly called Setanta [250]: that was Cúchulainn's first name. Now compare this with what Dr. Tylor mentions in the case of the Lapps, namely, that 'the future mother was told in a dream what name to give her child, this message being usually given her by the very spirit of the deceased ancestor, who was about to be incarnate in her [251].' If the mother got no such intimation in a dream, the relatives of the child had to have recourse to magic and the aid of the wise man, to discover the name to be given to the child.

Here let it suffice to say, that the similarity is so close between the Irish and the Lapp idea, and so unlike anything known to have been Aryan, that it is well worth bearing in mind. The belief in rebirth generally seems to fit as a part of the larger belief in the transmigration of souls which is associated with the teachings of the ancient druids, a class of shamans or medicine-men who were probably, as already hinted, not of Celtic or Aryan origin; and probably the beliefs here in question were those of some non-Aryan people of these islands, rather than of any Aryans who settled in them. This view need hardly be regarded as incompatible with the fact, that Lug's name, genitive Loga, would seem to have meant light, and that Lug was a sun-god, very possibly a Celtic sun-god: or more correctly speaking, that there was a series of Lugs, so to say, or sun-gods, called in ancient Spain, Switzerland, and on the banks of the Rhine, Lugoves [252]. For one is sorely tempted to treat this much as a rescue from the wreckage of the solar myth theory, as against those who, having regard mainly to Lug's professional skill and craft as described in Irish story, make of him a kind of Hermes or Mercury. In other words, we have either to regard a Celtic Lug as having become the centre of certain non-Celtic legends, or else to suppose neither Lug nor his name to be of Aryan origin at all. It is hard to say which is the sounder view to take.

The next question which I wish to suggest is as to the ethnology of the fairies; but before coming to that, one has to ask how the fairies have been evolved. The idea of fairies, such as Welshmen have been familiar with from their childhood, clearly involves elements of two distinct origins. Some of those elements come undoubtedly from the workshop of the imagination, as, for example, the stock notion that their food and drink are brought to the fairies by the mere force of wishing, and without the ministration of servants; or the notion, especially prevalent in Arfon, that the fairies dwell in a country beneath the lakes of Snowdon; not to mention the more general connexion of a certain class of fairies with the world of waters, as indicated in chapter vii. Add to this that the dead ancestor has also probably contributed to our bundle of notions about them; but that contains also an element of fact or something which may at any rate be conceived as historical. Under this head I should place the following articles of faith concerning them: the sallowness of their skins and the smallness of their stature, their dwelling underground, their dislike of iron, and the comparative poverty of their homes in the matter of useful articles of furniture, their deep-rooted objection to the green sward being broken up by the plough, the success of the fairy wife in attending to the domestic animals and to the dairy, the limited range generally of the fairies' ability to count; and lastly, one may perhaps mention their using a language of their own (p. 279), which would imply a time when the little people understood no other, and explain why they should be represented doing their marketing without uttering a syllable to anybody (p. 161).

The attribution of these and similar characteristics to the fairies can scarcely be all mere feats of fancy and imagination: rather do they seem to be the result of our ancestors projecting on an imaginary world a primitive civilization through which tradition represented their own race as having passed, or, more probably, a civilization in which they saw, or thought they saw, another race actually living. Let us recur for examples also to the two lake legends which have just been mentioned (p. 650): in both of them a distinction is drawn between the lake fairy's notion of bread and that of the men and women of the country. To the fairy the latter's bread appeared crimped or overbaked: possibly the backward civilization, to which she was supposed to belong, was content to support itself on some kind of unleavened bread, if not rather on a fare which included nothing deserving to be called bread at all. Witness Giraldus Cambrensis' story of Eliodorus, in which bread is conspicuous by its absence, the nearest approach to it being something of the consistency of porridge: see p. 270 above. Then take another order of ideas: the young man in both lake legends lives with his mother (pp. 3, 27): there is no father to advise or protect him: he is in this respect on a level with Undine, who is the protegee of her tiresome uncle, Kühleborn. Seemingly, he belongs to a primitive society where matriarchal ideas rule, and where paternity is not reckoned [253]. This we are at liberty at all events to suppose to have been the original, before the narrator had painted the mother a widow, and given the picture other touches of his later brush.

To speak, however, of paternity as merely not reckoned is by no means to go far enough; so here we have to return to take another look at the imaginary aspect of the fairies, to which a cursory allusion has just been made. The reader will possibly recall the sturdy smith of Ystrad Meurig, who would not reduce the notions which he had formed of the fairies when he was a child to conformity with those of a later generation around him. In any case, he will remember the smith's statement that the fairies were all women: see p. 245. The idea was already familiar to me as a Welshman, though I cannot recollect how I got it. But the smith's words brought to my mind at once the story of Condla Rúad or the Red, one of the fairy tales first recorded in Irish literature (p. 291). There the damsel who takes Condla away in her boat of glass to the realm of the Everliving sings the praises of that delectable country, and uses, among others, the following words, which occur in the Book of the Dun Cow, fo. 120:--

Ni fil cenel and nammá acht mná ocus ingena [254].

There is no race there but women and maidens alone.

Now what people could have come by the idea of a race of women only? Surely no people who considered that they themselves had fathers: it must have been some community so low in the scale of civilization as never to have had any notion whatsoever of paternity: it is their ignorance that would alone render possible the notion of a race all women. That this was a matter of belief in the past of many nations, is proved by the occurrence of widely known legends about virgin mothers [255]; not to mention that it has been lately established, that there are savages who to this day occupy the low place here indicated in the scale of civilization. Witness the evidence of Spencer and Gillen in their recently published work on The Native Tribes of Central Australia, and also what Frazer, author of The Golden Bough, says of a passage in point, in the former, as follows:--

'Thus, in the opinion of these savages, every conception is what we are wont to call an immaculate conception, being brought about by the entrance into the mother of a spirit apart from any contact with the other sex. Students of folklore have long been familiar with notions of this sort occurring in the stories of the birth of miraculous personages, but this is the first case on record of a tribe who believe in immaculate conception as the sole cause of the birth of every human being who comes into the world. A people so ignorant of the most elementary of natural processes may well rank at the very bottom of the savage scale [256].'

Nevertheless, it is to some population in that low position, in the remote prehistory of this country, that one is to trace the belief that the fairies were all women. It is to be regarded as a position distinctly lower than that of the Ultonians in the time of Cúchulainn; for the couvade seems to me to argue a notion of paternity--perhaps, in their case, as clear a notion of paternity as was possible for a community which was not quite out of the promiscuous stage of society.

The neo-Celtic nations of these islands consist, speaking roughly, of a mixture of the invading Celts with the earlier inhabitants whom the Celts found in possession. These two or more groups of peoples may have been in very different stages of civilization when they first came in contact with one another. They agreed doubtless in many things, and perhaps, among others, in cherishing an inherited reluctance to disclose their names, but the Celts as Aryans were never without the decimal system of counting. Like the French, the Celtic nations of the present day show a tendency, more or less marked, to go further and count by scores instead of by tens. But the Welsh are alone among them in having, in certain instances, gone back from counting by tens to counting by fives, which they do when they count between 10 and 20: for 16, 17, 18, and 19 are in Welsh 1 on 15, 2 on 15, 3 on 15, and 4 on 15 respectively; and similarly with 13 and 14 [257]. We have seen how the lake fairy reckoned by fives (pp. 8, 418) all the live stock she was to have as her dowry; and one otherwise notices that the fairies deal invariably in the simplest of numbers. Thus if you wish, for example, to find a person who has been led away by them, ten to one you have to go 'this day next year' to the spot where he disappeared. Except in the case of the alluring light of the full moon, it is out of the question to reckon months or weeks, though it is needless to say that to reckon the year correctly would have been in point of fact far more difficult; but nothing sounds simpler than 'this day next year.' In that simple arithmetic of the fairies, then, we seem to have a trace of a non-Aryan race, that is to say, probably of some early inhabitants of these islands.

Unfortunately, the language of those inhabitants has died out, so that we cannot appeal to its numerals directly; and the next best course to adopt is to take as a sort of substitute for their language that of possible kinsmen of a pre-Celtic race in this country. Now the students of ethnology, especially those devoted to the investigation of skulls and skins, tell us that we have among us, notably in Wales and Ireland, living representatives of a dark-haired, long-skulled race of the same description as one of the types which occur, as they allege, among the Basque populations of the Pyrenees. We turn accordingly to Basque, and what do we find? Why, that the first five numerals in that language are bat, bi, iru, lau, bost, all of which appear to be native; but when we come to the sixth numeral we have sei, which looks like an Aryan word borrowed from Latin, Gaulish, or some related tongue. The case is much the same with 'seven,' for that is in Basque zazpi, which is also probably an Aryan loan-word. Basque has native words, zortzi and bederatzi, for eight and nine, but they are longer than the first five, and appear to be of a later formation affecting, in common with sei and zazpi, the termination i. I submit, therefore, that here we have evidence of the former existence of a people in the West of Europe who at one time only counted as far as five. Some of the early peoples of the British Isles may have been on the same level, so that our notions about the fairies have probably been derived, to a greater or less extent, from ideas formed by the Celts concerning those non-Celtic, non-Aryan natives of whose country they took possession.

As regards my appeal to the authority of craniology, I have to confess that it is made with a certain amount of reservation, since the case is far less simple than it looks at first sight. Thus, in August, 1891, the Cambrian Archæological Association, including among them Professor Sayce, visited the south-west of Ireland. During our pleasant excursions in Kerry, the question of race was one of our constant topics; and Professor Sayce was reminded by what he saw in Ireland of his visit to North Africa, especially the hilly regions of the country inhabited by the Berbers. Among other things, he used to say that if a number of Berbers from the mountains were to be brought to an Irish village and clad as Irishmen, he felt positive that he should not be able to tell them from the Irishmen themselves, such as we saw on our rambles in Kerry. This struck me as all the more remarkable, since his reference was to fairly tall, blue-eyed men whose hair could not be called black. On the other hand, owing perhaps to ignorance and careless ways of looking at things around me, I am a little sceptical as to the swarthy long-skulls: they did not seem to meet us at every turn in Ireland; and as for Wales, which I know as well as most people do, I cannot in my ignorance of craniology say with any confidence that I have ever noticed vast numbers of that type. I should like, however, to see the heads of some of the singers whom I have noticed at our Eistedfodau at Cardiff, Aberdare, and Swansea, placed under the hands of an experienced skull-man. For I have long suspected that we cannot regard as of Aryan origin the vocal talent so general in Wales, and so conspicuous in our choirs of working people as to astonish all the great musicians who have visited our national festival. Beyond all doubt, race has not a little to do with the artistic feelings: a short-skull may be as unmusical, for example, as I am; but has anybody in this country ever known a narrow long-skull to be the reverse of unmusical? or has any one ever considered how few clergymen of the tall, fair-haired, blue-eyed type have been converted to the ritualistic and æsthetic movement in the Church of England?

As it seems to me that the bulk of the Welsh people would have to be described as short-skulls, it would be very gratifying to see those who are wont to refer freely to the dark-complexioned long-skulls of Wales catch a respectable number of specimens. I trust there are plenty to be found; and of course I do not care how they are taken, whether it be by an instantaneous process of photography or in the meshes of some anthropometric sportsman, like Dr. Beddoe. Let them be secured anyhow, so that one may rest assured that the type is still numerically safe, and be able to judge with one's own eyes how heads long and swarthy look on the shoulders of living Welshmen. We might then be in a position also to compare with them the prevalent description of fairy changelings; for when the fairies steal nice, blond babies, they usually place in their stead their own aged-looking brats with short legs, sallow skins, and squeaky voices. Unfortunately for me, all the adult changelings of whom I happen to have heard any account had died some years before I began to turn my attention to the population of Faery, with the exception, perhaps, of one whose name I obtained under the seal of secrecy. It was that of the wife of a farmer living near Nefyn, in West Carnarvonshire. It was whispered that she was a changeling, so I am inclined to regard her as no other than one of the representatives of the same aboriginal stock to which one might conjecture some of her neighbours also to belong; she ought to be an extreme specimen of the type. It is to be hoped that the photographer and his anthropometric brother have found her out in time and in good humour; but it is now many years since I heard of her.

To return again to the fairies, some of them are described as more comely and good-looking than the rest (pp. 83, 250), but the fairy women are always pictured as fascinating, though their offspring as changelings are as uniformly presented in the light of repulsive urchins; but whole groups of the fairy population are sometimes described as being as ugly of face as they were thievish in disposition--those, for instance, of Llanfabon, in Glamorganshire (p. 262). There is one district, however, which is an exception to the tenor of fairy physiognomy: it is that of the Pennant neighbourhood, in Carnarvonshire, together with the hills and valleys, roughly speaking, from Cwm Strallyn to Llwytmor and from Drws y Coed to Dolbenmaen. The fairies of that tract are said to have been taller than the others, and characterized by light or even flaxen hair, together with eyes of clear blue: see pp. 89, 93-7, 105-8. Nor is that all, for we are told that they would not let a person of dark complexion come near them (p. 96). The other fairies, when kidnapping, it is true, preferred the blond infants of other people to their own swarthy brats, which, perhaps, means that it was a policy of their people to recruit itself with men of the superior physique of the more powerful population around them. The supposed fairy ancestress of the people of the Pennant Valley bears, in the stories in point, such names as Penelope, Bella, Pelisha, and Sibi, while her descendants are still taunted with their descent--a quarrel which, within living memory, used to be fought out with fists at the fairs at Penmorfa and elsewhere. This seems to indicate a comparatively late settlement [258] in the district of a family or group of families from without, and an origin, therefore, somewhat similar to that of the Simychiaid and Cowperiaid (p. 67) of a more eastern portion of the same county, rather than anything deserving to be considered with the rest of the annals of Faery. Passing by this oasis, then, such snap-shot photographs as I have been able to take, so to speak, of fairyland cleared of the glamour resting on its landscape, seem to disclose to the eye a swarthy population of short stumpy men occupying the most inaccessible districts of our country. They appear to have cared more for soap than clothing [259], and they lived on milk taken once a day, when they could get it. They probably fished and hunted, and kept domestic animals, including, perhaps, the pig; but they depended largely on what they could steal at night or in misty weather. Their thieving, however, was not resented, as their visits were believed to bring luck and prosperity (p. 251). Their communities formed as it were islands, owing to the country round about them having been wrested from them by later comers of a more warlike disposition and provided with better weapons. But the existence of the scattered groups of the fairies was in no danger of coming to a violent end: they were safe in consequence of the superstitious beliefs of their stronger neighbours, who probably regarded them as formidable magicians, powerful, among other things, to cause or to cure disease as they pleased. Such, without venturing to refresh my memory by perusing what has been written about dwarf races in other parts of the world, are the impressions made on my mind in the course of analysing and sifting the folklore materials crowded into this volume. That applies, of course, in so far only as regards the fairies in their character of a real people as distinguished from them as creatures of the imagination. But, as I have no wish to earn the displeasure of my literary friends, let me hasten to say that I acknowledge the latter, the creatures of the imagination, to be the true fairies, the admiration of one's childhood and the despair of one's later years: the other folk--the aborigines whom I have been trying to depict--form only a sort of substratum, a kind of background to the fairy picture, which I should be the last man to wish to mar.

It is needless to say that we have no trace of any fairies approaching the minute dimensions of Shakespeare's Queen Mab; for, after all, our fairies are mostly represented as not extravagantly unlike other people in personal appearance--not so unlike, in fact, that other folk might not be mistaken for them now and then as late as the latter part of the fifteenth century. Witness the following passage from Sir John Wynne's History of the Gwydir Family, p. 74:--

'Haveing purchased this lease, he removed his dwelling to the castle of Dolwydelan, which at that time was in part thereof habitable, where one Howell ap Jevan ap Rys Gethin, in the beginning of Edward the Fourth his raigne, captaine of the countrey and an outlaw, had dwelt. Against this man David ap Jenkin rose, and contended with him for the sovreignety of the countrey; and being superiour to him, in the end he drew a draught for him, and took him in his bed at Penanmen with his concubine, performing by craft, what he could not by force, and brought him to Conway Castle. Thus, after many bickerings betweene Howell and David ap Jenkin, he being too weake, was faigne to flie the countrey, and to goe to Ireland, where he was a yeare or thereabouts. In the end he returned in the summer time, haveing himselfe, and all his followers clad in greene, who, being come into the countrey, he dispersed here and there among his friends, lurking by day, and walkeing in the night for feare of his adversaries; and such of the countrey as happened to have a sight of him and his followers, said they were the fairies, and soe ran away.'

But what has doubtless helped, above all other things, to perpetuate the belief in the existence of fairies may be said to be the popular association with them of the circles in the grass, commonly known in English as fairy rings. This phenomenon must have answered for ages the purpose for our ancestors, practically speaking, of ocular demonstration, as it still does no doubt in many a rustic neighbourhood.

The most common name for the fairies in Welsh is y Tylwyth Teg, 'the Fair or Beautiful Family'; but in South Cardiganshire we have found them called Plant Rhys Dwfn, 'the Children of Rhys the Deep' (pp. 151, 158), while in Gwent and Morgannwg they are more usually known as Bendith y Mamau, 'the Blessing of the Mothers' (p. 174). Our fourteenth century poet, D. ab Gwilym, uses the first-mentioned term, Tylwyth Teg, in poem xxxix, and our prose literature has a word corr, cor in the sense of a dwarf, and corres for a she dwarf. The old Cornish had also cor, which in Breton is written korr [260], with a feminine korrez, and among the other derivatives one finds korrik, 'a dwarf, a fairy, a wee little sorcerer,' and korrigez or korrigan, 'a she dwarf, a fairy woman, a diminutive sorceress.' The use of these words in Breton recalls the case of the cor, called Rhudlwm or else Eidilig, teaching his magic to Coll, son of Collfrewi: see pp. 326, 503, 505. Then we have uncanny dwarfs in the romances, such, for example, as the rude cor in the service of Edern ab Nud, as described in French in Chrétien's romance of Erec et Enide and in Welsh in that of Gereint vab Erbin, also the cor and corres who figure in the story of Peredur. The latter had belonged to that hero's father and mother till the break-up of the family, when the dwarfs went to Arthur's Court, where they lived a whole year without speaking to anybody. When, however, Peredur made his rustic appearance there, they hailed him loudly as the chief of warriors and the flower of knighthood, which brought on them the wrath of Cai, on whom they were eventually avenged by Peredur. In the case [261] of both Edern and Peredur we find the dwarfs loyally interested in the fortunes of their masters and their masters' friends. With them also the shape-shifting Menw, though not found placed in the same unfavourable light, is probably to be ranged, as one may gather from his name and his rôle of wizard scout for Arthur's men (p. 510). In the like attachment on the part of the fairies, which was at times liable to develop into devotedness of an embarrassing nature (p. 250), we seem to have one of the germs of the idea of a household fairy or banshee, as illustrated by the case of the ugly wee woman in the Pantannas legend (p. 188); and it seems natural to regard the interested voices in the Kenfig legend, and other stories of the same kind (p. 452), as instances of amalgamating the idea of a fairy with that of an ancestral person.

At all events, we have obtained something to put by the side of the instances already noticed of the fairy girl who gives, against her will at first, her services in the dairy of her captor (pp. 45, 87); of the other fairy who acts as a nurse for a family in the Pennant Valley, till she is asked to dress better (p. 109); and of Bwca'r Trwyn who works willingly and well, both at the house and in the field, till he has tricks played on him (pp. 593-6). To make this brief survey complete, one has to mention the fairies who used to help Eilian with her spinning (pp. 211-3), and not to omit those who were found to come to the rescue of a woman in despair and to assist her on the condition of getting her baby. The motive here is probably not to be confounded with that of the fairies who stealthily exchanged babies: the explanation seems in this case to be that the fairies, or some of the fairies, were once regarded as cannibals, which is countenanced by such a story as that of Canrig Bwt, 'Canrig the Stumpy.' At Llanberis the latter is said to have lived beneath the huge stone called y Gromlech, 'the Dolmen,' opposite Cwmglas and near the high-road to the Pass. When the man destined to dispatch her came, she was just finishing her dinner off a baby's flesh. There are traces of a similar story in another district, for a writer who published in the year 1802 uses the following words:--'There was lately near Cerrig y Drudion, in Merionethshire, a subterraneous room composed of large stones, which was called Carchar Cynric Rwth, i. e. "The Prison of Cynric Rwth," which has been taken notice of by travellers.' Cynric Rwth may be rendered 'Cynric the Greedy or Broad-mouthed.' A somewhat similar ogress is located by another story on the high ground at Bwlch y Rhiw Felen, on the way from Llangollen to Llandegla, and she is represented by the local tradition as contemporary with Arthur [262]. I am inclined to think the Cwmglas cromlech natural rather than artificial; but I am, however, struck by the fact that the fairies are not unfrequently located on or near ancient sites, such as seem to be Corwrion (pp. 57, 526), the margin of Llyn Irdyn (pp. 148, 563), Bryn y Pibion (pp. 212-4), Dinllaen (p. 227), Carn Bodüan (p. 227), on which there are, I am told, walls and hut foundations similar to those which I have recently seen on Carn Fadrun in the same district, Moedin camp (p. 245), and, perhaps, Ynys Geinon Rock and the immediate vicinity of Craig y Nos, neither of which, however, have I ever visited (p. 254). Local acquaintance with each fairy centre would very possibly enable one to produce a list that would be suggestive.

In passing one may point out that the uncanny dwarf of Celtic story would seem to have served, in one way or another, as a model for other dwarfs in the French romances and the literatures of other nations that came under the influence of those romances, such as that of the English. But the subject is too large to be dealt with here; so I return to the word cor, in order to recall to the reader's mind the allusion made, at p. 196, to a certain people called Coranneit or Coranyeit, pronounced in later Welsh Corániaid, 'Corannians.' They come in the Adventure of Llûd and Llevelys, and there they have ascribed to them one of the characteristics of consummate magicians, namely, the power of hearing any word that comes in contact with the wind; so it was, we are told, impossible to harm them. Llûd, however, was advised to circumvent them in the following manner:--he was to bruise certain insects in water and sprinkle the water on the Corannians and his own people indiscriminately, after calling them together under the pretence of making peace between them; for the sprinkling would do no harm to his own subjects, while it would kill the others. This unholy water proved effective, and the Corannians all perished. Now the magic power ascribed to them, and the method of disposing of them, combine to lend them a fabulous aspect, while their name, inseparable as it seems from cor, 'a dwarf,' warrants us in treating them as fairies, and in regarding their strange characteristics as induced on a real people. If we take this view, that Coraniaid was the name of a real people, we are at liberty to regard it as possible, that their name suggested to the Celts the word cor for a dwarf, rather than that cor has suggested the name of the Corannians. In either case, I may mention that Welsh writers have sometimes thought--and they are probably right--that we have a closely related word in the name of Ptolemy's Coritani or Coritavi. He represents the people so called as dwelling, roughly speaking, between the Trent and Norfolk, and possessed of the two towns of Lindum, 'Lincoln,' and Ratæ (p. 547), supposed to have been Leicester. There we should have accordingly to suppose the old race to have survived so long and in such numbers, that the Celtic lords of southern Britain called the people of that area by a name meaning dwarfs. There also they may be conjectured to have had quiet from invaders from the Continent, because of the inaccessible nature of the fens, and the lack of inviting harbours on the coast from the country of the Iceni up to the neighbourhood of the Humber. How far their territory extended inland from the fens and the sea one cannot say, but it possibly took in one-half of what is now Northamptonshire, with the place called Pytchley, from an older Pihtes Léa, meaning the Meadow of the Pict, or else of a man named Pict. In any case it included Croyland in the fens between Peterborough and the Wash. It was there, towards the end of the seventh century, that St. Guthlac built his cell on the side of an ancient mound or tumulus, and it was there he was assailed by demons who spoke Bryttisc or Brythonic, a language which the saint knew, as he had been an exile among Brythons. For this he had probably not to travel far; and it is remarkable that his father's cognomen or surname was Penwall, which we may regard as approximately the Brythonic for 'Wall's End.' That is to say, he was 'So-and-so of the Wall's End,' and had got to be known by the latter designation instead of his own nomen, which is not recorded, for the reason, possibly, that it was so Brythonic as not to admit of being readily reduced into an Anglian or Latin form. It is not quite certain that he belonged to the royal race of Mercia, whose genealogy, however, boasts such un-English names as Pybba, Penda, and Peada; but the life [263] states, with no little emphasis, that he was a man whose pedigree included the most noble names of illustrious kings from the ancient stock of Icel: that is, he was one of the Iclingas or Icklings [264]. Here one is tempted to perpetrate a little glottologic alchemy by changing l into n, and to suppose Iclingas the form taken in English by the name of the ancient people of the Iceni. In any case, nothing could be more reasonable to suppose than that some representatives of the royal race of Prasutagus and Boudicca, escaping the sword of the Roman, found refuge among the Coritanians at the time of the final defeat of their own people: it is even possible that they were already the ruling family there. At all events several indications converge to show that communities speaking Brythonic were not far off, to wit, the p names in the Mercian genealogy, Guthlac's father's surname, Guthlac's exile among Brythons, and the attack on him at Croyland by Brythonic speaking foes. Portions of the Coritanian territory were eminently fitted by nature to serve as a refuge for a broken people with a belated language: witness as late as the eleventh century the stand made in the Isle of Ely by Hereward against the Norman conqueror and his mail-clad knights [265].

Among the speakers of Goidelic in Ireland and the Highlands of Scotland the fairies take their designation chiefly from a word síd or síth (genitive síde or sída), which one may possibly consider as of a common origin with the Latin word sedes, and as originally meaning a seat or settlement, but it sooner or later came to signify simply an abode of the fairies, whence they were called in Medieval Irish aes síde, 'fairy folk,' fer síde, 'a fairy man,' and ben síde, 'a fairy woman or banshee.' By the side of síd, an adjective síde, 'of or belonging to the síd,' appears to have been formed, so that they are found also called simply síde, as in Fiacc's Hymn, where we are told that before the advent of St. Patrick the pagan tribes of Erin used to worship síde or fairies [266]. Borrowed from this, or suggested by it [267], we have in Welsh Caer Sidi, 'the Fortress of the Fairies,' which is mentioned twice in the Book of Taliessin [268]. It first occurs at the end of poem xiv, where we have the following lines, which recall Irish descriptions of Tír na nÓg or the Land of the Young:--

Ys kyweir vyg kadeir ygkaer sidi. Nys pla6d heint a heneint a uo yndi. Ys gwyr mana6yt a phryderi. Teir oryan y am tan agan recdi. Ac am y banneu ffrydyeu g6eilgi. Ar ffynnha6n ffr6ythla6n yssyd oduchti. Ys whegach nor g6in g6yn yllyn yndi.

Perfect is my seat in the fort of Sidi, Nor pest nor age plagues him who dwells therein: Manawydan and Pryderi know it. Three organs play before it about a fire. Around its corners Ocean's currents flow, And above it is the fertile fountain, And sweeter than white wine is the drink therein.

The wine is elsewhere mentioned, but the arrangement of the organs around a fire requires explanation, which I cannot give. The fortress is on an island, and in poem xxx of the Book of Taliessin we read of Arthur and his men sailing thither in his ship Prydwen: the poem is usually called the 'Spoils of Annwn,' and the lines in point run thus:--

Bu kyweir karchar g6eir ygkaer sidi. Tr6y ebostol p6yll aphryderi. Neb kyn noc ef nyt aeth idi. Yr gad6yn tromlas kywirwas ae ketwi. Arac preideu ann6fyn tost yt geni. Ac yt ura6t paraha6t ynbard wedi. Tri lloneit prytwen yd aetham ni idi. Nam seith ny dyrreith o gaer sidi.

Perfect was the prison of Gwair in Caer Sidi, Thanks to Pwyll and Pryderi's emissary. Before him no one entered into it, To the heavy, dark chain held by a faithful youth; And before the spoils of Annwn sorely he sang, And thenceforth remains he till doom a bard. Three freights of Prydwen went we thither, But only seven returned from Caer Sidi.

The incidents in these lines are mostly unintelligible to me, but the incarceration of Gweir or Gwair, together with other imprisonments, including that of Arthur in Caer Oeth and Anoeth (p. 619), are mentioned also in the Triads: see i. 50, ii. 7, 49, iii. 61. It is not improbable that the legend about Gwair located his prison on Lundy, as the Welsh name of that island appears to have been Ynys Wair, 'Gwair's Isle.' Pwyll and Pryderi did not belong to Annwn, nor did Pryderi's friend Manawydan; but the Mabinogi of Pwyll relates how for a whole year Pwyll exchanged crown and kingdom with Arawn king of Annwn, from whom he obtained the first breed of domestic pigs for his own people (pp. 69, 525).

In the lowlands of Scotland, together with the Orkneys and Shetlands, the Picts have to a certain extent taken the place of our fairies, and they are colloquially called Pechts. Now judging from the remains there ascribed to the Pechts, their habitations were either wholly underground or else so covered over with stones and earth and grass as to look like natural hillocks and to avoid attracting the attention of strangers. This was helped by making the entrance very low and as inconspicuous as possible. But one of the most remarkable things about these síds is that the cells within them are frequently so small as to prove beyond doubt, that those who inhabited them were of a remarkably short stature, though it is demonstrated by the weight of the stones used, that the builders were not at all lacking in bodily strength [269]. Here we have, accordingly, a small people like our own fairies. In Ireland one of the most famous kings of the fairies was called Mider of Brí Léith, where he resided in a síd or mound in the neighbourhood of Ardagh, in the county of Longford; and thither Irish legend represents him carrying away Étain, queen of Eochaid Airem, king of Ireland during a part of Conchobar MacNessa's time. Now Eochaid was for a whole year unable to find where she was, but his druid, Dalán, wrote Ogams and at last found it out. Eochaid then marched to Brí Léith, and began to demolish Mider's síd, whereupon Mider was eventually so frightened that he sent forth the queen to her husband, who then went his way, leaving the mound folk to digest their wrath. For it is characteristic of them that they did not fight, but chose to bide their time for revenge. In this instance it did not arrive till long after Eochaid's day [270]. I may add that Étain was herself one of the síde or fairies; and one of Mider's reasons for taking her away was, that she had been his wife in a previous stage of existence. Now it is true that the fairy Mider is described as resembling the other heroes of Irish story, in having golden yellow hair and bright blue eyes [271], but he differs completely from them in being no warrior but a great wizard; and though he is not said to have been of small stature, the dwarfs were not far off. For in describing the poet Atherne, who was notorious for his stinginess (p. 635), the story-teller emphasizes his words by representing him taking from Mider three of his dwarfs and stationing them around his own house, in order that their truculent looks and rude words might drive away anybody who came to seek hospitality or to present an unwelcome request [272], a rôle which recalls that of Edern ab Nud's dwarf already mentioned (p. 672). Here the Irish word used is corr, which is probably to be identified with the Brythonic cor, 'a dwarf,' though the better known meaning of corr in Irish is 'crane or heron.' From the former also is hardly to be severed the Irish corrguinigh, 'sorcerers,' and corrguinacht [273], or the process of cursing to which the corrguinigh resorted, as, for instance, when Néde called forth the fatal blisters on Caier's face (p. 632). The rôle would seem exactly to suit the little people, who were consummate magicians.

Let me for a moment leave the little people, in order to call attention to another side of this question of race. It has recently been shown [274] by Professor J. Morris Jones, of the University College of North Wales, that the non-Aryan traits of the syntax of our insular Celtic point unmistakably to that of old Egyptian and Berber, together with kindred idioms belonging to the southern shore of the Mediterranean Sea. He has thereby reduced to articulate speech, so to say, the physiognomical convictions of Professor Sayce (p. 665), to which the reader's attention has been called. To the linguistic argument he appends a statement cited from a French authority and bearing on the question of descent by birth, to the effect, that when among the Berbers the king dies or is deposed, as happens often enough, it is not his son that is called to succeed, but the son of his sister, as in the case of the historical Picts of Scotland down to the twelfth century or thereabouts. Here I would add, that my attention has been called by Professor Sayce to old Egyptian monuments representing the Libyan chiefs with their bodies tattooed, a habit which seems not to be yet extinct among the Touaregs and Kabyles [275]. Lastly, Mr. Nicholson has recently directed attention to the fact that some princes of ancient Gaul are represented with their faces tattooed on certain coins found in the west of France so far south as the region once occupied by the ancient Pictones. We have a compendious commentary on this in the occurrence of a word Chortonicum in a High German manuscript written before the year 814: I allude to the Wessobrunn Codex at Munich, in which, among a number of geographical names connected with Gaul and other countries, that vocable is so placed as to allow of our referring it to Poitou or to all Gaul as the country once of the ancient Pictones. The great German philologist Pott, who called attention to it, brought it at once into relation with Cruithne, plural Cruithni, 'the Picts of Britain and Ireland,' a word which has been explained at p. 281 above [276].

Now at last I come to the question, what pre-Celtic race or races make themselves evident in the mass of things touched on in this and the foregoing chapters? The answer must, I think, recognize at least two. First comes the race of the mound folk, consisting of the short swarthy people variously caricatured in our fairy tales. They formed isolated fractions of a widely spread race possessed of no political significance whatsoever; but, with the inconsistency ever clinging to everything connected with the fairies, the weird and uncanny folk emerging from its underground lairs seems to have exercised on other races a sort of permanent spell of mysteriousness amounting to adoration. In fact, Irish literature tells us that the síde were worshipped (p. 678). Owing to his faculty of exaggeration, combined with his inability to comprehend the little people, the Celt was enabled to bequeath to the great literatures of Western Europe a motley train of dwarfs and brownies, a whole world of wizardry and magic. The real race of the little people forms the lowest stratum which we can reach, to wit, at a level no higher, seemingly, than that of the present-day natives of Central Australia. Thus some of the birth stories of Cúchulainn and Étáin seem to have passed through their hands, and they bear a striking resemblance to certain notions of the Lapps (pp. 657-8). In fact, the nature of the habitations of our little people, together with other points which might be mentioned, would seem at first sight to betoken affinity with the Lapps; but I am warned by experts [277] that there are serious craniological difficulties in the way of any racial comparison with the Lapps, and that one must look rather to the dwarf populations once widely spread over our hemisphere, and still to be found here and there in Europe, as, for example, in Sicily. To come nearer our British Isles, the presence of such dwarfs has been established with regard to Switzerland in neolithic times [278].

The other race may be called Picts, which is probably the earliest of the names given it by the Celts; and their affinities appear to be Libyan, possibly Iberian. It was a warlike stock, and stood higher altogether than the mound inhabitants; for it had a notion of paternity, though, on account of its promiscuity, it had to reckon descent by birth (pp. 654-6). To it probably belonged all the great family groups figuring in the Mabinogion and the corresponding class of literature in Irish: this would include the Danann-Dôn group and the Lir-Llyr group, together with the families represented by Pwyll and Rhiannon, who were inseparable from the Llyr group in Welsh, just as the Lir group was inseparable from the Tuatha Dé Danann in Irish legend (pp. 548-9). The Picts made slaves and drudges of the mound-haunting race, but how far any amalgamation may have taken place between them it is impossible to say. Even without any amalgamation, however, the little people, if employed as nurses to their Pictish lords' children, could not help leaving their impress in time on the language of the ruling nationality. But it may be that the treatment of the Picts, by Scottish legend, as a kind of fairies really points to amalgamation, though it is not impossible that archæology may be able to classify the remains of the dwellings ascribed to the Pechts, that is, to assign a certain class to the warlike Picts of history and another to the dwarf race of the síds. A certain measure of amalgamation may also be the meaning of the Irish tradition, that when the Milesian Irish came and conquered, the defeated Tuatha Dé Danann gave up their life above ground and retired inside the hills like the fairies. This account of them may be as worthless as the story of the extermination of the Picts of Scotland: both peoples doubtless lived on to amalgamate in time with the conquering race; but it may mean that some of them retreated before the Celts, and concealed themselves after the manner of the little people--in underground dwellings in the less accessible parts of the country. In any case, it may well be that they got their magic and druidism from the dwellers of the síds. In the next place, it has been pointed out (pp. 550-1) how the adjective hên, 'old, ancient,' is applied in Welsh to several of the chief men of the Dôn group, and by this one may probably understand that they were old not merely to those who told the stories about them in Welsh, but to those who put those stories together in Goidelic ages earlier. The geography of the Mabinogion gives the prehistoric remains of Penmaen Mawr and Tre'r Ceiri to the Dôn group; but by its name, Tre'r Ceiri should be the 'Town of the Keiri,' a word probably referring to the Picts (pp. 279-83): this, so far as it goes, makes the sons of Dôn belong by race to the Picts. Lastly, it is the widely spread race of the Picts, conquered by the Celts of the Celtican or Goidelic branch and amalgamating with their conquerors in the course of time, that has left its non-Aryan impress on the syntax of the Celtic languages of the British Isles.

These, it is needless to say, are conjectures which I cannot establish; but possibly somebody else may. For the present, however, they cannot fail to suggest a moral, habitually ignored with a light heart by most people--including the writer of these words--that men in his plight, men engaged in studies which, owing to a rapid accumulation of fresh facts or the blossoming of new theories, are in a shifting condition, should abstain from producing books or anything longer than a magazine article now and then. Even such minor productions should be understood to be liable to be cast into a great bonfire lit once a year, say on Halloween. This should help to clear the air of mistaken hypotheses, whether of folklore and myth or of history and language, and also serve to mark Nos Calangaeaf as the commencement of the ancient Celtic year. The business of selecting the papers to be saved from the burning might be delegated to an academy constituted, roughly speaking, on the lines of Plato's aristocracy of intellect. Such academy, once in the enjoyment of its existence, would also find plenty of work in addition to the inquisitional business which I have suggested: it should, for example, be invested with summary jurisdiction over fond parents who venture to show any unreasonable anxiety to save their mental progeny from the annual bonfire. The best of that class of writers should be ordered by the academy to sing songs or indite original verse. As for the rest, some of them might be told off to gesticulate to the gallery, and some to administer the consolations of platitude to stragglers tired of the march of science. There is a mass of other useful work which would naturally devolve on an academy of the kind here suggested. I should be happy, if space permitted, to go through the particulars one by one, but let a single instance suffice: the academy might relieve us of the painful necessity of having seriously to consider any further the proposal that professors found professing after sixty should be shot. This will serve to indicate the kind of work which might advantageously be entrusted to the august body which is here but roughly projected.

There are some branches of learning in the happy position of having no occasion for such a body academical. Thus, if a man will have it that the earth is flat, as flat in fact as some people do their utmost to make it, 'he will most likely,' as the late Mr. Freeman in the Saturday Review once put it, 'make few converts, and will be forgotten after at most a passing laugh from scientific men.' If a man insists that the sum of two and two is five, he will probably find his way to a lunatic asylum, as the economy of society is, in a manner, self-acting. So with regard to him who carries his craze into the more material departments of such a science as chemistry: he may be expected to blow out his own eyes, for the almighty molecule executes its own vengeance. 'But,' to quote again from Mr. Freeman, if that man's 'craze had been historical or philological'--and above all if it had to do with the science of man or of myth--'he might have put forth notions quite as absurd as the notion that the earth is flat, and many people would not have been in the least able to see that they were absurd. If any scholar had tried to confute him we should have heard of "controversies" and "differences of opinion."' In fact, the worst that happens to the false prophet who shines in any such a science is, that he has usually only too many enthusiastic followers. The machinery is, so to say, not automatic, and hence it is that we want the help of an academy. But even supposing such an academy established, no one need feel alarmed lest opportunities enough could no longer be found for cultivating the example of those of the early Christians who had the rare grace to suffer fools gladly.

Personally, however, I should be against doing anything in a hurry; and, considering how little his fellows dare expect from the man who is just waiting to be final and perfect before he commit himself to type, the establishment of an academy invested with the summary powers which have been briefly sketched might, perhaps, after all, conveniently wait a while: my own feeling is that almost any time, say in the latter half of the twentieth century, would do better than this year or the next. In the meantime one must be content to entrust the fortunes of our studies to the combined forces of science and common sense. Judging by what they have achieved in recent years, there is no reason to be uneasy with regard to the time to come, for it is as true to-day as when it was first written, that the best of the prophets of the Future is the Past.

ADDITIONS AND CORRECTIONS

P. 81. I learn that the plural of bodach glas was in Welsh bodachod gleision, a term which Elis o'r Nant remembers his mother applying to a kind of fairies dressed in blue and fond of leading people astray. She used to relate how a haymaking party once passed a summer's night at the cowhouse (beudy) of Bryn Bygelyd (also Bryn Mygelyd), and how they saw in the dead of night a host of these dwarfs (corynnod) in blue dancing and capering about the place. The beudy in question is not very far from Dolwydelan, on the way to Capel Curig. A different picture of the bodach is given in Jenkins' Bed Gelert, p. 82; and lastly one may contrast the Highland Bodach Glas mentioned at p. 520 above, not to mention still another kind, namely the one in Scott's Waverley.

P. 130. To Sarn yr Afanc add Llyn yr Afanc, near Llandinam (Beauties of Wales, N. Wales, p. 841), and Bed yr Afanc, 'the Afanc's Grave,' the name of some sort of a tumulus, I am told, on a knoll near the Pembrokeshire stream of the Nevern. Mr. J. Thomas, of Bancau Bryn Berian close by, has communicated to me certain echoes of a story how an afanc was caught in a pool near the bridge of Bryn Berian, and how it was taken up to be interred in what is now regarded as its grave. A complete list of the afanc place-names in the Principality might possibly prove instructive. As to the word afanc, what seems to have happened is this: (1) from meaning simply a dwarf it came to be associated with such water dwarfs as those mentioned at p. 432; (2) the meaning being forgotten, the word was applied to any water monster; and (3) where afanc occurs in place-names the Hu story has been introduced to explain it, whether it fitted or not. This I should fancy to be the case with the Bryn Berian barrow, and it would be satisfactory to know whether it contains the remains of an ordinary dwarf. Peredur's lake afanc may have been a dwarf; but whether that was so or not, it is remarkable that the weapon which the afanc handled was a llechwaew or flake-spear, that is, a missile tipped with stone.

P. 131. With the rôle of the girl in the afanc story compare that of Tegau, wife of Caradog Freichfras, on whom a serpent fastens and can only be allured away to seize on one of Tegau's breasts, of which she loses the nipple when the beast is cut off. The defect being replaced with gold, she is ever after known as Tegau Eur-fron, or 'Tegau of the golden Breast.' That is a version inferred of a story which is discussed by M. Gaston Paris in an article, on Caradoc et le Serpent, elicited by a paper published (in the November number of Modern Language Notes for 1898) by Miss C. A. Harper, of Bryn Mawr College, U.S.: see the Romania, xxviii. 214-31. One of Miss Harper's parallels, mentioned by M. Paris at p. 220, comes from Campbell: it is concerning a prince who receives from his stepmother a magic shirt which converts itself into a serpent coiled round his neck, and of which he is rid by the help of a woman acting in much the same way as Tegau. We have an echo of this in the pedigrees in the Jesus College MS. 20: see the Cymmrodor, viii. 88, where one reads of G6ga6n keneu menrud a vu neidyr vl6ydyn am y von6gyl, 'Gwgon the whelp of Menrud (?) who was a year with a snake round his neck'--his pedigree is also given. In M. Paris' suggested reconstruction of the story (p. 228) from the different versions, he represents the maiden who is to induce the serpent to leave the man on whom it has fastened, as standing in a vessel filled with milk, while the man stands in a vessel filled with vinegar. The heroine exposes herself to the reptile, which relinquishes his present victim to seize on one of the woman's breasts. Now the appropriateness of the milk is explained by the belief that snakes are inordinately fond of milk, and that belief has, I presume, a foundation in fact: at any rate I am reminded of its introduction into the plot of more than one English story, such as Stanley Weyman's book From the Memoirs of a Minister of France (London, 1895), p. 445, and A. Conan Doyle's Adventures of Sherlock Holmes (London, 1893), pp. 199-209. In Wales, however, it is to a woman's milk that one's interest attaches: I submit two references which will explain what I mean. The first of them is to Owen's Welsh Folk-Lore, p. 349, where he says that 'traditions of flying snakes were once common in all parts of Wales,' and adds as follows:--'The traditional origin of these imaginary creatures was that they were snakes, which by having drunk the milk of a woman, and by having eaten of bread consecrated for the Holy Communion, became transformed into winged serpents or dragons.' The other is to the Brython for 1861, p. 190, where one reads in Welsh to the following effect:--'If a snake chances to have an opportunity to drink of a woman's milk it is certain to become a gwiber. When a woman happens to be far from her child, and her breasts are full and beginning to give her pain, she sometimes milks them on the ground in order to ease them. To this the peasantry in parts of Cardiganshire have a strong objection, lest a snake should come there and drink the milk, and so become a gwiber.' The word gwiber is used in the Welsh Bible for a viper, but the editor of the Brython explains, that in our folklore it means a huge kind of snake or dragon that has grown wings and has its body cased in hard scales: for a noted instance in point he refers the reader to the first number of the Brython, p. 3. It is believed still all over Wales that snakes may, under favourable circumstances, develop wings: in fact, an Anglesey man strongly wished, to my knowledge, to offer to the recent Welsh Land Commission, as evidence of the wild and neglected state of a certain farm, that the gorse had grown so high and the snakes so thriven in it that he had actually seen one of the latter flying right across a wide road which separated two such gorse forests as he described: surprised and hurt to find that this was not accepted, he inferred that the Commissioners knew next to nothing about their business.

Pp. 148, 170. With 'the spell of security' by catching hold of grass may perhaps be compared a habit which boys in Cardiganshire have of suddenly picking up a blade of grass when they want a truce or stoppage in a sort of game of tig or touchwood. The grass gives the one who avails himself of it immunity for a time from attack or pursuit, so as to allow him to begin the game again just where it was left off.

P. 228. Bodermud would probably be more correctly written Bodermyd, and analysed possibly into Bod-Dermyd, involving the name which appears in Irish as Diarmait and Dermot.

P. 230. Since this was printed I have been assured by Mr. Thomas Prichard of Llwydiarth Esgob, in Anglesey, that the dolur byr is more commonly called clwy' byr, and that it is the disease known in English as 'black quarter.'

Pp. 259, 268. I am assured on the part of several literary natives of Glamorgan that they do not know dâr for daear, 'ground, earth.' Such negative evidence, though proving the literary form daear to prevail now, is not to be opposed to the positive statement, sent by Mr. Hughes (p. 173) to me, as to the persistence in his neighbourhood of dâr and clâr (for claear, 'lukewarm'), to which one may add, as unlikely to be challenged by anybody, the case of harn for haearn, 'iron.' The intermediate forms have to be represented as daer, claer, and haern, which explain exactly the gaem of the Book of St. Chad, for which modern literary Welsh has gaeaf, 'winter': see the preface to the Book of Llan Dâv, p. xlv.

P. 290. It ought to have been pointed out that the fairies, whose food and drink it is death to share, represent the dead.

P. 291. For Conla read Connla or Condla: the later form is Colla. The Condla in question is called Condla Rúad in the story, but the heading to it has Ectra Condla Chaim, 'the Adventure of C. the Dear One.'

P. 294. I am now inclined to think that butch was produced out of the northern pronunciation of witch by regarding its w as a mutation consonant and replacing it, as in some other instances, by b as the radical.

P. 308. With the Manx use of rowan on May-day compare a passage to the following effect concerning Wales--I translate it from the faulty Welsh in which it is quoted by one of the competitors for the folklore prize at the Liverpool Eistedfod, 1900: he gave no indication of its provenance:--Another bad papistic habit which prevails among some Welsh people is that of placing some of the wood of the rowan tree (coed cerdin or criafol) in their corn lands (llafyrieu) and their fields on May-eve (Nos Glamau) with the idea that such a custom brings a blessing on their fields, a proceeding which would better become atheists and pagans than Christians.

P. 325. In the comparison with the brownie the fairy nurse in the Pennant Valley has been overlooked: see p. 109.

P. 331, line 1. For I. 42-3 read ii. 42-3.

Pp. 377, 395. With the story of Ffynnon Gywer and the other fairy wells, also with the wells which have been more especially called sacred in this volume, compare the following paragraph from Martin's Description of the Western Islands of Scotland (London, 1703), pp. 229-30: it is concerning Gigay, now more commonly written Gigha, the name of an island near the west coast of Kintyre:--'There is a well in the north end of this isle called Toubir-more, i. e. a great well, because of its effects, for which it is famous among the islanders; who together with the inhabitants use it as a Catholicon for diseases. It's covered with stone and clay, because the natives fancy that the stream that flows from it might overflow the isle; and it is always opened by a Diroch, i. e. an inmate, else they think it would not exert its vertues. They ascribe one very extraordinary effect to it, and 'tis this; that when any foreign boats are wind-bound here (which often happens) the master of the boat ordinarily gives the native that lets the water run a piece of money, and they say that immediately afterwards the wind changes in favour of those that are thus detain'd by contrary winds. Every stranger that goes to drink of the water of this well, is accustomed to leave on its stone cover a piece of money, a needle, pin, or one of the prettiest variegated stones they can find.' Last September I visited Gigha and saw a well there which is supposed to be the one to which Martin refers. It is very insignificant and known now by a name pronounced Tobar a veac, possibly for an older Mo-Bheac: in Scotch Gaelic Bëac, written Beathag, is equated with the name Sophia. The only tradition now current about the well is that emptying it used to prove the means of raising a wind or even of producing great storms, and this appears to have been told Pennant: see his Tour in Scotland and Voyage to the Hebrides, MDCCLXXII (Chester, 1774), p. 226:--'Visit the few wonders of the isle: the first is a little well of a most miraculous quality, for in old times, if ever the chieftain lay here wind-bound, he had nothing more to do than cause the well to be cleared, and instantly a favorable gale arose. But miracles are now ceased.'

P. 378. A similar rhyme is current in the neighbourhood of Dolgelley, as Miss Lucy Griffith informs me, as follows:--

Dolgelle dol a gollir, Daear a'i llwnc, dw'r 'n 'i lle.

Dolgelley, a dale to be lost; Earth will swallow it, and water take its place.

P. 394. With regard to wells killing women visiting them, I may mention a story, told me the other day by Professor Mahaffy after a friend whose name he gave, concerning the inhabitants of one of the small islands on the coast of Mayo--I understood him to say off the Mullet. It was this: all the men and boys, having gone fishing, were prevented by rough weather from returning as soon as they intended, and the women left alone suffered greatly from want of water, as not one of them would venture to go to the well. By-and-by, however, one of them gave birth to a boy, whereupon another of them carried the baby to the well, and ventured to draw water.

P. 418. As to Clychau Aberdyfi I am now convinced that the chwech and saith are entirely due to the published versions, the editors of which seem to have agreed that they will have as much as possible for their money, so to say. I find that Mrs. Rhys learnt in her childhood to end the words with pump, and that she cannot now be brought to sing the melody in any other way: I have similar testimony from a musical lady from the neighbourhood of Wrexham; and, doubtless, more evidence of the same sort could be got.

P. 443. For Llywelyn ab Gruffyd read Llywelyn ab Iorwerth.

Pp. 450-1. Some additional light on the doggerel dialogue will be found thrown by the following story, which I find cited in Welsh by one of the Liverpool Eistedfod competitors:--There is in the parish of Yspytty Ifan, in Carnarvonshire, a farm called Trwyn Swch, where eighty years ago lived a man and his wife, who were both young, and had twins born to them. Now the mother went one day to milk, leaving the twins alone in the cradle--the husband was not at home--and who should enter the house but one of the Tylwyth Teg! He took the twins away and left two of his own breed in the cradle in their stead. Thereupon the mother returned home and saw what had come to pass; she then in her excitement snatched the Tylwyth Teg twins and took them to the bridge that crosses the huge gorge of the river Conwy not very far from the house, and she cast them into the whirlpool below. By this time the Tylwyth Teg had come on the spot, some trying to save the children, and some making for the woman. 'Seize the old hag!' (Crap ar yr hen wrach!) said one of the chiefs of the Tylwyth Teg. 'Too late!' cried the woman on the edge of the bank; and many of them ran after her to the house. As they ran three or four of them lost their pipes in the field. They are pipes ingeniously made of the blue stone (carreg las) of the gully. They measure three or four inches long, and from time to time several of them have been found near the cave of Trwyn Swch.--This is the first indication which I have discovered, that the fairies are addicted to smoking.

P. 506. A Rhiw Gyferthwch (printed Rywgyverthwch) occurs in the Record of Carnarvon, p. 200; but it seems to have been in Merionethshire, and far enough from Arfon.

P. 521. In the article already cited from the Romania, M. Paris finds Twrch Trwyth in the boar Tortain of a French romance: see xxviii. 217, where he mentions a legend concerning the strange pedigree of that beast. The subject requires to be further studied.

P. 535. A less probable explanation of Latio would be to suppose orti understood. This has been suggested to me by Mr. Nicholson's treatment of the Llanaelhaiarn inscription as Ali ortus Elmetiaco hic iacet, where I should regard Ali as standing for an earlier nominative Alec-s, and intended as the Celtic equivalent for Cephas or Peter: Ali would be the word which is in Med. Irish ail, genitive ailech, 'a rock or stone.'

P. 545. We have the Maethwy of Gilvaethwy possibly still further reduced to Aethwy in Porth Aethwy, 'the Village of Menai Bridge,' in spite of its occurring in the Record of Carnarvon, p. 77, as Porthaytho.

P. 548. To the reference to the Cymmrodor, ix. 170, as to Beli being called son of Anna, add the Welsh Elucidarium, p. 127, with its belim vab anna, and The Cambro-British Saints, p. 82, where we have Anna ... genuit Beli.

P. 560. Two answers to the query as to the Llech Las are now to be found in the Scottish Antiquary, xv. 41-3.

P. 566. Caer Gai is called also Caer Gynyr, after Cai's father Cynyr, to wit in a poem by William Lleyn, who died in 1587. This I owe to Professor J. Morris Jones, who has copied it from a collection of that poet's works in the possession of Myrdin Fard, fo. 119.

P. 569. Here it would, perhaps, not be irrelevant to mention Caer Dwrgynt, given s. v. Dwr in Morris' Celtic Remains, as a name of Caergybi, or Holyhead. His authority is given in parenthesis thus: (Th. Williams, Catal.). I should be disposed to think the name based on some such an earlier form as Kair D6bgint, 'the Fortress of the Danes,' who were called in old Welsh Dub-gint (Annales Cambriæ, A. D. 866, in the Cymmrodor, ix. 165), that is to say 'Gentes Nigræ or Black Pagans,' and more simply Gint or Gynt, 'Gentes or Heathens.'

Pp. 579-80. The word banna6c, whence the later bannog, seems to be the origin of the name bonoec given to the famous horn in the Lai du Corn, from which M. Paris in his Romania article, xxviii. 229, cites Cest cor qui bonoec a non, 'this horn which is called bonoec.' The Welsh name would have to be Corn (yr) ych banna6c, 'the horn of (the) bannog ox,' with or without the article.

P. 580, note 1. One of the Liverpool Eistedfod competitors cites W. O. Pughe to the following effect in Welsh:--Llyn dau Ychain, 'the Lake of Two Oxen,' is on Hiraethog Mountain; and near it is the footmark of one of them in a stone or rock (carreg), where he rested when seeking his partner, as the local legend has it. Another cites a still wilder story, to the effect that there was once a wonderful cow called Y Fuwch Fraith, 'the Parti-coloured Cow.' 'To that cow there came a witch to get milk, just after the cow had supplied the whole neighbourhood. So the witch could not get any milk, and to avenge her disappointment she made the cow mad. The result was that the cow ran wild over the mountains, inflicting immense harm on the country; but at last she was killed by Hu near Hiraethog, in the county of Denbigh.'

P. 592. With trwtan, Trwtyn-Tratyn, and Trit-a-trot should doubtless be compared the English use of trot as applied contemptuously to a woman, as when Grumio, in Shakespeare's Taming of the Shrew, Act i, sc. 2, speaks of 'an old trot with ne'er a tooth in her head': the word was similarly used by Thomas Heywood and others.

P. 649. With regard to note 1, I find that Professor Zimmer is of opinion--in fact he is quite positive--that tyngu and tynghed are in no way related: see the Göttingische gelehrte Anzeigen for 1900 (No. 5), pp. 371-2.

P. 673. I am tempted to rank with the man-eating fairies the Atecotti, who are known to have been cannibals, and whose name seems to mean the ancient race. Should this prove tenable, one would have to admit that the little people, or at any rate peoples with an admixture of the blood of that race, could be trained to fight. Further, one would probably have to class with them also such non-cannibal tribes as those of the Fir Bolg and the Galiúin of Irish story. Information about both will be found in my Hibbert Lectures, in reading which, however, the mythological speculations should be brushed aside. Lastly, I anticipate that most of the peoples figuring in the oldest class of Irish story will prove to have belonged either (1) to the dwarf race, or (2) to the Picts; and that careful reading will multiply the means of distinguishing between them. Looking comprehensively at the question of the early races of the British Isles, the reader should weigh again the concluding words of Professor Haddon's theory, quoted on p. 684 above.

NOTES

[1] For most of my information on this subject I have to thank Mr. David Davies, editor of the South Wales Daily Post, published at Swansea.

[2] I am indebted for this information to Mr. J. Herbert James of Vaynor, who visited Kenfig lately and has called my attention to an article headed 'The Borough of Kenfig,' in the Archæologia Cambrensis for 1898: see more especially the maps at pp. 138-42.

[3] Here the Welsh has a word edafwr, the exact meaning of which escapes me, and I gather from the remarks of local etymologers that no such word is now in use in Glamorgan.

[4] See the Book of Aberpergwm, printed as Brut y Tywysogion, in the Myvyrian Archaiology, ii. 524; also Morgan's Antiquarian Survey of East Gower, p. 66, where the incident is given from 'Brut y Tywysogion, A. D. 1088.' It is, however, not in what usually passes by the name of Brut y Tywysogion, but comes, as the author kindly informs me, from a volume entitled 'Brut y Tywysogion, the Gwentian Chronicle of Caradoc of Llancarvan, with a translation by the late Aneurin Owen, and printed for the Cambrian Archæological Association, 1863': see pp. 70-1.

[5] For this also I have to thank Mr. Herbert James, who recently inspected the spot with Mr. Glascodine of Swansea.

[6] I do not know whether anybody has identified the spot which the writer had in view, or whether the coast of the Severn still offers any feature which corresponds in any way to the description.

[7] Supposed to be so called after a certain Tegid Foel, or 'Tegid the Bald,' of Penllyn: the name Tegid is the phonetic spelling of what might be expected in writing as Tegyd--it is the Latin Tacitus borrowed, and comes with other Latin names in Pedigree I. of the Cuneda dynasty; see the Cymmrodor, xi. 170. In point of spelling one may compare Idris for what might be expected written Idrys, of the same pronunciation, for an earlier Iudrys or Iudris.

[8] The translation was made by Thomas Twyne, and published in 1573 under the title of The Breuiary of Britayne, where the passage here given occurs, on fol. 69b. The original was entitled Commentarioli Britannicæ Descriptionis Fragmentum, published at Cologne in 1572. The original of our passage, fol. 57a, has Guynedhia and Llunclis. The stem llwnc of llyncaf, 'I swallow,' answers, according to Welsh idiom, to the use of what would be in English or Latin a participle. Similarly, when a compound is not used, the verbal noun (in the genitive) is used: thus 'a feigned illness,' in Welsh 'a made illness,' is saldra gwneyd, literally 'an indisposition or illness of making.' So 'the deuouryng of the Palace' is incorrect, and based on Llwyd's vorago Palatij instead of Palatium voratum.

[9] For other occurrences of the name, see the Black Book, fol. 35a, 52a, and Morris' Celtic Remains, where, s. v. Benlli, the Welsh name of Bardsey, to wit, Ynys Enlli, is treated by somebody, doubtless rightly, as a shortening of Ynys Fenlli.

[10] The meaning of this name is not certain, but it seems to equate with the Irish Fochard, anglicized Faughard, in County Louth: see O'Donovan's Four Masters, A. D. 1595; also the Book of the Dun Cow, where it is Focherd, genitive Focherda, dative Focheird, fo. 70b, 73b, 75a, 75b, 76a, 77a.

[11] This is sometimes given as Glannach, which looks like the Goidelic form of the name: witness Giraldus' Enislannach in his Itin. Kambriæ, ii. 7 (p. 131).

[12] See Choice Notes, p. 92, and Gerald Griffin's Poetical and Dramatic Works, p. 106.

[13] Failing to see this, various writers have tried to claim the honour of owning the bells for Aberteifi, 'Cardigan,' or for Abertawe, 'Swansea'; but no arguments worthy of consideration have been urged on behalf of either place: see Cyfaill yr Aelwyd for 1892, p. 184.

[14] For some of the data as to the reckoning of the pedigrees and branching of a family, see the first volume of Aneurin Owen's Ancient Laws--Gwyned, III. i. 12-5 (pp. 222-7); Dyfed, II. i. 17-29 (pp. 408-11); Gwent, II. viii. 1-7 (pp. 700-3); also The Welsh People, pp. 230-1.

[15] See the Book of the Dun Cow, fol. 99a & seq.

[16] For instances, the reader may turn back to pp. 154 or 191, but there are plenty more in the foregoing chapters; and he may also consult Howells' Cambrian Superstitions, pp. 123-8, 141-2, 146. In one case, p. 123, he gives an instance of the contrary kind of imagination: the shepherd who joined a fairy party on Frenni Fach was convinced, when his senses and his memory returned, that, 'although he thought he had been absent so many years, he had been only so many minutes.' The story has the ordinary setting; but can it be of popular origin? The Frenni Fach is a part of the mountain known as the Frenni Fawr, in the north-east of Pembrokeshire; the names mean respectively the Little Breni, and the Great Breni. The obsolete word breni meant, in Old Welsh, the prow of a ship; local habit tends, however, to the solecism of Brenin Fawr, with brenin, 'king,' qualified by an adjective mutated feminine; but people at a distance who call it Frenni Fawr, pronounce the former vocable with nn. Lastly, Y Vrevi Va6r occurs in Maxen's Dream in the Red Book (Oxford Mab. p. 89); but in the White Book (in the Peniarth collection), col. 187, the proper name is written Freni: for this information I have to thank Mr. Gwenogvryn Evans.

[17] It is right to say that another account is given in the Rennes Dindsenchas, published by Stokes in the Revue Celtique, xvi. 164, namely, that Laiglinne with fifty warriors 'came to the well of Dera son of Scera. A wave burst over them and drowned Laiglinne with his fifty warriors, and thereof a lake was made. Hence we say Loch Laiglinni, Laiglinne's Lake.'

[18] The Oxford Mabinogion, p. 224, and Guest's, i. 343.

[19] See Afanc in the Geiriadur of Silvan Evans, who cites instances in point.

[20] See the Revue Celtique, i. 257, and my Hibbert Lectures, pp. 92-3.

[21] The Four Masters, A.M. 3520.

[22] In another version Campbell had found it to be sand and nothing else.

[23] As to this incident of a girl and a supernatural, Campbell says that he had heard it in the Isle of Man also, and elsewhere.

[24] See the Revue Celtique, ii. 197. He was also called Labraid Longsech, and Labraid Longsech Lorc. The explanation of Labraid Lorc is possibly that it was originally Labraid Morc, and that the fondness for alliteration brought it into line as Labraid Lorc: compare Llûd Llaweraint in Welsh for Nûd Llaweraint. This is not disproved by the fact that Labraid Lorc's grandfather is said to have been called Loegaire Lorc: Loegaire Lorc and Labraid Lorc are rather to be regarded perhaps as duplicates of the same original.

[25] See my Arthurian Legend, p. 70; also Hibbert Lectures, p. 590.

[26] The original has in these passages respectively siblais a fual corbo thipra, 'minxit urinam suam so that it was a spring'; ar na siblad a fúal ar na bad fochond báis doib, 'ne mingat urinam suam lest it should be the cause of death to them'; and silis, 'minxit,' fo. 39b. For a translation of the whole story see Dr. O'Grady's Silva Gadelica, pp. 265-9; also Joyce's Old Celtic Romances, pp. 97-105.

[27] See the story in Dr. O'Grady's Silva Gadelica, pp. 292-311.

[28] See Stengel's edition of li Romans de Durmart le Galois (Tübingen, 1873), lines 4185-340, and my Arthurian Legend, pp. 68-9.

[29] See Williams' Scint Greal, pp. 60-1, 474-5; Nutt's Holy Grail, p. 44; and my Arthurian Legend, pp. 69-70.

[30] Bardoniaeth D. ab Gwilym, poem 183. A similar descent of Blodeuwed's appears implied in the following englyn--one of two--by Anthony Powel, who died in 1618: it is given by Taliesin ab Iolo in his essay on the Neath Valley, entitled Traethawd ar Gywreined, Hynafiaeth, a hen Bendefigion Glynn Ned (Aberdare, 1886), p. 15:--

Crug ael, carn gadarn a godwyd yn fryn, Yn hen fraenwaith bochlwyd; Main a'i llud man y lladwyd, Merch hoewen loer Meirchion lwyd.

It refers, with six other englynion by other authors, to a remarkable rock called Craig y Dinas, with which Taliesin associated a cave where Arthur or Owen Lawgoch and his men are supposed, according to him, to enjoy a secular sleep, and it implies that Blodeuwed, whose end in the Mabinogi of Mâth was to be converted into an owl, was, according to another account, overwhelmed by Craig y Dinas. It may be Englished somewhat as follows:

Heaped on a brow, a mighty cairn built like a hill, Like ancient work rough with age, grey-cheeked; Stones that confine her where she was slain, Grey Meirchion's daughter quick and bright as the moon.

[31] This comes from the late series of Triads, iii. 10, where Merlin's nine companions are called naw beird cylfeird: cylfeird should be the plural of cylfard, which must be the same word as the Irish culbard, name of one of the bardic grades in Ireland.

[32] For some more remarks on this subject generally, see my Arthurian Legend, chapter xv, on the 'Isles of the Dead.'

[33] See his Itinerarium Kambriæ, ii. 11 (p. 139); also my Celtic Britain, p. 68, and Arthurian Legend, p. 364.

[34] From the Myvyrian Archaiology of Wales, i. 302.

[35] I regard nid kywiw as a corruption of ni chywiw from cyf-yw, an instance of the verb corresponding to cymod (= cym-bod), 'peace, conciliation.' The preterite has, in the Oxford Bruts, A.D. 1217 (p. 358), been printed kynni for what one may read kymu: the words would then be y kymu reinald y bre6ys ar brenhin, 'that Reginald de Breos was reconciled with the king, or settled matters with him.'

[36] See the Book of Taliessin, poem xxx, in Skene's Four Ancient Books, ii. 181; also Guest's Mabinogion, ii. 354, and the Brython for 1860, p. 372b, where more than one article of similar capacity of distinguishing brave men from cowards is mentioned.

[37] See Dugdale's Monasticon, v. 672, where they are printed Dwynech and Dwynaur respectively.

[38] See my Hibbert Lectures, pp. 649-50.

[39] A full account of them will be found in a volume devoted to them, and entitled Roman Antiquities at Lydney Park, Gloucestershire, being a posthumous work of the Rev. W. Hiley Bathurst, with Notes by C. W. King, London, 1879. See also an article entitled 'Das Heiligtum des Nodon,' by Dr. Hübner in the Jahrbücher des Vereins von Alterthumsfreunden im Rheinlande, lxvii. pp. 29-46, where several things in Mr. King's book are criticized.

[40] See my Hibbert Lectures, pp. 122, 125.

[41] On this subject, see The Welsh People, especially pp. 54-61.

[42] Why our dictionary makers have taken into their heads to treat it as Tamesis I know not. The Welsh is Tafwys with a diphthong regularly representing an earlier long e or ei in the second syllable. There is, as far as I know, no reason to suppose Tafwys an invention, rather than a genuine vocable of the same origin as the name of the Glamorganshire river Taff, in Welsh Taf, which is also the name of the river emptying itself at Laugharne, in Carmarthenshire. Tafwys, however, does not appear to occur in any old Welsh document; but no such weakness attaches to the testimony of the French Tamise, which could hardly come from Tamesis: compare also the place-name Tamise near the Scheldt in East Flanders; this, however, may be of a wholly different origin.

[43] A more difficult version has been sent me by Dewi Glan Ffrydlas, of Bethesda: Caffed y wrach, 'Let him seize the hag'; Methu'r cryfaglach, 'You have failed, urchin.' But he has not been able to get any explanation of the words at the Penrhyn Quarries. Cryfaglach is also the form in Mur y Cryfaglach, 'the Urchin's Wall,' in Jenkins' Bed Gelert, p. 249. He informs me that this is the name of an old ruin on an elevated spot some twenty or thirty yards from a swift brook, and not far in a south-south-easterly direction from Sir Edward Watkin's chalet.

[44] For this I am indebted to Mr. Wm. Davies (p. 147 above), who tells me that he copied the original from Chwedlau a Thradodiadau Gwyned, 'Gwyned Tales and Traditions,' published in a periodical, which I have not been able to consult, called Y Gordofigion, for the year 1873.

[45] The meaning of the word mwthlach is doubtful, as it is now current in Gwyned only in the sense of a soft, doughy, or puffy person who is all of a heap, so to say. Pughe gives mwythlan and mwythlen with similar significations. But mwthlach would seem to have had some such a meaning in the doggerel as that of rough ground or a place covered with a scrubby, tangled growth. It is possibly the same word as the Irish mothlach, 'rough, bushy, ragged, shaggy'; see the Vision of Laisrén, edited by Professor K. Meyer, in the Otia Merseiana, pp. 114, 117.

[46] The account here given of the Cyhiraeth is taken partly from Choice Notes, pp. 31-2, and partly from Howells, pp. 31-4, 56-7, who appears to have got uncertain in his narrative as to the sex of the Cyhiraeth; but there is no reason whatsoever for regarding it as either male or female--the latter alone is warranted, as he might have gathered from her being called y Gyhiraeth, 'the Cyhiraeth,' never y Cyhiraeth as far as I know. In North Cardiganshire the spectre intended is known only by another name, that of Gwrach y Rhibyn, but y Gyhiraeth or yr hen Gyhiraeth is a common term of abuse applied to a lanky, cadaverous person, both there and in Gwyned; in books, however, it is found sometimes meaning a phantom funeral. The word cyhiraeth would seem to have originally meant a skeleton with cyhyrau, 'sinews,' but no flesh. However, cyhyrau, singular cyhyr, would be more correctly written with an i; for the words are pronounced--even in Gwyned--cyhir, cyhirau. The spelling cyhyraeth corresponds to no pronunciation I have ever heard of the word; but there is a third spelling, cyheuraeth, which corresponds to an actual cyhoereth or cyhoyreth, the colloquial pronunciation to be heard in parts of South Wales: I cannot account for this variant. Gwrach y Rhibyn means the Hag of the Rhibyn, and rhibyn usually means a row, streak, a line--ma' nhw'n mynd yn un rhibyn, 'they are going in a line.' But what exactly Gwrach y Rhibyn should connote I am unable to say. I may mention, however, on the authority of Mr. Gwenogvryn Evans, that in Mid-Cardiganshire the term Gwrach y Rhibyn means a long roll or bustle of fern tied with ropes of straw and placed along the middle of the top of a hayrick. This is to form a ridge over which and on which the thatch is worked and supported: gwrach unqualified is, I am told, used in this sense in Glamorganshire. Something about the Gwrach sprite will be found in the Brython for 1860, p. 23a, while a different account is given in Jenkins' Bed Gelert, pp. 80-1.

[47] This statement I give from Choice Notes, p. 32; but I must confess that I am sceptical as to the 'wings of a leathery and bat-like substance,' or of any other substance whatsoever.

[48] For more about her and similar ancestral personages, see The Welsh People, pp. 54-61.

[49] This seems to be the Goidelic word borrowed, which in Mod. Irish is written cnocc or cnoc, 'a hill': the native Welsh form is cnwch, as in Cnwch Coch in Cardiganshire, Cnwch Dernog (corrupted into Clwch Dernog) in Anglesey, printed Kuwgh Dernok in the Record of Carnarvon, p. 59, where it is associated with other interesting names to be noticed later.

[50] All said by natives of Anglesey about rivers and mountains in their island must be taken relatively, for though the country has a very uneven surface it has no real mountain: they are apt to call a brook a river and a hillock a mountain, though the majestic heights of Arfon are within sight.

[51] See pp. 13-16 of his essay on the Neath Valley, referred to in a note at p. 439 above, where Craig y Dinas is also mentioned.

[52] This is an interesting word of obscure origin, to which I should like our ingenious etymologists to direct their attention.

[53] See the Poetical Works of John Leyden (Edinburgh, 1875), p. 36 (Scenes of Infancy, part ii); also my Arthurian Legend, p. 18.

[54] I am indebted for the English story to an article entitled 'The Two Pedlar Legends of Lambeth and Swaffham,' contributed by Mr. Gomme to the pages of the Antiquary, x. 202-5, in which he gives local details and makes valuable comparisons. I have to thank Mr. Gomme also for a cutting from the weekly issue of the Leeds Mercury for Jan. 3, 1885, devoted to 'Local Notes and Queries' (No. cccxii), where practically the same story is given at greater length as located at Upsall Castle in Yorkshire.

[55] I have never been to the spot, and I owe these particulars partly to Mr. J. P. Owen, of 72 Comeragh Road, Kensington, and partly to the Rev. John Fisher, already quoted at p. 379. This is the parish where some would locate the story of the sin-eater, which others stoutly deny, as certain periodical outbursts of polemics in the pages of the Academy and elsewhere have shown. Mr. Owen, writing to me in 1893, states, that, when he last visited the dinas some thirty years previously, he found the mouth of the cave stopped up in order to prevent cattle and sheep straying into it.

[56] Mr. Fisher refers me to an account of the discovery published in the Cambrian newspaper for Aug. 14, 1813, a complete file of which exists, as he informs me, in the library of the Royal Institution of South Wales at Swansea. Further, at the Cambrians' meeting in 1892 that account was discussed and corrected by Mr. Stepney-Gulston: see the Archæologia Cambrensis for 1893, pp. 163-7. He also 'pointed out that on the opposite side of the gap in the ridge the noted cave of Owain Law Goch was to be found. Near the Pant-y-llyn bone caves is a place called Craig Derwydon, and close by is the scene of the exploits of Owain Law Goch, a character who appears to have absorbed some of the features of Arthurian romance. A cave in the locality bears Owain's name.'

[57] As in Llewelyn's charter to the Monks of Aberconwy, where we have, according to Dugdale's Monasticon, v. 673a, a Scubordynemreis, that is Scubor Dyn Emreis, 'Din-Emreis Barn,' supposed to be Hafod y Borth, near Bedgelert: see Jenkins' Bed Gelert, p. 198. In the Myvyrian, i. 195a, it has been printed Din Emrais.

[58] See Somer's Malory's Morte Darthur, xxi. v (= vol. i. p. 849), and as to the Marchlyn story see p. 236 above. Lastly some details concerning Llyn Llydaw will be found in the next chapter.

[59] The oldest spellings known of this name occur in manuscript A of the Annales Cambriæ and in the Book of Llan Dâv as Elized and Elised, doubtless pronounced Elissed until it became, by dropping the final dental, Elisse. This in time lost its identity by assimilation with the English name Ellis. Thus, for example, in Wynne's edition of Powell's Caradog of Llancarfan's History of Wales (London, 1774), pp. 22, 24, Elised is reduced to Elis. In the matter of dropping the d compare our Dewi, 'St. David,' for Dewid, for an instance of which see Duffus Hardy's Descriptive Catalogue, i. 119. The form Eliseg with a final g has no foundation in fact. Can the English name Ellis be itself derived from Elised?

[60] Boncyn is derived from bonc of nearly the same meaning, and bonc is merely the English word bank borrowed: in South Wales it is pronounced banc and used in North Cardiganshire in the sense of hill or mountain.

[61] The name occurs twice in the story of Kulhwch and Olwen: see the Mabinogion, p. 107, where the editors have read Ricca both times in 'Gormant, son of Ricca.' This is, however, more than balanced by Rita in the Book of Llan Dâv, namely in Tref Rita, 'Rita's town or stead,' which occurs five times as the name of a place in the diocese of Llandaff; see pp. 32, 43, 90, 272. The uncertainty is confined to the spelling, and it has arisen from the difficulty of deciding in medieval manuscripts between t and c: there is no reason to suppose the name was ever pronounced Ricca.

[62] This can hardly be the real name of the place, as it is pronounced Gwybrnant (and even Gwybrant), which reminds me of the Gwybr fynyd on which Gwyn ab Nûd wanders about with his hounds: see Evans' facsimile of the Black Book of Carmarthen, p. 50a, where the words are, dy gruidir ar wibir winit.

[63] Dugdale has printed this (v. 673a) Carrecerereryr with one er too much, and the other name forms part of the phrase ad capud Weddua-Vaur, 'to the top of the Great Gwydfa'; but I learn from Mr. Edward Owen, of Gray's Inn, that the reading of the manuscript is Wedua vawr and Carrecereryr.

[64] The MSS. except B have y 6ylva, which is clearly not the right word, as it could only mean 'his place of watching.'

[65] See Derfel Hughes' Llandegai and Llanllechid, p. 53. As to Drystan it is the Pictish name Drostan, but a kindred form occurs in Cornwall on a stone near Fowey, where years ago I guessed the ancient genitive Drustagni; and after examining it recently I am able to confirm my original guess. The name of Drystan recalls that of Essyllt, which offers some difficulty. It first occurs in Welsh in the Nennian Genealogies in the Harleian MS. 3859: see Pedigree I in the Cymmrodor, ix. 169, where we read that Mermin (Merfyn) was son of Etthil daughter of Cinnan (Cynan), who succeeded his father Rhodri Molwynog in the sovereignty of Gwyned in 754. The spelling Etthil is to be regarded like that of the Welsh names in Nennius, for some instances of which see § 73 (quoted in the next chapter) and the Old Welsh words calaur, nouel, patel, so spelt in the Juvencus Codex: see Skene, ii. 2: in all these l does duty for ll. So Etthil is to be treated as pronounced Ethill or Ethyll; but Jesus College MS. 20 gives a more ancient pronunciation (at least as regards the consonants) when it calls Cynan's daughter Ethellt: see the Cymmrodor, viii. 87. Powell, in his History of Wales by Caradog of Llancarfan, as edited by Wynne, writes the name Esylht; and the Medieval Welsh spelling has usually been Essyllt or Esyllt, which agrees in its sibilant with the French Iselt or Iseut; but who made the Breton-looking change from Eth to Es or Is in this name remains a somewhat doubtful point. Professor Zimmer, in the Zeitschrift für französische Sprache und Litteratur, xiii. 73-5, points out that the name is an Anglo-Saxon Ethylda borrowed, which he treats as a 'Kurzform für Ethelhild': see also the Revue Celtique, xii. 397, xiii. 495. The adoption of this name in Wales may be regarded as proof of intermarriage or alliance between an English family and the royal house of Gwyned as early as the eighth century.

[66] See the Brython for 1861, pp. 331-2, also Cymru Fu, p. 468, where Glasynys was also inclined to regard the Hairy Fellow as being Owen.

[67] I have never seen a copy, but Mr. Fisher gives me the title as follows: Prophwydoliaeth Myrdin Wyllt yn nghyda ber Hanes o'i Fywyd, wedi eu tynu allan o Lyfr y Daroganau ... Caerfyrdin ... Pris dwy Geiniog. It has no date, but Mr. Fisher once had a copy with the date 1847. Recently he has come across another versified prophecy written in the same style as the printed ones, and referring to an Owain who may have been Owen Lawgoch. The personage meant is compared to the most brilliant of pearls, Owain glain golyaf. The prophecy is to be found at the Swansea Public Library, and occurs in a seventeenth century manuscript manual of Roman Catholic Devotion, Latin and Welsh. It gives 1440 as the year of the deliverance of the Brytaniaid. It forms the first of two poems (fo. 37), the second of which is ascribed to Taliessin. Such is Mr. Fisher's account of it, and the lines which he has copied for me cling to the same theme of the ultimate triumph of the Kymry. Quite recently I have received further information as to these prophecies from Mr. J. H. Davies, of Lincoln's Inn (p. 354), who will, it is to be hoped, soon publish the results of his intimate study of their history in South Wales.

[68] Record of Carnarvon, p. 133, to which attention was called by me in the Report of the Welsh Land Commission, p. 648: see now The Welsh People, pp. 343-4, 593-4.

[69] Nor was Owen the only Welshman in the king of France's service: there was Owen's chaplain, who on one occasion distinguished himself greatly in battle. He is called in Froissart's text David House, but the editor has found from other documents that the name was Honvel Flinc, which is doubtless Howel, whatever the second vocable may have been: see Froissart, viii, pp. xxxviii, 69.

[70] As to the original destination of the flotilla, see Kervyn de Lettenhove's edition of Froissart (Brussels, 1870-7), viii. 435-7, where the editor has brought together several notes, from which it appears that Owen tried unsuccessfully to recruit an army in Spain, but that he readily got together in France a considerable force. For Charles V, on May 8, 1372, ordered the formation of an army, to be placed under Owen's command for the reconquest of his ancestors' lands in Wales, and two days later Owen issued a declaration as to his Welsh claims and his obligations to the French king; but the flotilla stopped short with Guernsey. It is not improbable, however, that the fear in England of a descent on Wales by Owen began at least as early as 1369. In his declaration Owen calls himself Evain de Gales, which approaches the Welsh spelling Ewein, more frequently Ywein, modern Ywain, except that all these forms tended to be supplanted by Owain or Owen. This last is, strictly speaking, the colloquial form, just as Howel is the colloquial form of Hywel, and bowyd of bywyd, 'life.'

[71] For the account of Owen's life see the Chroniques de J. Froissart publiées pour la Société de l'Histoire de France, edited with abstracts and notes by Siméon Luce, more especially vols. viii. pp. 44-9, 64, 66-71, 84, 122, 190, and ix. pp. 74-9, where a summary is given of his life and a complete account of his death. In Lord Berners' translation, published in Henry VIII's time, Owen is called Yuan of Wales, as if anybody could even glance at the romances without finding that Owen ab Urien, for instance, became in French Ywains or Ivains le fils Urien in the nominative, and Ywain or Ivain in régime. Thomas Johnes of Hafod, whose translation was published in 1803-6, betrays still greater ignorance by giving him the modern name Evan; but he had the excuse of being himself a Welshman.

[72] For copies of some of the documents in point see Rymer's Foedera, viii. 356, 365, 382.

[73] I have not been able to find a copy of this work, and for drawing my attention to the passage in Hanes Cymru I have again to thank Mr. Fisher. The pedigree in question will be found printed in Table I in Askew Roberts' edition of Sir John Wynne's History of the Gwydir Family (Oswestry, 1878); and a note, apparently copied from Miss Llwyd, states that it was in a Hengwrt MS. she found the identification of Owen Lawgoch. The editor surmises that to refer to p. 865 of Hengwrt MS. 351, which he represents as being a copy of Hengwrt MS. 96 in the handwriting of Robert Vaughan the Antiquary.

[74] This has already been undertaken: on Feb. 7, 1900, a summary of this chapter was read to a meeting of the Hon. Society of Cymmrodorion, and six weeks later Mr. Edward Owen, of Gray's Inn, read an elaborate paper in which he essayed to fix more exactly Yvain de Galles' place in the history of Wales. It would be impossible here to do justice to his reasoning, based as it was on a careful study of the records in point. Let it suffice for the present, however, that the paper will in due course appear in the Society's Transactions. Mr. J. H. Davies also informs me that he is bringing together items of evidence, which tend, as he thinks, to show that Miss Llwyd's information was practically correct. Before, however, the question can be considered satisfactorily answered, some explanation will have to be offered of Froissart's statement, that Yvain's father's name was Aymon.

[75] We seem also to have an instance in point in Carmarthenshire, where legend represents Owen and his men sleeping in Ogof Myrdin, the name of which means Merlin's Cave, and seems to concede priority of tenancy to the great magician: see the extinct periodical Golud yr Oes (for 1863), i. 253, which I find to have been probably drawing on Eliezer Williams' English Works (London, 1840), p. 156.

[76] For the Greek text of the entire passage see the Didot edition of Plutarch, vol. iii. p. 511 (De Defectu Oraculorum, xviii); also my Arthurian Legend, pp. 367-8. It is curious to note that storms have, in a way, been associated in England with the death of her great men as recently as that of the celebrated Duke of Wellington: see Choice Notes, p. 270.

[77] See my Arthurian Legend, p. 335. I am indebted to Professor Morfill for rendering the hexameters into English verse.

[78] They are produced here in their order as printed at the beginning of the second volume of the Myvyrian Archaiology of Wales, and the series or versions are indicated as i, ii, iii. Version ii will be found printed in the third volume of the Cymmrodor, pp. 52-61, also in the Oxford Mabinogion, pp. 297-308, from the Red Book of Hergest of the fourteenth century. The letter (a, b, c) added is intended to indicate the order of the three parts of the Triad, for it is not the same in all the series. Let me here remark in a general way that the former fondness of the Welsh for Triads was not peculiar to them. The Irish also must have been at one time addicted to this grouping. Witness the Triad of Cleverest Countings, in the Book of the Dun Cow, fol. 58a, and the Triad of the Blemishes of the Women of Ulster, ib. 43b.

[79] As to the names Drystan (also Trystan) and Essyllt, see the footnote on p. 480 above.

[80] This was meant to explain the unusual term g6rdueichyat, also written g6rdueichat, g6rueichyat, and gwrddfeichiad. This last comes in the modern spelling of iii. 101, where this clause is not put in the middle of the Triad but at the end.

[81] The editor of this version seems to have supposed Pendaran to have been a place in Dyfed! But his ignorance leaves us no evidence that he had a different story before him.

[82] This word is found written in Mod. Welsh Annwfn, but it has been mostly superseded by the curtailed form Annwn, which appears twice in the Mabinogi of Math. These words have been studied by M. Gaidoz in Meyer and Stern's Zeitschrift für Celtische Philologie, i. 29-34, where he equates Annwfn with the Breton anauon, which is a plural used collectively for the souls of the departed, the other world. His view, however, of these interesting words has since been mentioned in the same Zeitschrift, iii. 184-5, and opposed in the Annales de Bretagne, xi. 488.

[83] Edited by Professor Kuno Meyer (London, 1892): see for instance pp. 76-8.

[84] See Windisch's Irische Texte, p. 256, and now the Irish Text Society's Fled Bricrend, edited with a translation by George Henderson, pp. 8, 9.

[85] Windisch, ibid. pp. 99-105.

[86] See the Oxford Mabinogion, p. 196, and Guest's trans., i. 302, where the Welsh words a gol6ython o gic meluoch are rendered 'and collops of the flesh of the wild boar,' which can hardly be correct; for the mel in mel-uoch, or mel-foch in the modern spelling, is the equivalent of the Irish melg, 'milk.' So the word must refer either to a pig that had been fed on cows' milk or else a sucking pig. The former is the more probable meaning, but one is not helped to decide by the fact, that the word is still sometimes used in books by writers who imagine that they have here the word mel, 'honey,' and that the compound means pigs whose flesh is as sweet as honey: see Dr. Pughe's Dictionary, where melfoch is rendered 'honey swine,' whatever that may mean.

[87] Windisch's Irische Texte, p. 133, where laith lemnacht = Welsh llaeth llefrith, 'sweet milk.'

[88] Collfrewi was probably, like Gwenfrewi, a woman's name: this is a point of some importance when taken in connexion with what was said at p. 326 above as to Gwydion and Coll's magic.

[89] This reminds one of Geoffrey of Monmouth's Henvinus, whom he makes into dux Cornubiæ and father of Cunedagius or Cuneda: see ii. 12, 15. Probably Geoffrey's connecting such names as those of Cuneda and Dyfnwal Moelmud (ii. 17) with Cornwall is due to the fact, that the name of the Dumnonia of the North had been forgotten long before that of the Dumnonia to be identified with Devon and Cornwall.

[90] See the Oxford Mabinogion, p. 104, and the Oxford Bruts, p. 292.

[91] See the Oxford Bruts, pp. 299, 317, 345-6, 348, 384. I learn from Prof. Anwyl that Castell Penwedig is still remembered at Llanfihangel Genau'r Glyn as the old name of Castell Gwallter in that parish.

[92] See his note in Owen's Pembrokeshire, p. 237, where he also notices Aber Tarogi, and the editor's notes to p. 55.

[93] Mergaed for Mengwaed hardly requires any explanation; and as to Breat or rather Vreat, as it occurs in mutation, we have only to suppose the original carelessly written Vreac for Vreach, and we have the usual error of neglecting the stroke indicating the n, and the very common one of confounding c with t. This first-mentioned name should possibly be analysed into Mengw-aed or Menw-aed for an Irish Menb-aed, with the menb, 'little,' noticed at p. 510 below; in that case one might compare such compounds of Aed as Beo-aed and Lug-aed in the Martyrology of Gorman. Should this prove well founded the Mod. Welsh transcription of Menwaed should be Menwaed. I have had the use of other versions of the Triads from MSS. in the Peniarth collection; but they contribute nothing of any great importance as regards the proper names in the passages here in question.

[94] See the Oxford Mabinogion, pp. 41, 98, and Guest's trans., iii. 313.

[95] See Geoffrey's Historia Regum Britanniæ, vi. 19, viii. 1, 2; also Giraldus, Itinerarium Kambriæ, ii. 8 (p. 133).

[96] Itinerarium Kambriæ, ii. 9 (p. 136).

[97] Menw's name is to be equated with the Irish word menb, 'little, small,' and connected with the Welsh derivative di-fenw-i, 'belittling or reviling': it will be seen that he takes the form of a bird, and his designation Menw fab Teirgwaed might perhaps be rendered 'Little, son of Three-Cries.'

[98] Identified by Professor Kuno Meyer in the Transactions of the Cymmrodorion Society, 1895-6, p. 73, with a place in Leinster called Sescenn Uairbeóil, 'the Marsh of Uairbhél,' where Uairbhél may possibly be a man's name, but more likely that of a pass or gap described as Cold-mouth: compare the Slack or Sloc in the Isle of Man, called in Manx 'the big Mouth of the Wind.' The Irish name comes near in part to the Welsh Esgeir Oervel or Oerfel, which means 'the mountain Spur of cold Weather.'

[99] The word used in the text is ystyr, which now means 'meaning or signification'; but it is there used in the sense of 'history,' or of the Latin 'historia,' from which it is probably borrowed.

[100] In the original his designation is Gwrhyr Gwalstawt Ieithoed, and the man so called is in the Kulhwch credited with the mastery of all languages, including those of certain birds and quadrupeds. Gwalstawt, found written also gwalstot, is the Anglo-Saxon word wealhstód, 'an interpreter,' borrowed. The name Gwrhyr is possibly identical with that of Ferghoir, borne by the Stentor of Fionn mac Cumhaill's following. Ferghoir's every shout is said to have been audible over three cantreds. Naturally one who was to parley with a savage host had good reason to cultivate a far-reaching voice, if he wished to be certain of returning to his friends. For more about it see the footnote at p. 489 of my Hibbert Lectures.

[101] The original has Pelumyawc, p. 138, and the name occurs in the (Red Book) Bruts, p. 355, as Pelunyawc, and p. 411, as Pelunea(wc) between the commots of Amgoed and Velfrey. The identification here suggested comes from Mr. Phillimore, who has seen that Peuliniawc must be a derivative from the name Paulinus, that is of the Paulinus, probably, who is mentioned in an ancient inscription at Llandysilio. There are other churches called after Tysilio, so this one used to be distinguished as Llandysilio yn Nyfed, that is, Llandysilio-in-Dyfed; but the pronunciation was much the same as if it had been written Llandysilio yn Yfed, meaning 'Llandysilio a-drinking,' 'whereof arose a merrye jest,' as George Owen tells us in his Pembrokeshire, p. 9. It is now sometimes called Llandysilio'r Gynffon, or 'Llandysilio of the Tail,' from the situation of a part of the parish on a strip, as it were a tail, of Carmarthenshire land running into Pembrokeshire.

[102] This Aber Towy appears to have been a town with a harbour in 1042, for we read in Brut y Tywysogion of a cruel engagement fought there between Gruffyd ab Llewelyn and Howel ab Edwin, who, with Irish auxiliaries, tried to effect a landing. Not long ago a storm, carrying away the accumulation of sand, laid bare a good deal of the site. It is to be hoped that excavations will be made soon on the spot.

[103] See the Transactions of the Cymmrodorion, 1894-5, pp. 146-7. There are a good many clyns about South Wales, but our etymologists are careful to have them in most cases written glyn, 'a glen.' Our story, however, shows that the word came under the influence of glyn long ago, for it should be, when accented, clûn, corresponding to Irish cluain, 'a meadow.' We have it as clun in Clun Kein in the Black Book, p. 34b, where I guess it to mean the place now called Cilcain, 'Kilken' in Flintshire, which is accented on the first syllabic; and we have had it in y Clun Hir, 'the Long Meadow,' mentioned above at p. 22.

[104] Cas Llychwr, 'Loughor Castle,' is supposed to involve in its Llychwr, Llwchwr, or Loughor, the name of the place in the Antoninus Itinerary, 484, 1, to wit Leucarum; but the guttural spirant ch between vowels in Llychwr argues a phonetic process which was Goidelic rather than Brythonic.

[105] Llwydawc Gouynnyat would seem to mean Llwydawc the Asker or Demander, and the epithet occurs also in the Kulhwch in the name Gallcoyt Gouynynat (Mabinogion, 106), to be read doubtless G. Gouynnyat, 'G. who asks or demands': possibly one should rather compare with Go-uynnyat the word tra-mynyat, 'a wild boar': see Williams' Seint Greal, pp. 374, 381. However, the epithets in the Twrch Trwyth story do not count so far as concerns the place-names derived.

[106] Other instances of the like shortening occur in words like cefnder, 'a cousin,' for cefnderw, and ardel, 'to own,' for ardelw. As to Amman, it enters, also, into a group of Glamorganshire place-names: witness Aber Amman and Cwm Amman, near Aberdare.

[107] It should perhaps be looked for near Brechfa, where there is a Hafod Grugyn, and, as I am told, a Garth also which is, however, not further defined. For it appears that both Brechfa and Cayo, though now in Carmarthenshire, once belonged to Keredigion: see Owen's Pembrokeshire, p. 216. But perhaps another spot should be considered: J. D. Rhys, the grammarian (p. 22 above), gives in the Peniarth MS. 118 a list of caers or castles called after giants, and among them is that of Grugyn in the parish, he says, of 'Llan Hilar.' I have, however, not been able to hear of any trace of the name there, though I should guess the spot to have been Pen y Castell, called in English Castle Hill, the residence of Mr. Loxdale in the parish of Llanilar, near Aberystwyth.

[108] I have re-examined the passage, and I have no doubt that the editors were wrong in printing Gregyn: the manuscript has Grugyn, which comes in the last line of column 841. Now besides that the line is in part somewhat faint, the scribe has evidently omitted something from the original story, and I guess that the lacuna occurs in the first line of the next column after the words y llas, 'was killed,' which seem to end the story of Grugyn.

[109] Those who have discovered an independent Welsh appellative wy meaning water are not to be reasoned with. The Welsh wy only means an egg, while the meaning of Gwy as the name of the Wye has still to be discovered.

[110] This name also occurs in a passage quoted in Jones' Brecknock, ii. 501, from a Carte MS. which he treats as relating to the year 1234: the MS. is said to be at the Bodleian, though I have not succeeded in tracing it. But Jones gives Villa de Ystraddewi, and speaks of a chapel of St. John's of Stradtewi, which must have been St. John's Church, at Tretower, one of the ecclesiastical districts of Cwm Du: see also p. 497. The name is probably to be treated as Strad or Strat d'Ewe.

[111] A river may in Welsh be briefly called after anybody or anything. Thus in North Cardiganshire there is a stream called Einon, that is to say 'Einion's river,' and the flat land on both sides of it is called Ystrad Einon, which looks as if one might translate it Einion's Strath, but it means the Strath of Einion's river, or of the stream called Einon, as one will at once see from the upper course of the water being called Blaen Einon, which can only mean the upper course of the Einon river. So here yw is in English 'yew,' but Ystrad Yw and Llygad Yw have to be rendered the Strath of the Yew burn and the Eye of the Yew burn respectively. It is moreover felt by the Welsh-speaking people of the district that yw is the plural of ywen, 'a single yew,' and as there is only one yew at the source somebody had the brilliant idea of making the name right by calling it Ywen, and this has got into the maps as Ewyn, as though it were the Welsh word for foam. Who began it I cannot say, but Theophilus Jones has it in his History of the County of Brecknock, published in 1809. Nevertheless the name is still Yw, not Ywen or Ewyn, in the Welsh of the district, though Lewis gives it as Ywen in his article on Llanvihangel-Cwm-Du.

[112] For exact information as to the Gaer, the Yw, and Llygad Yw, I am indebted chiefly to the courtesy of Lord Glanusk, the owner of that historic strath, and to the Rector of Llansantffread, who made a special visit to Llygad Yw for me; also to Mr. Francis Evans, of the Farmers' Arms at the Bwlch, who would be glad to change the name Llygad Yw into Llygad dan yr Ywen, 'the Source beneath the Yew-tree,' partly on account of the position 'of the spring emanating under the but of the yew tree,' and partly because there is only a single yew there. Theophilus Jones complained a century ago that the Gaer in Ystrad Yw had not attracted the attention it deserved; and I have been greatly disappointed to find that the Cambrian Archæological Association has had nothing to say of it. At any rate, I have tried the Index of its proceedings and found only a single mention of it. The whole district is said to teem with antiquities, Celtic, Roman, and Norman.

[113] Theophilus Jones, in his Brecknockshire, ii. 502, describes Miarth or Myarth as a 'very extensive' camp, and proceeds as follows:--'Another British camp of less extent is seen on a knoll on Pentir hill, westward of the Rhiangoll and the parish church of Cwmdu, above a wood called Coed y Gaer, and nearly opposite to the peak or summit called Cloch y Pibwr, or the piper's call.' This would probably be more accurately rendered the Piper's Rock or Stone, with cloch treated as the Goidelic word for a stone rather than the Brythonic word for a bell: how many more clochs in our place-names are Goidelic?

[114] The Twrch would seem to have crossed somewhere opposite the mouth of the Wye, let us say not very far from Aust; but he escapes to Cornwall without anything happening to him, so we are left without any indication whether the story originally regarded Kernyw as including the Penrhyn Awstin of the Coll story given at p. 503.

[115] For this suggestion I am indebted to the Rev. Dr. Gaster in the Cymmrodorion's Transactions for 1894-5, p. 34, and also for references in point to M. Cosquin's Contes Populaires de la Lorraine, i. 134, 141, 152. Compare also such Gaelic stories as that of the Bodach Glas, translated by Mrs. Mackellar, in the Celtic Magazine, xii. 12-6, 57-64.

[116] In some native Welsh words we have an option between a prefix ym and am, an option arising out of the fact that originally it was neither ym nor am, but m, for an earlier mbi, of the same origin as Latin ambi and Greek amphi, 'around, about.' The article, its meaning in the combination in banbh being forgotten, would fall under the influence of the analogy of the prefix, now am or ym, so far as the pronunciation was concerned.

[117] Possibly the benwic was thrown in to correct the reckoning when the redactor discovered, as he thought, that he had one too many to account for: it has been pointed out that he had forgotten that one had been killed in Ireland.

[118] It is just possible, however, that in an older version it was named, and that the place was no other than the rock just above Ystrad Yw, called Craig Lwyd or, as it is said to be pronounced, Craig Llwyd. If so, Llwyd would seem to have been substituted for the dissyllable Llwydog: compare the same person called Llwyt and Llwydeu in the Mabinogion, pp. 57, 110, 136.

[119] The name is well known in that of Llanrhaiadr yn Mochnant, 'Llanrhaiadr in Mochnant,' in the north of Montgomeryshire.

[120] Between Colwyn Bay and Llandudno Junction, on the Chester and Holyhead line of railway.

[121] I have discussed some of the traces of the Goidels in Wales in the Arch. Camb. for 1895, pp. 18-39, 264-302; 1899, pp. 160-7.

[122] In fact the genitive Grúcind occurs in the Book of Leinster, fo. 359a.

[123] The sort of question one would like to ask in that district is, whether there is a spot there called Bed y Rhyswyr, Carn y Rhyswyr, or the like. The word rhyswr is found applied to Arthur himself in the Life of Gruffyd ab Cynan, as the equivalent probably of the Latin Arthur Miles (p. 538 below): see the Myvyrian Archaiology, ii. 590. Similarly the soldiers or champions of Christ are called rys6yr crist in the Welsh Life of St. David: see the Elucidarium and other Tracts (in the Anecdota Oxoniensia), p. 118.

[124] Rudvyw Rys would be in Modern Welsh Rhudfyw Rys, and probably means Rhudfyw the Champion or Fighter, as Rhys is likely to have been synonymous with rhyswr. The corresponding Irish name was Russ or Ross, genitive Rossa, and it appears to come from the same origin as Irish ross, 'a headland, a forest,' Welsh rhos, 'moorland, uncultivated ground.' The original meaning was presumably 'exposed or open and untilled land'; and Stokes supposes the word to stand for an early (p)ro-sto- with sto of the same origin as Latin sto, 'I stand,' and as the English word stand itself. In that case Ros, genitive Rossa, Welsh Rhys, would mean one who stands out to fight, a prostatês, so to say. But not only are these words of a different declension implying a nominative Ro-stus, but the Welsh one must have been once accented Ro-stús on the ending which is now lost, otherwise there is no accounting for the change of the remaining vowel into y. Other instances postulating an early Welsh accentuation of the same kind are very probably llyg, 'a fieldmouse,' Irish luch, 'a mouse'; pryd, 'form,' Irish cruth; pryf, 'a worm,' Irish cruim; so also with ych, 'an ox,' and nyth, 'a nest,' Irish nett, genitive nitt, derived by Stokes from nizdo-, which, however, must have been oxytone, like the corresponding Sanskrit nidhá. There is one very interesting compound of rhys, namely the saint's name Rhwydrys, as it were Redo-rostus to be compared with Gaulish Eporedo-rix, which is found in Irish analysed into rí Eochraidhi, designating the fairy king who was father to Étáin: see Windisch's Irische Texte, p. 119. Bledrws, Bledrus, as contrasted with Bledrys, Bledris, postulate Goidelic accentuation, while one has to treat Bledruis as a compromise between Bledrws and Bledris, unless it be due to misreading a Bledruif (Book of Llan Dâv, pp. 185, 221-2, and Arch. Camb. for 1875, p. 370). The Goidelic accent at an early date moved to first syllables, hence cruth (with its vowel influenced by the u of a stem qurt) under the stress accent, became, when unstressed, cridh (from a simplified stem crt) as in Noicride (also Nóicrothach, Windisch, ibid., pp. 259, 261, 266) and Luicridh (Four Masters, A.D. 748), Luccraid, genitive Luccraide (Book of Leinster, 359f), Luguqurit- in Ogam.

[125] These operations cannot have been the first of the kind in the district, as a writer in the Archæologia Cambrensis for 1862, pp. 159-60, in extracting a note from the Proceedings of the Society of Antiquaries (series II, vol. i. p. 10) relative to the discovery of the canoe, adds a statement based on the same volume, p. 161, to the effect that 'within half a mile of Llyn Llydaw there are the remains of a British town, not marked in the ordnance map, comprising the foundations of numerous circular dwellings. In some of them quantities of the refuse of copper smeltings were found. This town should be visited and examined with care by some of the members of our Association.' This was written not far short of forty years ago; but I am not aware that the Association has done anything positive as yet in this matter.

[126] According to Jenkins' Bed Gelert, p. 300, the canoe was subsequently sold for a substantial price, and nobody seems to know what has eventually become of it. It is to be hoped this is not correct.

[127] See Holder's Alt-celtischer Sprachschatz, s. v. Litavia.

[128] For these notes I am indebted to Williams' Dictionary of Eminent Welshmen, and to Rees' Welsh Saints, pp. 187, 191; for our Paulinus is not yet recognized in the Dictionary of Christian Biography. His day was Nov. 22.

[129] There are two other inscriptions in South Wales which contain the name Paulinus, one on a stone found in the neighbourhood of Port Talbot in Glamorgan, reading Hic iacit Cantusus Pater Paulinus, which seems to imply that Paulinus set up the stone to the memory of a son of his named Cantusus. The other, found on the site of the extinct church of Llanwrthwl, near Dolau Cothi in Carmarthenshire, is a remarkable one in a kind of hexameter to the following effect:--

Servatur fidæi patrieque semper amator Hic Paulinus iacit cultor pientisimus æqui.

Whether we have one or two or three Paulini in these inscriptions I cannot say. Welsh writers, however, have made the name sometimes into Pawl Hên, 'Paul the Aged,' but, so far as I can see, without rhyme or reason.

[130] Since I chanced on this inscription my friend Professor Lindsay of St. Andrews has called my attention to Plautus' Asinaria, 499 (II. iv. 92), where one reads, Periphanes Rhodo mercator dives, 'Periphanes a wealthy merchant of Rhodes'; he finds also Æsculapius Epidauro (Arnobius, 278. 18), and elsewhere Nepos Philippis and Priscus Vienna.

[131] See Stokes' Patrick, pp. 16, 412.

[132] This will give the reader some idea of the pre-Norman orthography of Welsh, with l for the sound of ll and b for that of v.

[133] The softening of Cafall to Gafall could not take place after the masculine corn, 'a horn'; but it was just right after the feminine carn, 'a cairn.' So here corn is doubtless a colloquial corruption; and so is probably the t at the end, for as llt has frequently been reduced to ll, as in cyfaill, 'a friend,' from the older cyfaillt, in Medieval Irish comalta, 'a foster brother or sister,' the language has sometimes reversed the process, as when one hears hollt for holl, 'all,' or reads fferyllt, 'alchemist, chemist,' for fferyll from Vergilius. The Nennian orthography does not much trouble itself to distinguish between l and ll, and even when Carn Cabal was written the pronunciation was probably Carn Gavall, the mutation being ignored in the spelling, which frequently happens in the case even of Welsh people who never fail to mutate their consonants in speaking. Lastly, though it was a dog that was called Cafall, it is remarkable that the word has exactly the form taken by caballus in Welsh: for cafall, as meaning some sort of a horse, see Silvan Evans' Geiriadur.

[134] An instance or two of Trwyd will be found in a note by Silvan Evans in Skene's Four Ancient Books of Wales, ii. 393.

[135] For more about these names and kindred ones, see a note of mine in the Arch. Cambrensis, 1898, pp. 61-3.

[136] See my Hibbert Lectures, pp. 398-401.

[137] See the Black Book of Carmarthen in Evans' facsimile, p. 47b; Thomas Stephens' Gododin, p. 146; Dent's Malory, preface, p. xxvi; and Skene's Four Ancient Books of Wales, ii. 51, 63, 155.

[138] See the Göttingische gelehrte Anzeigen for 1890, p. 512.

[139] See De Courson's Cartulaire de l'abbaye de Redon, pp. 163, 186.

[140] See Reeves' note to the passage just cited in his edition of Adamnan's Vita, pp. 6, 7.

[141] Here possibly one might mention likewise Gilmin Troetu or Troeddu, 'Gilmin of the Black Foot,' the legendary ancestor (p. 444) of the Wynns of Glyn Llifon, in Carnarvonshire. So the name might be a shortening of some such a combination as Gilla-min, 'the attendant of Min or Men,' a name we have also in Mocu-Min, 'Min's Kin,' a family or sept so called more than once by Adamnan. Perhaps one would also be right in regarding as of similar origin the name of Gilberd or Gilbert, son of Cadgyffro, who is mentioned in the Kulhwch, and in the Black Book, fo. 14b: at any rate I am not convinced that the name is to be identified with the Gillebert of the Normans, unless that was itself derived from Celtic. But there is a discrepancy between Gilmin, Gilbert, with unmutated m and b, and Gilvaethwy with its mutation consonant v. In all three, however, Gil, had it been Welsh, would probably have appeared as Gill, as indicated by the name Gilla in the Kulhwch (Oxford Mabinogion, p. 110), in which we seem to have the later form of the old name Gildas. Compare such Irish instances as Fiachna and Cera, which seem to imply stems originally ending in -asa-s (masculine) and -asa (feminine); and see the Journal of the Royal Society of Antiquaries of Ireland, 1899, P. 402.

[142] An article in the Rennes Dindsenchas is devoted to Liath: see the Rev. Celtique, xvi. 78-9. As to Celtchar, genitive Celtchair, the name would seem to have meant 'him who is fond of concealment.' The Mabinogi form of the Welsh name is Llwyt uab kil coet, which literally meant 'Ll. son of (him of) the Retreat of the Wood.' But in the Twrch Trwyth story, under a slightly different form of designation, we appear to have the same person as Llwydeu mab kelcoet and Llwydeu mab kel coet, which would seem to mean 'Ll. son of (him of) the Hidden Wood.' It looks as if the bilingual story-teller of the language transition had not been able to give up the cel of Celtchar at the same time that he rendered celt by coet, 'wood or trees,' as if identifying it with cailt: witness the Medieval Irish caill, 'a wood or forest,' dative plural cailtib, derivative adjective caillteamhuil, 'silvester'; and see Windisch's Irische Texte, p. 410, s. v. caill.

[143] Windisch's Irische Texte, p. 217, and the Book of the Dun Cow, fo. 47b.

[144] There has been a good deal of confusion as to the name Llyr: thus for instance, the Welsh translations of Geoffrey of Monmouth make the Leir of his Latin into Llyr, and the personage intended is represented as the father of three daughters named Gonerilla, Regan, and Cordeilla or Cordelia. But Cordelia is probably the Creurdilad of the Black Book, p. 49b, and the Creidylat of the Kulhwch story (the Oxford Mabinogion, pp. 113, 134), and her father was Llûd Llawereint (= Irish Nuada Airgetlám) and not Llyr. Then as to the Leir of Geoffrey's Latin, that name looks as if given its form on the strength of the legr- of Legraceaster, the Anglo-Saxon name of the town now called Leicester, of which William of Malmesbury (Gesta Pontificum, § 176) says, Legrecestra est civitas antiqua in Mediterraneis Anglis, a Legra fluvio præterfluente sic vocata. Mr. Stevenson regards Legra as an old name of the Soar, and as surviving in that of the village of Leire, spelled Legre in Domesday. It seems to point back to a Legere or Ligere, which recalls Liger, 'the Loire.'

[145] I say in that case, as this is not quite conclusive; for Welsh has an appellative llyr, 'mare, æquor,' which may be a generalizing of Llyr; or else it may represent an early lerio-s from lero-s (see p. 549 below), and our Llyr may possibly be this and not the Irish genitive Lir retained as Llyr. That, however, seems to me improbable on the whole.

[146] Here it is relevant to direct the reader's attention to Nutt's Legend of the Holy Grail, p. 28, where, in giving an abstract of the Petit saint Graal, he speaks of the Brân of that romance, in French Bron, nominative Brons, as having the keeping of the Grail and dwelling 'in these isles of Ireland.'

[147] The Dôn and Llyr groups are not brought into conflict or even placed in contact with one another; and the reason seems to be that the story-teller wanted to introduce the sons of Beli as supreme in Britain after the death of Brân. Beli and his sons are also represented in Maxen's Dream as ruling over Britain when the Roman conqueror arrives. What is to be made of Beli may be learnt from The Welsh People, pp. 41-3.

[148] These things one learns about Lir from the story mentioned in the text as the 'Fate of the Children of Lir,' as to which it is right, however, to say that no ancient manuscript version is known: see M. d'Arbois dc Jubainville's Essai d'un Catalogue de la Litérature épique de l'Irlande, p. 8.

[149] See Skene's Four Ancient Books of Wales, ii. 303, also 108-9, where the fragment of the poem as given in the Book of Taliessin is printed. The line here quoted has been rendered in vol. i. 286, 'With Matheu and Govannon,' which places the old pagan Gofannon in rather unexpected company. A few lines later in the poem mention is made of a Kaer Gofannon: where was that? Skene, in a note on it (ii. 452), says that 'In an old list of the churches of Linlithgow, printed by Theiner, appears Vicaria de Gumanyn. The place meant is probably Dalmeny, on the Firth of Forth, formerly called Dumanyn.' This is interesting only as showing that Gumanyn is probably to be construed Dumanyn, and that Dalmeny represents an ancient Dún Manann in a neighbourhood where one already has Clach Manann, 'the stone of Manau,' and Sliabh Manann, 'Mountain of Manau' now respectively Clackmannan and Slamannan, in what Nennius calls Manau Guotodin.

[150] This occurred unrecognized and, therefore, unaltered by the scribe of the Nennian Pedigree no. xvi in the Cymmrodor, ix. 176, as he found it written in an old spelling, Louhen. map. Guid gen. map. Caratauc. map. Cinbelin, where Caradog is made father of Gwydion; for in Guid-gen we seem to have the compound name which suggested Gwydion. This agrees with the fact that the Mabinogi of Math treats Gwydion as the father of Llew Llawgyffes; but the pedigree itself seems to have been strangely put together.

[151] See Bertrand's Religion des Gaulois, pp. 314-9, 343-5, and especially the plates.

[152] The Oxford Mabinogion, pp. 40-3; Guest's Mabinogion, iii. 124-8.

[153] See Louis Leger's Cyrille et Méthode (Paris, 1868), p. 22.

[154] See Pertz, Monumenta Germaniæ Historia Scriptorum, xii. 794. The whole passage is worth quoting; it runs thus: Erat autem simulacrum triceps, quod in uno corpore tria capita habens Triglaus vocabatur; quod solum accipiens, ipsa capitella sibi cohærentia, corpore comminuto, secum inde quasi pro tropheo asportavit, et postea Romam pro argumento conversionis illorum transmisit.

[155] See The Welsh People, pp. 56-7.

[156] The Oxford Mabinogion, p. 147; Guest's Mabinogion, ii. 398.

[157] This may have meant the 'Blue Slate or Flagstone'; but there is no telling so long as the place is not identified. It may have been in the Pictish district of Galloway, or else somewhere beyond the Forth. Query whether it was the same place as Llech Gelydon in Prydyn, mentioned in Boned y Saint: see the Myvyrian Archaiology, ii. 49.

[158] The story of Kulhwch and Olwen has a different legend which represents Nynio and Peibio changed by the Almighty into two oxen called Ychen Banna6c: see the Oxford Mabinogion, p. 121, also my Arthurian Legend, p. 304, and the remarks which are to follow in this chapter with respect to those oxen.

[159] For the story in Welsh see the Iolo MSS., pp. 193-4, where a footnote tells the reader that it was copied from the book of 'Iaco ab Dewi.' From his father's manuscript, Taliesin Williams printed an abstract in English in his notes to his poem entitled the Doom of Colyn Dolphyn (London, 1837), pp. 119-20, from which it will be found translated into German in the notes to San-Marte's Geoffrey of Monmouth's Historia Regum Britanniæ, pp. 402-3.

[160] Oxford Bruts, p. 213: compare p. 146, together with Geoffrey's Latin, vii. 3, x. 3.

[161] See Kölbing's Altenglische Bibliothek, the fifth volume of which consists of Libeaus Desconus, edited by Max Kaluza (Leipsic, 1890), lines 163, 591, and Introduction, p. cxxxxiv. For calling my attention to this, I have to thank my friend, Mr. Henry Bradley.

[162] Malory's Morte Darthur, i. 27: see also i. 17-8, 28; ii. 6, 8-9.

[163] See Evans' Autotype Facsimile, fo. 33a: could the spot so called (in the Welsh text argel Ardudwy) be somewhere in the neighbourhood of Llyn Irdyn (p. 148), a district said to be rich in the remains of a prehistoric antiquity? J. Evans, author of the North Wales volume of the Beauties of England and Wales, says, after hurriedly enumerating such antiquities, p. 909: 'Perhaps in no part of Britain is there still remaining such an assemblage of relicks belonging to druidical rites and customs as are found in this place, and the adjacent parts.'

[164] As to Rion, see Gaston Paris and Ulrich's Merlin (Paris, 1886), i. 202, 239-46. Other instances will readily occur to the reader, such as the Domesday Roelend or Roelent for Rothelan, in Modern Welsh Rhudlan; but for more instances of this elision by French and Anglo-Norman scribes of vowel-flanked d and th, see Notes and Queries for Oct. 28, 1899, pp. 351-2, and Nov. 18, p. 415; also Vising's Étude sur le Dialecte anglo-normand du xije Siècle (Upsala, 1882), p. 88; and F. Hildebrand's article on Domesday, in the Zeitschrift für romanische Philologie, 1884, p. 360. According to Suchier in Gröber's Grundriss der rom. Philologie, i. 581, this process of elision became complete in the twelfth century: see also Schwan's Grammatik des Altfranzösischen (Leipsic, 1888), p. 65. For most of these references, I have to thank my friend and neighbour, Mr. Stevenson of Exeter College.

[165] It comes from the same Llwyd MS. which has already been cited at pp. 233-4: see the Cambrian Journal for 1859, pp. 209-10.

[166] I notice in the maps a spot called Panylau, which is nearer to Llyn Gwynain than to Llyn y Dinas.

[167] See Morris' Celtic Remains, s. v. Serigi, and the Iolo MSS., p. 81.

[168] The Iolo MSS., p. 81, have Syrigi Wydel son of Mwrchan son of Eurnach Hen.

[169] See Triads, ii. 12, and the Mabinogion, p. 301: in Triads, i. 72, iii. 86, instead of Solor we have Doler and Dolor.

[170] See the Oxford Mabinogion, pp. 125-8.

[171] Evans' Autotype Facsimile, fo. 48a; see also my preface to Dent's Malory, p. xxvii; likewise p. 457 above.

[172] See my Lectures on Welsh Philology, pp. 377-9; and, as to the Caer Gai tradition, the Arch. Camb. for 1850, p. 204, and Morris' Celtic Remains, p. 63. I may add as to Llanuwchllyn, that the oldest inhabitants pronounce that name Llanuwllyn.

[173] I cannot discover that it has ever been investigated by the Cambrian Archæological Association or any other antiquaries. Compare the case of the neighbouring site with the traces of the copper smeltings mentioned in the note on p. 532 above. To my knowledge the Cambrians have twice failed to make their way nearer to the ruins than Llanberis, or at most Llanberis Pass, significantly called in Welsh Pen Gorffwysfa for the older name Gorffwysfa Beris, 'Peris' Resting-place': thus we loyally follow the example of resting set by the saint, and leave alone the archæology of the district.

[174] The subject has been discussed at length by Mr. Jacobs, in a note to the legend, in his Celtic Fairy Tales, pp. 259-64; and quite recently by Mr. D. E. Jenkins in his Bed Gelert (Portmadoc, 1899), pp. 56-74.

[175] Professor J. Morris Jones, to whom I am indebted for the particulars connected with these names, informs me that the local pronunciation is Drónwy; but Mrs. Rhys remembers that, years ago, at Amlwch, it was always sounded Darónwy. The Professor also tells me that Dernog is never made into Dyrnog: the Kuwgh of the Record is doubtless to be corrected into Knwgh, and probably also Dornok into Dernok, which is the reading in the margin. Cornewe is doubtless the district name which we have still in Llanfair y'Nghornwy, 'St. Mary's in Cornwy': the mill is supposed to be that of Bodronyn.

[176] The Book of Llan Dáv has an old form Cinust for an earlier Cingust or Congust. The early Brythonic nominative must have been Cunogústu-s and the early Goidelic Cúnagusu-s, and from the difference of accentuation come the o of Conghus, Connws, and the y of the Welsh Cynwst: compare Irish Fergus and Welsh Gurgúst, later Gurúst (one syllable), whence Grwst, finally the accented rwst of Llanrwst, the name of a small town on the river Conwy. Moreover the accentuation Cúnogusi is the reason why it was not written Cunogussi: compare Bárrivendi and Véndubari in one and the same inscription from Carmarthenshire.

[177] Such as that of a holding called Wele Dauid ap Gwelsantfrait, the latter part of which is perversely written or wrongly read so for Gwas Sant Freit, a rendering into Welsh of the very Goidelic name, Mael-Brigte, 'Servant of St. Bridget.' This Wele, with Wele Conus and Wele More, is contained in the Extent marginally headed Darronwy cum Hameletta de Kuwghdernok.

[178] This comes in Triad i. 49 = ii. 40; as to which it is to be noted that the name is Catwallawn in i and ii, but Caswallawn in iii. 27, as in the Oxford Mabinogion.

[179] Serrigi, Serigi, or Syrigi looks like a Latin genitive torn out of its context, but derived in the last resort from the Norse name Sigtrygg-r, which the Four Masters give as Sitriucc or Sitriug: see their entries from 891 to 1091. The Scandinavians of Dublin and its neighbourhood were addicted to descents on the shores of North Wales; and we have possibly a trace of occupation by them in Gauell Seirith, 'Seirith's holding,' in the Record of Carnarvon, p. 63, where the place in question is represented as being in the manor of Cemmaes, in Anglesey. The name Seirith was probably that written by the Four Masters as Sichfraith Sichraidh (also Serridh, A. D. 971), that is to say the Norse Sigræd-r before it lost the f retained in its German equivalent Siegfried. We seem to detect Seirith later as Seri in place-names in Anglesey--as for example in the name of the farms called Seri Fawr and Seri Bach between Llandrygarn and Llannerch y Med, also in a Pen Seri, 'Seri's Knoll or Hill,' at Bryn Du, near Ty Croes station, and in another Pen Seri on Holyhead Island, between Holyhead and Llain Goch, on the way to the South Stack. Lastly Dugdale, v. 672b mentions a Claud Seri, 'Seri's Dyke or Ditch,' as being somewhere in the neighbourhood of Llanwnda, in Carnarvonshire--not very far perhaps from the Gwyrfai and the spot where the Iolo MSS. (pp. 81-2) represent Serrigi repulsed by Caswallon and driven back to Anglesey, previous to his being crushed at Cerrig y Gwydyl. The reader must, however, be warned that the modern Seri is sometimes pronounced Sieri or Sheri, which suggests the possibility of some of the instances involving rather a form of the English word sheriff.

[180] See my Hibbert Lectures, pp. 546-8.

[181] The case with regard to the extreme south of the Principality is somewhat similar; for inscriptions in Glamorgan seem to bring the last echoes there of Goidelic speech down to the seventh century: see the Archæologia Cambrensis for 1899, pp. 160-6.

[182] See Evans' Report on MSS. in the Welsh Language, p. 837, where the Welsh is quoted from p. 131 of the Peniarth MS. 134.

[183] See my Arthurian Legend, p. 70.

[184] See the Revue Celtique, ii. 197-9, where Dr. Stokes has published the original with a translation and notes; also p. 435 above.

[185] The gentlemen to whom I am chiefly indebted for the information embodied in the foregoing notes are the following four: the Rev. John Jones of Ystad Meurig, Professor Robert Williams of St. David's College, the Vicar of Llandewi Brefi, Mr. J. H. Davies of Cwrt Mawr and Lincoln's Inn (p. 354); and as to the 'wild cattle' story of Llyn Eidwen, Mr. J. E. Rogers of Aber Meurig is my authority.

[186] So I had it many years ago from an old woman from Llangeitho, and so Mr. J. G. Evans remembers his mother repeating it; but now it is made into Llan Dewi Brefi braith, with the mutations disregarded.

[187] See the Archæologia Cambrensis for 1868, p. 88.

[188] See ib. p. 87. I have ascertained on the best authority the identity of the present owner of the horn, though I have not succeeded in eliciting from him any reply to my inquiries. I conclude that there is something wrong with the postal service in my native county.

[189] Several passages bearing on the word bannog have been brought together in Silvan Evans' Geiriadur. He gives the meaning as 'high, lofty, prominent, conspicuous.' The word is derived from ban, 'a summit or peak,' plural bannau, so common in the names of hills and mountains in South Wales--as in y Fan in Carmarthenshire, Bannwchdeni (p. 22) in Breconshire, Pen y Bannau near Pont Rhyd Fendigaid in Cardiganshire, Bannau Brycheiniog and Bannau Sir Gaer, the mountains called in English the Beacons of Breconshire and Carmarthenshire respectively. In North Wales we have it possibly in the compound Tryfan, which the mapsters will have us call Tryfaen; and the corresponding word in Scotch Gaelic appears in such names as Ben Nevis and the like, while in Irish the word benn meant a horn or peak. I am, nevertheless, not at all sure that Ychen Bannog meant horned oxen or even tall and conspicuous oxen; for there is a Welsh word man, meaning a spot or mark (Latin menda), and the adjective was mannawc, mannog, 'spotted, marked, particoloured.' Now in the soft mutation all four words--ban, bannog, and man, mannog--would begin with f = v, which might help to confusion between them. This may be illustrated in a way from Williams' Seint Greal (pp. 88-92), where Gwalchmai has a dream in which he sees 150 bulls with spots or patches of colour on them, except three only which were 'without any spot in the world' (neb ryw vann or byt), or as it is also put 'without spot' (heb vann). This word vann, applied to the colour of the bulls, comes from the radical form mann; and the adjective was mannawc or mannog, which would mean spotted, particoloured, or having patches of colour. Now the oxen of Welsh legends are also sometimes called Ychen Mannog (pp. 131-2), and it is possible, that, whichever way the term is written, it should be interpreted to mean spotted, marked, or particoloured oxen. I take it also that Llan Dewi Frefi fraith was meant as synonymous with Llan Dewi Frefi fannog, which did not fit the rhyme. Lastly, the Dyfed use of the saying Fel dau ych bannog, 'Like two Bannog oxen,' in the sense of 'equal and inseparable companions' (as instanced in the Geiriadur), sounds like the antithesis of the passage in the Kulhwch (Mabinogion, p. 121). For there we have words to the following effect: 'Though thou shouldst get that, there is something which thou wilt not get, namely the two oxen of Bannog, the one on the other side of the Bannog mountain and the other on this side, and to bring them together to draw the same plough. They are, to wit, Nynio and Peibio, whom God fashioned into oxen for their sins.' Here the difficulty contemplated was not to separate the two, but to bring them together to work under the same yoke. This is more in harmony with the story of the mad quarrel between the two brother kings bearing those names as mentioned above.

[190] See the Revue Celtique, iii. 310, after Gruter, 570, 6.

[191] An important paper on the Tarvos Trigaranus, from the pen of M. Salomon Reinach, will be found in the Revue Celtique, xviii. 253-66; and M. d'A. de Jubainville's remarkable equations are to be read in the same periodical, xix. 245-50: see also xx. 374-5.

[192] This, we are told, was a stone with a hollow in it for pounding corn, so as to separate the husks from the grain; and such a stone stood formerly somewhere near the door of every farm house in Scotland.

[193] The editor here explains in a note that 'this was a common saying formerly, when people were heard to regret trifles.'

[194] I have heard of this belief in Wales late in the sixties; but the presence was assumed to be that of a witch, not of a fairy.

[195] The word twt, 'tidy,' is another vocable which has found its way into Wales from the western counties of England; and though its meaning is more universally that of 'tidy or natty,' the term gwas twt, which in North Cardiganshire means a youth who is ready to run on all kinds of errands, would seem to bring us to its earlier meaning of the French tout--as if gwas twt might be rendered a 'garçon à tout'--which survives as tote in the counties of Gloucester and Hereford, as I am informed by Professor Wright. Possibly, however, one may prefer to connect twt with the nautical English word taut; but we want more light. In any case one may venture to say that colloquial Welsh swarms with words whose origin is to be sought outside the Principality.

[196] See Folk-Lore for 1889, pp. 144-52.

[197] Ibid. for 1891, p. 246, where one will find this rhyme the subject of a note--rendered useless by a false reference--by Köhler; see also the same volume, p. 132, where Mr. Kirby gives more lines of the rhyme.

[198] See Choice Notes from 'Notes and Queries,' p. 35.

[199] A number of instructive instances will be found mentioned, and discussed in his wonted and lucid fashion, by Mr. Clodd in his Tom Tit Tot, pp. 80-105.

[200] The Welsh spelling is caws pob, 'baked (or roasted) cheese,' so called in parts of South Wales, such as Carmarthenshire, whereas in North Wales it is caws pobi. It is best known to Englishmen as 'Welsh rabbit,' which superior persons 'ruling the roast' in our kitchens choose to make into rarebit: how they would deal with 'Scotch woodcock' and 'Oxford hare,' I do not know. I should have mentioned that copies of the Hundred Mery Talys are exceedingly scarce, and that the above, which is the seventy-sixth in the collection, has here been copied from the Cymmrodor, iii. 115-6, where we have the following sapient note:--'Cause bobe, it will be observed, is St. Peter's rendering of the phrase Caws wedi ei bobi. The chief of the Apostles apparently had only a rather imperfect knowledge of Welsh, which is not to be wondered at, as we know that even his Hebrew was far from giving satisfaction to the priests of the capital.' From these words one can only say that St. Peter would seem to have known Welsh far better than the author of that note, and that he had acquired it from natives of South Wales, perhaps from the neighbourhood of Kidwelly. I have to thank my friend Mr. James Cotton for a version of the cheese story in the Bodleian Library, namely in Malone MS. 19 (p. 144), where a certain master at Winchester School has put it into elegiacs which make St. Peter cry out with the desired effect: Tostus io Walli, tostus modo caseus.

[201] See Choice Notes from 'Notes and Queries,' pp. 117-8.

[202] For instance, when Cúchulainn had fallen asleep under the effect of fairy music, Fergus warned his friends that he was not to be disturbed, as he seemed to be dreaming and seeing a vision: see Windisch's Irische Texte, p. 208; also the Revue Celtique, v. 231. For parallels to the two stories in this paragraph, see Tylor's first chapter on Animism in his Primitive Culture, and especially the legend of King Gunthram, i. 442.

[203] See Mr. Gomme's presidential address to the Folk-Lore Society, printed in Folk-Lore for 1892, pp. 6-7.

[204] See Sale's preliminary discourse to his translation of the Koran, § iv.

[205] Perhaps we may regard this as the more Goidelic account of Blodeuwed's origin: at any rate, traces of a different one have been noticed in a note at p. 439 above.

[206] One version of it is given in the Myvyrian Archaiology, i. 176-8; and two other versions are to be found in the Cymmrodor, viii. 177-89, where it is suggested that the author was Iolo Goch, who flourished in the fourteenth century. See also my Arthurian Legend, pp. 57-8.

[207] See also the notes on these passages, given in San-Marte's edition of Geoffrey, pp. 219, 463-5, and his Beiträge zur bretonischen und celtisch germanischen Heldensage (Quedlinburg and Leipsic, 1847), p. 81.

[208] See Choice Notes, pp. 69-70.

[209] See Wood-Martin's Pagan Ireland (London, 1895), p. 140.

[210] See Choice Notes, p. 61, where it is also stated that the country people in Yorkshire used to give the name of souls to certain night-flying white moths. See also the Athenæum, No. 1041, Oct. 9, 1847.

[211] For this also I am indebted to Wood-Martin's book, p. 140.

[212] See the Book of the Dun Cow, fo. 198, and Windisch's Irische Texte, pp. 136-45. An abstract of the story will be found in the Hibbert Lectures on Celtic Heathendom, p. 502.

[213] See the Book of the Dun Cow, fo. 129a-133a; Windisch's Irische Texte, pp. 117-33, more especially pp. 127-31; also my Arthurian Legend, pp. 29-33.

[214] See the Book of Taliessin, poem vii, in Skene's Four Ancient Books of Wales, ii. 136-7; also poem viii, p. 137 et seq.

[215] Some account of this process will be found in Elton's Origins of English History (London, 1882), p. 33, where he has drawn on Martin's Description of the Western Islands of Scotland, published in 1703: see pp. 204-5.

[216] For one or two instances of the nomenclature in question, see pp. 76-7 above.

[217] Sywedyd is probably a word of Goidelic origin: compare Irish súi, 'a sage,' genitive súad, and derivative súithe, 'wisdom.' Stokes suggests the derivation su-vet, in which case súi = su-vi, for su-viss = su-vet-s, and sú-ithe = suvetia, while the Welsh sywedyd is formally su-vetios or su-vetiios. Welsh has also syw, from súi, like dryw, 'a druid,' from Goidelic drúi. Syw, it is true, now only means elegant, tidy; but Dr. Davies of Mallwyd believed its original signification to have been 'sapiens, doctus, peritus.' The root vet is most probably to be identified with the wet of Med. Welsh gwet-id, 'a saying,' dy-wawt, 'dixit,' whence it appears that the bases were vet and vat, with the latter of which Irish fáith, 'a poet or prophet,' Latin vates, agrees, as also the Welsh gwawd, 'poetry, sarcasm,' and in Mod. Welsh, 'any kind of derision.' In the Book of Taliessin syw has, besides the plurals sywyon and sywydon (Skene, ii. 142, 152), possibly an older plural, sywet (p. 155) = su-vet-es, while for súithe = su-vetia we seem to have sywyd or sewyd (pp. 142, 152, 193); but all the passages in point are more or less obscure, I must confess.

[218] See the Book of Taliessin, in Skene's Four Ancient Books of Wales, ii. 130-1, 134, 142, 151-2, 155.

[219] As, for instance, in the account given of Uath mac Imomain in Fled Bricrenn: see the Book of the Dun Cow, fo. 110b, and Windisch's Irische Texte, p. 293.

[220] The Book of the Dun Cow, fo. 77a, and the Book of Leinster, fo. 75b: compare also the story of Tuan mac Cairill in the Book of the Dun Cow, fo. 16b, where the Tuatha Dé Danann are represented as Tuatha Dee ocus Ande, 'the tribes of gods and not-gods,' to whom one of the manuscripts adds a people of legendary Ireland called the Galiúin. See the story as recently edited by Professor Kuno Meyer in Nutt's Voyage of Bran, ii. 291-300, where, however, the sense of § 12 with its allusion to the fall of Lucifer is missed in the translation. It should read, I think, somewhat as follows:--'Of these are the Tuatha Dee and Ande, whose origin is unknown to the learned, except that they think it probable, judging from the intelligence of the Tuatha and their superiority in knowledge, that they belong to the exiles who came from heaven.'

[221] See Evans' Black Book of Carmarthen, fo. 33b; also the Mabinogion, pp. 104, 306. The Irish lucht cumachtai would be in Welsh literally rendered llwyth cyfoeth, 'the cyfoeth tribe or host,' as it were. For cyfoeth, in Med. Welsh, meant power or dominion, whence cyfoethog, 'powerful,' and holl-gyfoethog, 'almighty'; but in Mod. Welsh cyfoeth and cyfoethog have been degraded to mean 'riches' and 'rich' respectively. Now if we dropped the prefix cum from the Irish cumachtai, and its equivalent cyf from the Welsh cyfoeth, we should have lucht cumachtai reduced to an approximate analogy to llwyth Oeth, 'the Oeth tribe,' for which we have the attested equivalent Teulu Oeth, 'the Oeth household or family.' Oeth, however, seems to have meant powerful rather than power, and this seems to have been its force in Gwalchmai's poetry of the twelfth century, where I find it twice: see the Myvyrian Arch., i. 196b, 203a. In the former passage we have oeth dybydaf o dybwyf ryd, 'I shall be powerful if I be free,' and in the latter oeth ym uthrwyd, 'mightily was I astonished or dismayed.' An-oeth was the negative of oeth, and meant weak, feeble, frivolous: so we find its plural, anoetheu, applied in the story of Kulhwch to the strange quests on which Kulhwch had to engage himself and his friends, before he could hope to obtain Olwen to be his wife. This has its parallel in the use of the adjective gwan, 'weak,' in the following instance among them:--Arthur and his men were ready to set out in search of Mabon son of Modron, who was said to have been kidnapped, when only three nights old, from between his mother Modron and the wall; and though this had happened a fabulously long time before Arthur was born, nothing had ever been since heard of Mabon's fate. Now Arthur's men said that they would set out in search of him, but they considered that Arthur should not accompany them on feeble quests of the kind: their words were (p. 128), ny elli di uynet ath lu y geissa6 peth mor uan ar rei hynn, 'thou canst not go with thy army to seek a thing so weak as these are.' Here we have uan as the synonym of an-oeth; but Oeth ac Anoeth probably became a phrase which was seldom analysed or understood; so we have besides Teulu Oeth ac Anoeth, a Caer Oeth ac Anoeth, or fortress of O. and A., and a Carchar Caer Oeth ac Anoeth, or the Prison of Caer O. and A., which is more shortly designated also Carchar Oeth ac Anoeth, or the Prison of O. and A. A late account of the building of that strange prison and fortress by Manawydan is given in the Iolo MSS., pp. 185-6, 263, and it is needless to point out that Manawydan, son of Llyr, was no other than the Manannán mac Lir of Irish literature, the greatest wizard among the Tuatha Dé or Tuatha Dé Danann; for the practical equivalence of those names is proved by the Book of the Dun Cow, fo. 16b. For further details about Oeth and Anoeth, Silvan Evans' Geiriadur may be consulted, s. v. Anoeth, where instances are cited of the application of those terms to tilled land and wild or uncultivated land. Here the words seem to have the secondary meanings of profitable and unprofitable lands, respectively: compare a somewhat analogous use of grym, 'strength, force,' in a passage relating to the mutilated horses of Matholwch--hyt nad oed rym a ellit ar meirch, 'so that no use was possible in the case of the horses,' meaning that they were of no use whatever, or that they had been done for: see the Oxford Mabinogion, p. 29, and Lady Charlotte Guest's, iii. 107, where the translation 'and rendered them useless' is barely strong enough.

[222] It is right, however, to state that M. d'A. de Jubainville's account of the views of Erigena is challenged by Mr. Nutt, ii. 105.

[223] For instance, by Silvan Evans in his Geiriadur, where, s. v. dihaed, he suggests 'unmerited' or 'undeserved' as conveying the sense meant.

[224] The reader will find them quoted under the word Druida in Holder's Alt-celtischer Sprachschatz: see also M. Alexandre Bertrand's Religion des Gaulois, especially the chapter entitled Les Druides, pp. 252-76, and Nutt's Voyage of Bran, ii. 107-12.

[225] See Valerius Maximus, ii. 6. 10.

[226] See the Book of the Dun Cow, fo. 68a.

[227] Notably Johannes Schmidt in Kuhn's Zeitschrift, xxiii. 267, where he gives the following gradations of the stem in question:--1. anman; 2. anaman; 3. naman; 4. naman.

[228] See Clodd's Tom Tit Tot, p. 97.

[229] Tom Tit Tot, p. 89.

[230] The Oxford Mabinogion, p. 100.

[231] The Oxford Mabinogion, p. 35.

[232] As to Irish, I would not lay much stress on the question 'What is your name?' being put, in a fourteenth or fifteenth century version of the French story of Fierabras, as ca hainm tú?--literally, 'what name art thou?' see the Revue Celtique, xix. 28. It may be mentioned here that the Irish writers of glossaries had a remarkable way of appearing to identify words and things. Thus, for instance, Cormac has Cruimther .i. Gædelg indi as presbyter, which O'Donovan (edited by Stokes) has translated, p. 30, as 'Cruimther, i. e. the Gaelic of presbyter': literally it would be rather 'of the thing which is presbyter.' Similarly, Cormac's explanation of the Irish aiminn, now aoibhinn, 'delightful,' runs thus in Latin, Aimind ab eo quod est amoenum, 'from the word amoenus,' literally, 'from that which is amoenus.' But this construction is a favourite one of Latin grammarians, and instances will be found in Professor Lindsay's Latin Language (Oxford, 1894), pp. 26, 28, 42, 53. On calling his attention to it, he kindly informed me that it can be traced as far back as Varro, from whose Lingua Latina, vi. 4, he cites Meridies ab eo quod medius dies. So in this matter, Irish writers have merely imitated their Latin models; and one detects a trace of the same imitation in some of the Old Welsh glosses, for instance in the Juvencus Codex, where we have XPS explained as irhinn issid crist, 'that which is Christ,' evidently meaning, 'the word Christos or Christus.' So with regia, rendered by gulat, 'a state or country,' in celsi thronus est cui regia caeli; which is glossed issit padiu itau gulat, 'that is the word gulat for him' = 'he means his country': see Kuhn's Beiträge, iv. 396, 411.

[233] Some instances in point, accompanied with comments on certain eminently instructive practices and theories of the Church, will be found in Clodd's Tom Tit Tot, pp. 100-5.

[234] For some instances of name-giving by the druid, the reader may consult The Welsh People, pp. 66-70; and druidic baptism will be found alluded to in Stokes' edition of Coir Anmann, and in Stokes and Windisch's Irische Texte, iii. 392, 423. See also the Revue Celtique, xix. 90.

[235] See The Welsh People, more especially pp. 71-4, where it has been attempted to discuss this question more at length.

[236] See Stokes' Cormac's Glossary, translated by O'Donovan, p. 87, and O'Curry's Manners and Customs of the Ancient Irish, ii. 218-9.

[237] See Mind for 1893, p. 390: the review is by Mr. A. T. Myers, and the title of the book noticed is La Pathologie des Émotions, Études physiologiques et cliniques, par Charles Féré, médecin de Bicêtre (Paris, 1892).

[238] See Frazer's Golden Bough, i. 9, where a few most instructive instances are given.

[239] See Guest's Mabinogion, iii. 255, where, however, Dôn is wrongly treated as a male.

[240] One has, however, to admit that the same agency may also mar the picture. Since the above was written I have read in Stokes' Festschrift, pp. 7-19, a very interesting article by L. Chr. Stern, in which he discusses some of the difficulties attaching to the term Tuatha Dé Danann. Among other things he suggests that there was a certain amount of confusion between Danann and dána, genitive of dán, 'art or profession'--the word meant also 'lot or destiny,' being probably of the same origin as the Latin donum, in Welsh dawn, which means a gift, and especially 'the gift of the gab.' But it would invert the natural sequence to suppose any such a formula as Tuatha Dé Dána to have preceded Tuatha Dé Danann; for why should anybody substitute an obscure vocable Danann for dána of well-known meaning? Dr. Stern has some doubts as to the Welsh Dôn being a female; but it would have been more satisfactory if he had proved his surmise, or at any rate shown that Dôn has nothing to do with Danann or Donann. I am satisfied with such a passage in the Mabinogi of Math as that where Gwydion, addressing Math, describes Arianrhod, daughter of Dôn, in the words, dy nith uerch dy ch6aer, 'thy niece daughter of thy sister': see the Mabinogion, p. 68, and, for similar references to other children of Dôn, consult pp. 59 and 65. Arianrhod is in the older Triads, i. 40,