''To the Gaelic language a great proportion of the names of places in the neighbourhood, and indeed through the whole of Fife, may unquestionably be traced All names of places b^linning with Bal, Ck>l or Cul, Dal, Drum, Dun, Inch, Inver, Auchter, Kil, Kin, Glen, Mon, and Strath, are of Gaelic origin. Those b^iinning with Aber and Pit are supposed to be Pictish DIALECTIC DIFFERENCES INDICATED BT IT. 149 names, and do not oocar beyond the tenitory which the Pkta are thought to have inhabited" Chalmers states it still more broadly and minutely. He says — ** Of those words which form the chief oompoanda in many of the Celtic names of phices in the LowUnds, some are exdmiYely British, as Aber, LUn, Gaer, Pen, Cors, and othws; some are common to both British and Irish, as Cam, Craig, Crom, Bra^ Dal, Eaglis, Glas, Inis, Binn, Bos, Strath, Tor, Tom, Olen; and many more are significant only in the Scoto-Irish or Gaelie, as Ach, Aid, Ard, Aird, Auchter, Bar, Blair, Ben, Bog^ daoh. Cony, Cul, Dun, Drum, Fin, Olao, Inyer, Kin, Kil, Knoc, Laig^ Lnig^ Lag, Logic, Lead, Letter, Lon, Loch, Meal, Pit, Pol, Stron, Tullach, Tullie, and others."
This attempt at classification is, however, ex- ceedingly inaccurate. Two of the words in the first class, Llan and Caer, are common to both British and Irish; and a large portion of the third class are signi- ficant in pure Irish, as well as in the Scoto-Irish or Gaelic. No attempt is made to show, by the geo- graphical distribution of these words, in what parts of the country the respective elements prevaiL The most popular view of the subject, and that which has recently been most insisted in, is the line of demarcation between a Cymric and a Gaelic population, supposed to be indicated by the occurrence of the words Aber and Inver. This view has been urged with great force by Kemble, in his J nglo-SdXons; but I may quote the recent work of Mr. Isaac Taylor, on words and places, as containing a fair statement of the popular view of the subject:— 150 CELTIC TOPOGRAPHY OF SCOTLAND: ^To establish tho point that tlie Picts, or the nation, what- ever was its name, tliat held central Scotland, was Cymric, not Graelic, we may refer to the distinction already mentioned between Ben and Pen, Ben is confined to the west and north; Pen to the east and south. Inver and Aher are also useful test-words in dlBcriminating between the two branches of the Celts. Tho difference between the two words is dialectic only; the etymology and the meaning is the same — a confluence of waters, either of two riven or of a river with the sea. Aber occurs repeatedly in Brittany, and is found in about fifty Welsh names, as Aberdare, Abeipivenny, Abei^le, Aberystwith, and Barmouth, a corruption of Abermaw. In England wo find Aberiord in Yorkshire, and Berwick in Northumberland and Sussex; and it has been thought that the name of the Humber is a corruption of the same root InveTf the Erse and Gaelic forms, is common in Ireland, where Aber is unknown. Thus, we find places called Inver in Antrim, Donegal, Mayo, and Invermore in Galway and in Mayo. In Scotland the Invers and Ahers are distributed in a curious and instructive manner. If we draw a line across the map from a point a little south of Inveraray to one a little north of Aberdeen, we shall find that (with very few exceptions) the Invers lie to the north of the line and the Ahers to the south of it This line neariy coincides with the present southern limit of the Gaelic tongue^ and probably also with the ancient division between the Picts and the Scots."
Nothing can be more inaccurate than this state- ment. Ben is by no means confined to the west and north; and as examples of Pen he refers^ among others, to the Penthmd Hills, Pentland being a Saxon word, and corrupted from Pectland; and to Pendriech in Perthshire, which is a corruption from Pittendriech. So fiir firom Inver being common in Ireland, it is very rare. The Index locorum of the Annals of the Four Masters shows only six instances. On the other hand, Aber is not unknown in Ireland. It certainly DIALECTIC DIFFEREKCES IKDICATBD BT IT. 151 existed formerly to some extent in the north of Ireland; and Dr. Reeves produees four instances near Ballyshannon.
The statement with regard to the distribution of Aber and Inver in Scotland here is, that there is a lino of demarcation which separates the two words — ^that^ with few exceptions, there is nothing but Invers <m one side of this line, nothing but Abers on the other; and that this line extends firom a point a little south of Inveraray to a point a Uttle north of Aberdeen. This is the mode in which the distribution of these two words is usually represented, but nothing can be more perfcctiy at variance with the real state of the case. ^ South of this line there are as many Invers as Abeis. In Perthshire, south of the Highland line, there aio nine Abers and eight Invers; in Fifeshire, four Abers and nine Invers; in Forfar, eight Abers and eight Invers: in Aberdeenshire, thirteen Abers and twenty- six Invers. Again, on the north side of this suppo^ line of demarcation, where it is said that Invers alone should be found, there are twelve Abers, extending / across to the west coast, till they terminate with Aber- crossan, now Applecross, in Eoss-shire. In Argyll- ^ shire alone there are no Abers. The true picture of the distribution of these two words is — ^in Argyllshire, luvers alone; in Inverness and Boss shires, Invers and Abers in the proportion of three to one and two to one; and on the south side of this supposed line, Abers and Invers in about equal proportions.
Again he says, quoting Chalmers, *' The process of 1 152 CELTIO TOPOGRAPHY OF SCOTLAKD: change is Bhown hy an old charter, in which King David grants to the monks of May 'Inverin qui fuit Aberin.' So Abcmethy became InYemethy, although the old name is now restored.^ In order to produce the antithesis of Inverin and Aberin, one letter in this charter has been altered. The charter is a grant of " Petneweme et Inverin quae fuit Averin; ** and I have the authority of the first charter antiquary in Scotland for saying that this construction is im- possible: ^ qu8d fuit '* does not, in charter Latin, mean ** which was^" but ** which belonged to,** and Averin was the name of the previous proprietor of the lands. Abemethy and Invemethy are not the same place, and the former never lost its name. Invemethy is at the junction of the Nethy with the Earn, and Abemethy 18 a mile further up the river.
When we examine these Abers and Invers more closely, we find, 1^^, that in some parts of the country they appear to altemate, as in Fife — Inverkeithing, Aberdour, Inveiyne, Abcrcrombie, Invcrlevin, and so forth; 2d, That some of the Invers and Abers have the same specific terms attached to them, as Abemethy and Invemethy, Aberachill and Inverochill, Abercrumbye and Invercrumbye, Abergeldie and Invergeldie; and 3d, That the Invers are always at the mouth of the river, close to its junction with another river, or with the sea; and the Abers usually a little distance up the river where there is a ford. Thus Invemethy is at the mouth of the Nethy; Abemethy a mile or two above. These and other fietcts lead to the conclusion that they DIAXiECnC DIFFERENCES INDICSATED BY IT. 153 are part of the saine nomenclature, and belong to the same period and to the same people.
When we look to the south of the Forth, howeyer, we find this remarkable circumstance that in Aynhire, Kenfrew, and Lanarkshire, which formed the posses- sions of the Stratbclyde Britons, and were occupied by a British people till as late a period as the' more northern districts were occupied by the Picts, there are no Abers at alL What we have, therefore, is the Scots of Argyll with nothing but Invers^ the Picts with sUl, Abers and Invers together, and the Strathclyde Britons with no Abers.
As a mark of discrimination between races this criterion plainly breaks down, and the words them- selves contain no sounds which, from the different phonetic laws of the languages, could afford an indica- tion of a dialectic difference. The truth is, that there were three words expressive of the junction of one stream with another, and all formed from an old Celtic word, Ber, signifying water. These were A ber, Inver, and Conber (pronounced in Welsh cummer, in Gaelic cumber). These three words were originally common to both branches of the Celtic as derivations firom one common word. In old Welsh poems we find not only Aber as a living word in Welsh, but Ynver likewise.* and Dr. Reeves notices an Irish document in which Applecross or Appurcrossan is called Conber Crossaiu Ynver, however, became obsolete in Welsh, just as Cummer or Cumber and Aber became obsolete in Irish; * Tnver uccun twice in the Book of Talieniii 154 CELTIC TOPOGRAPHY OF SCOTLAND: but we have no reason to know that it did so in Pictish. In the Pictish districts, therefore, the Abers and Invers were deposited when both were living words in the language. When the Scots settled in Argyll, Aber had become obsolete in their language, and Inver was alone deposited, and in Strathclyde both words seem to have gone into desuetude.
In the same manner Dwfr or Dwr^ is quoted as a word for water, peculiar to the Welsh form of Celtic, and an invariable mark of the presence of a British people, but the old form of this word in Scotland was Dobair, as appears fix)m the Book of Deer, where Aber^ dour is written Ahher^hoir^ and in Cormac's Glos- sary of the old Irish, Dchoir is given as an old Irish word for water. In another old Irish glossaiy we have this couplet: — ^ Bior and An and Dobar, The three names of the water of the world" These words, therefore, form no criterion of differ- ence of race, and to judge by them is to fall into the mistake of the phonetic etjnnologists — ^viz. to apply to old names^ as the key, the present spoken language, which does not contain words which yet existed in it in its older form.
In order to make generic terms a test of dialect, they must be words which contain sounds affected differently by the different phonetic laws of such dialects — such as Pen, Owynn, Gwem, and Owydd which all enter copiously mto Welsh topography, and the equivalents of which in the Gaelic dialects DIALECTIC DIFFKREKCE8 INDICATED BT IT. 155 are Ceann, Fionn^ Feam, and Fiodh. Such generic terms afford a test bj which we can at once determine whether the Celtic topography of a country partakes most of the Cymric or the Gaelic character. The earliest collection of names in North Britain is to be found in Ptolemy's Geography in the second century, but we know too little of the origin of his names, whether they were native terms, or names applied by the invaders, to obtain from them any cer- tain result After Ptolemy, the largest collection of names in Great Britain is in the work of the anony- mous geographer of Ravenna, a work of the seventh century. The exact localities are not given, but the names are grouped according to the part of Britain to which they belong. Those which commence the topo- graphy of Scotland are placed under this title: — ^Iterum sunt civitates in ipsa Britannia qusB recto tramite de una parte in alia, id est^ de oceano in oceano existunt, ac dividunt in tertia portione ipsam Britan- niam." They commence with the stations on the Roman wall between the T}me and the Solway, and then proceed northwards. Among these we find two names together, Tadoriton and Maporiton, and as Tad and Map are Cymric forms for father and son, we have no doubt that here we are on the traces of a Cymric population. The next group is arranged under this head: — " Iterum sunt civitates in ipsa Britannia recto tramite una alteri conexse, ubi et ipsa Britannia pins angustissima de oceano in oceano esse dinoecitnr.* This part of Britain, which is plus anguatimmaf is the 156 CELTIC TOPOGRAPHY OF SCOTLAND: isthmus between the Forth and the Clyde, and in pro- ceeding with the names northwards we come to one called CindocellunL Tlie Ocelli Montes were the OchiUsy and here the Gaelic form of Kin is equally un- mistakable. When we apply to the present topography tihe testing words Pen, Gwynn, Gwem, and Gwydd, the Gaelic equivalents of which are Kin, Fionn, Feam, and Fiodh, we find that^ with one exception, Pen, though frequent south of the Forth, where there was a British population, docs not occur north of the Forth, while it IB full of Kins, and Gwynn, Gwem, and Gwydd occur only in their Gaelic equivalents.
Such then being the aspect in which the question really presents itself, it becomes important, with a view to ethnological results, to ascertain more closely the geographical distribution of the generic terms over Scotland, and in order to show this I have prepared a table of such distribution. The generic terms are taken from the index to the Scottish Record of Retours; and as this record relates to properties, and not to mere natural objects, the generic terms they contain are to a great extent confined to names of places connected with their poesessionby man.and more readily affected by changes in the population. For the purposes of comparison, I have framed a list of generic terms con- tained in Irish topography fix>m the index to the Annals of the Four Masters, and of those in Welsh topography firom a list in the Cambrian Register. I have divided Scotland into thirteen districts, so as to •how the local character of the topography of each DIALECnO DIFFEBEN0B8 IMDKUTED BY Tt. 167 part of Scotland, and opposite each generic term in Scotch topography is marked— 1^ if it oocnrs in Ireland, and how often; 2d, if it occotb in Wales; and Sd, I have marked the number of times it Occur in each district of Scotland from the Index of Betoma.
On examining this table, it will be seen that there ore five terms peculiar to the districts occupied hy the Picts. These are Auchter, Pit^ Pitten, For, and Fnu Now none of these five terms are to be found in Welsh topography at all, and For and Fin are ohnotaalj Gaelic forms.
It is necessary, however, in examining these terms^ which may be called Pictish, to ascertain their old form. Auchter appears to be the Gaelic UadUer^ upper; and as such we have it in Ireland, and in the same form, as in Scotland Ochtertire^ in Ireland Uach* tertirc. It does not occur in Wales.
The old form of Fit and Pitten, as appears from the Book of Deer, is PettCf and it seems to mean a portion of land, as it is conjoined with proper namos^ as Pette MacGamait, Pette Malduib. But it also appears connected with Gaelic specific terms, as Pette an Mulenn, the Pette of the MiU, and in a charter of the Chartulary of St Andrews, of the church of Migvie, the terra ecclesiiB is said to be vocatus Petten- taggart — *' an taggart ** being the GaeUc form of the expression " of the priest* The old forms of For and Fin are Fothuir and Fothcn« The old form of Forteviot is Fothuir-. tabaicht^ and of Finhaven is Fothen^vin. ^ The firat 158 CELTIC TOPOORAPIIY OF SCOTLAND: of these woidsy however, discloses a very remark- able dialectic difference. Fothair becomes For, as Fothuir-tabacht is Forteviot; Fothuir-duin is For- dun; but Fothuir likewise passes into Fetter, as Fothoiresach becomes Fetteresso; and these two forms are fonnd side by side, Fordun and Fetteresso being adjacent parishes. The form of For extends bom the Forth to the Moray Firth— that of Fetter from the Esk, which separates Forfar and Kincardine, to the Moray Firth.
An examination of some other generic terms will disclose a perfectly analogous process of change. The name for. a river is Amhuin. The word is the same as the Latin Amnis. The old Gaelic form is Amnin, and the m, by aspiration, becomes mh, whence Amhuin, pronounced Avon. In the oldest forms of the lan- guage the consonants are not aspirated, but we have these two forms, both the old unaspirated form and the more recent aspirated form, in our topography, lying side by side in the two parallel rivers which bound Linlithgowshire — the Amend and the Avon. There is also the Amend in Perthshire. We know fiom the Fictish Chronicle that the old name was Aman, and the Avon, with its aspirated m, is men- tioned in the Saxon Chronicle. It is a further proof that Inver is as old as Aber in the eastern districts, that we find Aman in its old form conjoined with Inver in the Fictish Chronicle in the name ^ Inveraman.''
In Dumbartonshire we find the names Lomond and Leven together. We have Loch Lomond and Ben DIALBCTIO DIFFBIIENOES INDICATED BT FT. 159 Lomond, with the river Leven flowing oat of the loch through Strathleven; but we have the same names in connection in Fifeshire, where we have Loch Leven with the two Lomonds on the side of it^ and the xiTer Levcn flowing from it through Strathleven. This recurrence of the same words in connection would be. unaccountable, were it not an example of the same thing. Leven comes from the Gaelic Leamhan^ signi- fying an dm-tree, but the old form is Leoman, and the m becomes aspirated in a later stage of the language and forms Leamhan^ pronounced Leven. Here the old form adheres to tiie mountain, while the river adopts the more modem.
A curious illustration of two different terms lying side by side, which are derived from the same word undergoing different changes, will be found in Forfar- shire, where the term Llan for a church appears^ as in Lontrcthin. It is a phonetic law between Latin and Celtic, that words beginning in the former with pi jv^. are in the latter H The word Planum, in Latin signifying any cultivated spot^ in contradistinction from adesert spot^ and which, according toDucange, came to signify Cimiterium, becomes in Celtic Llan, the old meaning of which was a fertile spot» as well as a churcL In the inquisition, in the reign of David L, into the possessions of the See of Glasgow, we find the word in its oldest form in the name Planmichael, now Carmichael; and as we find BaUin corrupted into Ban, as Ballindoch becomes Bondoch, so Plan becomes cor- rupted into Pan, and we find it in this form likewise 160 CELTIC TOPOORAPHY OF SCOTLAND: in Forfarshire, Ponmure and Panbride. In the Lothians and the Merse this word has become Long, as in Longnewton and Longniddrie.
The Celtic topography of Scotland thus resembles a palimpsest^ in which an older form is found behind the more modem writing. I shall not lengthen this chapter by going through other examples. The exist- ence of the phenomenon is sufficiently indicated by those I have brought forward, and I shall conclude by stating shortly the results of this investigation.
Ist, In order to draw a correct inference firom the names of places as to the ethnological character of the people who imposed them, it is necessary to obtain the old form of the name before it became corrupted, and to analyse it according to the philological laws of the language to which it belongs.
2d, A comparison of the generic terms affords the best test for discriminating between the different dialects to which they belong, and for this comparison it is necessary to have a correct table of their geo- graphical distribution.
3d, Difference between the generic terms in different parts of the country may arise from their belonging to a different stage of the same language, or from a capricious selection of different synonyms by separate tribes of the same race.
Ath, In order to afford a test for discriminating between dialects, the generic terms must contain within them those sounds which are differently affected by the* phonetic laws of each dialect DlALEOnO DIFFJSRSN0B8 INDICATED BY IT. 161 5th, Applying this test» the generic tenns do not show the existence of a Cymric language north of the Forth.
6th, We find in the topography of the north-eist of Scotland traces of an older and of a more recent form of Gaelic — ^the one preferring labials and dentals^ and the other gutturals; the one hardening the coDr sonants into tenues — the other softening them by aspiration; the one having Abers and Inyen — and the other having Inyers alone; the one a low Gkielic dialect — ^the other a high Gktelic dialect; the one I concehre the language of the Ficts — the other that of the Scots.* * The KilMtaiite of theie Uivm efaapten bit aluiidx i^ptaivd is a different shape in the ArdUtol^^ Cfam^rami, and the laii is the TransaeiionM of iki Bdjfol Socitij. Ttmj wen writtn with a vkw to this work.
CUMBRIA AND THE MEN OF THE NOBTH.
The distxicts comprehended at an early period under the name of Cumbria were of considerable extent; and, as its name indicates, occupied by a C}anric popolatioiL Joceline, who wrote about the year 11 80, in his life of Kentigem, states that the limits of his bishopric were coextensive with those of the '* regio CambrenaiSy* and extended &om the Roman wall to the '^flumen Fordense;'' but it originally extended even further south than this, for Joceline was judging by the extent of the diocese of Glasgow, and Carlisle and the district surroimding it had, after the Norman Con- quest of England, been formed into an earldom, and in 1132 erected into the diocese of Carlisle. In a document printed in the lolo MSS., the extent o£many of the old Welsh districts is given, and the district of Teymllwg is said to have extended £ix)m Aerven to Argoed Derwennydd — that is, to the Forest upon the Derwent. This river, which falls into the Western Sea at Workington, now divides the diocese of Chester from that of Carlisle; and as soon as we pass the Derwent, dedications of churches to Kentigem com- mence. The district south of the Derwent had very early come under the power of the kings of Northr 166 CUMBRIA AND THE MEN OF THE NORTH.
umbcrlandy and the independent states of the Cymry probably extended &om the Derwent and from Stan- more to the Clyde, including Westmoreland (with the exception of Kendal), and the central districts in Scotr land, of Teviotdale, Selkirk, and Tweeddale. It com- prehended what afterwards formed the dioceses of Glasgow and Carlisle; and its C}rmric population appears as a distinct people, even as late as the battle of the Standard, in 1130, where they formed one of the battalions in King David's army, consisting of the Combrenses and Tevidalenses.
They appear to have been composed of numerous small states under their petty kings.
There is a document in one of the Hengwrt MSS., transcribed about 1300, with the title of BorUicd Gwifr y Oogledd, or Grcnealogies of the Men of the Korth — a name used to designate these Northern Cymry. It gives the pedigrees of twelve families, and they fall into three groups — one consisting of six fiunilies, whose descent is traced from Ceneu, son of Coel; • the second, of five families descended fix)m Dy&wal Hen, or the aged, grandson of Macsen Guledig, the Soman Emperor; and the third, of one family cmmected with the north, apparently through the female line. The first group again falls into two branches respectively derived fix)m two sons of Ceneu, son of Coel, Gorwst Ledlwm, and Mar or Mor. To Mcichion Gul, the son of Gorwst Ledlwm, are given two sona — Cynvarch, the father of Urien and Elidir Lydanwyn, father of Llywarch Hen. To Garthwys CUMBRIA AND THE MEN OF THE NOHTH. 167 or Artiiwys, son of Mor, are given four sons — Ceidiaw, the father of Gwenddolew, Nudd, and Gov; EUvir Gosgorddvawr, or of the large retinue» the father of Gwrgi and Peredur; Pabo Post Prydain, or the pillar of Britain, the father of Sawyl Benuchel, Dunawd Yawr, and Carwyd; and Cynvelyn, the grandfather, hj hig son Cynwyd Cynwydion, of Qyddno Eiddyn, Cyiuai Genhir, Cadrod Calchvynydd, and Cynvdyn DrwagL - The second group, consisting of the descendants of Djrfnwal Hen, also falls into four branches, descended of four sons of Dyfiiwal Hen: — Cedig, father of Tudwal Tudclud, the father of Rydderch Had, Senyllt^ fSEkthcr of Nudd Hael, and Servan, father of Mordav; Gar- "wynwyn, father of Caurdav, father of Gwyddno Garanhir; Aeddan Vradog; and Gorwst Briodawr, father of Elidr Mwynvawr.
The genealogies annexed to Nennius in 977 do not greatly differ from this. In the first group of families descended from Cocl they add the pedigrees of two additional families — that of Gwallawg ap Leenawg and of Morcant. In the second group, the most important variation is that the descent of Dyfnwal Hen, the common ancestor, is not brought from Macsen Gulcdig but from a Caredig Guledic, whose pedigree is taken back to a Confer the Rich; and that the descent of the later kings of Strathclyde from Dy&wal Hen is given.
Adding, therefore, the two additional families de- scended from Coel, we have eight in the first group, and five in the second — in all, thirteen; and the fol- lowing tables will show their connection: — o g o o o GO o g Eh s I I I' I s I 51 N^ CT i I •c I ^-H-ih -I I .1 9 0.
It isy of course, not maintained that tliesc gene- alogies are, strictly speaking, historical, and that e^ich link in the pedigree represents a real person; but they are valuable as conveying a general idea of the period and tribal connection of these " Gwyr y Gogledd/' or Hen of the NortL The thirteen families no doubt re- presented as many petty states in Cumbria; and in the two groups we can see the mixture of two races — the provincial Roman and the native C3rmric — and the small septs into which they were respectively divided.
There are indications, derived from their names, their history, and from local tradition, which con- nect most of these families with localities within the limits of Cumbria. Beginning with the first group, Ajnrshire — divided into the three districts of Cuningham, Kyle, and Carrick — seems to have been the main seat of the families of the race of Coel, from whom indeed the district of Coel, now Kyle, is said traditionally to have taken its name. There is every reason to believe that Boece, in filling up the reigns of his phantom kings with imaginary events, used local traditions where he could find them; and he teUs us " Kyi dcin proxima est vel Coil potius nominata, a Coilo Britannorum rege ibi in pugna caeso;^ and a circular mound at Coilsfield, in the parish of Tarbolton, on the highest point of which are two large stones, and in which sepulchral remains have been found, is pointed out by local tradition as his tomb. He likewise connects two of his early kings with this part of the country. These are Caractacus and CUMBRIA AND THE MEN OF THE NORTH. 171 Corbrcdus Galdus^ son of his brother CorbreduB. He identifies the first with the British king Coractacus, and the second with Galgacus, who fought against Agricola; but he says of them — " Horum quss de Carataoo^ Gor- bredo ac Oaldo Scotorum rcgibus, his voluminibus me- morise dedimus, nonulla ex nostris annalibus, at longe uberiora ex Comelio Tocito sunt deprompta.^ While adapting the events from Tacitus, he likewise made use of native traditions. His Caratacus is obviously the name Caradawg; and his Oaldus I believe to be taken from Owallawg ap Lleenawg. It is curious that these two warriors of the " Gwyr y Gogledd" should have tbo same relationship of uncle and nephew. Now he says that in Carrick, one of the three divisions of A3rrshire9 and lying to the south of Kyle, ** erat civitas tum maxima a qua Caractani regie vidctur nomen sortita. In ea Caratacus natus, nutritus, cducatus." Of Galdus or Gwallawg he says that, on his death, *' Elatum est corpus...in vicino campi ut vivens mandaverat, est conditum ubi omatissimum ci monumentum patrio more, immensis ex lapidibus est crcctum." Symson, in his Description of OdUo- twiy, written in 1684, says — " In the highway between Wigton and Portpatrick, about three miles westward of Wigton, is a plaine called the Moor of the Standing Stones of Torhouse, in which there is a monument of ^ three large whinstones, called King Oaldus's tomb surrounded, at about twelve feet distance, with nineteen considerable great stones, but none of them so great as the three first mentioned, erected in a dicum-