Glen Discovery in GlenLyon
Discovery
About us
History
Contact

Watson's Celtic Placenames of Scotland

GENERAL SURVEY OF SCOTLAND NORTH OF FORTH


Original pagination [pp] from W.J. Watson, History of the Celtic Placenames of Scotland, 1926
(reprinted 1993 by BIRLINN, Edinburgh, ISBN 1 874744 06 8).

[N.B. To return from the endnotes to the main text, click on your browser's 'back' button.

[206]We have now surveyed in a general way the country south of the Wall between Forth and Clyde from the linguistic point of view; there remains something to say on the position north of the Wall, in what is called Pictland or in Gaelic Cruithentuath. Bede informs us that in his day the Firth of Forth divided the territory of the Angles from that of the Picts, while of old - long before his time - the Britons were divided from the Picts on the west by the Firth of Clyde. As in Bede's time and long before it the chief seat of the Britons of Strathclyde was at Dumbarton, on the north side of the estuary, it is not to be supposed that their territory ceased there; it probably extended round about Loch Lomond - the Lennox district - and into Menteith, while their linguistic influence may have extended well beyond the bounds of their territory.

In the Irish semi-historical literature there is an account of how Cruithneachán mac Lochit maic Cinge (or Inge) went from the sons of Míl, i.e. the Gael of Ireland, with the Britons of Fortriu to fight against the Saxons, and won land for themselves, namely Cruithentuath, and stayed among the Britons. This, says a writer in the Book of Ballymote, took place in the time of Erimon, that is to say, not long after the Gael arrived in Ireland. We are further told that the newcomers cleared a swordland (claideamlir) for themselves among the Britons (itir Breatnaib), first in the Plain of Fortriu, thereafter in Magh Circin, i.e., the Mearns and Angus. Elsewhere the tale informs us that the work of clearance was done by the Gael under compulsion (ar éigin), for when the ancestors of the Gael were engaged in their wanderings on the Continent, eighteen [207] warriors of Scythia joined them, and it was prophesied that once the Gael had settled down, they would provide lands for their friends and allies. (1) This tale appears to have elements that are very old: it may refer to a time when the Irish joined the people of Fortriu in making war not upon the Saxons but upon the Romans. Its immediate purpose is to explain the presence of Cruithnigh in the midlands of Scotland, which it does by asserting that they were a colony from Ireland, presumably from Ulster, and unhistorical as this explanation may be, it would appear reasonable enough to an Irish writer who was familiar with the tradition of the expulsion of the men of Ulster, by whom we are to understand the Cruithnigh of Ireland, by Cormac in the third century. Much more remarkable is the part assigned to the Gael of clearing land for the Cruithnigh in Alba under compulsion, actually doing forced service as mercenaries, or at least as subordinates, to the Cruithnigh, who in Ireland were vassals of the Gael. This is a matter which the historian would certainly not have invented. It is an unpleasant fact which he feels to demand explanation, and he does explain it by the ingenious device of making the compulsion arise out of an ancient obligation backed by a prophecy: the thing was, in fact, a debt of honour which the Gael were destined to repay. The historical significance of the tale is its recognition of early settlements of Gael in Fortriu and Magh Círcinn - not independent, but subject to the Picts, and serving with the Picts as mercenaries, both south of the Wall and apparently north of it. In this it agrees with other accounts, as we shall see.

This is the only instance in Irish literature, so far as known to me, where express mention is made of 'Britons' north of the Wall. It is, however, worth noting that the Latin Life of St. Serf makes Adamnan assign to the saint 'the land of Fife, and from the hill of the Britons to the [208] mount which is called Ochil.' (2) The position of the 'hill of the Britons' is uncertain, but it was north of Forth. Some further evidence of their presence is the occurrence of Breatan in place-names, such as Clach nam Breatann, 'the stone of the Britons,' in Glen Falloch at the head of Loch Lomond; Balbretane, now Balbarton, 'the Britons' stead,' in Fife; Drumbarton, 'the Britons' ridge,' in Aberdeenshire. Along with this, there are the place-names and traditions involving the name of Arthur. The Gael carried the tale of Diarmid to Scotland, locating the scene of his hunting of the Boar and his tragic death in many parts of the North, as for instance at Beinn Laghail (Ben Loyal) in Sutherland, in Kintail of Ross-shire, in Brae Lochaber, and in Perthshire. Similarly the post-Roman Britons took with them wherever they went the tale of Arthur, and as place-names and traditions connected with the Diarmid saga are a sure sign of the presence of the Gael, so Arthurian names and legends are a sign of the presence of the Briton.

The best-known 'Arthurian locality' is Arthur's Seat, Edinburgh. North of the Wall on the West are Suidhe Artair, Arthur's Seat, Dumbarton, on the right bank of the Leven; Beinn Artair (the Cobbler), at the head of Loch Long; Aghaidh Artair, 'Arthur's Face,' a rock on the west side of Glenkinglas, in the same district, with the likeness of a man's profile; Sruth Artair, Struarthour 1573 (RMS) in Glassary, Argyll. In the East there are Arthurstone near Cupar Angus, Arthur's Cairn, Arthouriscairne 1595 (RMS), apparently on the south side of Bennachie, Aberdeenshire; Arthurseat in Aberdeenshire ; and Suidhe Artair, Suiarthour 1638 (Ret.), now Suidhe, in Glenlivet, Banffshire. There is, or rather was, also Arthur's Oven, in 1293 Furnus Arthuri, described in 1723 as between the house of Stenhouse (Larbert) and the water of Carron, 'an old building in form of a sugar loaf, built without [209] lime or any other mortar.' (3) It may well be that these are not all connected with the British hero, but most of them probably are so connected, and it is particularly suggestive to find an 'Arthur's Seat' at the head of Glenlivet.

South of the Wall the Old British names have passed into Welsh, though, as we have seen, Gaelic forms of some of them come from Old British direct. There is some evidence that shows very clearly that north of the Wall also a similar process took place. Old British uxellos, high, is represented by Gaelic uasal, Welsh uchel. The latter occurs thrice in the south in Ochiltree (Ouchiltre, 1282, (RPSA)); Ugheltre, 1304, (Bain's Cal.); Uchiltrie, 1406, (RMS), Welsh Ucheldref, 'Highstead,' which occurs in Wales now in that form and as Ucheldre. North of the Wall we have the Ochil Hills in the counties of Perth and Stirling; an ancient Irish tract records that St. Serf's special seat was at Culross between Mount Ochel and the Firth of Forth. (4) There are also Rossie Ochill, in Forteviot parish, and Catochil, Catoichill 1507 (RMS), near Strathmiglo in Fife, in which the first part may be W. cat, a bit, piece, fragment, ? 'high-part.' In the north, the river Oykel between Ross and Sutherland is in G. Oiceil (genitive), on record Strath-ochell, 1490 (RMS), Kill-ochell, 1582, ib. (for Kyle-, G. caol), Strath-okell, 1582,., and so on.ib The Norse form is Ekkjall, Ekkjallsbakki, 'Oykel-bank,' (Orkn. [210] Saga). (5) The name is not Gaelic, nor is it Norse, but it may go back ultimately to E.Celtic uxellos, which had become something like uckel by the time at which the Norsemen took it over as Ekkjall. If so, Ekkjallsbakki represents Ptolemy's 'Ripa Alta,' 'high bank.' (6)

Another test is the treatment of E.Celtic vo, under, which is in Gaelic fo, in Welsh go, O.W. guo. This is the first part of Gogar in Midlothian, 'ecclesia de Goger ' (Reg. Dunf.), Nethergoger, 1335 (Bain's Cal.), Coger, 1336, ib., Gogare, 1392 (RMS) ; there is another Gogar near Menstrie, Stirling. The second part of both is probably cor, as in W. ban-gor, 'the upper row of rods ; a coping, battlement ' ; Irish cor, a setting as, in cleth-chor, 'wattle-setting.' Gogar would thus mean 'a small setting or cast,' with reference to some physical feature such as a small spur or eminence or piece of land ; the Gaelic equivalent is fochar, as in Fochar Maigi, 'small cast (spur) of the plain' (BB 139 a 16).

E.Celt. ver, an intensive prefix, as in Welsh gor, O.W. guor ; in Gaelic for. It probably appears in Gourdon, Gordoun 1587 (RMS), on the coast of Kincardineshire, with a hill 400 feet high, meaning 'great fort,' like Gordon in Berwickshire.

The first part of Gospartie, Strathmiglo, Gospertie 1507 (RMS), may be W. gwas, an abode, dwelling; the second may be a Gaelic extension of W. perth, a copse : 'wood dwelling.' This would correspond to G. Fas na Coille, Fasnakyle, for W. gwas is G. fas.

E.Celt. -ct- becomes in Welsh -th-, as in Vectis, W. Gueith, Isle of Wight. The river Nethan in Lanarkshire, twelfth century Neithan, is for an early Nectona, 'pure one.' When British names were taken over into Gaelic they were often [211] gaelicized by the addition of the Gaelic suffix -ach, -ech, to the name itself or to a shortened form of it. Thus Abur-nethige of the Pictish Chronicle, now Abernethy near Perth, has as its second part the genitive of a nominative Nethech or Neitheach (fem.), which is gaelicized either from Neithon directly, or from a British river name from the same root. With it goes Abernethy on Speyside, in Gaelic now Obar Neithich, proving that the change of E.Celt. or Old British -ct- into -th- took place not only in Perthshire but also north of the Grampians. Another example of the same change occurs in the Aberdeenshire river name Ythan, which is the same as the Welsh Ieithon of Radnor and Shropshire, for an early Iectona, 'talking one,' from the root seen in W. iaith, language. For the idea we may compare such stream names as Briathrachán, 'wordy one,' Balbhág, 'little dumb one.' These names, Nethy and Ythan, cannot be explained from Gaelic, and they show that they were taken over into Gaelic at a period when Old British had passed into the stage corresponding to Early Welsh.

Aberdeen is Abberdeon in the Book of Deer, now in G. Obar Dheathan, with dh silent, as it is regularly after obar and inbhir - the th is used merely to divide the syllables. Here Deon, Deathan, represents E.Celt. Devona a river goddess name, formed from devos (deivos), a god, in accordance with O.Ir. phonetics, at a period when E.Celt. (for ei) was still preserved, as it is in Adamnan and in O.Ir. of the eighth century. Corresponding to this, we have in early records Aberden, e.g. c. 1180 (Chart. Lind.). Aberdeen means 'the aber of the river Don' for in Gaelic the Don is Deathan, rising in Coire Dheathan, and flowing through Srath Deathan, Strathdon. Corresponding to the modern Don are the early record forms Aberdon, e.g. 1172 (Ch. Inch.), 'Simone archidiacono de Aberdoen,' 'Roberto decano de Aberdeen,' 'Matheo de Aberdoen,' c. 1202 (Ch. Lind.). 'Don, 'Doen,' did not arise from G. Deon, Deathan: Deon and Doen represent different linguistic traditions of a common original, the former Gaelic, the latter British. In Welsh the word for god is duw, for an older hypothetical [212] duyw, from deivos; Devona, would be in O.W. Duion. In Breton 'god' is doúe, M.Br. doe. It is some such form that is represented by Don, and also by the river Doon in Ayrshire: both have the same forms in early records Don, Done, Doyne - and both have as sister stream the Dee (Dva). Here then we have in effect a doublet, Gaelic Deathan (-deen), British or 'Pictish' Don, parallel to the other doublets G. Foirthe, W. Gwerid, the Forth, and G. Tatha, W. Tawy, the Tay.

In Nennius, Loch Lomond is 'stagnum Lumonoy'; the chapter headings have 'De magno lacu Lummonu, qui Anglice vocatur Lochleuen in regione Pictorum ' ; 'Of the great lake Lummonu, which is called in English Loch Leven, in the region of the Piets.' The name in modem Welsh form would be Llyn Llumonwy; it is from llumon, a chimney, a beacon, as in PumIumon, Plynlymon. The tale of Kulhwch and Olwen recounts how 'Kai and Bedwyr sat on a beacon cairn on the summit of Plynlymon.' (7) The ending -oy, later -wy, is that in Cornwy, later Kernwy, Cornwall, from Cornavia, 'horn-land.' Thus Lomond (llumon) is primarily the beacon hill, Ben Lomond; Lumonoy was the district at its base. The Lomond Hills in Fife are, of course, also 'beacons,' and one has only to look at the peaks of the East and West Lomond to see how well they were suited for that purpose. In Gaelic, Loch Lomond is Loch Laoiminn, also Loch Laomuinn, and these forms are not from the Welsh Lumon, but from an older loimmon-, whence both the Welsh and the Gaelic forms; compare G. fraoch, W. grug, heather; the base is that of G. laom, blaze. This, then, is another instance of a doublet.

Other survivals of British in the north, such as aber, a confluence, monadh, hill ground, carden, a copse, the stream names Peffer and Calder, etc., will be noticed later.

But north of the Wall, though proof exists that the old British language, or whatever dialect of it was spoken [213] there, passed into the stage corresponding to Old Welsh, the vast majority of the names are either pure Gaelic or have been gaelicized. Many of the Gaelic names may have been given at a fairly late period, but not a few are old. As has been pointed out, certain of the northern names, as well as of those in the south, give evidence of having been taken over into Gaelic at a time when the old values of the letters were still distinct enough to admit of their being treated according to Gaelic phonetics. It will thus be desirable to summarize the evidence for traces of early Scottic or Gaelic influence in the north.

Skene's position is that the very first Scottic settlement in this country took place about A.D. 500, and he commits himself further to the statement that 'there is no reason to suppose that prior to [A.D.] 360 a single Scot ever set foot in North Britain.' (8) Now it would be rather extraordinary if it were true that North Britain had remained a terra clausa as regards Ireland up to this date, but it is not true. We have the authority of Tacitus that when Agricola was in North Britain, he held discourse with an Irish ruler (regulus), who had been forced to leave Ireland on account of civil war and had fled to Britain and thrown himself on Agricola's protection, apparently with some hope of being restored by help of the Romans. (9) It is not likely that this prince came unattended. The occasion may have been the revolt of the aitheach tuatha, or vassal tribes, which is traditionally placed just about this period. This was probably not the first Irish prince who had to flee to Britain, and he was certainly by no means the last. Again in A.D. 297 the orator Eumenius states that the Britons were accustomed to Picts and Irish (Hibernis) as enemies. These references are by contemporaries of the events.

Irish tradition refers to still earlier times. It claims that Reachtaidh Righdhearg, 'of the red arm,' king of Ireland about B.C. .100, and Labhraidh Loingseach, who came about one hundred years later, were two of the Gaels [214] who beat down the lordship of Alba. (10) It has always to be kept in view that by Alba the older writers mean Britain, by no means necessarily North Britain or Scotland. According to Keating, Reachtaidh's exploits were in the north: he imposed a tribute on the Cruithnigh. Tradition is more definite as to Labhraidh Loingseach, who, as we have seen, is recorded to have made expeditions against Tiree, Skye, the people of Orkney, and other places, presumably on the West of Scotland, which cannot now be identified. We may also note the romantic tale of Deirdre and the sons of Uisliu (later Uisneach), who fled to Alba from the wrath of Conchobhar, king of Ulster. 'It was Manannan, son of Agnoe,' according to an old account, 'who settled the sons of Uisneach in Alba. Sixteen years they were in Alba, and they took possession of Alba from Man northwards. And it was they who expelled the three sons of Gnathal, son of Morgann, namely Iatach and Triatan and Manu (Mani, YBL) Lámhgharbh from that territory, for it was their father who had sway over that land, and it was the sons of Uisneach who slew him.' (11) Later versions connect them specially with Cowal and Lorne.

Conn Céadchathach (A.D. 177-212) is credited with having fought a battle in Kintyre. (12)

Later comes the settlement of Cairbre Riata, son of Conaire, son of Mogh Lámha, otherwise - and earlier - known as Eochaidh Riata and Fiacha Riata ; later Riata becomes Rioghfhada (Rigfota), 'long-armed.' His father Conaire was son-in-law of Conn Céadchathach (Conaire cliamain Cuinn), whom he succeeded as king of Ireland; Conaire was killed in the beginning of the third century; Joyce puts his reign from 212 to 220. After Conaire's death, Cairbre and his two brothers settled in Munster, and thereafter owing to a famine led his people to Ulster, where they settled in the northern part of Antrim, known thenceforward as Dál Riata, 'Riata's share.' A section of them, [215] however, still led by Riata, crossed over to Alba and settled on the north side of the Clyde estuary, apparently in Cowal, possibly in Lennox. (13) For the latter part of this account the authority is mainly Bede, who, as is well known, was an extremely careful writer. His authority, again, can hardly have been other than Adamnan, who, as Bede himself informs us, visited Ceolfrid's monastery in his own time. (14) If this was the case, Bede's statement is of great weight, for if any man was in a position to know the facts, that man was Adamnan. At the very least it shows that in the opinion of Bede's informant there had been in the West a settlement of Scots from Ireland long anterior to that by the sons of Erc in the beginning of the sixth century. This settlement also was called Dál Riata; its kings - when it came to have kings in the sixth century - were of the race of Conaire, and the connection between the two Dál Riatas was very close. So close indeed was it, that at one time there was clearly a desire on the part of the Irish Dál Riata to secede to Dál Riata in Alba. Had this happened, the position would have been similar to that of Ulster now, but the problem of the relations between Dál Riata in Ireland and the king of Ireland on the one hand, and Dál Riata in Alba on the other, was settled amicably at the Convention of Druim Ceata in A.D. 575. (15) The settlement in Ulster, it may be remarked, [216] was doubtless connected with the operations conducted by Conn and his successors against the Cruithnigh or Dál nAraide of Ulster, for Dál Riata must have been a 'swordland' wrested from them. Cairbre himself is stated to have been slain later by one of the Dál nAraide. (16)

Lughaidh mac Con, who became king of Ireland in A.D. 250, was a contemporary of Cairbre Riata and fought against him when Cairbre was attempting to avenge his father's murder. In consequence he was exiled, and going to Alba in the usual manner of exiles he stayed there for seven years and gained friends and supporters, including Béinne Briot, 'son of the king of Britain.' With these he returned to Ireland, fought and defeated Art, son of Conn, and became king in his stead. (17) Among his chief exploits were a voyage to Alba and four great victories over the men of Orkney. (18)

Lughaidh is said to have married a daughter of Béinne Briot and to have had three sons, who were known as the three Fothads, a name which suggests connection with Fothudáin, the Gaelic form of Votadini. Fothad Canann, one of the three, was a famed leader of Fiana (rígféinnid), and is said to have taken possession of lands in Alba. It is interesting to note that Keating derives the house of MacCailín (Argyll) from Fothad Canann, and that the official genealogies have Béinne Briot as one of MacCailin's ancestors. One of the lost 'chief tales' was entitled Longes Fothaid, 'the sea expedition (or exile) of Fothad.' (19) The two other Fothads reigned jointly for one year after Cairbre of the Liffey and fell in the battle of Ollarba against the Fiana.

In the earlier part of the fourth century the three brothers called Conla, later Colla, are prominent figures. They were sons of Eochaidh Doimhlein, son of Cairbre of the Liffey, son of Cormac, and their mother was Aileach, daughter [217] of Fubthaire of Hi. The eldest, Colla Uais, was king of Ireland for four years (327-331), when he and his brothers with three hundred men had to flee to Alba to escape the vengeance of the son of the previous king, who had been slain by them. They stayed in Alba for three years, doing military service with their kinsmen. (20)

Criomhthann Mór, son of Fidach, is mentioned as one of the three who crushed the sovereignty of Alba. He is mentioned in the important passage in Cormac's Glossary which says, 'when great was the power of the Gael in Alba (Britain), they divided Alba between them into districts and each knew the residence of his friend, and not less did the Gael dwell on the east side of the sea than in Scotia (Ireland), and their habitations and royal forts were built there. Whence is named Dinn Tradui, i.e. the triple-fossed fort of Crimthann the Great, king of Ireland and of Alba to the sea of lcht (the English Channel), and hence also is Glasimpere or Glastonbury of the Gael . . . they continued in this power till long after the coming of Patrick (A.D. 432).' But Criomhthann's activities in Britain were probably confined to the south-west of England, and there is nothing to show that he was ever in Scotland.

His successor was Niall Nóigiallach,' of nine hostages,' (379-405), so called because he held a hostage from each of the five provinces of Ireland and four from Britain; elsewhere the four are said to have been from Alba, the Saxons, the Britons, and the Franks. (21) Of him it is said among other things, 'many shall be his deeds on Druim Alban,' i.e. the water-shed between the east and the west of Scotland. (22)

Dathí, who succeeded Niall and reigned from 405 till 428, was regarded as a mighty conqueror, and the tradition is that he was killed by lightning at the Alps. His reign, as we have seen, was inaugurated by a great raid on Strathclyde, but he is also credited - and this is notable - with a [218] battle in the cast of Scotland, the battle of Magh Crcin or Gerginn, i.e. the Mearns and Angus. (23)

In this connection it is proper to note the statements regarding an early settlement in the eastern midlands, which proceeded from Munster. The two great ruling families of that province were the heads of the Dál nCais and the Eoganacht, descended respectively from Cormac Cas and Eoghan, sons of Oilill Olum, a contemporary of Art, son of Conn. Towards the end of the fourth century, Lughaidh, king of Munster, who belonged to the Eoghanacht branch, took occasion to banish his son Corc, who went to Alba. The romantic tale of his adventures there is contained in the Book of Leinster (287 b 1), but the beginning is missing. On his arrival in Alba he was caught in a snow-storm, was six days without food, and when at the point of death was found by the king's chief poet, Gruibne, whom he had once befriended. Gruibne resuscitated him, and observed an Ogham written on his shield, requesting the king of Alba if Corc came to him by day, to behead him before night, if he came by night, to behead him before day. Gruibne amended the Ogham so as to read that Corc was to receive the king's daughter in marriage, which ultimately he did. He stayed in Alba till three sons were born to him, after which he left with wife and sons and much treasure. (24)

Such is the outline of the romance, which in this form may not be much older than the date of the Book of Leinster itself (c. 1150). Gruibne and Corc, however, are twice mentioned in relation to each other in Cormac's Glossary, and the second of these passages contains some words of what purports to be Gruibne's welcome (fáilte) to Corc, a version of which is given in the Book of Leinster. (25) [219] Cormac mac Culinan, king of Munster and archbishop of Cashel, fell in battle in A.D. 903. The other great work ascribed to him is styled the 'Psalter of Cashel' (Saltair Caisil), from which extracts are often made by the later MSS. Among these extracts is an account of the genealogies of the descendants of Eber, including Corc, contained in four of the great MSS., namely, Rawl. B 502, the Book of Leinster, the Book of Ballymote, and the Book of Húi Maine. (26) These all agree in stating that Corc had seven sons, the mother of one of whom was Mongfinn, daughter of Feradach Finn Fechtnach, and this son was ancestor of the Eoganacht of Magh Gerginn. The three first MSS. style Feradach 'king of Cruithentuath' (Pictland), and give the name of Mongfinn's son as Cairbre Cruithnechkin, 'Cairbre the little Pict,' or rather 'Pict-sprung' ; the fourth makes Feradach 'king of Alba,' and calls the son Maine. (27) Further, Rawl., BB, and Húi Maine add that from the Eoganacht of Magh Gerginn came Oengus, king of Alba. This Oengus was most probably the son of Fergus, who died in 761, and is styled by Tighernach 'king of Alba' under A.D. 759; there was, however, another king of the same name who died in 832, king of Fortriu.

Corc, son of Lughaidh, is, of course, a perfectly historical character. He became king of Munster, and his grandson Oengus, son of Natfraoch, was baptized by Patrick. That he was actually the founder of the Eoghanacht of Magh Gerginn may be true or it may not. What is certain is that there was a branch of this great Munster family there, and that already in the ninth century it was reckoned to be of old establishment. It may have been these [220] Eoghanacht who helped to clear the 'swordland' among the Britons of Magh Gerginn. Their name appears to survive in Balmackewan, Baile mac Eoghain, 'stead of the sons of Eoghan,' in the parish of Marykirk, and the name Cairbre is found in Drumforber, Drumquharbir 1539 (RMS), 'Cairbre's ridge,' in the adjoining parish of Laurencekirk. Near it is Conveth, the old name of the whole parish representing Early Irish coindmed, modem coinmheadh, 'free quartering, billeting' ; the district would have been so named because it bore the special burden of quartering the household troops of the lord. Dundee is in Gaelic Dùn Dèagh, which seems to mean 'Fort of Daig(h),' for the genitive of Daig is Dega, in modern spelling Deagha. Daig, meaning 'fire,' was a rather uncommon Irish name, though Aed, 'fire,' was very common. One of Corc's sons was named Daig.

The Irish authorities state that the nobles of Lennox were of the same origin as the Eoghanacht. (28) Muiredach úa Dálaigh, the well-known Irish bard who flourished A.D. 1213 and spent part of his life in Scotland, wrote a poem in honour of Alún, son of Muiredach of Lennox, which has been printed by Skene. (29) In this poem he refers to the coming of Corc to Alba and his marriage to the daughter of Feradach, whom he calls Leamhain. Corc and Leamhain had a son named Maine, from whom were descended the rulers of Lennox. Keating says: 'Maine Leamhna (i.e. Maine of Leamhain or Lennox), son of Corc, went from Ireland to Alba, and there occupied territory which is called Magh Leamhna' (the plain of Leamhain or [221] Leven). (30) The Book of Húi Maine also, as we have seen, makes Maine a son of Corc. That the Gaelic lords of Lennox believed in this descent is indicated by the fact that Alún or Alwyn, the second Earl of Lennox who appears on record, had a son named Corc. (31) Maine, however, was the grandson, not the son, of Corc; his father was Cairbre, and from him were descended two saints, Cummine Fota and Faithlenn. (32) In the poem by Muiredach úa Dálaigh, the Mormaer of Lennox is styled rí Bealaigh, 'king of Bealach,' in English now Balloch, at the lower end of Loch Lomond. In early times Balloch was the seat of the lords of Lennox, and it is notable that close beside it is Tullichewan, Tulach Eoghain, 'the hill of Eoghan.'

Munster, the homeland of the Eoghanachta, was the great centre and source of the Ogham cult, as Professor John MacNeill has shown in his foundational monograph on the Irish Ogham inscriptions. (33) The total number of known Ogham inscriptions is about 360, of which about 300 are Irish, and of the Irish Oghams five-sixths belong to the counties of Kerry, Cork, and Waterford. Kerry has about 120, Cork about 80, Waterford about 40. In Britain 26 have been found in Wales, 5 in Devon and Cornwall, 1 in Hampshire. Scotland has 1 in Gigha, Fife, Kincardine, Perth, Moray, Sutherland, and Caithness respectively ; 3 in Aberdeenshire, 2 in Orkney, 4 in Shetland. 'The distribution of the British Oghams,' says Professor MacNeill, 'clearly corresponds to the region of Gaelic, or, as it was then called, Scottish influence in the period that followed the withdrawal of the Roman legions from Britain.' But it may have begun even earlier. He further [222] gives good reasons for believing that the Ogham cult was reckoned as distinctively pagan and was banned by the Church, and that as a consequence it was arrested by the growing power of Christianity. The distribution of Oghams in Scotland is so sporadic as to indicate clearly that the cult had no centre here, and that it did not endure. The absence of Oghams on the west is due to the fact that the west was settled from the north of Ireland and especially from Ulster, where the Ogham cult never found a footing. The solitary instance in Gigha may be compared with the sporadic instances in Ulster. On the other hand, the presence of Oghams on the east indicates early Gaelic influence from Munster, and confirms in a striking manner the traditions which we have been relating. There remains the question, why there are no Oghams in Lennox, if early settlers from Munster found their way there also. The answer may be that, granting the truth of the tradition, the conditions were different there from those in the northeast: the Christian influence which we know to have existed in the Vale of Leven before the time of Coroticus would have been sufficient to prevent the use of Oghams. Saint Patrick was born about A.D. 365; his father was a deacon and his grandfather was a presbyter.

Here may be noted the distribution of the term cathair, gen. cathrach, a circular stone fort. In Ireland, says Joyce, there are more than three hundred townlands and towns whose names begin with this term, 'all in Munster and Connaught except three or four in Leinster - none in Ulster.' (34) With us it is very rare, and on the west it does not occur. (35) On the east the furthest north instance which I have met is Corncattrach, for Coire na Cathrach, 'Corrie of the Cathair,' adjacent to Shanquhar, for Sean-chathair, 'old fort,' Gartly, Aberdeenshire. In Forfarshire there is Stracathro, in 1212 Stracatherach (Johnston), for Srathcathrach: the cathair may have been the White Caterthun, or possibly Dunlappie (? for Dùn-lapaigh 'Fort of the [223] of the bog'), the name of a parish now included in Stracathro. The White and the Brown Caterthun are powerful stone forts of the broch type in the adjacent parish of Menmuir; the first part of the name is doubtless cathair, but the second part is obscure to me. It is worth noting that on the south side of the White Caterthun is the Gallows Hillock.

'The ville of Catherlauenoch called Tullibardine is on early record (36) ; the name is for Cathair Leamhnach, 'Elmfort,' and is now represented by Carlownie Hill on the south border of Auchterarder parish, Perthshire. Another instance in Perthshire is Kathermothel, for Cathair Mhaothail, 'Muthil Fort,' the old Gaelic name for the important Roman camp of Ardoch. (37) Both of these names came to be used as names of districts.

A seat of the ancient Mormaers of Lennox was called Cathair, probably with some distinctive or qualifying term which has not come down to us. Near it was the place of execution, referred to in a charter as furcas nostras de Cather, 'our gallows of Cather '; elsewhere as furcas nostras del Cathyre, 1370 (RMS), 'of the Cathair.' It was in the parish of Kilmaronock, Dumbartonshire, and the name survives in Catter, where, as we are informed by the Old Stat. Account, 'there is a large artificial mound of earth, where in ancient times courts were held; near to which the Duke (or rather Mormaer) of Lennox had a place of residence. There is not now the smallest vestige of the building.' (38) Near Catter House is Drumquhassle, for Druim (an) Chaisil, 'Ridge of the Cashel,' or circular stone fort. The furcae seem to be commemorated in Crosshill, on Catter Muir. With Catter may be compared Cadder, Kirkintilloch, the site of a Roman fort. 'Foresta de Passelet et Senecathir' appears in 1226 (Theiner).

[224] The period we have been considering was one of great activity and enterprise among the Gael of Ireland. That they made settlements in North and South Wales (Venedotia or Gwynedd and Demetia or Dyfed) is well known. The Deisi, who were expelled from Meath in the third century, went to Dyfed; long afterwards they were known to the Irish as the race of Crimthann (cenél Crimthaind). (39) The Gael of Gwynedd and Anglesey were driven out by the sons of Cunedda in the fifth century. The Irish settlements in the south-west of England are referred to in the extract given from Cormac's Glossary. In both regions their presence is attested by Ogham inscriptions. During the same period they were familiar with North Britain, and made settlements among the Picts in the midlands of Scotland ; it would appear, in fact, that they were rather welcomed, and helped materially to stiffen the native struggle for independence as well as to join in the raids on Roman Britain. The Irish Nennius says that the northern Wall was constructed against the Gael and the Cruithnigh. (40) The tenor of Patrick's Epistle to Coroticus of Strathclyde also makes it clear, as Professor Bury has seen, that the Scots who shared in the booty of Christian captives taken from Ireland were located in Scotland. As in Wales a noble family is recorded to have sprung from the exiled Deisi, so in Scotland there is reason to believe that the mightiest king of the Picts was sprung from a Munster family which had settled in the Mearns about A.D. 400. It would be indeed remarkable if the Irish inroads on Britain in the fourth century were not accompanied by settlement north of the Wall of Antonine.

The coming of the sons of Erc soon after A.D. 500 and the establishment of the Scottish kingdom of Dál Riata were events of first-rate importance ; the facts are stated by Skene and need not be repeated. Not less important [225] was the introduction of Christianity by Columba into the north and west; according to one authority Columba also visited the valley of the Tay. (41) There are several indications that the Scots of Dál Riata attempted to acquire power in the midlands. Aedán's father, Gabrán, it will be remembered, is said to have made an expedition to Forth, and Aedán himself is styled sub-king or prince of Forth. The Irish life of St. Berach says that Berach came to Aedán's fort, and that Aedán offered up the fort to Berach, even Eperpuill, which is Berach's monastery (cathair) in Alba. Eperpuill, as the Reverend C. Plummer says correctly, is Aberfoyle, on the Forth; the October market that was wont to be held there was called Féill Barachan (Bearchán), and near Aberfoyle there are what appear to be the remains of a fort. (42) There is the further incident of Aedán's battle in the Mearns recorded by Tighernach and referred to by Adamnan. Aedán's son and successor, Eochaidh Buidhe, is styled 'king of the Picts' in AU, quoting from the ancient Book of Cuanu. The descendants of Eochaidh's son, Conall Cearr, are styled 'the men of Fife ' (fir ibe, for Fhíbe). (43)

We may here notice the occurrence of one or other of the various names for Ireland which are found in Scotland. In Latin, Ireland is Scotia or Hibernia. Adamnan shows a marked preference for the former, using Hibernia and the adjective Hiberniensis fifteen times, while he uses Scotia, Scoticus, Scotice, sixty times, according to Skene's index. As Scotland became more and more gaelicized, and especially when a Gaelic king came to rule over both Picts and Scots, the term Scotia began to be applied to Pictland, particularly, as Skene points out, to the lowland part east of Drumalban and north of the Firth of Forth. This usage, which resulted in the term altogether ceasing to denote [226] Ireland, is already indicated when Dicuil, about A.D. 825, writes of Ireland as nostra Scotia, in contrast with the other Scotia in Britain. Scotia is, of course, entirely a literary term; the terms in actual use among the Gaelic people were Eire or Eriu, Banba, Fodla, and Ealg or Ealga, and of these the first was the most usual. The first three were really names of ancient local goddesses, associated with different parts of Ireland; the last is traditionally said to have been used in the time of the Firbolg, who were pre-Milesian. We might naturally expect to find the literary (Latin) use of Scotia as applied to part of Scotland to be parallelled by a use of the native terms; in other words we might expect to find Eire or its equivalents applied to the Gaelic parts of Pictland. Further, the literary Scotia may be expected to have arisen as the reflex of the vernacular term, which would naturally have been in use first. There are reasons to believe that this was actually the case and that at one time Pictland stood a chance of being called Eire, though in the event the old native name of Alba won the day as the equivalent of Scotia.

The tract De Situ Albanie, ascribed by Skene to about 1165, ends with the statement that Fergus, son of Erc, was the first of the seed of Conaire to become king of Alba, 'that is, from the mount Brunalban to the sea of Ireland and to Innse Gall. Thereafter kings of Fergus' seed reigned in Brunalban or Brunhere up to Alpin, son of Eochaidh.' (44) Here we have Brunalban instead of the regular Drumalban, with Brunhere as an alternative name. Now whether brun is written in error for drum, or, as is more likely, represents the Welsh bryn, a hill, it is clear that the second part of Brunhere is Eire, uninflected (as it would be if the compound is Welsh), and that we are offered 'Mount of Eire' as an alternative for 'Mount of Alba' or 'Ridge of Alba.' Had this form persisted, we should now be speaking of Bràghaid Éireann, 'Breaderin ' instead of Bràghaid Alban, 'Breadalbane.' The inference [227] is that at one time the district, or part of the district, east of the watershed was called Eire.

St. Fillan (Faolán), son of Oengus, son of Natfraoch, son of Corc of Cashel. (of the Eoghanacht of Munster), is designated 'of Ráth Érenn in Alba.' (45) This place has been assumed, without any proof, to have been at or near St. Fillan's at the lower end of Loch Earn, but no such name exists there, either now or on record, though the Old Stat. Acc. of Comrie mentions Dùn Fhaoláin, 'Fillan's Hill' (or Fort)' with St. Fillan's Well on the top of it. Ràth Éireann is, as I believe, still extant; it appears on record as 'Raterne in the earldom of Stratherne,' 1488 (RMS); Raterne, 1466 ib., now Rottearns, in the parish of Ardoch. As this is quite outside the valley of the Earn, the meaning must be 'Rath of Eire' - a district. Again in the 'Prophecy of Berchan,' Girig or Grig, who is recorded to have slain and succeeded Aed, son of Kenneth mac Ailpin, fought a battle 'on the fields of Eire ' (ar bhrughaibh Éirenn). (46) Elsewhere we are told that he died at Dundurn, a fort near the lower end of Loch Earn. This is certainly on the river Earn, but in view of such expressions as brug Banba, 'land of Banba (Ireland),' brugh Bretan, 'land of the Britons,' etc., the probability is that we have to do with the name of a district. Drummondernoch, Drummenerinoch 1595 (RMS) between Comrie and Crieff, is in Gaelic Drumainn Eireannach, 'Drummond of Eire.'

We may note further an instance connected with the Irish Church settlement at Glastonbury in Somersetshire. In 971 king Edgar granted certain privileges to Glastonbury and in the record mention is made of a place called 'Bekeria, which is called parva Hibemia (little Ireland).' It is described as being in insulis, 'among the islands,' that is to say the low insulated lands or Inches near the Abbey. (47) [228] This is a case of transference, not indeed of the name of Ireland itself, but of the name of a small island called Bec Ériu, now Beggery Island, in Wexford harbour. The latter got its name, according to the Irish account, from the bishop Ibar, to whom Patrick said in anger, 'thou shalt not be in Eriu.' 'Eriu,' replied Ibar, 'shall be the name of the place in which I am wont to be,' and he settled in the isle which was so named thenceforward. (48) These names are both connected with religious establishments.

In North Wales, which was once occupied by the Gael, there is a closely parallel instance. There, at the head of Afon Lledr in Carnarvonshire, is Llyn Iwerddon, 'Lake of Ireland,' corresponding exactly to our Loch Éireann, Loch Earn. Lower down, not far from the Falls of the Conway river, is a hill or place called Iwerddon, Ireland. About midway between the two is Dolwyddelan, 'Gwyddelan's Meadow,' where Gwyddelan is a derivative of Gwyddel, a Gael.

There can thus be no doubt that the name Éire was used as a district name in the parts of Britain where the Gael settled ; I believe that Strathearn means 'Ireland's Strath,' not 'Strath of the river Earn,' and that Loch Earn (Loch Éireann, now Loch Éir) means 'Ireland's Loch.' It is not without significance that the purely Gaelic name Strathearn came to displace in part the gaelicized British name Fortriu as the name of the province between Tay and Forth. The south-western part of Fortriu became the Earldom of Menteith, the remainder became the Earldom of Strathearn. Strathearn of old apparently came all the way to the Firth of Forth, for Cuilennros (Culross), 'Holly-point,' is said to be in it. There is a fort called Dunearn in Burntisland parish, Fife.

Marching with Strathearn was the district and Earldom of Athol. In the Book of Deer Atholl is spelled Athótla (for Athfhotla) in Tigernach's annals it is Athfhotla (genitive); the corresponding entry in AU (739) has Athfoithle. The Norse form is Atiotlar, answering, says Professor Craigie, [229] to the Scottish Athwotle, Athodel, etc., from an original Athfodla. From these forms it is clear that Atholl represents ath, in the sense of Latin re- denoting repetition, and Fótla, later Fódla, Fódhla, Ireland; the name means 'New Ireland.' Late forms are Abhuill (genitive), in a bardic elegy on Sir Duncan Campbell who died in 1631, and Afall (genitive) in the latter part of the same century. (49) The present form is usually Athall for all cases, but the Rev. C. M. Robertson tells me that Blàr an Abhaill (Awaill) is heard for Blair in Athol, and that the Athol men are na h-Abhaillich (Awaillich). (50)

North of the Grampians Éire occurs thrice. The Register of Moray records 'the church of Eryn, with the chapel of Inuernarren,' i.e. of Nairn, and about 1140 King David confirmed to the monks of Urquhart a grant of 'Pethenach near Eren, and the shieldings of Fathenachten.' Pethenach is now, I think, Penick, in Gaelic a' Pheighinneag, 'the little pennyland,' near Auldearn church. Fathenachten is probably Fornighty in Ardelach, the Gaelic of which is Achadh-ghoididh (? -ghoide, from goid, theft), apparently a different name for the same place. This Erin was therefore in Nairnshire, and appears to have been the old name for Auldearn parish. (51) In any case it is the name of a district, and Allt Éireann, Auldearn, means 'Ireland's Burn.'

We learn that the Abbey of Kinloss possessed a toft in Inverness, Eren, Forres, Elgin, and Aberdeen. (52) Elsewhere separate mention is made of Invereren, the land of the prepositura of Invereren, and one toft in Eren, all belonging to Kinloss. Invereren is the lower part of the river Findhorn. It was also the name of the old village of Findhorn, at the mouth of the river, which was swept away by the storm of 1701. Prepositura became in Scots 'Grieveship,' and Lachlan Shaw says 'below Mundole, on the [230] side of the river (Findhorn), is the Grieship.' The name survives in Greshope. Elsewhere the term is found in connection with places of strength, e.g. Cromarty, Forres, Cullen, all of which had castles, and it is more than likely that the prepositura of Eren is to be connected with the Castle of Eren referred to in a charter of William i. in 1185 as 'castellum meum de heryn.' (53) The place called Eren in which the toft was situated would have been near this; it was most probably in fact the same as the old village of Invereren. The names Cullerne and Earnhill, near the mouth of Findhorn, meaning 'nook of Eren' and 'hill of Eren,' are further in favour of Eren having the name of a district. The name Findhorn itself is the dative-locative of Fionn-Éire, 'white Ireland,' and doubtless refers to the white sands of the estuary. The river is in Gaelic Éire, Uisge Éire, and its strath is Strath Éireann, Strathdearn, 'Ireland's strath.' Near Dulsie Bridge is a large fort called Dùn Éireann, 'Ireland's fort,' a name which may indicate that the district extended well inland. Near the head of Strathdearn is a remote little district called 'the Coigs ' or 'Fifths,' of which it is said 'tha cóig cóigimh an Éirinn is tha cóig cóigimh an Srath Éireann, ach is fearr aon chóigeamh na h-Éireann na cóig cóigimh Srath Éireann,' 'there are five-fifths in Ireland and five-fifths in Ireland's strath, but better is one-fifth of Ireland than the five-fifths of Ireland's strath.' The saying is old.

Deveron is on record Douern and Duffhern, meaning apparently 'Black Éire,' as distinguished from 'White Éire,' or Findhorn, but unfortunately the name does not survive in Gaelic. It can scarcely be mere chance that has put Banbh, Banff at its mouth, and, as we have seen, Banba was a name for Ireland. It is also notable that a stream near Deveron on the west is called Boyne, while the patron saint of Boyndie parish, near Banff, was Brendan the Voyager. Tolachherene 1242 (Ant. A. and B.), 'hill of Éire,' appears to have been near Deveron.

Elg or Ealg has gen. Elgga (LL 45 a 42, 81 b 41) or Eilgi (LL 377 b 1 6) ; dat. Eilgg (LL 49 b 44) ; Kuno Meyer has [231] compared Druimm nElgga in Munster (LL 198 b 4) with Druimm nAlban. (54) According to one authority, it means 'pig' (muc); others make it 'noble' (uasal). (55) With us it occurs as the name of a district in the west of Invernessshire, commonly known as Gleann Eilge, Glenelg. I am assured, however, that the old people considered Eilg to be the name of the district, and 'Glenelg' to be 'the glen of Eilg.' This is borne out by the bardic poetry, where we find such expressions as iath Eilge, 'the region of Eilg'; fear finn Eilge 'the lord of fair Eilg.' (56) A hill on the border between the parish of Glenelg and that of Glenshiel in Ross is an Cruachán Eilgeach, 'the Rick of Eilg.' A Glenelg man, however, is Eilginneach, formed probably on the analogy of deilginneach, shingles, from dealg, a prickle, which seems to be a byform of deilgneach, prickly. The form 'Glenelgenie' which appears in Robertson's Index, is curious, if authentic.

Eilginn, Elgin, has been explained by Kuno Meyer as for Eilgín, 'Little Ireland,' a diminutive from Elg. This explanation is made the more attractive by the fact that a certain quarter of Elgin, and by no means a recent one, is actually called 'Little Ireland.' (57) The difficulty is that the diminutive in -ín, which is common in Irish, is rare with us, and when it does occur - as in cailin, a girl - it does not usually double the final consonant. Eilginn looks like a locative case, and it may have been formed on the analogy of Éirinn and Albainn, which are themselves formed on the analogy of Mumhain(n), Munster. Elgin used to be called in Gaelic Eilginn Muireibh, 'Elgin of Moray,' to distinguish it from some other place of like name, probably Elg of Glenelg.

Banba is connected with banb, now banbh, a sucking pig; [232] she was probably a swine goddess. Kuno Meyer did not hesitate to regard both Banff on Deveron and Bamff near Alyth, in Perthshire, as the equivalents of Banba, both meaning Ireland. (58) To these may be added 'Banff with the fulling mill,' 1582, 1587 (RMS), part of the churchlands of Arbuthnot in Kincardineshire, now apparently obsolete. It is true that Banff is Banb in the Book of Deer and Banbh in modern Gaelic - one syllable. On the other hand, banbh, a sucking pig, is not appropriate - one might say it is impossible - as the name of a place or district ; the Isle of Muck is not a parallel, for it is Eilean nam Muc, Helantmok in Fordun, 'isle of swine.' Besides these there are Banbhaidh, Banavie, near Fortwilliam, Loch Banbhaidh near Loch Shin, Allt and Gleann Banbhaidh in Athol, and Banevyn (Reg. Arbr.), now Benvie in Forfarshire. All these might represent Banbha, so far as phonetics go; compare Munlochy for bun-locha, Dalarossie for Dail Fhearghusa. I think, however, that the Athol Banvie is a stream-name like Mucaidh, 'pig-burn,' on Loch Tay; the others may be stream names or place-names like Tarvie, 'bull-place,' from tarbh, a bull. The Welsh for banbh is banw, a young pig, and in Montgomeryshire there are confluent streams called Twrch, Boar, and Banw (on maps Banwy). A Welsh writer says of Twrch, 'many rivers forming deep channels or holes into which they sink into the earth, and are lost for a distance, are so called. A small brook called Banw in the parish of Llan Vigan, meaning 'a little pig,' has been said to be of this family'. (59) In Ireland banbh appears commonly as an element of names, but Joyce gives no instance of it used by itself alone or as a stream name. In Wales several streams are called Twrch, none with us.

Yet another name for Ireland is Fál, and we have Dunfàil, Dunphail, near Forres, but this may equally well come from fàl, a hedge, palisade.

There are some instances of Ireland itself as a place-name, [233] e.g. in Menmuir parish, Forfarshire, where it is adjacent to Rome: Rowme, Ireland, and Corsbank, 1536 (RMS). This may be a translation, but there is no proof.

To these indications of settlement must be added the influence of the Scoto-Irish Church through its monasteries established in various parts of Pictland. It is clear that whatever may have been the relations of the Church among the Picts to the civil authority, there was a close connection between that Church and the Irish Church, that its personnel was largely Gaelic and that Picts went to Ireland for training. It is highly significant that in the list of rulers and clerics who signified assent to Adamnan's Law of the Innocents, Scotland is represented by Bruide, king of Cruithentuath or Pictland, Curitan, bishop of Rosmarkie, and Ceti, bishop of Hí. Clerics and nobles would have been among the first to learn Gaelic. An interesting glimpse of the transition stage between Pictish and Gaelic is given in the tradition of St. Manirus recorded in the Aberdeen Breviary, where we are told that in consequence of the difficulty caused by diversity of language among the people, Manirus, being excellently skilled in both languages, went to labour at Crathie in Braemar. He was probably a native Pict who had been trained in Ireland.

The difference, so far as it exists, between the place-names of the west and those of the east is due to the historical circumstances. The west was largely and continuously peopled from Ireland and at an early stage became a Gaelic kingdom; for many of the people and for all the ruling class Gaelic was no acquired language, but their native speech. Consequently the change of language was comparatively rapid and general, and the old names were largely displaced. On the east, where settlement was sporadic and the rulers were chiefly Picts, the change was much more gradual. More of the old names were retained or adapted to Gaelic, many were doubtless translated in whole or in part.

In the west again, and also in the north from Caithness to Beauly, strong Norse influence caused a great displacement of old names. In the Long Island, from the Butt of [234] Lewis to Barra Head, Norse must long have been the predominant language, if not the only one. Here there are indeed many Gaelic names, but all the more important names are Norse and the Gaelic names are of comparatively recent type, being phrase names, e.g. Allt na Muilne, 'the burn of the mill.' In the Inner Hebrides names of an older type are found, such as Sligeachán, 'shell-place,' Draighneach, 'blackthorn-place,' in Skye. Caithness, part of the old Pictish homeland, must have been mainly Norse-speaking for centuries. Sutherland and, in a less degree, Ross are shot through with Norse names.

Bilingualism is no new thing in Scotland. It existed when Celtic began to displace some older tongue before the Christian era; that older tongue had probably displaced one still older. In Galloway, Strath Clyde, and the north Gaelic took the place of British or Pictish. In the far north and the west Gaelic or Pictish was for a time eclipsed by Norse, which in its turn was displaced by Gaelic. The circumstances under which those changes took place can never be known in the way we know how Gaelic is now being displaced by English. Two points, however, are to be noted ; the first is that change of language does not necessarily mean change of race: the second is the somewhat remarkable readiness with which languages do change.

Some of the features more or less distinctive of the names of the region north of Forth and east of Drumalban may now be mentioned.

One of the most striking is the frequency of certain terms of British origin, such as pett, a piece or portion or share of land, anglicized as Pit-; monadh, a mountain, hillground; preas, a copse. To these may be added gronn, a mire or marsh, found from Forth to Beauly, and possibly fother, fetter, with about the same range.

Among hill terms barr, an eminence, is very rare; tulach, tilach is very common. Sliabh in the sense of 'mountain' probably does not occur. Carn, which usually means 'a heap of stones,' 'congeries lapidum,' is often used in the sense of a high, rocky hill; we may compare O. Welsh carn explained as rupes, a rock, crag, cliff. Tom regularly [235] means a rounded hillock or hill, a mound, like Welsh tomen ; in Irish tom means usually a bush, thicket, as sometimes also with us, as tom luachrach, a clump of rushes.

The great native unit of land was the dabhach or dobhach, fem., a Gaelic term meaning it large tub or vat, the largest of vessels in use; I have heard it applied to the large tub in a smuggling bothy. The secondary meaning of 'large measure of land' is peculiar to Scotland. But the davach (60) was not so much a measure of land as of 'souming,' i.e. it was an amount of land reckoned to support so many head of stock. Burdens on land - càin, tax; coinmheadh or 'conveth,' dues of maintenance, and probably the military obligations of feacht, 'expedition,' and sluaghadh, 'hosting' - were assessed on the basis of the davach. (61) In 1772, in the roll of the Eastern Company of the Strathspey Volunteers, the men were entered according to the davochs of the parish. (62)

The term occurs by itself as Dauch or Doch, usually, however, followed by a descriptive term, as Dabhach na Creige, Dochcraig, 'davoch of the rock,' Dabhach Phùir, ' Dochfour, 'davoch of pasture,' and so on. It is rare at the end of names but there are Phesdo, Fasdawach 1443 (RMS), 'firm davoch,' in Kincardineshire, Fendoch, Findoch 1542 (RMS), 'white davoch,' near Crieff, and Gargawach, 'rough davoch,' in Lochaber. Leth-dabhach, a [236] half davoch, is anglicized Lettoch. Lettoch in the Black Isle is Haldach in 1527, Haddoch in 1611, forms which clearly explain Haddo, in 1538 Haddauch, in Aberdeenshire to mean 'half -davoch.' Trian, a third part, is rather rare; Trian-a-phùir, Trinafour, 'pasture-third,' is in Glen Erichdie, Struan; Blaeu has Trien high up on Black Esk. Ceathramh, a fourth part, a quarter, is probably the fourth part of a davoch, or it may be of a half-davoch, in such names as an Ceathramh Ard, Kerrowaird, 'the high quarter,' an Ceathramh Gearr, Kerrowgair, 'the short quarter,' near Inverness, and Kirriemuir. Ochdamh, the eighth part, occurs on the Kyle of Sutherland as Ochtow : 'the wester bovate vulgo the Ochtow,' 1589 (RMS). Auchtogorm in Moray is 'the green octave.' Achterblair near Carrbridge is ochdamh a' bhlàir, 'the octave of the moor.' In Argyll there are Ochtofad, 'the long octave,' Ochtomore, 'the big octave,' Ochtavullin, 'octave of the mill.'

Other denominations also occur, though as a rule they do not appear on the map. Thus on the south side of Loch Tay there are Marg na Crannaig, Marg Mhór, Marg Bheag, Marg na h-Àtha, Marg an t-Sruthain, Merkland of the Crannag, Big and Little Merkland, Merkland of the Kiln, Merkland of the Brook. Baile Mac Neachdain, Balmacnaughton, on the same side, is known as an dà fhichead sgillinn, 'the forty shilling land' ; near Ardtalnaig is an deich sgillinn, 'the ten shilling land' ; Ardeonaig is an fhichead sgillinn, 'the twenty shilling land.' In the east end of Fortingal there is an dà mharg dheug, 'the twelve merk land.' (63) In Sutherland there is Loch Merkland, and I have heard a field in Altas, Sutherland, called am plang, 'the plack.'

Of a number of old personal names which occur, the following may serve as examples. I do not mean to imply that the names are by any means peculiar to this region.

Abhartach, whence the name of the Irish sept hUi Abartaich (Rawl. B. 502, 159 b), appears in Rossawarty, [237] 1508 (RMS), 'Abhartach's cape,' now Roshearty near Fraserburgh. Hence too, Dùn Àbhartaigh, Dunaverty, 'Abhartach's Fort,' on the Mull of Kintyre, besieged by Sealbhach in 710. (64) Some Ailpin has given name to Rathelpin (RPSA), now Rathelpie, 'Alpin's rath,' in Fife; there is also Cairnelpies (Ret.) in Banffshire.

The old Irish name Breasal, 'warrior,' is seen in Donibristle, Donibrysell and Donybrisle in the twelfth century, for dùnadh Breasail, 'Breasal's fortress.' (65) From Bruadar comes Drumbroider in Stirlingshire, 'Bruadar's ridge'; compare Ballybroder in Ireland (Joyce); Tillybrother in Aberdeenshire may be 'Bruadar's hill.' From Brandubh, 'raven-black,' comes Loch Branduibh, Loch Brandy, near Milton of Clova, Sìthean Druim Mhac Bhranduibh, 'fairy knoll of the ridge of the sons of Brandubh,' near Onich, Argyll, and probably Brandy Burn in Fife. Gleann Buichead, Glen Buchat in Aberdeenshire, is 'Buichead's glen'; compare Dún Buchad or Dún Buichead in Ireland. Frìth Bhàtair, Freevater in Ross, is 'Walter's forest,' probably from Walter Leslie, earl of Ross.

O.Ir. Coirpre, later Cairbre, gives Drumforber, 'Cairpre's ridge,' in Kincardineshire, in 1539 Drumquhariber (RMS). Cathalán is probably the second part of Petcathelin, Pethkathilin of early twelfth century (Chart. Lind.), now Pitcaithly, 'Cathalán's share.' From Cormac, 'chariot-lad' (corbmac) come Balcormock 1428 (RMS), near Abercrombie, and Balcormock 1485, ib. near Lundie in Fife, both now Balcormo; another Balcormo in Forfar, so spelled in 1489 (RMS), may be the same.

Fearchar gives Glenferkar, Glenferkaryn (Reg. Arbr.), now Glenfarquhar, 'Ferchar's glen,' in Kincardineshire. From the female name Finnghuala, 'white-shoulder' or 'white-shouldered,' comes Finella's Den in Kincardineshire.

[238] Gartán gives Srath Ghartáin, Strathgartney on Loch Katrine, 'Gartán's strath,' also Làirig Ghartáin, 'Gartán's pass,' off Glen Etive.

Muirgheal, 'sea-white,' a lady's name, gives Rathmuriel (RPSA), now Murriel in Inch parish, Aberdeen, 'Muriel's rath.' In Glen Nant, near Taynuilt, Dùn Muirgil, otherwise Dùn Meirghil is probably ' Muriel's fort.'

Mac Gille Eoin, 'son of St. John's lad,' is seen in Balmakgillona, Balmacgillon, c.1200 (Chart. Inch.), now Bellyclone, 'stead of the sons of St. John's lad,' near Inchaffray. The name is now Mac Gille Eathain, Maclean. Cupermaccultin, 1150 (Reg. Dunf.), is 'Cupar of the sons of Ultán' ; it is Cupermaculty, 1415 (Bamff Chart.), Cultirmacowty, 1525, ib., now Couttie near Coupar-Angus. Dùn Mac Tuathail, a fort on the eastern end of Drummond Hill, Aberfeldy, is 'fort of the sons of Tuathal'; Inchtuthil, on Tay, is 'Tuathal's meadow'; Auchtertool in Fife may be 'Tuathal's upland,' for in Ireland Tuathal is anglicized Toole. Dùn Mac Glais, Dunmaglass, at the head of Strath Nairn, is 'fort of the sons of Glas,' and Beinn Mac Duibh, Ben MacDhui in the Cairngorm range, is 'hill of the sons of Dubh' (black). (66) Ard Mac Maoin, on the north side of Loch Katrine, is 'height of Maon's sons' (maon, dumb). Baile Mac Cathain, Balmacaan in Glen Urquhart, is 'stead of the sons of Cathan.' Tomerail, opposite Killiecrankie station, is Tom Mhic Réill, 'Macneil's knoll'; similarly a great stone on Boreland Farm, Loch Tay, is Clach Mhic Réill, 'MacNeil's stone,' and a ferry on Tay below Ballinluig is Bàta Mhic Réill, 'MacNeil's Boat,' all with the not uncommon dialectic change of initial n to r.

Maol-domhnaich, 'Sunday's lad,' lit. 'Sunday's servant,' occurs in Petmuldonych, 1504 (RMS), 'Muldonych's share,' near Struan; the name survives in the burn name allt Phit 'al-domhnaich. (67) Morgan, a British name, appears in [239] Tillymorgan, Aberdeen ; Ramornie, Ramorgany 1512 (RMS), in Fife, seems to be for Rath Morganaigh, ' rath of (the) Morganach,' i.e. of a man of Clan Morgan of Aberdeenshire. (68)

Nechtán or Neachdán gives Dunnichen, of old Dunnechtyn (Reg. Arbr.), in Forfarshire, 'Nechtan's fort,' repeated in Badenoch as Dunachton; the former is considered to have been the scene of the great battle of Dún Nechtain in 685 (AU and Tighernach), where Brude, king of the Picts, routed the Angles. Bunachton in Stratherrick is Both Neachdain, 'Nechtan's hut.'

Selbach or Sealbhach, 'rich in possessions,' appears in Belhelvie, Aberdeen, 'Selbach's stead'; compare M'Kelvie.

The old name for the men of Ulster was Ulaid, gen. Ulad, and this is probably found in Rathillet in Fife, spelled Radhulit before 1200, Rathulit 1528 (RMS), for ráth Ulad, 'rath of the Ulstermen.' The later term for an Ulsterman is Ultach, whence Dùn nan Ultach, Downanultich 1539 (RMS), 'the Ulstermen's fort,' in Kintyre; compare Barnultoch in Wigtownshire.

In the west also personal names are fairly common, and as a rule easy to recognize. Glen Finnan in Inverness-shire is Gleann Fhionghuin, 'Fingon's glen'; compare MacPhionghuin, 'Mackinnon.' Glen Masan in Cowal is 'Massan's glen.'

One other point may be mentioned here, namely 'eclipsis '.

In Irish Gaelic a final n is in certain cases regularly carried forward to the next word if that word is closely connected with the preceding word. If the second word begins with a vowel, the final n of the preceding word is sounded and written before the vowel; if the second word begins with a consonant, the final n disappears before s, assimilates to l, m, n, r, and combines in various ways with the other consonants. In the case of c, l, p (tenues), nc, nt, np are sounded as g, d, b, and are written in modem Irish as gc, dt, [240] bp; in the case of g, d, b (mediae), ng, nd, nb are sounded as ng (as in English anger), n, m, and are written now ng, nd, mb ; lastly, nf is sounded as bh, i.e. as English v, and written now bhf. Thus 'in Ireland' is i nEirinn; 'of the blades' is na lann (for na nlann) ; 'of the dogs' is na gcon, pronounced na gon ; 'of the strings' is na dtéad, pronounced na déad ; 'of the teeth' is na ndéad, pronounced na néad; 'of the men' is na bhfear, pronounced na bhear. This process is commonly called 'eclipsis'; the corresponding process in Welsh is called 'the nasal mutation.'

In modern Scottish Gaelic eclipsis survives only in a few isolated instances, but it was once general here also. In the place-names we meet it as the result (1) of the preposition in or an, 'in'; (2) of the genitive plural of the article, van ; (3) of a neuter noun in the nominative singular. As to range, it is not uncommon in the east, rarer in the west, and very rare or non-existent in the Isles. As has been mentioned, it occurs in Lothian and in Galloway.

The record forms of names sometimes show eclipsis where the current forms do not. Gylltalargyn, 1203/24 (RM), is now Cill Taraghlain, Kiltarlity, 'Talorgan's church'; initial g of the old spelling is due to the influence of the preposition in the phrase i gCill Talorgain, which would be in constant use. Similarly we find Gillepedre, 1362 (RMS), for what is now Cill Pheadair, Kilpeter in Strathbrora ; and Gillecallumkille, 1566 (Orig. Paroch.), for Cill Chaluim Chille, 'St. Columba's church,' in the same strath. Cill Chriosd, 'Christ's church,' in Mull is Gilzacrest in 1496 (RMS); Cill Chaomhaidh, 'Caomhi's church,' near Logierait, Gilliquhamby 1558 (RMS), is anglicized now as Killichangie. Initial b under the influence of the same preposition becomes m. Mochastir 1452 (RMS), Mouchester 1524, ib., Moucastell 1579, ib., is now Both Chaisteil, Bochastle, 'hut of the castle' or 'chester,' near Callander, with reference to a Roman encampment which is now almost obliterated. Bonn-sgaod (also -sgaoid), Bonskeid near Pitlochry, is Monskeid, 1511 (RMS), and Both-reithnich, Borenich, 'bracken hut,' on Loch Tummel, is Montramyche (read Montrainyche), [241] 1508 (RMS). Bunchrew, 'near (the) tree,' (69) Inverness, is Monchrwe 1507, Moncrew 1510, Munchrow 1511 (RMS). The form Airdendgappil, 1351 (Reg. Lenn.), is for Ard na gCapull, 'cape of the horses' or 'of the mares,' now Ardincaple on the Gare Loch - an instance of the genitive plural of the article. A good example of the influence of a neuter nominative singular is Mogomar, 1500 (RMS), now Magh Comair, 'plain of (the) confluence,' anglicized as Mucomir, in Lochaber at the junction of Spean and Lochy ; Mogomar is for Magh gComair, a survival from the period when magh was neuter.

I shall now give examples of eclipsis in names that are still current in their eclipsed form.

(1) With the preposition in: Moness at Aberfeldy is for i mbun eas, 'near waterfalls,' lit. 'at waterfalls' foot.' Monessie in Lochaber is for i mbun easa, 'near waterfall,' the final -ie representing the old genitive sg. ending -a. The waterfall is Eas Chlianáig, whence the Glaistig took the stones to build Kennedy's house at Lianachan. (70) Munlochy is for i mbun locha, 'at loch foot,' or 'near the loch,' -y being for the old genitive sg. ending. Benderloch in Lorn, for beinn eadar dà loch, 'hill between two lochs,' is in Gaelic Meadarloch, where m, for mb, is all that is left of beinn in unstressed position. Muckairn in Lorne, Mocarne 1527 (Orig. Paroch.), is spelled Bo-càrna and Bu-càrna by the Gaelic poet Alexander MacDonald about 1750, (71) apparently for Both-càrna, '(?) hut of flesh'; this form, if correct, would readily become Mo-càrna after in.

(2) After nan, the genitive pl. of the article: Achnagairn, Beauly, is for achadh na gcarn, 'field of the cairns.' Achnagullan, Oykel, is for achadh na gcuilean, 'field of the whelps'; Amat, close by at the junction of Eunag and [242] Oykel, used to be called Àmad na gCuilean 'Amat of the whelps,' to distinguish it from Àmad na h-Eaglaise, Amat of the church, and Àmad na' Tuath, 'Amat of the laity,' in Strathcarron. Allt na gCealgach, on the road from Lairg to Lochinver, is 'the burn of the deceitful men,' in Macfarlane Alt Gellagach and Aldene-Gealgigh. (72)

Balnagore in Easter Ross is for baile na gcorr, 'stead of the cranes.' Balnaguard near Aberfeldy is baile na gceard, 'stead of the artificers.' Bada na Bresoch, Forfarshire, is for bad na bpreasach, 'spot of the copses.' Cairnagad in Aberdeenshire is 'wildcats' cairn.' Dalnavert in Badenoch and near Aberfeldy is for dail na bhfeart, 'dale of the graves.' Between Pitlochry and Glen Briarachan is Dail na gCarn, Dalnagairn, 'dale of the cairns.' Dail na bhFàd in Glen Briarachan is 'dale of the sods'; compare Blairnavaid in Stirlingshire. Féith na gCeann 'bog of the heads,' in Kirkmichael parish, Perthshire, is anglicized as Finegand. High up on Ben Lawers is Lochan na gCat, 'the wildcats' lochan.' Loinveg in Braemar is in Gaelic lòn na bhfiodhag, 'meadow of the bird-cherry trees.' (73)

(3) Of nouns originally neuter the most common in place-names are allt, a height, a burn, ceann, a head, comar, a confluence, druim, a ridge, dun, a fort, gleann, a glen, inbhear, a confluence, loch, a loch, magh, a plain, but very slight traces remain of their influence. Dundurcus, Elgin, might be explained on the theory that the second part is turcais, 'boar-place,' formed from torc like tarbhais, Tarves, from tarbh, a bull ; t becoming d, after dun. But there seems to be no other example of eclipsis after dun. Druingowdrom, Kildrummy, is probably for druim gcolldroma, 'ridge of hazel-ridge,' and Kingoldrum, Kincaldrum, 1505 (RMS), in Forfarshire means 'head of hazel-ridge,' from coll, call, hazel. Inbhir nAllt, Invernauld on the Kyle of Sutherland, means 'confluence of cliffs': the stream which [243] enters the Kyle here flows through a precipitous gorge; this old meaning of allt is seen, e.g., in an t-Allt Grànda, 'the ugly precipice,' the great gorge near Evanton in Ross-shire, and in an t-Allt Mór, the precipice near Inverfarigaig. Similarly we have Cumbernauld for comar (O.Ir. combor) nallt, where the meaning may be 'confluence of brooks.'

Monzie near Crieff is Mugedha, 1226/34 (Chart. Inch.), Muyhe, 1226/34 (Reg. Inch.), Moythethe, 1282 (Chart. Inch.), now in Gaelic Magh-eadh. The second part is probably for eadha, earlier etho, the genitive sg. of O.Ir. ith, corn, later iodh, as in G. iodh-lann, a cornyard; the old name would thus be mag n-elho, 'plain of corn,' represented very well by the old spellings, which however rather strangely omit the n preserved in the English form. Monzievaird in Strathearn, near Monzie, is Muithard 1200 (Chart. Inch.), Monewarde 1203, ib., Moytheuard 1234, ib. ; Moeghauard c. 1251 (Skene, P.S.), also Morgoauerd (read Mog-), in Gaelic now magh bhàrd, correctly translated 'bardorum campus,' 'plain of the bards,' in the Cronicon Elegiacum (c. 1270). As the old form would be strictly mag mbard (for mag nbard), pronounced magh mard, which could not result in 'Monzie-vaird,' it is probable that the latter has been influenced by the neighbouring 'Monzie.'





1. I have here made a synthesis of several accounts which are to be found in Skene, P.S., pp. 45, 319, 328 (BB 43 b 49 fcs.); Leathar Gabhála, p. 233 (ed. John MacNeill and R. A. S. MacAllister); BB 19 a 37.

2. 'Habitent terram Fif, et a monte Britannorum usque ad montern qui dicitur Okhél' (Skene, P.S., p. 416). The Register of St. Andrews mentions a place called Munobretun, apparently in Fife.

3. Macfarlane, Geog. Coll., i. p. 330. It is said to have been 19 feet 6 inches in internal diameter, avid 22 feet in height to the round opening at the top, which was 11 feet 6 inches in diameter. The building was circular and rounded towards the top. (Pennant, ii. p. 228.)

4. Tract on the Mothers of the Irish Saints, BB 212 a-214 b: 'Alina (? Alnia) ingen rig Cruithnech mathair Seirb maic Proic rig Canandan Egipti 7 is e sin in sruth senóir congeibh Cuillenn-ros i nSraith Érenn i nComgellaib etir sliab n-Ochel 7 mur (sic) nGiudan'; 'Alma, daughter of the king of the Cruithnigh, was the mother of Serb, son of Proc, king of the Canaanites of Egypt; and he is the sage senior who set up (the religious establishment of) Cuillenn-ros in Strath Earn in Comgellaib between Mount Ochel and the sea of Giudan' - BB 214 a 19. The parallel passage from the Book of Lecan is printed by Skene in Celtic Scotland, ii. p. 258 n., where S. Serf's mother's name is given as Alma, and Culross is said to be hi Sraith hIrend hi Comgellgaib.

5. Compare Mons Okhél, p. 208.

6. Rev. C. M. Robertson, Celt. Rev., i. p. 93. In equating Ekkjall with a late form of uxellos the difficulties are (1) initial c ; in view of G. a, it seems not unlikely that c may be an error; (2) kk for ch: Norse often makes G. ch into k, e.g. Búkan for Buchan, but there is no example of ch becoming kk; that Norse kk is the correct reading is proved by G. c in Oiceil, for a single k of Norse would become g in Gaelic.

7. 'Standard,' which occurs in Carrick, Galloway, and Dumfriesshire as a hill name, is probably a translation of W. lluman, a standard, confused with llumon, a beacon.

8. Celtic Scotland, iii. p. 125.

9. Agricola. c. 24.

10. The third was Criomhthan Mór, son of Fidach : 'triffiur dochomairt fiaith nAlban do Gáidelaib i . Rechtaid Rigderg et Labraid Loingsech et Crimthand Mór mac Fidaig'; Rawl. B 502, 148 h 12.

11. BB 258 a 58 sqq. ; YBL 178.

12. Celt. Zeit., iii. p. 461.

13. The position of Cairbre's settlement in Munster is stated thus: 'robí dano tír Cairbri Rigfhoda in tír itáit Ciarraige Luachra 7 Orbraige Droma Imnocht, conad win dolotar i nAlbain. Gabsat Corco Duibne in tír atat.' 'Now Cairbre Rigfhoda's land was the land wherein are the Ciarraige Luachra and the Orbraige of Druim Imnocht, and it was thence that they went to Alba. The Corco Duibne took possession of the land wherein they are (now)' ; BB 140 a 49. North Kerry and Orrery in Cork are the corresponding places now.

14. Adamnan visited Aldfrid, king of Northumbria, twice, in 686 and 688, and it was probably in 686 that he visited Jarrow; he was then sixty two and Bede was thirteen. Bede's information was doubtless got later. Adamnan died in 704.

15. The question at issue has been much misunderstood, but certainly had reference to the political position of Dál Riata in Ireland, not to that of Dál Riata in Alba. The Irish king had threatened to drive the Dál Riata of Ulster and the poets across sea (tafunn Dáil Riata dar muir 7 tafunn na n-cess. - Rawl. 502 B, 96 a 15).

16. LL 38 b, foot.

17. LL 140 a 27; 288 a.

18. 'a longus co h-íath nAlpan. Forbrisfi cethri mórcatha for túatha Orca,' Celt. Zeit., iii. 461.

19. For the Fothads, see LL 190 b; BB 164 b; Rawl. B 502, 155 b; Fianaigecht (Todd Lect., xvi.), p. 4 sqq. ; Cóir Anmann, 220.

20. Rawl. B 502, 141-2 ; LL 333 a 8. Celt. Zeit., viii. 317 seq.

21. Rawl. B 502, 81 a 35; 136 b 23.

22. 'bíet ile a gluinn ar Druim nAlpuind ' (read nAlpund) ; Celt. Zeit., iii. 463.

23. YBL 192 b 25 sqq.

24. Compare 'Conall Corc and the Corco Luigde '; Anecdota, iii. 57. A tale almost identical is told of the Emperor Conrad ii. (d. 1039) ; it is found in Gesta Romanorum and in Wyntoun. The Ogham motif is, of course, as old as the tale of Bellerophon.

25. In Sanas Cormaic (no. 688): 'Gorn. i. gai-orn. i. gae orcne. i. aithinde, unde dixit Gruibne oc fáilte fri Core: immicuirithar gurna gair.' The words here quoted occur in the LL version thus: '(ditnech anaill Chuirc) immaluritar carnd gáir' (LL 287 a 24). 'Ditnech ' is corrected to 'ditdech' in the MS. (fcs.). This retoric therefore was old when Sanas Cormaic was composed; it was probably contained in the Psalter of Cashel.

26. For a transcript of the relative part of Húi Maine. 1 am indebted to Professor T. F. O'RahiIly.

27. See p. 221.

28. 'Ic hEber condrecait na secht nEoganachta 7 Lemnaig Alban ' ; 'at Eber meet the seven Eoghanachts and the Lennoxmen in Alba'; LL 318 b 42. 'Eber nero is da claind-sein Dáil Cais . . . 7 Eoganacht Caisil . . . 7 Leamnaigh i nAlbain '; 'as to Eber, of his descendants are the Dáil Cais . . . and the Eoghanacht of Cashel. . . . and the Lennoxmen in Alba' ; BB 41 b 36.

29. Celtic Scotland, iii. 454, from MacFirbis' Book of Genealogies. Another copy, for a transcript of which 1 am indebted to Professor T. F. O'Rahilly, is in R.I.A., MS. 23 L 17 fo. 15 a. This copy is anonymous, but R.I.A., MS. 23 D 4 (p. 93) gives the first ten quatrains under the name of Muireadhach Albanach, as Professor O'Rahilly informs me.

30. Irish Text Soc., ii. p. 386.

31. Register of Lennox.

32. Rawl. 502 B, 00 g, 91 d; LL 351 c 1, c 24; LB 19 a, 19 b. The genealogies are: (1) Cummine Fota mac Fiachna maic Fiachrach maic Duach maic Maine maic Cairpre maic Cuircc maic Lugdach maic Ailella Flainn Big maic Ailella Flainn Móir; (2) Faithlenn mac Aeda Damain maic Crimthaind maic Cobthaich maic Duach larlaithi maic Maithne (Maine) maic Cairpre maic Cuircc maic Luigdech (Rawl.).

33. Proceedings of Royal Irish Academy, vol. xxvii. sect. C, no. 15 (1909).

34. Irish Names of Places, i. p. 261.

35. Except in Wester Ross, where it means 'a fairy seat', like sìthean.

36. Charters of Inchaffray, p. lxxv. (Scot. Hist. Soc.).

37. Chart. of Lindores, 27, etc., 243 (Sect. Hist. Soc.). The district called Kathermothel included Fedal and Cotken, the latter of which appears later (Retours) as the moor of Ardoch.

38. Compare Macfarlane, Geog. Coll., i. p. 353: 'the house of Easter Catter closs by which ther was once ane old castle belonging to the ancient Earles of Lennox.'

39. Kuno Meyer, Expulsion of the Dessi; Y Cymmrodor xiv. (1901) where their genealogy is given.

40. p. 64; it was called Clad na Muicc, 'the pig's fosse' also 'doronad clad accu dar in i-insi ri hucht Cruithnech 7 Góedel' ; Laud 610, 90 a 1 (quoted in Kuno Meyer's Contribb. s.v. clad).

41. Amra Coluim Cille and notes.

42. Irish Lives, i. p. 35; ii. p. 327.

43. A. O. Anderson, Early Sources, cliv. p. 149 ; Skene, P.S., p. 315, n.7. Conall Cearr is also given as Conadh Cearr.

44. Skene, P.S., p. 137; A. O. Anderson, Early Sources, p. cxvii.

45. 'Foelan mac Oengusa mic Natfraich, Raith Erend i nAlbain 7 o Chill Faelan i Laigis'; note on Félire of Oengus, June 20, in Leb. Breac, p. 90.

46. Skene, P.S., p. 88; A. O. Anderson, Early Sources, i. p. 367 (where Skene's reading and translation are corrected).

47. From M'Clure's British Place-Names, p. 205.

48. Félire of Oengus (Stokes), p. 119.

49. Rev. Celt., ii. p. 180.

50. For -thfh- becoming bh compare Srath-thamhuinn for Srath-athfhinn, Strath Avon in Banffshire (Inverness Gael. Soc. Trans., xxiv. p. 165). In thamhuinn the th is carried over from srath; cf. Bohespick, etc.

51. Lawrie, Early Scottish Charters, p. 442.

52. Records of the Monastery of Kinloss.

53. Charters of Inchaffray.

54. Zur kelt. Wortkunde, no. 42.

55. 'Ealga. i. Eiriu. i. ealg ainm do mhuic isin teen-gaidhlig'; 'Ealga, i.& Eriu; in Old Gaelic ealg means 'pig': Coir Anmann, 243. 'Inis Ealga. i. oilean uasal'; 'Inis Ealga 'means' noble island,' Keat, i. p. 98.

56. Adv. Lib. MS. LII 43 a; XXXIX 32. The poem is to Macleod.

57. The Town Clerk of Elgin informs me that the origin of this term is quite unknown.

58. Zur keltischen Wortkunde, no. 42.

59. Archiv f. Celt. Lexik., iii. p. 45.

60. A full discussion of the davach and its equivalents is given by Skene, Celtic Scotland, iii. p.223. It was reckoned equal to 1 tirung or ounce-land or to 4 ploughgates or to 20 pennylands. A ploughgate or carrucate was 8 oxgangs, so that the davach was equal to 32 oxgangs; but this proportion may not have been invariable, for in MacGill's Old Ross-shire and Scotland, ii. p. 31 there is 'that western oxgate or fourth part of the half davoch toun and lands of Newnakill.' Pennant in his Tour of 1772 (p. 314) says of Loch Broom: 'Land is set here by the Davoch or half Davoch; the last consists of ninety-six Scotch acres of arable land, such as it is, with a competent quantity of mountain and grazing ground. This maintains sixty cows and their followers; and is rented for fifty-two pounds a year. To manage this the farmer keeps eight men and eight women servants, and an overseer.'

61. It appears from the Gaelic entries in the Book of Deer that the king, the mormaer, and the toiseach had, cacti of them his own rights in respect of these burdens, e.g. each was entitled to dues of maintenance.

62. In the Shadow of Cairngorm, p. 339.

63. For this and much other information about Perthshire I am indebted to Mr. Alexander Campbell of Boreland, Loch Tay.

64. 'Obsessio Aberte apud Selbacum ' (AU). There seems no doubt that Dunaverty is meant, though dún is omitted and the natural spelling would be (Dún) Ábartaig.

65. O. Welsh bresel means 'war' (Zeuss, Gram. Celt., p. 135). The name might thus be claimed as British, but compare Ballybrassil, Ballybrazil, Clonbrassil, ete. - Joyce, vol. iii.

66. It may be 'Macduff's peak,' for Macduff (MacDuibh), Earl of Fife, held much land in that neighbourhood in early times.

67. I was informed by the late Mr. John Whyte that in Skye the name Maol-domhnaich used to be given to a boy whose maintenance was provided for by the Sunday's collection.

68. The Mackays of Sutherland were. known as Clanna Morgainn, also as Morganaigh or Morganaich, the plural of Morganach. - Rel. Celt., ii. pp. 176, 260.

69. For i mbun chraoibhe ; here the phrase i mbun or a mbun is used idiomatically as often, in the sense of 'close to,' in the same way as we say 'an cois na fairge,' 'near the sea,' lit. 'at foot of the sea.'

70. This well-known tale is too long to tell here; see Carmina Gadelica, ii. p. 287, another good English version is given by Mr. Donald A. Mackenzie in his Elves and Heroes.

71. Celtic Review, iv. p. 303 ; v. p. 226.

72. So called according to tradition because in a dispute about marches, the witnesses on one side swore that they wore standing on the soil of a certain estate, having first put some of it inside their brogues.

73. For this I am indebted to Mr. F. C. Diack.