Skip to main content

Full text of "Archaeologia Cambrensis : a record of the antiquities of Wales and its Marches and the journal of the Cambrian Archaeological Association"

See other formats

This is a digital copy of a book that was preserved for generations on library shelves before it was carefully scanned by Google as part of a project 
to make the world's books discoverable online. 

It has survived long enough for the copyright to expire and the book to enter the public domain. A public domain book is one that was never subject 
to copyright or whose legal copyright term has expired. Whether a book is in the public domain may vary country to country. Public domain books 
are our gateways to the past, representing a wealth of history, culture and knowledge that's often difficult to discover. 

Marks, notations and other marginalia present in the original volume will appear in this file - a reminder of this book's long journey from the 
publisher to a library and finally to you. 

Usage guidelines 

Google is proud to partner with libraries to digitize public domain materials and make them widely accessible. Public domain books belong to the 
public and we are merely their custodians. Nevertheless, this work is expensive, so in order to keep providing this resource, we have taken steps to 
prevent abuse by commercial parties, including placing technical restrictions on automated querying. 

We also ask that you: 

+ Make non-commercial use of the files We designed Google Book Search for use by individuals, and we request that you use these files for 
personal, non-commercial purposes. 

+ Refrain from automated querying Do not send automated queries of any sort to Google's system: If you are conducting research on machine 
translation, optical character recognition or other areas where access to a large amount of text is helpful, please contact us. We encourage the 
use of public domain materials for these purposes and may be able to help. 

+ Maintain attribution The Google "watermark" you see on each file is essential for informing people about this project and helping them find 
additional materials through Google Book Search. Please do not remove it. 

+ Keep it legal Whatever your use, remember that you are responsible for ensuring that what you are doing is legal. Do not assume that just 
because we believe a book is in the public domain for users in the United States, that the work is also in the public domain for users in other 
countries. Whether a book is still in copyright varies from country to country, and we can't offer guidance on whether any specific use of 
any specific book is allowed. Please do not assume that a book's appearance in Google Book Search means it can be used in any manner 
anywhere in the world. Copyright infringement liability can be quite severe. 

About Google Book Search 

Google's mission is to organize the world's information and to make it universally accessible and useful. Google Book Search helps readers 
discover the world's books while helping authors and publishers reach new audiences. You can search through the full text of this book on the web 

at |http : //books . google . com/ 





l^xittcetiin %xbretmt}s. 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Jir4h»«I*fiia l^ambrrnsis. 



Cntnhrian lrr|Hlogirnl l00oriation. 






Digitized by VjOOQ IC 



Digitized by VjOOQ IC 


The Town of Holt, in County Denbigh • 

Tre'r Ceiri . . . • 

Report on the Excavations carried ont at 
TreV Ceiri in 1906 . 


A. N. Palmer 1 
W* Boyd Dawkins 35 

Llansaint .... 

Epigraphic Notes .... 

Cambrian Archseological Association : 
Sixtieth Annual Meeting 

Report on the Excavations at Coelbren . 

Roman Remains at Cwmbrwyn, Carmar- 
thenshire .... 

Geological Notes on Roman Remains at 
Cwmbrwyn, Carmarthenshire 

Cambrian Archsaological Association : 
Routes of the Excursions 

Notes on Eglwys Cymmyn, Parc-y-Ceryg 
Sanctaidd, and Llandawke 

Carmarthen in Early Norman Times 

The Capel Mair Stone 

The Town of Holt, in County Denbigh : 
its Castle, Church, Franchise, and 
Demesne .... 

St. Peter's Church, Carmarthen . 

Harold Hughes 88 

Rev. G. E. Evans 63 

John Rhys 66 


Col. W. LI. Morgan 129 

John Ward 175 

T. C. Cantrill 209 


G. G. T. Treherne 257 

J.E.Lloyd 281 

John Rhys 298 

A N. Palmer 311 

T. E. Brigstocke 335 

Digitized by VjOOQ IC 



Notes on the East Window of the Church 

of All Saints, Gresford . Rev. E. A. Fishboume 352 

The Early Settlers of Carmarthen Professor An wyl 361 

The Town of Holt, in Connty Denbigh : 
its Castle, Church, Franchise, and 
Demesne . . . A. N. Palmer 389 

Rbviiws ....... 254, 435 

Abohjeologioal Notes and Qdrribs . . 358,437 

Obituary : 

J. Romilly Allen, P.S.A. .... 441 

Digitized by VjOOQIC 


JANUARY, 1907. 




(Continued from p. 240.) 

The Lords and Charter of Holt. 

It has been said already that at the time when Domesday 
Booh was compiled, the site of Holt, with the "manors*' 
of Gresford, Allington, Hoseley, Sutton, and Eyton, 
were entered under the Cheshire hundred of "Exestan." 
They had then long been English, as the names of 
nearly all the townships within the area testify, al- 
though it does not follow that the underlying Welsh 
population had been displaced. But soon alter Domes- 
day ^ this district became annexed to the principality of 
Powys, being included in the new commote of Merf'ord 
(which, with that of Wrexham, was known in English 
as " Bromfield"), and the very lords of land became 
Welsh.^ We have to assume that the newly-formed 
commote (cymwd) or rhaglotry was Welsh, not in lan- 
guage only, but in customs, tenures, and feeling, with 

1 I have dealt wiih this question at length in my Huf-nr^ of the 
Townships of the Old Parish of Gresford, 

^a SBB., VOL. VUf I 

Digitized by VjOOQIC 


an increasing tendency, however, aftei^ a while, to 
imitate and adopt English customs and methods of 
government. Still, we must believe that the new 
oommote of the early twelfth century starii^d as a 
fully-fledged organized Welsh community. 

But, perhaps, before we go any[ further, it may be well 
to intimate tnat Bromfield, having become two Welsh 
commotes (those of Merford and Wrexham),^ did not 
remain continuously in the possession of the Princes of 
Powys. The Earls of Chester kept alive their claim to 
the district, and, according to the " Chronicle of St. 
Werburgh," Earl Hugh Cy veilioc " took the whole of 
Bromfield on Whit Monday, June 13th, 1177" : a state- 
ment which shows, at any rate, that he did not hold it 
before. Mr. J. E. Lloyd, M.A., of Bangor, also calls my 
attention to the fact that in the seventh year of 
Henry II (1101), under "the land of the Earl of 
Chester," are mentioned ** Hodesleu" (Hoseley) and the 
" Castellum de Wristlesha" (castle of Wrexham).^ The 

^ There is plenty of other evidenoo which might be offered in 
proof of the statement that " BromBeld," roughly speaking, desig- 
nated the commotes of Wrexham and Merford, but the following 
extract from the Thirty-Sixth Annual Report of the Deputy -Keeper of 
tlie Public Records must here suffice :— "On 18 Feb., 139 J, Richard II 
issued a writ to the Justice of BromBeld and Yale, for delivery to 
John Hope, of the office of Serjpant of the Peace, as well within the 
raglory [rhaglotry], courts, and bailiwicks of Wrexham and Mer- 
ford, within the lordship of Brom6eld, as in the raglory, courts, and 
bailiwick of Yale, which are called the office of Pensithith [Penceis- 
iaeth p]. Here the two commotes, or rhaglotries, of Wrexham and 
Merford, each with its courts, etc., are said to be in the lordship of 

'-* The entry occurs in Vol. IV of the Pipe Roll Society's publica- 
tions, where the account is given thus : — 

" Robert' de Monte Alto et Sim' fili' Will'i redd* comp' 

In lib'at* Castellanor* de Hodeslea . . xvi/t. xviiix. 

Et in lib'at' CastelVi de Wristlesha . . xvi/i. xviii.«<." 

And in the next year the entry is "... Castell* de Hodeslea" . . . 
and **.... Castellnnor' de Wris . . . ." So that there were at 
this time not only castles at Wrexham and Hoseley, but mention 
was made of castellans or castle-keepers at each place. Whether 

Digitized by VjOOQ IC 


explanation of this seems to be that the Welsh were 
not left undisturbed ; and Bromfield, after being inter- 
mittently under the Earls of Chester and Princes of 
Powys, was in all probability afterwards formally ceded 
to the last named, in return for the help which they 
often rendered to the English king against other 
Welsh princes. Certain it is, that Bromfield was a part 
of Powys Fadog throughout the greater part of the 
twelfth and thirteenth centuries, and that its popula- 
tion was Welsh. 

To this must be added that when Bromfield came 
under the lordship of the Warrennes, the commote of 
Yale was added to it. This is necessary to be said, so 
that the phrase "lord of Bromfield and Yale" may be 
intelligible. It also remains to be remarked that the 
Welsh organization of the three commotes (Merford, 
Wrexham, and Yale) constituting the new Anglo- 
Norman lordship, with the officials, customs, and dues 
of the same, continued long after those commotes were 
enclosed, so to say, in a feudal shell. Except the 
newly-founded town of Holt and its franchise, the 
whole of the chapelry of the same — Hewlington and 
" the five townships of Isycoed"* — was subject to this 
organization and to these customs. 

The proof of the statements made in the preceding 
paragraph is partly to be found in the recoras* of the 
proceedings of the court of the bailiwick of Wrexham, 
which for the years 1339 and 1340 have been preserved. 
The word " bailiwick" here does not denote merely the 

the castellans of the Earl of Chesf^r were in actual continnoas 
possession of the two castles named is another qaestion, and in the 
text the best explanation which suggests itself is given. The occu- 
pation by the Earl's officials appears to have been intermittent. 
Hoseley was the twin township of Merford. 

^ These five townships were Ridley, Sutton, Datton Difiaeth, 
Dntton y brain, and Caeca Dntton. 

* These records were copied in 1887, at my suggestion, and the 
cost defrayed out of Griffith's Fund, at the disposal of the Corpora- 
tion of Wrexham. The transcript has been placed in the Reference 
Department of the Wrexham Free Library. 

Digitized by VjOOQ IC 


town of Wrexham, but the group of townships, forming 
a commote, to which Wrexham gave its name. At 
this time, it would seem that, although Holt was 
already founded, the two great courts, or " tourns,*' of 
the year for the whole lordship were held at Wrexham. 

On October 24th, 1339, at Wrexham, the community 
ot Bromfield and Yale, the burgesses of Villa /leonum 
(Holt) and the men of Minora excepted, granted an aid 
of 200 marks to the Earl Warrenne ; and if they paid 
not, the raglots of Wrexham, Merford, and Yale were 
to make a levy on their goods. Here we see that the 
three commotes, each with its raglot, are distinguished. 
Then, at the great court held at Wrexham in May, 
1340, the township of Morton complains that Ken' ap 
Codblawd (Cynwrig ap Codblawd, or Codflawd) and 
Eign ap Ririt (Einion ap Rhirid) collected eleven 
hobetts of oats for the raglot's horse beyond the right 
measure ; and the township of '* Dynulle" (Dinhinlle) 
complains also that the same two persons, evidently 
servants of the raglot, come daily to the houses of the 
lord's bondsmen '*ad westand'' — quartering themselves, 
that is, as guests upon them, or demanding from them 
the due known as " gwestfa." In June, 1340, the 
raglot of the bailiwick of Wrexham ** presented" four 
pitchforks as " waifs" taken in his bailiwick : they were 
valued at 2s., of which 3d, went to the raglot, 4oJ. to 
the ringild, and the remaining 17d. to the lord. At 
Michaelmas, 1340, Eign ap Mad (Einion ap Madoc) 
complains that Adaf apEigno(Addaf ap Einion) took a 
cow from him for 12d, yearly, ** in aid of the Welsh 
forester," for which he was not liable, and the case was 
referred to the council of the lord. These are some of 
the Welsh customs from which the charter of Holt 
delivered the burgesses of the town. 

The two following entries in these records are also 
typical. At Wrexham, in November, 1339, Ken' ap 
Jor' ap Ken' (Cynwrig ap lorwerth ap Cynwrig) died ; 
that is, the fact of his death was presented, and Hova 
uad Mad' (Hwfa and Madoc), his sons, came into court, 

Digitized by VjOOQ IC 


and sought the heredity of their father, and it was 
granted to them, the right of anyone whatever [therein] 
being saved, and they pledged a heriot, which was 
7s. 6d, Then, in September, 1340, the death of Llewel' ap 
Edden Voil (Llewelyn ap Ednyfed Foel), freeman, was 
presented, and Griff, Mad', and Llew' (Grifl&th, Madog, 
and Llewelyn) sons of the said Llewelyn, as next 
heirs, came and sought the heredity of their father, and 
it was granted to them, the right of anyone whatever 
[therein] being saved, and they pledged a heriot, which 
was 7s. 6d.f and made fealty. Here we recognise gavel- 
kind in its Welsh form. 

I have also seen the accounts of Richard de Parys 
from Michaelmas, 1377, to Michaelmas, 1378; and 
herein he mentions the sum of £10, at which the issues 
of the custom of " amobr" in Bromfield and Yale were 
farmed yearly to John Wilde and Morgan le Yonge. 

Add to all this that the names of freemen, as well 
as of bondsmen, in Bromfield conformed in 1339 and 
1340 almost exclusively to the Welsh type of nomen- 

Thus, if about fifty years after Bromfield came into 
the possession of the Warrennes, it was so predomi- 
nantly Welsh in custom, tenure, and in the names of 
its inhabitants, the two commotes (Merford and Wrex- 
ham) which composed it must have been still more 
Welsh in the respects named when the Warrennes 
acquired them, and for some time before that date. 
And it was in this territory that Holt was founded as 
an English town, for English burgesses only, who were 
to be free from subjection to Welsh customs. 

The commote of Merford, as adjoining Cheshire, and 
containing within it the borough of Holt, would in all 
probability be the first to yield to English influences; 
yet it would seem to have yielded very slowly, and the 
evidence of deeds, a critical examination of the pedigrees 
of free families, and a careful scrutiny of the relics of 
tenure by kindred within it, clearly prove that family 
holdings, according to Welsh custom, must have lasted 

Digitized by VjOOQ IC 

6 THE TOWN OF floLt, 

there to a very late date. The charter* of Henry VII, 
in the twentieth year of his reign, to the tenants of 
Bromfield and Yale, assumes the continuance in that 
year of the system of Welsh family holdings, as well as 
of other Welsh customs and dues, throughout the 
lordship. But it was always open to say that, in fact, 
this system of tenure by kindreds and these customs 
and dues had then fallen into disuse, and that the 
charter did but confirm existing facts, recognise 
changes which had taken place, and make practices 
illegal for which former custom might be pleaded ; or 
at best was but a replica for Bromfield and Yale of 
charters granted about the same time to other more 
distinctively Welsh lordships, wherein such practices 
and customs did then actually exist ; and, in short, that 
the extreme eastern part of Bromfield, bordering upon 
Cheshire, was at the beginning of the sixteenth 
century practically Anglicised in the respect named. 
But Mr. Edward Owen has unearthed at the Record 
Office recently, a survey of Bromfield and Yale, taken in 
the twenty-third year of Henry VII, which shows, 
among other things, that a portion at least of Allington, 
a township adjoining Holt, was still parted into the 
*' gavells" \gafaelion) of the sons of Ithel, and a portion 
of Sutton Isycoed parted out into the '*gwelys" 
(gwelyau) of the sons of Elidur, Ithel ap Eunydd, and 
Elidur ap Rhys Sais being the Welshmen who, accord- 
ing to tradition, wrested the supremacy of this district 
from the Anglo-Normans, and settled in it. A "gafael " 
(holding) and a " gwely " (bed) may be taken for our 
present purpose as one, and as denoting the holding of 
a kindred, subject to Welsh law and custom. We find 
Button y brain, moreover, described in the same 
survey as " of the progeny of Edonowyn," and read of 
the " gavell " of " Madoc ap Gorgene [Gwrgeneu] de 

^ A translation of this grant, or charter, was printed in 1885 in 
A| pendix IV to mj History of Ancient Tenures of Land in the 
Ma7*ches of North Wales, and more recently in vol. xix of F Cymviro* 
doi, together with a copy of the Latin text. 

Digitized by VjOOQ IC 

tN COUNTY DENblGti. 7 

Hewlyngton," Hewlyngton beiug a part of the present 
township of Holt, and Sutton and Dutton within the 
old chapelry of the same town. All these places, 
moreover, are mentioned as being in the bailiwick of 
Merford. This is not the place to comment upon other 
statements of the Survey ^ but I may be permitted to 
indicate in this brief form the economic condition of the 
area adjoining the town of Holt, on its Welsh side, in 
the twenty-third year of Henry VII. 

Now, let us go back to the time of the last prince of 
Powys Fadog, whose castle was Dinas Bran, and to the 
circumstances under which Bromfield and Yale, from 
being parts of a petty Welsh princedom, became a 
lordship marcher, held of the English king. 

After the death of Madog ap Grruffydd ap Madog, of 
Dinas Bran, who had sworn fealty to the King of 
England, Edward I acted on the whole in the most 
just fashion, according to his notions of feudal right. 
On the 10th December, 1277, the King informed Roger 
L'EIstrange that he had appointed Gruffydd ap lorwerth 
to keep justice in the lands formerly of Madog of 
Bromfield, ** according to the law and custom of those 
parts," aiid pay the issues thereof to Margaret, who had 
been the wife of the said Madog, for the maintenance 
of Llewelyn and Gruffydd, his sons, according to the 
counsel of the Bishop of St. Asaph. The King had, 
however, already exacted homage from the lads, and 
appointed Boger L'Estrange guardian of the lands of 
the said Madog of Bromfield, so far as the preservation 
of peace and punishment of malefactors were concerned. 
And Edwards care of the boys and of their mother 
and grandmother extended until January, 12J§, and 
doubtless later, when suddenly Llewelyn ap Gruffydd, 
Prince of Gwynedd, not without provocation, broke the 
truce, and in conjunction with his brother David, the 
King's sworn vassal, stormed the castles of Aber- 
ystwyth and Hawarden, and attacked those of Rhudd- 
lan and Flint. 

Digitized by VjOOQ IC 

triE fOWN OF riOLf, 

Grufiydd ap Madog ap Gruffydd MaelQr.=T=-Emma Audley. 


Madog ap Qniflfydd, of=pMargaret. Llewelyn, Gruffydd Fychau, Owen, a 

of Chirk of Yale and clerk ; died 

and Nan- Qlyndwrdwy. young, 


Bromfield and Dinas 
Bran, alias Madog 
Fychan ;i died 1277. 

Llewelyn. Gruffydd. 

Both died young. 

* N.B. — This Madog ap Gruffydd was otherwise called in the English rolls not 
only " Madoc de Bromfield," but also " Madoc Fychan," or " Madoc Vachan." 
Now this last name was opposed utterly to the Welsh system of personal nomen- 
clature. And it appears as though the King treated " Vachan" as a sort of sur- 
name for the sons of Gruffydd ap Madog : for not merely was Gruffydd ap 
Gruffydd so named, which would be usual, but also his brother, Llewelyn ap 
Grufl^dd, of Chirk. 

The rebellion, in which Llewelyn and Gruffydd, 
brothers of Madog of Bromfield, were concerned, failed. 
The boys themselves died. It is alleged by late writers 
that they were drowned in 1282, under Holt Bridge 
(not, probably, then built), by the King's express 
orders. But this is one of those stories for which 
there is, so far as I can make out, no real evidence. 
Certainly, the lads died most opportunely from Edward's 
standpoint ; and the King, encouraged by the death 
of Llewelyn ap Gruflfydd, and determined on the 
settlement of Wales, granted on the 7th October, 
1282, to John de Warrenne, Earl of Surrey, the land 
of Bromfield and the Castle of Dinas Bran, which 
Gruffydd and Llewelyn, sons of Madog Fychan, held 
by themselves or their tutors at the beginning of the 
war, together with the land of Yale which Gruffydd 
Fychan ap Gruffydd, the King's enemy, had held, 
reserving the Castle and land of Hope. To Roger 
Mortimer, junior, were also granted Chirkland, etc., 
the lands of Llewelyn Fychan, another brother of 
Madog ap Gruffydd. 

Be it noted that in the grant no other castle than 
Dinas Bran is mentioned in the two " lordships," as we 
may now call them. There had been, as we have 

Digitized by VjOOQ IC 


seen, castles of some sort at Wrexham and Hoseley 
[Merford], but these must have been of little account, 
or dismantled, and we do not yet read of any castle 
of Holt 

John de Warrenne, Earl of Surrey, the first Anglo- 
Norman Lord of Bromfield and Yale, sub-granted to 
his son, William de Warrenne, for life, the territories 
just named, and seisin was delivered to him at Wrex- 
ham, on Thursday next after the Feast of St. Peter ad 
Vincula,* in the twelfth year of King Edward, 1284. 
But this William soon after died, and his father resumed 
possession. An inquisition after the death of the said 
William was held on Thursday next before the Feast of 
St. George,* in the fifteenth year of Edward, 1287, in 
which inquisition again no other castle than Dinas 
Bran within Bromfield and Yale is named. 

John, Earl of Surrey, to whom the lordships in 
question were first given, died 27th September, 1304, 
and was succeeded by his grandson of the same name, 
the third Anglo-Norman Lord of Bromfield and Yale, 
if we reckon his father William in the succession. 

Mr. Edward Owen has drawn my attention to an 
entry on the Patent KoU of 5 Edward II (part 1, 
membrane 6), 6th December, 1311, of an inspeximus 
and confirmation of a charter (in French), dated the 
Vigil of the Nativity of Our Lady [7th September, 
1308], by John de Warrenne, Earl of Surrey, granting 
in fee to John de Wysham 400 acres of land in the 
waste of the land of Bromfeud in the little Hem, 
and between the little Hem and Kaemaur, between 
Kaemaur and Pulle, between PuUe and Iwen [? Y 
Waun] Uchaf and Lidiate [? Llidiart] and the river 
Alom, and thence behind the Esk to the little Hem, to 
hold by service of a knight's fee, attending twice a 
year at the Castle of " Chastellion," finding in time of 
war a man-at-arms with a caparisoned horse to remain 
in the Castle of " Chastellion " for forty days at his 

^ Feast of St. Peter ad Viucula, 1st August. 
2 Feast of St. George, 23rd April. 

Digitized by VjOOQ IC 

to Tlifi TOWN Ot HOLt, 

expense, and rendering a rent of £10 sterling a year; 
and also granting to him the right of fishing at all 
seasons of the year in the river Alonn, the witnesses 
being William Paynel, Maddock ap Lewelyn, and Roger 
de Rysinge, parson of Hawarden, etc. " Chastellyon" is 
doubtless Castrum Leonum, Castle Lions, and is pro- 
bably intended to represent Holt Castle, but the place- 
names occurring in the charter are rather puzzling and 

The second John de Warrenne, Lord of Bromfield and 
Yale, having no children by his wife, Joan de Barre, 
granted on Thursday after the Feast of Saints Peter 
and Paul,* 9 Edward II, 1316, all his lands, including 
those in Wales, with the castles of Dinas Bran and 
Holt, to the King {Powys Fadog, vol. i, p. 365). The 
Castle of Holt was thus certainly built before 1316, 
and perhaps before 1308; and, as Mr. Edward Owen 
tells me, on 1st January, 1319, licence was given by 
the King to Thomas, Earl of Lancaster, to hold various 
castles, towns, and manors in Yorkshire, the Castles of 
" Dynas Bran and Caerleon," Bromfield and Yale, and 
other lands in Wales, for the life of John de Warrenne, 
Earl of Surrey [Duchy of Lancaster Records — Royal 
Charters]. This licence is very interesting, firstly, 
because it gives the first record of the connection of the 
House of Lancaster with Bromfield and Yale ; and, 
next, because it supplies a Welsh name [Caerlleon] for 
Holt Castle, the existence of which, in Chapter I of this 
history — written before the end of 1905 — I ventured to 

The said second John de Warrenne married, secondly, 
Isabel de Howland, and died 30th June 1347 ; and 
there is an account of the expenses of the officers of the 
Prince staying at Ciistrum Leonum from 9th July to 
6th August, 1347. 

Which one of the first three Lords of Bromfield and 
Yale, of the Warrenne family, built Castrum Leonum 
(Holt Castle) has not been yet ascertained, but the 
1 Feast of St. Peter and St. Paul, 29th June. 

Digitized by VjOOQ IC 

In county DENBIGH. ll 

claim made by the Joneses of Chilton (near Shrewsbury) 
that their ancestor possessed a castle at Holt upon the 
site of which the Warrennes erected their later building 
(see Arch. Camh., vol. 1875, p. 92) cannot be main- 
tained. The reasons for placing a castle on this site, 
after the grant of Bromfield and Yale to John de 
Warrenne, are obvious. Dinas Bran, besides being set 
on a high hill and most difficult of access, was out of 
the way, so to speak ; whilst Holt Castle, commanding 
as it did the chief passage from Cheshire to Bromfield, 
was easy of approach ^rom England, and what it lacked 
in strength of natural position could easily be made up 
artificially by the depth and breadth of its moat, and 
the strength of its walls and towers. The only signal 
disadvantage of Holt, as the new head of the two 
lordships, was its situation on the easternmost border 
of Bromfield, and remote therefore from the western 
parts of Yale. 

The second John de Warrenne of Bromfield dying 
without legal issue, his next heir in blood was Richard 
Fitzalan, Earl of Arundel, son of his sister Alice de 
Warrenne, by her husband Edmund Fitzalan, Earl of 
Arundel ; and on the 24th October, 1353, this Richard 
Fitzalan did homage to Edward III for Bromfield and 
Yale as immediately subject to the Crown, in the 
presence and with the consent of Edward, Prince of 
Wales. The precise manner in which Richard Fitz- 
alan came into possession of the lordship of Brom- 
field and Yale is difficult to follow. Certain it is 
that the later Fitzalans based their title to Bromfield 
and Yale on a fine levied in Easter, 1366, in the Court 
of the Lord king, between the aforesaid Richard, Earl 
of Arundel, and Alianor [Plantagenet], his wife, 
daughter of Henry, late Earl of Lancaster, complainants, 
and John, Duke of Lancaster [John of Gaunt] and 
others, deforciants, wherein the said Earl Richard re- 
cognised " Dynas Bran," Castrum Leonis, and the lands 
of Bromfield, Yale, and " Wrightesham" [Wrexham] to 
be the right of the said Duke and others, as of the gift 

Digitized by VjOOQ IC 


of the said Richard, the said Duke reconveying the 
properties to Richard and Alianor, with remainders to 
Richard de Arundel, junior, and Elizabeth his wife, 
and the heirs of their bodies, the aforesaid lands, held 
of the king, being worth 300 marks yearly. 

This Richard, Earl of Arundel, who died about 
1375, was succeeded by Richard Fitzalan, his son. Earl 
of Arundel, the fifth English lord of Brom field and 
Yale, who married for his first wife, Elizabeth, daughter 
of Humphry, Earl of Hereford and Essex. He had 
also Chirkland, and built in 1392 a bridge of stone, one 
of the predecessors of the present " Newbridge," " be- 
tween the domains of Bromfield and Chirk" [Thirty- 
Sixth Annxuil Report of the Deputy-Keeper of Public 
Records, Apx. ii), Robert Fagan, the builder of St. 
Asaph cathedral church, being the chief mason there. 
The second Earl Richard dabbled much in politics, and 
was one of the five lords appellant in the " Wonderful 
Parliament" of 1388. But his turn came at the end of 
1397, when Richard II for a time got the upper hand, 
and the Earl was condemned and executed the same 
day (21st September, 1397). Then, on 29th September, 
1397, the King granted the custody of the bridge and 
passage of Holt " between the Duchy of Chester and 
Holt Castle" {ibid), to Thomas Cholmondeley,^ the said 
Thomas to answer for all the value of the same exceed- 
ing five marks yearly. On the same day, he granted 
the office of ** porterwyk" of Holt Castle to Ralph atte 
Piatt for life ; on the day before, John MoUington and 
John Tranmoll (or Tranmore) were appointed foresters 
of Bromfield and Yale, to receive the same fees as John 
Dekka, late forester there, had ; an office which, the 
next year, was given to John Cholmondeley. On the 
ninth day of the Parliament of the same year, Castle 
Lyons, Bromfield and Yale, Chirk Castle and Chirkland, 

^ There must have been some delay in the handing over of this 
oflSce to Thomas Cholmondeley, for on the 15th September, 1398, 
David Holbach, Vice- Justice of Bromtield and Yale, was ordered to 
give livery of the same to the said Thomas. 

Digitized by VjOOQ IC 


Oswestry Castle and hundred, and the Eleven Towns to 
the said castle belonging, were annexed to the earldom 
of the principality of Chester : an enactment which must 
soon after have been either abrogated or neglected. And 
on 28th January, 139f , William le Scrope, Earl of Wilt- 
shire, was granted the office of Justice of Chester and 
North Wales, and of all the lordships late of Richard, 
Earl of Arundel, in those parts, for life, a grant which 
was augmented on 1st July, 1398. Richard II was 
himself at Holt Castle on 8th August of the year last 
named. But the Earl of Wiltshire did not enjoy his 
honours long. 

In the first year of Henry IV, Thomas, son of 
Richard Fitzalan, junior, Earl of Arundel, was restored 
to the estate which his father had formerly held, 
becoming Lord of Brom field, Yale, Chirk, Oswestry, 
Clun, etc. On the 20th February, 140f, he entered 
into an indenture by which he engaged to serve Henry, 
Prince of Wales, afterwards Henry V, for life, by sea 
and by land, in peace and in war, receiving for such 
service 260 marks yearly. 

It was this Thomas, Earl of Arundel and Surrey, 
who granted to the burgesses of Holt, in November, 
1411, a charter known to us by an " Inspeximus," of 
the fifth year of Queen Elizabeth. I made in 1892 a 
rough transcript (partly in pencil) of this charter, and 
intended, as soon as I could command the time, to com- 
pare the transcript with the Latin manuscript at Holt. 
But no opportunity of doing so presented itself until 
the beginning of May, 1905, when it appeared, on 
making enquiry, that the charter could nowhere be 
found, having somehow disappeared during the three 
or four years preceding, and all searches after it have 
proved futile. Further, the Record Office was able to 
yield no help, Jis Mr. Edward Owen, upon examination 
of the Patent and Close Rolls, found that the Holt 
charter had not been enrolled. There is a copy of the 
charter at the British Museum among the Harleian 
MSS. (vol. 2058, ff. 25 and 26) ; but this copy is not 

Digitized by VjOOQ IC 


merely a somewhat unsatisfactory one, but the end of 
it, about a third or fourth of the whole, is wanting. I 
therefore print in Appendix I to this chapter my own 
transcript of 1892, furnished with notes. It does not 
seem necessary to supply any translation, that made 
by Mr. W. H. Hewlett in 1848 (printed in vol. iv, 
pp. 927-9 of Report of the Welsh Land Commission) 
being adequate. But I scarcely need say how much it 
were to be wished that it had been possible to collate 
the tmnscript, either with the original, so unfortunately 
lo8t» or with a good copy of it. 

This charter deserves to be read and carefully con- 
sidered. We first notice that, in the preamble, earlier 
charters and ratifications are mentioned as having been 
granted to the burgesses of Lyons by the ancestors and 
progenitors of Earl Thomas. And we may picture to 
ourselves the prominent features of the town, franchise, 
and lands of Holt in 1411, then differing very little, 
doubtless, from the state of things at the time of 
its foundation about a century before. 

The burgesses, who were English, enjoyed their 
liberties in respect of their burgages. Each burgage 
stood across its own curtilage,* or courtyard, and had 
appurtenant to it certain acres of free land, also, in 
many cases, certain other acres of land formerly belong- 
ing to the lord 8 demesne. A rent of one shilling each 
was due to the lord for every curtilage, for every 
burgage built thereon, and for every free acre pertaining 
to it, and two shillings were payable for every acre 
that had been in demesne. The burgesses were also 
subject to " reliefs," or payments of double one year's 
rent at their deaths, by their heirs or assigns, and liable 
to furnish each one fit man for forty days in the year, 
for the defence of the castle in time of war until the 

^ Many of these cartilages still remain, especially on the east side 
of Chnrch Street, with the cottages within them representing the 
old burgages. Some of the free acres, lying in long narrow strips, 
are also to be seen west of Vicarage Lane and Green Street, and 
north of Wrexham Road. Each acre contained 2,115 statute acres. 

Digitized by VjOOQ IC 


town should be walled, and then to find a fit man for 
the defence of the town, for every burgage there. They 
had also to grind their corn at the lord's mill. The 
seneschal, or high steward, held, in the lord's name, two 
general courts (courts leet) at Michaelmas and Easter, 
to which the burgesses owed suit. There was a com- 
mon weekly market on Friday, and there were two 
fairs in the year. The burgesses were empowered to 
elect yearly one mayor, one coroner, two bailiffs, being 
English, and to hold the lord's courts every three 
weeks (the courts baron), determining all manner of 
transgressions, debts, felonies, covenants, pleas of land, 
etc., within the liberty of the town, according to the 
common law^ to keep their own prison, which might be, 
and I may add was, within the castle, and maintain the 
assize of victuals. They had also liberty to make 
English burgesses, to have common of pasture at Com- 
mon Wood* for their cattle, to possess a common pinfold 
or pound, and were to be subject in no way to Welsh 
customs, or to the authority of Welsh officials. They 
were free to dig coals and turves at Coedpoeth and 
Brymbo, and to carry them away for use in their 
dwelling-houses. Many of the burgesses had ovens of 
their own, and there was the common oven besides, 
which was still in existence in 1544, and even as late 
as 1620. No one could sell beer that was not brewed 
within the said town. 

In a writ of livery, dated 26th July, 1416, Thomas, 

^ This makes it clear that a Lord Marcher was mach more 
dependent than is generally supposed upon the central government. 
When he wished to grant a charter, he had to go Brst to the King, 
who conld impose any conditions which seemed to him desii*able 
and possible. The common law of England was administered at 
Holt, and the Lordship Marcher of Bromfield and Yale was always 
very mnch subject to the authority of the Crown. 

2 The Holt or Wood must have been partially cleared at the 
laying out of the town, to make room for the same, to furnish 
timber for building the burgages, and to provide good arable land ; 
and it would appear by this time that what had been at first reserved 
as " The Common Wood" had also been stripped, probably for fuel 
and repairs of bouses, and was become a pasture. 

Digitized by VjOOQ IC 


Earl of Arundel, is said to have died on Monday next 
before the Feast of All Saints last past, which means 
that he died at the end of October, 1415. He had a 
rather troublous time, many of his tenants having 
joined Owen Glyndwr, for which tenants he afterwards 
procured a pardon from the King. Moreover, it would 
seem that in his manor of Hewlington, just outside the 
franchise of Holt, now part of the township of the 
same, and certainly elsewhere within his lordship of 
Bromfield and Yale, the country was wasted by Owen's 
adherents, and houses were destroyed; so that the 
stewards had to grant the lands to such as would take 
them at a lower rent than was formerly paid for the 
same (see my Ancient Tenures of Land in the Marches 
of North Walesy p. 30). 

Altogether, we get the impression that Earl Thomas 
was a very fine sort of a man compared with the 
ordinary Lord Marcher of the time. He died without 
children surviving, and Henry V assigned to his widow, 
Beatrix of Portugal, as dower, certain possessions of 
the deceased lord. We learn what these lands were 
from the inquisition taken in Pentecost week, in the 
eighteenth year of Henry VI, after the death of 
Beatrix, on 23rd October, 1437. This inquisition has 
been printed on pp. 385-388, vol. i, of Powys Fadog, 
and I extract therefrom all that concerns Holt, Hew- 
lington, and what is now the parish of Isycoed. The 
said Countess Beatrix had, among other things, " a 
third of the gaol within the Castle Leonis, by the 
name of the Castle of Holt, with free ingress and 
egress, and safe custody of prisoners, and also the third 
part of a house called ' The Chekers,' ^ within the said 
Castle; also the third part of all houses outside the 
ward of the Castle. Also ... a certain stable for five 
horses next the court-house^ and near the ditch of the 
said Castle ; also the third part of a garden, together 

^ The Exchequer Tower. 

2 The Welsh court-hoase, or court-hoQse of the two lordships. 

Digitized by VjOOQ IC 


with a pasture called ' LeQuarrer,*^ adjoining the same; 
also the manor of Hewlington,^ the ringildry of Iscoed, 
and the park of Merselej." The jury declared the 
third part of the gaol to be of no value : that is, as 
bringing in no income, beyond repairs and custody of 
prisoners, and the third part of ** The Cheker '' and of 
the houses outside the ward of the Castle, also of no 
value. The stable was valued at 6s. 8d. yearly, and 
the third part of the garden and the pasture called 
'* The Quarrer," Ss. id. yearly, The site of the manor 
[house ?J of ** Heulyngton" was worth nothing. The 
rents of assize of the same manor were £6, and there 
were in it [assigned to the Countess] thirty-two acres 
of arable land at 2d. an acre, six acres of meadow at 6d. 
an acre, and forty acres of pasture at ^. an acre. The 
ringildry of Iscoed was worth £10 yearly, and the 
Park of Merseley 10^., beyond the custody and sus- 
tenance of the deer. 

As Thomas Earl of Arundel died without heirs 
male surviving, his estates were divided, subject to 
the aforesaid dower, among his three sisters, or among 
their children or grandchildren in right of them. These 
sisters were Elizabeth, wife of Thomas Mowbray, Duke 
of Norfolk ;' Joan, wife of William Beauchamp, Lord 
Abergavenny ; and Margaret, wife of Sir Roland 
Lenthall, knight, all of whom were still living on the 
20th July, 1416. The inheritors of the three portions 
after the death of the Countess Beatrix were (1) John 
Mowbray, son of Thomas Mowbray, Duke of Norfolk ; 
(2) Elizabeth, wife of Sir Edward Neville, and daughter 
of Richard, Earl of Worcester, who was the son of 
Joan, Lady Abergavenny ; and (3) Edmund, son of 
Sir Roland and Margaret Lenthall. I cannot explain 
how the Lenthalls dropped out of the inheritance, but 

^ The quarry forming part of the moat whence the stone was 
hewed to baild the Castle. 

^ HewlingtoD will be described in a later chapter. 

» This Thomas, Dnke of Norfolk, was appointed, in 1397 or 1398, 
Justice of Bromfield, Yale, Chirk, Oswestry, etc., but was soon after 
banished from the kingdom. 

6th sbr., vol. VII. 3 

Digitized by VjOOQ IC 


the lordship of Bromfield and Yale, in which we are 
here alone interested, soon became held in two equal 
and undivided parts or moieties : one belonging to the 
Mowbrays of Norfolk, and the other to the Nevilles, 
heirs of the Beauchamps.^ I expect the Mowbrays and 
Nevilles bought out the Lenthalls' share. 

On the fourth day of the Parliament of 17 Edward IV 
(1477), it was declared that Richard, the King's second 
son, was to be Duke of York and Norfolk, Earl Marshal, 
Warrenne, and Nottingham, and to marry Anne, 
daughter and heir to John, late Duke of Norfolk, the 
said Anne being then but six years old ; and if she 
should die without issue, the said Richard, Duke of 
Norfolk, should have, by consent of Elizabeth, Duchess 
of Norfolk (widow of the said John, Duke of Norfolk), 
** for the terme of his life, the halvendale (that is, the 
moiety) of the Castell, Towne, Lordship and Maners 
of Dynesbran [of the] Castell, Lordshipp, and Towne 
of Lyons [and of] the Lordship, Maners, and Londes of 
Heulyngton, Bromefield, Yale, Wraxham, and Almore, 
with their appurtenaunces, in the Marche of Wales," etc. 

This Richard, Duke of York, was one of the two 
young princes afterwards murdered in the Tower. His 
marriage was never consummated, and one of the 
above-named moieties, or **halvendales," of Bromfield 
and Yale became vested in the Crown. At a date 
which I cannot specify with precision, the other moiety 
— that of the Nevilles — became vested in the Crown 

Certain it is that on 10th December, 1484, the whole 
of Bromfield and Yale, " late of John, Duke of Norfolk, 
and Sir George Neville, knight," was granted by 
Richard III to Sir William Stanley (see the grant set 
out in Arch. Camb., 1882, pp. 150 and 151).^ Never- 

1 On the 14th October, 1467, John, Dake of Norfolk, and Sir 
Edward Neville, Lord of Abergavenny, held the two moieties. 

^ Many manors or townships are mentioned in this grant, bat 
all of them, except " Sonford and Osseleston,*' were in Bromfield 
and Tale, 

Digitized by VjOOQ IC 


theless, in the fourth year of Heniy VII (1488), Sir 
William Stanley only petitioned to continue to enjoy 
what was practiiidly the moiety of the lordship, although 
he seems to have been allowed to retain the whole. Of 
this brilliant and unfortunate knight, I shall speak again 

Before proceeding further it may be desirable to 
make an additional explanation. The lordship of 
Bi'omfield and Yale, as already has been said, was 
made up of the commote of Yale and of Bromfield 
(Maelor Gymraeg, that is, Welsh Maelor)^ Bromfield 
including two commotes — those of Wrexham and Mer- 
ford.* Part of Merford was lost temporarily to the 
lordship of Hope (Eston or Estyn), and part alienated 
permanently thereto. The other two commotes re- 
mained intact, except in respect of those lands held in 
them by the Abbot of Valle Crucis and the Bishop ot 
St. Asaph, and those other lands forming outlying 
parts of Maelor Saesneg {English Maelor) and Hope- 
dale. And each commote or rhaglotry held at first, 
after the grant to John de Warrenne, Earl of Surrey, 
its courts within its own bounds. The courts of Mer- 
ford* were held at or near Merford (at Yr Orsedd Goch, 
that is, Rossett, very probably), those of Wrexham at 
Wrexham, and those of Yale at some spot within 
Yale. This certainly appears to have been the case 
in 1339 and 1340. But by 1467 (see Record of 

^ In the commote of Wrexham, Valle Crucis had that part of 
Wrexham called Wrexham Abbot, and that part of Stansty called 
Stansty Abbatia. The extensive lauds in Tale belonging to the 
Abbej need not be here enumerated. John L'Estrange held in 
1386, as lord of Maelor Saesneg, besides Abenbary Feohan and a 
part of Erbistock (both of which have only been attached to Brom- 
field in onr own time) " the town of Button." The townships of 
Merford and Hoseley were annexed to Flintshire in the thirty-third 
year of Henry VIII. A part of the township of Bodidris belongs 
to Maelor Saesneg and the county of Flint; and in Yale is ths 
manor of Llandegla, which belonged to the bishopric of St. Asaph. 

^ It must be remembered that in both these cases I am speaking 
of eommote, not of toumship or manor courts, as is explained at the 
end of the paragraph. 


Digitized by VjOOQ IC 


Proceedings of the Lordship of Bromfield and Yale, 
printed in ArehoBologia Cambrensis, 1847 and 1848) 
all general enquiries were held, at any rate, at " the 
common place next Castrum Leonis," that is to say, 
at the Welsh court-house in the yard of Holt Castle ; 
and I believe that the ordinary courts, namely, those 
two courts of the year afterwards called ''courts 
leet" of the three commotes, were kept there also, 
thouqfh perhaps at first separately. The court held 
at Wrexham on the 8th October, in the sixth year 
of Edward IV, 1466, was the court of the township 
or manor of Wrexham Regis, for the appointment of 
two bailiffs and an escheator for the town, and for 
other business, and not a commote court. In the 
borough books of Holt in 1860, the courts leet for 
that year profess to be those for Bromfield, but they 
were really those for Holt only, separate courts leet 
being held the same year at Wrexham Regis for that 
manor, and a few years before at Marford,^ for Marford 
and Hoseley. The truth is, that no courts for the 
whole lordship have been kept for at least two and 
a-half or three centuries. They were already dis- 
continued in 1620. 

At the lordship court held next Holt Castle on 
19th October, 1467, to which court the inhabitants and 
tenants of the rhaglotries of Wrexham, Merford, and 
Yale were summoned separately, the jurors for the 
rhaglotry of Merford presented Richard Baz [Richard 
Bach, Little Richard\ for unlawfully, and without 
licence, carrying away certain stones near the lord's 
court-house at Castrum Leonura, to the value of 105. ; 
an entry of great value, because it shows, firstly, that 
Holt was reckoned to be within Merford rhaglotry, 
and, next, that the court-house for the whole lordship 
was now established next Holt Castle. Then, turning 

1 <* Marford " is the modern spelling and pronunciation of the 
older ** Merford." But the older spelling is preserved to an astonish- 
ingly late date in the township rate books and elsewhere after the 
pronunciation had changed. 

Digitized by VjOOQ IC 


to the general presentments, we find the old Welsh 
commote oflScials, or some of them, still holding their 
place— *Hhe Maist' Forest" [Pen-fforestwr], '' the Ser- 
jaunt " [of the Peace or '* Pencais "], Raglow " [Rhaglaw 
or Raglot], etc. ; and that ** fyre silv' " [Treth dan, a 
fee for taking firewood from the lord's woods], was 
still levied under the supervision of the master-forester. 
John ap David ap leuan, of Wrexham, also declared 
himself to be not under advowry : a statement from 
which we gather that there were then other persons 
who lived in that state — persons, that is to say, who 
having no inherited landed rights in the lordships, not 
even being nativi or servile tenants of land, were 
nevertheless under the protection of the lord, or of 
the larger free tenants. They were said to be in 
advowry (in advocariA), and were called in Welsh 
" arddelwyr,'' in English ** arthelmen," and in Latin 
" advocarii.*' It is probable that most of them were 
craftsmen or small tradesfolk. 

Spite of all the evidence of Welsh survivals thus 
provided, it is impossible to read the account of the 
doings of the lordship Michaelmas court without per- 
ceiving how rapidly English methods of procedure and 
administration were ousting Welsh methods at this 

The grant of Bromfield and Yale, on 10th December, 
1484, by Richard III to Sir William Stanley, one of 
the knights of his body, has already been referred to. 
Already, on the 12th November, in the year preceding, 
had the same king appointed him Chief Justice of 
North Wales. He was also Chamberlain of Chester, 
and Constable of North Wales. He was second son 
to the first Lord Stanley, and brother to Thomas 
Stanley, first Earl of Derby and Lord of Hopedale, 
which Thomas married, for his second wife, Margaret, 
widow of Edmund Tudor, mother of Henry VII. Sir 
William Stanley, of Holt, must be distinguished from 
Sir William Stanley, of Hooton, Cheshire, with whom 
he is sometimes confounded. He was descended, on his 

Digitized by VjOOQIC 


mother's side, from Richard Fitzalan, Earl of Arundel, 
Lord of Bromtield and Yale, etc., of whom I have 
already spoken, his grandmother having been Elizabeth, 
widow of Thomas Mowbray, Duke of Norfolk. He 
was also Lord of Ridley, in Cheshyjp. This splendid 
knight, as is well known, decided the issue of the 
battle of Bosworth, placing the crown upon the head 
of Henry, Earl of Richmond, and practically making 
him Henry VII of England. Many of his followers, 
or brothers-in-arms, were doubtless men from this 
neighbourhood. John ap Elis Eyton, whose tomb still 
stands in Ruabon church, was certainly at Bosworth. 
The Chevalier Lloyd and others assert that the new 
king granted Bromfield, Yale, and Chirkland, to Sir 
William for his achievement, or (must we say?) treachery 
at the famous battle above-named ; but the knight of 
Holt had, as we have seen, Bromfield and Yale, at any 
rate, before. He enriched Holt Castle, it is said, with 
the spoils of Bosworth field; but, however that may be, 
he was one of the richest subjects in the kingdom, and 
thus excited the envy and suspicion of the King, whose 
meanness saw in the splendour of Sir William a pretext 
for getting rid of one to whom he stood under such in- 
convenient obligations ; so he was charged with being 
in active sympathy with Perkin Warbeck, was con- 
victed, and executed on. Tower Hill, 16th February, 
149 1, all his possessions escheating to the King. He 
had a son, William Stanley, to whom, on 19th November, 
1489, the reversion of the Constableship of Holt Castle 
had been granted (and who married Joan, daughter of 
Sir JeflFrey Massie, of Tatton, Cheshire), and a daughter, 
Jane, who married Sir John Warburton, knight. The 
arms borne by Sir William Stanley, of Holt, were 
these : — 1, argent ^ on a bend azure, three bucks' heads 
caboshed or (Stanley) ; 2 or on a chief indented azure, 
three plates (Lathom) ; barry of six or and azure, a 
canton ei^ne (Goushill) ; and 4 gules, a lion rampant 
or (Fitzalan). I owe the description of this coat to 
H. E. J. Vaughan, Esq. 

Digitized by VjOOQ IC 

IN COtTNTV bftNBlGfi. 23 

Sir William Stanley being executed, and his estates 
escheated, Bromfield, Yale, and Chirkland reverted to 
the Crown. Henry VII himself stayed at Holt Castle, 
17th July, 1495. 

Soon after Sir William's execution, in the twenty-first 
year of Henry Vlf , we find the aftemamed officials of 
Holt Castle and its dependencies named : — Lancelot 
Lothar, constable (with a yearly fee of £10) ; John 
Roydon, interpreter (with fee of £2) ; Lancelot Lothar, 
custodian of warrants (with fee of £3 0^. lOd.); John 
Puleston, coroner (with fee of £2) ; John Aimer, 
attorney of the lord king (with fee of £6) ; David ap 
leuan ap Deicws, clerk of the court (with a fee of £3) ; 
David ap Ithel, custodian of the garden (with fee of 
£1 6^, 8d.) ; Thomas Tarleton, keeper of the castle 
park (with fee of £3 Os. lOd,) ; John Pickering, door- 
keeper of the castle (with fee of £3 0^. 10c?.) ; and 
Geoffrey Legh, parker of Mersley Park (with fee of 
£3 Os. lOd.) ; and the afternamed officers of the whole 
lordship : — Hugh Porter, serjeant of the country (with 
fee of £4) ; David ap Howel, approver (with fee of 
£3 Os. lOd,); and John Puleston, sen., cnief forester 
(with fee of £3). And it is to be noted as to the three 
last-named that, instead of one officer of each kind for 
each commote, only one of each kind is mentioned for 
the whole lordship, and no raglot or ringild is named. 
It may also be further remarked that, although the 
fees recorded as given were low, even considering the 
high purchasing power of money at that time, there 
were not merely perquisites attached to most of the 
offices, but chances for the holders of them for acquiring 
leases of demesne land on favourable terms, and ex- 
erting influence in other ways. 

The extent or survey of Bromfield and Yale (of the 
twenty-third year of Henry VII) already referred to 
(page 6), must next be briefly discussed. Villa Leonum 
(Holt) is again described as being in the "balUua," 
that is, in the bailiwick or commote of Merford. From 
the complete list supplied me by Mr. Edward Owen, of 

Digitized by VjOOQ IC 

24 tHE TOWN O*- flOLt, 

the tenants of Holt, given in Appendix II, and taken 
from the said survey of twenty-third Henry VII, it 
will be seen how many of the tenants* names were 
Welsh. This only bears out other evidence available, 
which shows that at this time, while Holt was fulfilling 
its purpose of Anglicising, in some respects^ the adjoining 
parts of Wales, it was itself being partly Cymricised by 
the inflow of Welsh people into it. 

On the 21st April, 1512, Henry VIII granted the 
receivership and stewardship of feromfield and Yale, 
etc., to Sir Charles Brandon, afterwards Duke of Suffolk, 
and his brother-in-law ; and from the Duke's accounts 
made up from Michaelmas, 1519, to Michaelmas, 1520, 
we get the names of the bailiffs at that time of Holt, 
and of those of the manors adjoining. 

Thomas Knyght, deputy of Thomas ap David ap 
Gruffith (who, being a Welshman, could not s^rvi^, 
was one of the two bailiffs of Holt (the King's bailiff;, 
and collectors of rents, farms, perquisites of courts, etc., 
and Jeffrey Baker was the other,* Thomas Prestland, 
being bailiff of Hewlington, William Main wey ring, 
bailiff of Ridley, Richard Roydon of Isycoed, and 
Edward ap David ap lolyn of Cobham Isycoed. 

On the 14th April, 1519, Henry VIII granted 
further powers to the same Duke of Suffolk, in whose 
accounts for that year we get the names of the officers 
attached to Holt Castle, and to the lordship of Brom- 
tield and Yale, which we may compare with the names 
given on page 23 as those of the corresponding officers 
in the twenty-first year of Henry VII : 

The Dake of Saffolk, as seneschal or steward . 

The same [apparently for his deputy, who was, 

as we know, Sir John Chilston, knight] . 

Lancelot Lothar, constable of Castmm Leonum 

^ Here, perhaps, may be given the names of the men — John de 
Aldeford and Richard de Wodehay — who were bailiffs of Holt from 
Michaelmas, 1377, to the Michaelmas following. Mr. Edward Owen 
also tells me that Ffilkin del ChambV and Thomas Alenis were 
bailiffs in 1388-1389, 

Digitized by VjOOQ IC 











£ ». d. 








1 6 8 

3 10 





6 8 





John and Thomas Wren, the King's auditors 
there .... 

John Ejton, interpreter, or " latym' ". 

Thomas Thelton, castodian of warrants 

John Pnleston, jun', coroner . 

John Aimer, King's attorney there 

Edward ap Rees, clerk of the coart there 

David ap Ithel, castodian of the King's garden 
there .... 

John Pnleston, jnn", parker of the Castle park 
there .... 

John Pekerjnge, doorkeeper of Castrum Leo 
nam • • . , 

William Aimer, parker of M'shlej [Mersley] 
Morgan ap lolyn, crier of the coart . 
Hugh Porter, serjeant of the country . 
David ap Howel, approver 
John Paleston, senior, chief forester 

Other interesting items appear in these accounts. 

The tenants of Mochnant and Cynlleth, parts of 
Chirkland, held, among other *' illicit opinions," that 
although they were bound to guard the seneschal when 
he went into their region to nold the two great courts 
of the year, they were in no way bound to guard any 
deputy seneschal. So, twice in the eleventh year of 
Henry VIII, Sir John Chilston rode from Holt to 
Chirkland to overawe the tenants there, condemning 
them in the sum of forty marks (£26 135. 4d), and 
charging the expenses of his 100 men-at-arms. But 
the tenants of Bromfield and Yale appear to have held 
a similar " illicit opinion /' or, at any rate, they were 
charged with withholding {de retinacone) somewhat, 
and were summoned to appear at Chester before the 
Commissioners of the King. Whereupon Sir John, for 
his own protection and for the King s dignity, when he 
went to Chester to attend divers courts there, placed 
about himself all the officers of the country and othei's, 
to the number of forty, and got the tenants of Bromfield 

Digitized by VjOOQ IC 


and Yale fined 100 marks (£66 135. id), for want of 
respect to his deputy seneschalship, sgid for their 
" illicit opinions/' 

We find also recorded in the accounts of the afore- 
said year the names of four felons hanged in that year 
— four felons at 205. a piece being charged: William 
ap John ap Howel Fychan, John ap Howelap Llewelyn, 
Sander Ley, and Maurice ap Evan. 

On the house of Castrurn Leon um £17 195. lid. were 
in the same year expended, and on the house of the 
common bakehouse 35. 2d. 


(See pp. 13-15.) 

Transcript of Charter made in 1411, by Thomas, Earl of Arundel, 
to the Burgesses of Holt, and confirmed 1st March, 156§, 
by Queen Elizabeth.^ 

Elizabeth Dei gratia Aiiglie FFrancie et Hibernie Rcgina, 
Fidei defensor, etc., omnibus ad quos p'sentes Pre p^uen'ient 
Salt'm Inspeximus quandam Cartam Thome nup' Comitis 
Arundell et Surr' d'ni de Bromfeld et Yale Burgensib's Ville 
sue leonu' in marchia Wallie eor^ hered* et Successorib's Anglic' 
fact et sigillo ip'ius nup Comitis vt dicit^ sigillat in hec v'ba, 
Omnibiba Xr'i tidelib's p'sentem Cartam inspecturis Thomas 
Comes Arundell et Surr d n's de Bromfeld et Yale Salt*m in 
d'no Sciatia qd cum Villa n'ra leouu' in marchia Wallie tam ex 
concessioue diu'sor' Antecessor' n'ror' et p'genitor' p* diu'sas 
Cartas et ratifaco'es Burgensib's Ville n*re p*dict' eor' hered* et 
Successorib's Anglicis fact' q'm ex possessione antiqua de diu'sis 
lib'tatib's et ffranchesiis Ville et Burgo mercatorio p'tinentib's 
priuilegiata extitit et adhuc existat, videPt qd Burgenses ville 
n're p'dict' eor heredes et Successores Anglici h'eant et teneant 
om'ia burgagia Curtilagia t'ras et ten' sua qui ex antique iure 
hereditario tenuerunt de nob' et Antecessorb's n'ris infra villam 
n'ram p'dict et lib'tatem eiusdem, h'end' et tenend' om'ia p'dict 
burgagia, Curtilagia t'ras et ten' p'dict cum om'ib's suis p'tinen* 
eisdem Burgensib's n'ris eor' hered' et Assign' Anglicis de nob' 

^ All niarka of contraction in the original charter arc replaced in 
this traubcript by simple apostrophes. 

Digitized by VjOOQ IC 

tN COtJin'Y DENMGH. 2^ 

hered' et Assign' n'ris lib'e quiete et in pace imp'p'm. Reddendo 
nob' hered' et A^ign' n'ris An'uatim videl't quiPt d c'or Burgen- 
siu* eor' hered' et Assign' Anglicor' p' quol't burgagio duodecim 
denarios et p* quol't Curtilagio duodecim donarios et p' qual't 
acra t're lib'o burgagio eor' p'tinen' duodecim denarios et p' 
qual't acra t're que solebat esse in D'nico n'ro duos Solidos 
Argenti An'uatim, faciendo eciam An'uatim sectam ad duas 
Cur' n'ras gen'ales Ville n're pdict' videl't ad p'x Cur' post 
festum Sc'i Mich'is Arch'i et ad Cur' p'x post festum Pasche et 
q d quil't heres h'mo'i Burgensiu' hered aut Assign' suor* post 
mortem Antecessoris sui dabit nob' et heredib's n'ris duplum 
redd'us sui unius Anni no'ie releuii^'sui Et q'd quil't ip'or 
Burgensiu' heredum et successor' suor' p' tempus guerre p' quol't 
burgagio suo infra villam n'l-am p'dict' et lib'tatem eiusdem 
scituat ad suos custos p'prios inueniet vnu' ho'i'em defensibilem 
ad custodiam et defensionem Castri n'ri leonu' p' quadraginta 
dies An'uatim quousq'e dict^ Villa n'ra sit muro incluso et eadem 
Villa existen' sic inclusa extunc quil't eor'dem Burgensiu' here- 
dum et Successor' suor* inueniet vnu' ho'i'em defensibilem ad 
custod' et defensionem Ville n're p'dict p' quol't burgagio suo 
ibidem. Et si contingat aliquem d'c'or' Burgensiu' heredum et 
Successor' suor' in h'mo'i custodia et defensione Castri neu Ville 
n're p'dict' sic inclus' p' seip'm in p'sona p'pria vel aliu' ho'i'em 
defensibilem no'ie suo in forma sup'dict' defic'e extunc bene 
liceat nob' et heredib's n'ris quodl't Burgagiu' cuiusl't burgensiu' 
p'dict' heredum ac Successor' suor' p' quo d'c'urn s'uiciu' custod 
si debita non fiat in n'ras manus seisire et retinere quousq'e de 
illo s'uicio sic deficiente cum Arreragiis eiusdem si que fu'int 
nob' et heredib's n'ris plene satisfiat' et p'soluat'. Et q'd p'dict 
Burgenses n'ri eor' heredes h'eant et possideant com'une merca- 
tum die ven's qual't septimana cum duab's nundinis consuetis 
infra villam n'rara p'dict' saluis nob' et heredib's n'ris tolnet' et 
al* consuetudinib's [de pred'cis] nundinis et mercatis ab antique 
nob' p[tine]n et consuet. Gonceasimvs edara eisdem Burgen- 
sib's n'ris eor' hered' et Assign' Anglicis p' nob' et heredib's 
n'ris q'd ip'i h eant & libere eligere valeant vnu' discretum virum 
ut maiorem vnu' Coronatorem duos s[ub balliujos Burgenses 
Anglicos [sing'lis] An[nis] infra villam n'ram p'dict et q'd ip'i 
maior et Balliui teneant Cur' n'ras infra villam n'ram p'dict de 
tr'b's Septimanis in tres Septimanas more consuet' et h'eant 
plenam potestatem ad audiend' & t'minand' omni'od transgres- 
siones debita conuenco'es felon' pl'ita t'rar' et ten'tor' ac singula 
alia pl'ita et contractus quecumq'e fact' tarn infra lib' ville n're 
p'dict q'm ext' s'c'd'm formam legis co'is et ad faciend* et 
exequend' omni'od attachiamenta et execuco'ey que ad Cur' n'ras 

Digitized by VjOOQ IC 


p'tinent p' sup'uisum Senescalli n'ri ibidem p* 'tempore existen/ 
Et q*d h'eant prisonam suam infra villam n*iam p'dict cum cus- 
todia om'i illor qui attachiat seu arrestati f u*iiit infra lib'tetem 
ville n*re p'dict ad sectam n'ram seu alicuiuscumq'e. Saluis 
semp' nob et heredib's n'ris omui'od eschapiis p'quisic'o'ibs Cur' 
Escaetis forisflf'c'uris, vagis, stragis, tinib's redempco'ibs am'cia- 
mentis et aliis exitib's et p'ficuis inde infra villam n*ram p'dict 
em gentib's. Et q'd Assise de Victualib's fact' p' ministros n'ros 
p' maiorem et BaUiuos ville n're diet teneant' et conservent' sup* 
forisf turam n'ram. Volumus tamen q'd Senescallus D'nii n'li 
p'dict teneat de Anno in Annu^ duas Cur' n'ras gen'ales ville n'ra 
pMict' more consuet. Ooncessimus eciam eisdem Burgensib's n'ris 
eor' hered' Assign' Auglicis q'd impp'm quieti sint de omni'od 
[the] loniis, lastagiis passagiis pontagiis stallagiis taillagiis et de 
©m'i'b's consuetudinib's p' totam t'ram et potestatem n'ram tarn 
in Anglia q'm in Wallia et March' Wallie. Et q'd maior, Balliui 
et Burgenses ville n're antedict' h'eant lib'am et plenam potesta- 
tem eligendi et faciend* Burgenses Anglicos eis acceptabiles 
vsuros et congaudendos omni'od* franchesiis et lib'tatib's ac aliis 
lib'ris consuetudinib's sicut p'dict Burgenses n'ri usi sunt et 
gauisi imp'p'm. Et qM nullus qui non sit in Buigensem accep- 
tatus p' maiorem Balliuos et Co'itatem Burgensiu* ville n're 
p'dict* aliqua lib'tate burgensiali cont' voluntatem maioris et 
Burgensiu' pMict' infra villam p'dict nee lib'tatem eiusdem 
quomodoPt gaudeat nee vtat. Et q'd licet eisdem Burgensibs' 
n'ris distringere in burgo n'ro p'dict' debi tores suos forinsecos et 
extraneos p' Victualib's eis venditis infra lib'tatem ville n're 
p'dict ad primam empco'em. Et q'd p'dict Burgenses n'ri eor' 
heredes et Successores ac tenentes quicumq'e inii*a lib'tatem 
ville n're p'dict vel ext' residentes imp'p'm quieti sint de 
omni'od' consuetudinib's Amobrogior'^ Advocarar'^ feod' Cou- 
stabularior' n'ror' Castror' ac de om'ibs feod' liagloti^ Kingildi* 
finiu' am'ciamentor' ac omni' alior consuetudinu' infra D'nia 
n'ra Wallie et March' Wallie qualitercumq'e em'gen'. Con- 
cessimus eciam q'd si p'fati Burgenses aut eor* aliqui seu eor' 
tenentes infra t'ram et potestatem n'ram testati decesserint vel 
intestati, nos nee heredes n'ri bona seu Catalla ip'or' confiscari 
non faciemus quin eor' heredes seu executores ip'a h'eant quate- 

^ Amobr, a fee due to the lord on the marriage or violation of a 
woman in his lordship (see p. 5). 

^ Advocarii, persons living in the condition of advowiy (see p. 211). 

^ The raglot {rhaglaw) was the chief admiuistrative officer of a 
Welsh commote (see pp. 4 and 21). 

^ The riiigild {rhingyll) was the raglot's bailili" or apparitor (seg 
p. 4). 

Digitized by VjOOQ IC 


nus diet* bona et Catalla ip'or' defunctor' fuisse constiterit dum 
tamen de d'c'is heredib's aut executorib's noticia aut f[ama] 
suflBcient Veat. Concessimus eciam p'dict' Burgensib's n'ris eor' 
heredib's et Successorib's tenentib's et seruientib's suis infra 
villam n'ram residen' q'd nuUiis ip*or' de cet'o impletet' nee 
occ'onet^ sup' aliquib's Appellis rectis Iniuriis t^nsgressionib's 
debitis criminiVs calumpniis accusamenciis et ind'tamentis 
seu aliquib's aliis contractib's aut r[ebus] eis impositis aut 
imponend' vbicumq'e locor* f c'is seu qualit'cumq'e em'gentib's 
nisi solomodo coram maiore et Balliuis Ville n're p'dict et p* 
iudic'm et t'minaco'em ip'or maioris et Burgensiu* Anglicor* tan- 
turn et non Wallicor' conuiucendis nisi res ille tanganc nos vel 
heredes u'ros. Concessimus eciam. eisdem Burgensib's eor' hered' 
et Successorib's q'd cum aliquis extraneus seu alius quicumq'e 
sup* lib'm ten'tum vel lib'am t'ram cuiuscumq'e Burgensis hered' 
aut Assign' suor' infra villam n'ram p'dict' vel lib'tatem eiusdem 
decesserit idem decedens si non sit Burgensis ville n're p'dict 
dabit Burgensi illi sup' cuius t'ram seu ten'tum decesserit melius 
animal suu' no'i'e herietti sui et deficiente h'mo'i Animal optimu' 
Catallu' suu' no'i'e principalia p'dict. Concessimzis eciam eisdem 
Burgensi b's n'ris eor' hered' Assign' Anglicis q'd p' t'nsgressionib's 
seu forisfcuris, s'uenciu' seu Tenenciu' suor* Catalla vel bona 
sua in manib's suis inuent' seu alicubi locor' p' ip'os s'uientes 
aut Tenentes deposita quatenus [ip'i] Burgenses h'mo'i bona sua 
esse suflScient* p'bare pot'int non amittent. Concessimus eciam 
eisdem Burgens' n'ris eor' hered' et Successori'bs q'd nuUi de 
cet'o liceat p'sentare aliquam penam sup' aliquem Burgensem 
ville n're p'dict p' quacumq'e causa sed inde p'seq'at v'sus eum 
p* acco'em in Cur'ville n're p'dict. Concessimus eciam Bur- 
gensib's nMs q'd nullus minister aut Ballius n'r quicumq'e nee 
hered' n'ror' ingrediat villam p'dict' nee lib^tatem 'eiusdem nee 
in aliquo . . . se intromittat sup' Burgenses seu ho'i'es quo- 
scumq'e p'd'car' ville et lib'tatis seu eor' aliquem de aliqua 
quereki occ'one t'nsgessionis seu alia re quacumqe infra d'c'am 
villam seu lib'tatem eiusdem em'gen nisi in def'cu' maioris 
Balliuor^ et Burgensiu' diet* ville. Concessimus eciam eisdem 
Burgensib's n'ris eor' hered' et successorib's q'd liceat singulis 
Balliuis n'ris ville n're pMict' distringere om'es et singulos 
debitores ex parte n'ra in eor' on'e existentes tam p* totum 
D'miniu' n'rum de Bromfeld et Yale q'm infra lib'tatem ville 
n're p'dict* p* quibuscuniq^e finibs am'ciamentis redempcoi'bs seu 
aliis exitib's aut p'ficuis in Cur Ville n're p'dict quovismodo 
em'gen' et h'mo'i districco'es fugare usq'e in co'e punfaldum 
ville n're antedict' et ibidem ip'as districco'es retinere quousqe 
de singulis denariis iu eor' o'ne debitis plene nob' satisfc'm fuit. 

Digitized by VjOOQ IC 


Ooncessimvi eciam eisdem Bufgensib's n'ris p*dict' eor* hered* et 
Assign' Anglicis ac Tenentib's suis q*d ip'i h'eant co'i'am pasture 
in quadem parcella t're voc' le Comon Woode p* om'ib's Averiis 
suis infra villam n'ram p'dict' leuantib's et cubantib's cum lib*o 
ingressu et Egressu eisdem Tenend* eisdem Burgensib's n'ris 
eor* hered et Assign' Anglicis ac tenentib's suis in sep'ali om'ibus 
temporib's anni sine contradiccVe n'ri hered' vel ministror' 
n'ror' quor'cuinq'e inip'p'm. Concessimus eciam eisdem Bur- 
gensib's n'ris eor' et Assign' Anglicis ac tenentib's suis q'd bene 
liceat eisdem et eor' cuiPt om'ia Au'ia forinseca sen extranea 
infra pasturam p'dict' pasturancia cap'e et fugare usq'e in co'e 
punfaldum ville n're p'dict' et ip'a retinere quousq'e debite 
emende fiant eisdem Burgensib's n'ris h'mo'i t'nsgressione et 
pasturaco'e fact' in eor' pastura sup'dict'. Volumua eciam q'd 
o'es Burgenses n'ri p'dict' eor' hered et Assign' infra villam 
n'ram residen' qui non h'eant p'pr'm fumu' q'd in co'i fumo 
n'ro eiusdem ville furnirS debeant Soluend' diet furni occupanti 
p' quol't buscello london vnu' obulum et sic singulis buscellis 
tantum. Concessimus eciam Burgensib's n'ris eor' hered' et 
Assign' Anglicis et 'eor' Tenentib's Ville n're p'dict' licentiam 
fodiendi capiendi et lib'e cariand' carbones marinas et turbas 
in vastis n'ris de Coitpoeth et Brinbawe et in om'ib's aliis vastis 
et locis ubi alii Tenentes n'ri Anglici aut Wallici carbones et 
turbas fodiunt p' eor' focale' in suis manc'oib's infra villam 
p'dict ad lib'am voluntatem sine contradicc'o'e n'ri hered' vel 
niinistror' quor'cumq'e. Concessimtis eciam p' nob et heredib's 
n'ris q'd si aliquis Burgensis hered' sen Assign' suor' Anglicor' 
attachiari [arrestjari sen iud'care contigit Nos nee hered* n'ri 
non capiend' aliquem finem [sen] redempco'em de ip'o nee 
manucaptorib's suis licet ip'm contigit . . . p'nos aut aut ministros 
n'ros sub manucapco'e libera. Concessimus eciam antedict' 
Burgensib's n'ris eor' hered' et Assign' q'd de cet'o nulli liceat 
tenenciu' n'ror D'nii n'ri antedict' Burgensib's n'ris infra duo 
miliaria p'x ville n're antedict' residen trah .... c'uisiam Salopie 
sen Cestrie sen aliqua alia victualia infra p'cinctu[m] n'r'm' 
p'dict' in p'iudic'm et nocimentu* ville n're antedict' nee aliquam 
aliam c'uisiam p't cuisia infra villam n'ram leonu' brasiatam 
uendere sub pena sex Solidor octo denarior' unde una medietas 
nob'et heredib's n'ris et alia medietas burgensib's n'ris ip'am penam 
for'factam p'sentantib's. Concessimus eciam eisdem Burgensib's 
n'ris eor' hered' et Successorib's q'd nuUus 'eor de cet'o cogi sen 
compelli debeat p' nullum ministrum seu s'uientem n'r'm nee 
heredum n'r'or aliquem equu' ip'or' Burgensiu' nee alicuius ip'or 
ad vsum liuius ministris n'ri vel s'uientis sui sine alt'ius cuius- 
cumq'e absq'e mera voluntate sua accomodare uec locare nisi 

Digitized by VjOOQ IC 


solomodo ad sp'ialem vsum n'ram et hered' n'ror' p' Balliuos 
n'ros eiusdem . ville taatum. Volumus edam et concessimus p' 
nob' et heredib's n'ris eisdem Burgensib's n'ris et eor* herediVs 
et Successorib's q'd Balliui ville n're p'dict h'eant Cam'am seu 
prisonam vnam infra Castrum n'r*m leonu' p' arrestatis sine 
attachiatis infra villam n'ram p'dict* seu infra lib'tatem eiusdem 
secure conseruand* ex delib'aco'e Constabularii Castri p'dict qui 
p* tempore fu*it si et quando necessitas hoc requirat cum 
r'onabili ingressu et egressu p' eisdem visitand' ministrand' ac 
delib'and*. P'ta taraen q'd ip*i Balliui om'ino h'eant om'ia on'a 
salue custod' h'mo'i imprisouat ibidem et q*d i'pi Ballii nob' 
respondeant de om'ibs eschapiis et malef'cis eor'dem si que 
p' eosdem vel eor' aliqnem infra seu de Castro n'ro p'dict in 
futur' fieri contigit absq'e quocumq'e on'e ip'o constabular' seu 
aliis ministris aut sui'entib's n'ris Castri p'dict' p' eisdem im- 
prisonatis ex parte nostra nullatenus imponend'. Et q'd o'es et 
singuli Burgenses ville . n're p'dijct' eor' heredes et successores 
molabunt om'i' blada et brasia sua ad niolendina n'ra infra 
D'n'm n'r'm ibidem ad vicessimam mensuram. Et nos vero 
p'fat Thomas Comes et hered n'ri om'es & singulas lib'tates et 
fFranchesias sup'dict' debitis Burgensib's n'ris p'dict' eor' hered' 
et successorib's plene vtend* possidend' et congaudend' Waranti- 
zabim's et imp'p'm defendem'. In Cuius rei testimonium huic 
presenti Carte n're Sigillum n'r'm fecim's appo'm Hiis testihus 
Rob'to morley milite senescall' hospicii n'ri Joh'e Bourley, Dauid 
holbach, Joh'e [W]ele tunc Senescall' n'ro de Bromfeld et Yale 
WilFs Eyman et multis aliis. Dat apud Castrum n'r'm leonu' 
die lune p'x ante festum s'c'i Andree Ap'li'^ Anno Eegni Regis 
henrici Quarti post conquestum t'ciodecimo, Et hoc om'ib's quor' 
inte'st innotescim' p' p'sentes. Teste me ip'a apud Westm' 
Primo die marcij Anno Eegni n'ri Quinto. 



(See p. 24.) 

Names of Tenants of Holt and their Eents, in the 23rd 
year of Henry yil. 

£ 8, d. 
Thomas Crewe . . . 3 4 10 

Heirs of Jankyn Hugenson . . . 16 

Jankjn dene, for land late Mathew Morgaiint . 1 6 

^ Feast of St. Andrew, ,30th November. 

Digitized by VjOOQ IC 



Richard Lakeii . . • • 

Tenents of the laud of Richard Alford . 

Robert Alford, with I2d. for land late of John 

Maunsell .... 
The same Robert, for half an acre of meadow 
Thomas Alford .... 
The same Thomas, for 23 acres late of Richard 

Alford, and 2d, for a parcel of waste for a 

garden made thereon . 
The community there [Yillaf ibm] . 
William Wodey . 
Margaret Roden . 
leuan ap leuan • 
Richard Do' 
Thomas Dour 
John Sendr' 

leu'n ap D'd ap lorwerth . 
David ap lollyn ap " hyllynne" [Heilin] 
Howell " Gouz" [Goch « the red]] 
Richard Grone 

Thomas Knyght and Elena Goz [the red] 
Richard ap Atha, with 2t. for Jonet Aleyne 
William ap Atha, jun', for one parcel of waste 
John Stockley 
Robert Davyessone, for land late of Agnes 

Elena Wayte (18«. 2d,) and Jankyn Pate, sen' 

Executors of Geoffrey ap Dicus 
Margaret relict of William ap Grono . 
Joan Pomfret 
Richard Phelypp, with 2«. 2d. for John MauncelFj 

Jankyn leche and his partners [et soc' sui] 
Thomas Bach 
The same Thomas, for one tenement and five 

acres late of Richard Baskervylle 
James Bath [1 Bach] and William Bath [) BachJ 

for Harayng's late land 
Richard Griffithson, and 5«. for land late 

Ralph fflecher 





































7 10 


















Digitized by VjOOQ IC 



RandolffBaoh .... 
Thomas Gierke, for one burgage late of Randolff 

aforesaid . . 

Jankjn Edjnson . 

Jankjti Pate senior, and Alice his mother 
Richard Pate 
Heirs of David Gronoson 
Thomas Roden 
The same Thomas, for two burgages late of 

Jankyn Wodall, for term of years 
William Roden. . 
Robert Glover 
Richard Bach 
The same Richard for two burgages, the land late 

of the aforesaid Jankyn Wodall, for term of 


William ap Dicus, for two burgages of the afore- 
said land, for term of years 
John James 
Thomas Glover . 
William Crewe, with 12«. lie?, for the land late 

of Margaret Compane . 
William Says, for the land late of Hoell Baron 
Thomas Wodall . 
William Wodall . 
William Hortone, with 2i. Sd. for the land of 

Mawde verch lorwerth 
Heirs of William Hansone . 
Heirs of John Bach 
William Brereton 
Llewelyn ap Howel " de franch" [of the franchise] 
John Huchon [? Hut^heon] . 
Heirs of John Crewe, for land late of Mawt 

[Margaret], relict of William ap Atha 
Jankyn Wylde . 
Tenants of land late of Agnes Stokley [Stockley] 

formerly paying 6«. 8c^., now only 
Thomas ap Davy ap Duyo [Deio] 
William le Wyld, for J an acre of land 
John Almor 

David ap Jankyn ap Madoc 
6th 8KR., VOL. vn. 

£ «. 





1 3 








2 17 


1 5 







2 1 

1 4 

16 10 

5 2 

13 4 


10 IH 

18 6 


18 7 


2 2J 

9 6 

6 2 



4 4 

3 4 


I 6 




Digitized by VjOOQ IC 



Master John Kyffyn [Vicar of Gresford] 

Heirs of John Gray, diaplain 

John Pulesdon 

William and Richard, sons of Richard Buklej 

Morgaunt [Morgan] Massy . 

John Hogge 

Total rental, £40 U$. 2|d. 

The total rental just given does not quite correspond with the 
sum of the items, but the diflference is so slight as to be im- 

£ «. 








Digitized by VjOOQ IC 





Pbofbsbor W. BOYD DAWKINS, D.So., Oxon., F.RS. 

Before the results of the exploration of TreV Ceiri in 
June last are laid before the Cambrian Association by 
the Exploration Committee, it is not inopportune for me 
to define, as far as may be, its relation to other forti- 
fied villages, and its place in the history of Wales. 

It is one of many similar villages, occupying a com- 
manding position for purposes of defence, in the neigh- 
bourhood, such as Pen-y-gaer, some two miles to the 
east of Llanaelhaiarn, ana Garn Bodfean, about a mile 
to the south of Nevin, containing the remains of rude 
stone huts, called by the inhabitants of the district 
** cyttiau Gwyddelod " — the huts of the Goidels. This 
popular attribution to the Goidels — the conquerors of 
the aboriginal Iberic Welsh, who in their turn had to 
submit to the mastery of the Brythons — is in my 
opinion true. They are probably the dwellings of the 
Welsh prehistoric Goidels, and have no necessary con- 
nection with the Irish Goidels, who were undoubtedly 
in close touch with this, as well as with other districts 
in Wales, in the historic period. 

Similar fortified villages abound elsewhere in Wales, 
as for example at Dinas Maen Mawr, near Pen Maen 
Mawr, all having the same characters, where the stone 
for wall- and hut-building was ready to hand, split into 
convenient blocks by the frost of untold centuries. 
Their entrances are narrow, and sometimes slanting ; 
and in one case, in the Pen-y-Gaer, some two miles 
south-west of the Roman fort of Caerhun, on the Conway, 
the approaches are rendered diflBcult by a chevaux- 
de-frise of blocks of stone, with one end planted in the 


Digitized by VjOOQ IC 

36 trb'r oeiri. 

ground, to prevent a rush. I am unable to detect any 
such arrangement in the masses of tumbled blocks on 
the slopes of Trer Ceiri, that rest at their natural 
angle of repose. On this point I cannot agree with the 
eminent archaeologists who have taken the view that 
this method of defence was used by the inhabitants of 
Tre'r Ceiri. 

The class of fort to which TreV Ceiri belongs is 
amply represented in Somerset by Worlebury, near 
Weston-super-Mare, and the line of similar forts on the 
Mendip Hills, and by many in Devon and Cornwall. It 
is also met with in Ireland, and especially in the Western 
Isles which shield the coast from the Atlantic storms, 
in the Arran Isles, ofiF Galway, and in Inis Murray, off 
Donegal. In these they are preserved in singular 
perfection. In Dun iEngus, North Arran, the slabs of 
limestone favour a more stable construction than the 
polygonal blocks forming the walls of Tre'r Ceiri, and 
there is clear evidence of a chevaux-de-Jrise. It is also 
worthy of note that in the same island — at Baile-na- 
Sean — are upwards of forty primitive houses, described 
by Mr. Kinahan as : — 

!a) Cloghauns, with beehive roofs. 
6) Cnochauns, with roofs covered with earth. 
c) Fosleach, with flag walls. 

{d) Ointigh, with roofs made of other materials than 

The Welsh *'cyttiau" belong to one or other of these 
groups, and are therefore appropriately assigned to the 

This class of fort is proved by the remains found in 
various places to have been occupied at various periods, 
mnging from the Bronze Age into the Prehistoric 
Iron Age, and well into the historic period. The 
bronze sickle found in Dun iEngus proves that it was 
used in the Bronze Age ; while bronze pins with orna- 
mentation of the Prehistoric Iron Age indicate that it 
was occupied at that time, and a bronze ring with 
cable decoration that it was not without inhabitants in 

Digitized by VjOOQ IC 

tre'r oeiri. 37 

the fifth century after Christ. In the cashel on Inis 
Murray we have very thick rough stone walls, with 
narrow entrances, surrounding a group of monastic 
remains, including three small chapels, strangely inter- 
mingled with the prehistoric cloghauns, of which the 
circular ** school-house" is an example, and also with 
soiiteri'ains, or covered ways. 

This class of fort in England is clearly proved by 
the result of the exploration of Worlebury, to belong 
to the Prehistoric Iron Age. Here the inhabitants 
belonged to the aboriginal Iberic stock, the ancestors 
of the Silures of the north side of the Bristol Channel. 
Equally good evidence is presented by the brooch 
found in the excavations of 1903 atTrer Ceiri, that it 
also belongs to the Prehistoric Iron Age. It may, 
however, have been — and probably was — used in later 
times by the Goidels of the district, whenever the 
country was being harried, for purposes of defence. 

Digitized by VjOOQ IC 




Since the year 1903, when thirty-two of the "cyttiau" 
in Tre'r Ceiri were examined^ the work of exploration 
lay in abeyance till 1906. An account of the 1903 
excavations, written by the Rev. S. Baring-Gould and 
Mr. Robert Bumard, is published in ArchcBologia Cam- 
hrensis for 1904. At tne meeting of the Committee 
of the Cambrian Archaeological Association, held at 
Shrewsbury on August 14th, 1905, it was resolved 
*'that Professor W. Boyd Dawkins be asked if he 
would kindly consent to the excavations at Tre'r Ceiri 
being carried out under his direction, with the assist- 
ance of Colonel W. LI. Morgan and Mr. Harold 
Hughes." Professor Boyd Dawkins very kindly con- 
sented to undertake the work. 

Through the assistance of Mr. D. R. Daniel, of Four- 
crosses, eight labourers were obtained, and work was 
commenced on June 5th, 1906, and continued till 
June 16th. 

Unfortunately, Professor Boyd Dawkins was called 
to London before the excavations were completed, and 
Colonel Morgan was unable to be present during the 
whole fortnight. On the other hand, most valuable 
assistance was given by Mr. Charles E. Breese, who 
devoted several days to the work. I was present 
during the whole time occupied by the excavations. 

The workmen employed were Griffith Jones and 
William Dobson, of Fourcrosses; William Owen, Griffith 
Griffith, Jahn Evans, David Owen, and H. Oliver, of 
Douglas Hill, Bethesda ; and John G. Jones, of Pant 
yr Avon, Bethesda. 

Digitized by VjOOQ IC j 

Harold Hughes w^ 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


Altogether, thirty-two " cyttiau" were examined (the 
same number as in 1903), and two sites on the Bwlch 
bdfow the south-west entrance. The two main en- 
trances were cleared sufficiently to ascertain their 
plans,' 643 far as the dilapidated state of the walling 
would allow. 

For general notes on the form, grouping, sizes, and 
construction of the " cyttiau" in TreV Ceiri, the nature 
of the subsoil, the stone employed, and the water- 
supply, the reader should refer to Mr. Baring-Gould's 
and Mr. Burnard's account in the 1904 volume of the 

The accompanying plan is practically confined to the 
space enclosed within the inner walls of defence. It is 
intended as a key-plan only to the sites excavated. 
Many details require correction. The excavations have 
laid bare walls and doorways, and enabled the outlines 
of many " cyttiau" to be followed accurately, when pre- 
viously it was only possible indefinitely to trace their 
, general conformation. The corrected measurements 
have not yet been taken, and the outworks have not 
been surveyed, with the exception of those immediately 
outside the south-west entrance. It has, however, 
been considered that much interest and value will be 
added to this Report by the provision of a plan indi- 
cating clearly the position of each hut examined. 

In 1903, over one hundred of the ''cyttiau" were 
numbered by Mr. Baring-Gould. Only those " C3rttiau" 
excavated in 1903 and in 1906 are marked with figures 
on the plan. The ** cyttiau" examined on the present 
occasion follow the original numbering of 1903. 

The details of the work carried out in 1906 are given 
below. It has been thought advisable to arrange the 
list in consecutive order of numbering, rather than 
according to the order of date on which the sites were 

Digitized by VjOOQ IC 

40 ttEPORT Oi^ tMeI excavations CARUIBi) OUT 

Sites Excavated in 1906, together with Details 
OF " Finds." 

The numbers in the following list correspond with 
those of huts on key-plan : — 

21 (a) Several "pot-boilers." 
(b) Four small pebbles. 

37 (a) Left lower jaw-bone of a horse (the teeth complete and 

fragments of the jaw-bone). 

(b) Tibia of horse. 

(c) Fragment of a leg-bone of another animal. 

(d) A stone " rubber." 

(e) A white pebble. 

(f ) Two " pot-boilers." 

(g) Charcoal. 

38 (a) Eight fragments of corroded iron, which, pieced together, 

are illustrated in Fig. 1. The total length of the 
remaining portions is 10| ins. Sections are given of 
the iron at five different points. The iron is socketted 
at one end and at the other, apparently, was leaf- 
shaped. The remains are probably those of a leaf- 
shaped socketted lance-head. 

(b) Anteria dorsal of a colt. 

(c) Eib of sheep or goat. 

(d) Charcoal. 

(e) Two small fragments of black pottery. 

41 (a) Portions of a bronze torque or armlet (gold-plated). The 

remains of this article are illustrated in Fig. 2, page 42. 
They consist of three portions: a piece of a curved 
bronze bar ; three solid bronze beads, with the remains 
of a bar, on which they ai'e threaded, firmly joined to- 
^^ether by corrosion ; and one bronze bead, of similar 
design, pierced through the centre. The rod or bar is 
decayed, but the diameter appears to have been about 
^ths of an inch. 

The internal diameter of the circle formed by the 
curved bar would have been 4^^ ins., but that within 
the bronze beads would only have been about S^ths 
ins. The above measurements are calculated from the 
curve of the existing segment, and are based on the 
supposition that the ring, when complete, formed a 
true circle, and that it was threaded for its entire cir- 

Digitized by VjOOQ IC 



cumference through the bead - shaped ornaments. It 
should, however, be noted that the beads may have 
been carried only round a portion of the circle, as, for 





at M 





o . .• 

f^ < Z 
^^ a 


i ♦» ® 



example, in the case of the bronze beaded torque, 
from Lochar Moss, Dumfriesshire, now in the British 
Museum, and the beaded torque, from Mowroad, near 
Rochdale, both illustrated in Mr. liuniilly Allen's 

Digitized by VjOOQ IC 



Celtic Art in Fagan and Christian Times, In the 
latter example, half the torque is beaded ; in the 
former, between half and three-quarters. There is a 
nick round one end of the existing portion of the 
Tre'r Ceiri bar. The three connected beads, although 
much decayed in places, retain the remnants of thick 
gold-plating. The beads differ in size. Each is shaped 
into eight bulbous divisions. Fig. 2 shows a side and 
an end view of this cluster. 

Dr. Kennedy J. P. Orton has examined the curved 
centre bar and beads, and confirms the impression, 


Fig. 2. — PoriioDB of Torque or Armlet. Hut No. 41. 
Scale, I linear. 

conveyed by their appearance, that the main metal is 
bronze, and that it has been covered with gold. With 
regard to the curved bar, a core of unchanged metal 
is still present. The single bead is much decayed, 
and therefore retains no sign of the gold-plating. 

It may be noted that amongst the ''finds" at Chastel- 
coz was a " bronze necklace bead." 
41 (b) Remains of an iron loop, in two fragments. External 
diameter of loop about 1 J ins. (see Fig. 3, page 41). 
(c) A small part of tlie base and a small fragment of a red 
earthenware vessel. The surface and material is of a 
sandy consistency. External and internal surfaces, 
dull red ; interior of material, groy. 

Digitized by VjOOQ IC 

AT TRE^R OaiRI IN l906. 


41 (d) Two small fragments of black pottery, giving the section 

of a rim of a vessel (see Fig. 5). 

(e) Charcoal. 

(f) •' Pot-boilers.*' 

(g) Burnt stones, 
(h) Small pebble. 

(i) A portion of a bone, which appears to have been burnt, 
(j) Teeth and jaw-bone of horse. 

42 (a) Two small fragments of iron. 

(b) Stone (? pounder). 

(c) White pebble. 

Fig. 5.— (a and b). Fragment of Rim of Black Pot. Hut No. 45a. 
(c). Fragment of Rim of Black Pot. Hut No. 41. 
Scale, I linear. 

45 (d) Stone " pounder." 

45a (a) A small ribbed bead of blue-glazed porcelannic paste 
(see Fig. 4, page 41), where it is illustrated in two 
positions. The surface may, perhaps, be more correctly 
described as "bulbous," rather than "ribbed." The 
surface formation has a slightly spiral appearance, the 
bulbous construction inclining in wavy curves from left 
to right. As the surface is much worn, the ornamental 
design is rendered somewliat indistinct. The external 
diameter is J in. The ribs of the blue beads discovered 
at Tre'r Ceiri in 1908^ and illustrated in Archosologia 

Digitized by VjOOQ IC 


Camhrensis for 1904, are vertical, thus diflfering from 
the example found in Hut 45a. 
45a (b) Several " pot-boilers/' 

(c) Large fragment of the rim of a black pot. The inner 
and outer facings have black coatings on red material, 
with an inner core of grey colour. The outer surface 
is smooth, the inner slightly coarse. A bold moulding 
runs round the lip. The external surface is divided 
into zones by narrow bands (see Fig. 5, page 43). 

46 This Hut drew blank. 

47 (a) Sixteen '* pot-boilers," 

(b) Half a " rubber.'* This stone appears to have served the 

double purpose of a " rubber " and " pounder." 

(c) A fragment of bone. 

Pig. 6.— Hone. Hut No. 62. 

Plan, Section, and Side Elevation of Stone Rubber. Hut No. 47. 
Scale, } linear. 

48 (a) Many small fragments of black pottery. 

(b) Many " pot-boilers." 

(c) Charcoal. 

(d) A small circular stone ball, just over ^ in. in diameter. 

(e) Rotten black matter (probably decayed earthenware) 

48A(a) Small fragments of iron. 

(b) Remains of bone. 

(c) Stone, probably used as a " pounder.'' 

49 (a) Fragment of bone. 

54 This Hut was drawn blank. 

Digitized by VjOOQ IC 



55 (a) Fragment of iron. 

56 (a) Charcoal. 

(b) Frao^ment of iron. 

(c) Pebbles. 

57 (a) An irregular mass of metal, about IJ ins. by 1 in. by J in. 

A specimen of this material was submitted to Dr. 
Kennedy J. P. Orton, Professor of Chemistry at the 
University College of North Wales, Bangor. He 
reports : — " It consisted mainly of lead, with a trace 
of iron, encrusted, of course, with chalk, etc. There 
appeared to be no tin, zinc, or copper.** 
(b) Two fn^ments of a bronze plate (see Fig. 7). The 
surfaces are much decayed. The upper edge, as drawn, 
of the larger plate, is slightly curved downwards. 

Fig. 7. — Fragments of Bronee Plates. Hut No. 57. 
Scale, { linear. 

while the left-hand edge is curved upwards and turned 
over. The smaller fragment, which is in an advanced 
stage of decay, appears to retain indications of orna- 
mentation. There are the remnants of two slightly 
raised bosses, which appear to be the remains of a 
concentric circle of bosses surrounding a small circle, 
the pierced half of the latter remaining, and visible 
on the lower side of the drawing. Whether it was 
originally a pierced circle or raised boss cannot be 
ascertained. Outside the bosses are three sunk dots. 
Jaw-bone of sheep, 
(d) Fragment of " pot-boiler.** 

58 (a) Fragment of bone of ox, and one other small bone, 
(b) Several fragments of pottery, much decayed, of a yellow- 
ish tint ; in its present state very fragile. Similar to 
the remains of the Mortarium found in Hut 60. 


Digitized by VjOOQ IC 


69(a) "Pot-boaer." 

(b) Charcoal 

(c) Eight humerus of ox (Bos'langifrans), 

60 (a) Many fragments of a Mortarium. Fig. 8 gives a restored 
section through the vessel, and the detail of the boldly- 
moulded lip to a laiger scale. Where cleanly broken, 
the material is of a yellow colour. The surface, how- 
ever, has a drabby appearance, with a tinge of dull 
red. Portions are stained black, probably due to the 
peaty deposit accumulated on the floor-level of the 
Hut, in which they were imbedded. The interior is 
sprinkled with fragments of quartz, which are more 
numerous at the bottom, and gradually decrease in 
number upwards, till they cease below the sinking 

Pig. 8.— Mortarium. Hut No. 60. Scale, i linear. 
Detail of Rim, | linear. 

carried round the vessel about 1 in. vertically below 
the rim. The quartz fragments are worn down as if 
from continued use. The diameter of the base 
appears to have been about 4J ins., and the full 
external diameter across the rim a little over 13 ins. 

(b) A « rib-bone." 

(c) "Pot-boiler." 

(d) Teeth of ox. 

61 (a) The base, and several fragments of a red earthenware 
vessel. A plan of the base, a side elevation, and a 
sketch of the lower portion and a detail of some other 
fragments, are given in Figs. 9 and 10. The diameter of 
the base is § in. The formation of the vessel is spiral. 
The clay is worked on a curve, which continually 
recedes as it rises upwards from the centre of the 
bottom of the vessel, about which it revolves. The 

Digitized by VjOOQ IC 

AT TRE'r CEIRl IN 1 906. 


external face appears to have been slightly glazed, and 
finished to a terra-ootta surface. 

61 (b) CharooaL 

(c) A small fragment of '' metal " (iron). 

(d) White quartz " pot-boiler.*' 

(e) Tooth of ox. 

62 (a) A hone or whetstone. 7| ins. long (see Fig. 6, page 44). 
(b) Thirteen small pebbles. 

62a(a) Pot-boiler. 

Fig. 9.— Base and Side Elevation of Earthenware VesseL Hut No. 61. 
Scale, 3 linear. 


Fig. 10. — Fragments of Red Earthenware Veaael. Hut No. 61. 

63 (a) A bronze pin, in the form of a sickle (see Fig. 11, page 48). 

Although much corroded, it bore indications of gold- 
plating. The *' sickle'* shape may be accidental, 
(b) Pebble. 

64 (a) Fragment of leg-bone of horse or ox. 

65 (a) Humerus of ox. 

(b) Three " pot-boilers." 

(c) Two small pebbles. 

Digitized by VjOOQ IC 


66 This Hut drew blank. 

67 (a) Fragments of black pottery, including a portion of the 

rim of a vessel of identically the same detail as that 
found in Hut 45a, illustrated in Fig. 5. 

(b) Tooth of ox. 

(c) A white pebble. 

68 This Hut was drawn blank. 

69 (a) Five pieces of black pottery, including a fragment of a 

rim of a vessel. . The section of the latter, although 
differing slightly from, resembles that of rim found in 
Hut 41, illustrated in Fig. 5, rather than that of the 
45 (A) Hut. 

Fig. 11.— Bronze Pin, GoHplated. Hut No. 68. 
Scale, I linear. 

(b) An iron article, of uncertain use, consisting of a disk. 

about 3^ ins. in diameter, on a stem. The total length 
of disk and stem is 1 ft. OJ in. The surfaces are much 
corroded. Fig. 12 illustrates one face and a side view. 
The ironwork of the disk bulges out on either face, 
but to what extent it originally did so it is impossible 
to say. On one face, for the greater area, and on the 
other side, in patches, the ironwork has split and 
corroded away, revealing, apparently, a flat disk or 
plate, forming the core of the superimposed metal. 
The original section of the stem appears to have been 
rectangular. The disk may have had flat faces. The 
superimposed metal, in that case, would entirely be 
the result of corrosion. 

(c) A fragment of a tooth of ox. 

Digitized by VjOOQ IC 

AT trb'b ceibi in 1906. 




o -^ 



6th seb,, vol. vit, 

Digitized by VjOOQ IC 


70 (a) Charcoal. 

71 This Hut drew blank. 

72 (a) Three " pot-boUers." 

(b) Bone of ox. {Bos-langifrons,) 

At the initiation of Mr. Breese, two sites were ex- 
cavated on the Bwlch below the south-west entrance. 
The first site was about 500 ft. in a south-westerly 
direction from the entrance, and the second site about 
360 ft. south-south-west of the first. Each site con- 
sisted of an approximately rectangular space, about 
15 ft. long by 5 ft. 6 ins. wide, roughly paved with rude 
stone blocks. Mr. Breese suggested that the rough 
stones, evidently artificially arranged, might indicate 
the sites of graves. 

The detailed results are as below : — 

Site 1 (a) At a distance of 7 ft. 6 ins. from the base, or south- 
south-west end, and 1 ft. from the east-south-east side, 
and at a depth of 2 ft. 6 ins. below the surface, the 
metal article illustrated in Figs. 13, 14, and 15 was 
discovered. The drawings give a sketch of the object, 
a plan looking downwards, and a side elevation. It 
was seated in the position shown, on a small, rough, local 
stone, measuring about 3 ins. by 3 ins. by 1 in. The article 
is circular on plan. From the broadest part it contracts 
to a narrower neck by means of a concave sweep, 
slightly bulged towards the narrower part, and ter- 
minates with a knop, with a circle of raised dots round 
the widest part. The top of the knop has been battered 
in. It is, therefore, doubtful as to the manner of its 
termination. The article is hollow. With reference 
to the metal, it has been submitted to Dr. Orton, who 
reports that it " consists mainly of lead, but contains 
also a certain amount of tin. It may be called a sort 
of pewter." The use the object served is uncertain, 
but the design is, to a certain degree, suggestive of that 
of the pommel of a sword-hilt. In the British Museum 
" Guide to the Antiquities of the Bronze Age," p. 77, 
two swords found in Northumberland, witli remains of 
leaded pommels, are referred to. In Arcfuxologia 
Cambrensis for 1905. p. 144, Fig. 25, is an illustration 
of a Late-Celtic bronze ornament, from Seven Sisters, 

Digitized by VjOOQ IC 

AT trb'r obiri in 1906. 


near Neath, resembling this object to the extent that 
the two may have served similar purposes. The fol- 
lowing is the description of the object by Mr. Romilly 
Allen, in the letterpress: — "A bronze finial, shaped 
something like the umbo of a shield. It has three 

Fig. 1 3. — Pewter Object. 
Site 1, on Bwlch. 

Fig. 14. -Plan of Pewter Object. 

Site 1. on Bwlch. 

Scale, ] linear. 



Fig. 16. — Side Elevation of Pewter Object. 
Site 1, on Bwlch. 
Scale, { linear. 

Fig. 16.— Iron Object. 

Site 1, on Bwlch. 

Scale, 4 linear. 

rivet-holes for fixing it on to something." It does not 
state whether it is solid or hollow. With reference to 
the material of the " pommel," in the " Report on the 
Exploration of Moel Trigam," in Archceologia Cambren- 
sis for 1900, amongst the " finds" in Hut-site 21, 

Digitized by VjOOQ IC 


mention is made of half a finger-ring, polished, and 
'' composed of some hard white mineral substance, 
(b) At a distance of 4 ft. from the base and below the east- 
south-east boundary, at a depth of 2 ft. 6 ins. below 
the surface, an iron article of horse-shoe shape, in size 
and outline exactly suitable for a modern heel-plate 
of a boot, and with the five usual oblong piercings for 
attachment to the leather, was discovered (see Fig. 16). 
The main difference between a modern heel-plate and 
this object is that the section is rounded off on one face, 
and the ends are likewise rounded, while the modern 
generally has angular edges and terminations. With 
regard to the date of this object, and how it arrived 
in the position mentioned above, I will not venture an 
opinion. This object was found after the excavation 
had been left open for the niglit 

Site 2. Although considerable time was devoted to the excava- 
tion of this site, no object was discovered. The whole 
space, however, had not been explored by June 16th. 
when work ceased, and the site had to be abandoned. 
Below the surface, large portions of both Sites 1 and 2 
were composed of small-sized stones, bearing the 
appearance of having been filled in by man. 

Some of the more important " finds" were unearthed 
after Professor Boyd Dawkins had left. Sketches, 
however, were forwarded to him, and he very kindly 
wrote, expressing his opinion, as below : — ** The bronze 
object (beaded) is distinctly of prehistoric Iron Age, 
and is probably a torque, or armlet, as you suggest. 
The glass or porcelain bead — some of these found in 
Glastonbury, in the prehistoric Iron lake village. 
The "pommel" is not very far removed from one dis- 
covered in prehistoric Iron Age fort at Hod." The 
black pottery and the iron ladle-shaped object, he 
writes, would belong to the same period. The iron 
** heel-plate," he believes, is modern. 

All the objects may be said to have been found on 
the true floors of the huts, though, in a few instances, 
the construction was so rude that it was difficult to 
determine with exactitude the ancient floor-levels. 

With reference to the fragments of bronze plate, 

Digitized by VjOOQ IC 

At TRB*R CElttl IN 1906. 53 

found in 57 b, Mr. C. W. Dymond, in his work on 
Worlebury,^ describes and illustrates some pieces found 
in that stronghold. A slightly curved plate, If ins. 
long, and ^ ins. wide, with piercings at one end, he 
considers to have probably formed part of an accoutre- 
ment. The concave side had been left dull, but the 
convex side, which is now patinated, had been burnished. 
Two other pieces of bronze plate, with curved edging, 
are pronounced to be remains of binding. 

With regard to the mass of lead found in the same 
hut, it may be noted that a lump of lead, about the 
size of a walnut, was found in one of the pits at 

Dr. Orton reports, with reference to the analyses of 
the metals referred to in the list of the various ** finds,'* 
that we are indebted to Miss M. 6. Eki wards, who 
worked under his supervision. 

In the following summary of the 1903 and 1906 
"finds," the kindred objects are grouped together. 
The figures in the second column refer to the numbers 
of the Sites. 

Summary of " Finds." 


1903 8 Blue-glazed porcellanic paste. 

10 Blue-glazed porcellanic paste. 

13 Blue glass. 

1906 45a Blue-glazed porcellanic paste. 


1903 8 Triskele. 

10 Fibula. 

1906 41 Torque or armlet. 

57 Fragments : bronze plate. 

63 Sickle-shaped pin. 


1903 3 Small fragments of pointed iron. 
6 Combined adze and hammer. 
6 Part of blade. 

^ Worleburyy an Ancient Strtmghold in the County of Sonursety by 
Charles William Dymoud, F.S.A. 1902, p. 122. ^ Ibid,, p. 81. 

Digitized by VjOOQ IC 


1903 7 Fragment, 3 ins. long. 

7 Crescent-shaped piece. 

10 King, 2 ins. diameter. 

14 (? Strike-a-light). 

19 Fragment. 

25 Fragment. 

26 NaiL 

82 (? Bill-hook). 
1906 38 (? Leaf-shaped spear-head). 

41 Half of loop. 

42 Fragments. 
48a Fragments. 

55 Fragments. 

56 Fragments. 
61 Fragments. 

69 Ladle-shaped article. 


1906 57 Mass of lead. 

In Bwlch, 
Site 1 (? Pommel of sword-hilt), pewter. 


1903 3 About a dozen pieces, dark pottery. 

12 Small fragment of rim. 

13 Small fragment red pottery, slightly orna- 

16 Two small pieces, black pottery. 
50 Part of bottom of earthenware vessel. 
86 I'iny fragments, red pottery. 
1906 38 Two fragments, black pottery. 

41 Part of bottom, and other fragments of red 

41 Fragment of rim, black pottery. 
45a Large fragment of rim of black pot. 
48 Fn^ments, black pottery. 
58 Fragments pottery, yellow. 

60 Remains of Mortarium. 

61 Red pot, base and fragments. 

67 Piece of rim and fragments, black pottery. 
69 Five fragments, black pottery, including rim. 


1903 4 Two spindle- whorls. 
5 One spindle-whorl. 
7 One spindle-whorl. 

Digitized by VjOOQ IC 

AT tre'r ceiri in 1906. 55 


1903 23 Fragment, bone-comb. 


1906 37 "Rubber.^' 

42 (?*Tounder.") 
45 •' Pounder." 

47 Half "rubber.'* 

48 Stone ball, ^ in. diameter. 
48a (?" Pounder.") 

62 Hone. 


1903 5 Pieces teeth of ox. 

6 Fragments, bone and teeth of ox. 

7 Fragments of bone. 
19 Fragments of bone. 
24 Fragments of bone. 
86 Tooth of ox. 

1906 37 Teeth and jaw-bone, horse. 

37 Tibia, horse. 

37 Fragment, leg-bone, other animal. 

38 Anterior dor^, colt. 
38 Bib, sheep or goat 
41 Calcined bone. 

41 Teeth and jaw-bone, horse. 
47 Fragment, bone. 
48a Fragment, bone. 

49 Fragment, bone. 

57 Jaw-bone, sheep. 

58 Fragment, bone of ox. 

59 Bone of ox {Bos-loiig-lfrons), 

60 Eib-bone. 

60 Teeth of ox. 

61 Tooth of ox. 

64 Fragment, leg-bone, horse or ox. 

65 Humerus, ox. 
67 Teeth, ox. 

69 Fragment, tooth of ox. 
79 Bones, ox. 


1903 3, 4, 17, 18, 25, and 87. 


1906 21, 37, 41, 42, 45a, 47, 48, 56, 57, 60, 61, 62, 

62a, 63, 65, 67, and 79. 

Digitized by VjOOQ IC 



1903 1,4,5, 11, 12, and 50. 

1906 37, 38, 41, 48, 56, 59, 61, and 70. 

The following Huts di'ew blank : — 
1903 2, 9, 15, 22, 75, 88, 90, and 92. 

1906 46, 54. 66, 68, 71, and site 2— on Bwlch. 

It will be seen from the above summary that by far 
the greater number of bones and teeth which have 
been found are those of the ox {Bos-longifrons). The 
other animal remains which can be determined are those 
of the horse, sheep, and (?) goat. 

It may be instructive to compare this list with bones 
discovered elsewhere. At Worlebury, in the pits, the 
" finds" include those of a pig, ox {Bos-longifrons), horse, 
deer, goat, water-fowl, and small birds.^ At Cadbury 
( Wincanton), those of " the Bos-longifrons, deer and 
swine are noted.^ At Maiden Castle, in the pits, many 
bones were found, especially those of the red deer.' At 
Walton-down, amongst other bones, horses' teeth are 
mentioned.* At Mount Caburn, " the animal remains 
were ox (Bos-longifrons), pig {Sus scrofa), horse {Equus 
caballus), goat {Capra hircns), sheep (Ovis aries), with 
occasional bones of roe-deer and badger ; also the 
scapula of a rabbit, the leg and spur of a fighting cock, 
and part of the bone of a dog."*^ 

The only fortified hill-village in the neighbourhood 
of TreV Ceiri where, to my knowledge, any exploration 
has been undertaken, is that on the top of Garn Bod- 
fean, near Nevin. Several of the "cyttiau" were ex- 
plored, in 1904, by the Hon. Frederick G. Wynn. He 
wrote at the time, informing me that he had *' found 
spindle- whorls and sling-stones, but no implements," 
and '^no certain hearth.'* During our stay at TreV 
Ceiri, we visited Garn Bodfean on a Saturday afternoon, 
when our men were "off work." We were shown a 

» Worlthury, p. 80. 2 /^-^.^ p gg^ 

•' Ibid., p. 83. •» Ibid., p. 84. ^ Ibid,, p. 85. 

Digitized by VjOOQ IC 

At trb'k ceiri in i906. 57 

stone hammer which had been found, and inspected the 
huts which had been excavated. The camp is as im- 
portant as — even if not more so than — Tre'r Ceiri. As 
a plantation covers the top of the hill, the general idea 
of the plan is difficult to appreciate without a lengthy 
inspection. The huts are more circular and detached 
than those at TreV Ceiri. This may possibly be 
accounted for by the space being less limited. " Finds" 
of further interest than those already discovered should 
repay careful excavation. Although it is probable that 
the camp belongs to the same period as Tre'r Ceiri, 
the *' finds," mentioned by Mr. Wynn in themselves 
are insufficient to establish the fact. 

The Entrances. 

The south-west entrance, through the inner en- 
compassing wall, was cleared sufficiently to ascertain 
its plan. The pathway rises rapidly through the thick- 
ness of the wall; the entrance is slightly curved. 
The north-western or left-hand wall approaching the 
camp retains, to a certain height, its original face ; it 
is concave on plan. The south-eastern or right-hand 
wall is not so easy to trace ; the inner portion has 
practically been destroyed. There are, however, in 
this position, some stones low down, which, in all pro- 
bability, indicate the outline of the wall ; though, on 
the other hand, they may form portions of a rough 
paved way. The outer part of this wall is convex, but 
it appears to have changed to concave, with an ogee 
sweep, further in. The uncertain line of wall is in- 
dicated by a broken line on the plan (Fig. 17 (a)). The 
entrance at certain points evidently was not more 
than about 3 ft. in width ; the narrowest, or inner end, 
apparently narrowed down to 2 ft. ; though, as men- 
tioned above, it is impossible to definitely ascertain the 
plan at this point. It is quite possible the entrance, 
in any case the inner portion, was roofed over with 
rude slab lintels, in a similar manner to the existing 

Digitized by VjOOQ IC 



so-called ** sally-port *' in the northern part of the 
enclosing wall. 

Fig. 17.— (a) South- West Entrance. 

(b) North- West Entrance. 

(c) North- West Entrance through Outer Wall. 

The passage through the north-western entrance, in 
the inner encompassing wall, has been lengthened by 
extending the masonry inwards for about 20 ft. beyond 

Digitized by VjOOQ IC 

AT tre'r ceiri in 1906. 59 

the inner face of the rampart wall. On the plan (Fig. 
17 (b)), the entrance through the inner wall is shown 
with reversed hatching to that through the extended 
walling. The north-western, or left-hand, wall is 
slightly concave. The south-eastern wall slopes in 
sharply towards the north-western ; it is irregularly 
concave, and has a slight bulge in the middle; the 
gateway narrows from a width of 12 ft. at the entrance 
to 2 ft. at the inner end. The outline of a short 
length of the face of the inner portion of the left-hand 
wall could not be traced, owing to the dilapidated state 
of the masonry. This small section is shown by means 
of a broken line on the plan ; the pathway rises rapidly 
through the thickness of the wall. 

The plan of the entrance, through the outer wall, 
of the road or pathway leading up to this entrance is 
given in Fig. 17 (c). It differs considerably from the 
entrances through the inner wall ; the least width is 
at the entrance to the passage, where it measures a 
little under 8 ft. ; the passage widens out internally 
to about 20 ft. There is a peculiar bulge in the left- 
hand wall, at the entrance to the passage, which con- 
siderably contracts the width. At a distance of 1 ft. 
11 ins. from the outer face of the right-hand wall is a 
hole, roughly measuring 1 ft. in width, 1 ft. 1 in. in 
height, and 1 ft. 4 ins. in depth ; its position is indi- 
cated on the plan. It was probably intended as a 
mortise-hole, to receive the end of a balk of timber, in 
connection with some sort of wooden barrier across the 

The general scheme of this outer entrance is more 
adapted to resist a pressure from within than from 
without. Might it possibly be intended for the in- 
gress of cattle and other animals which might not usually 
be admitted into the inner enclosure ? The barrier, in 
that case, might be intended rather as a protection to 
keep the cattle within bounds than an impediment to 

The description of the entrances by Mr. E. L. Barn- 

Digitized by VjOOQ IC 


well, in a former volume of the Journal,^ appears to be 
utterly inaccurate and unintelligible. 

An examination of the north-east wall was made at 
one point, to ascertain its true construction. The wall 
in this direction retains its parapet and " banquette," 
or ** chemin de ronde." The measurements of the wall, 
at the point examined, are : Internal height of " ban- 
quette" from present internal ground-level, 4 ft. 4 ins. ; 
width of *' banquette," 4 ft. ; height of parapet above 
** banquette," 3 ft. ; width of parapet, 4 ft. 4 ins. ; 
visible height of parapet, externally, 9 ft, 6 ins. The 
last measurement, however, was taken, from the screes 
formed of fallen debris^ from the top of the wall. At 
another point, not far distant, the external height ot 
the wall measured 11 ft. 

The wall is of single construction — that is, it was car- 
ried up to the ** banquette" in one operation. The outer 
portion was then raised above the level, to form the 
parapet. In this respect, it differs from the construc- 
tion at Worlebury, where a massive inner wall was first 
raised to the required height, and afterwards inde- 
pendent terraces or platforms, varying in number, each 
about 4 ft. lower than the one within, raised as con- 
tinuous buttresses against it. These terraces are chiefly 
external ; but, in some places, the hinder part of the 
rampart was raised by similar degrees. Mr. Dymond 
estimates that the width of the inner wall, or core, 
measured across the top, was seldom less than 6 ft. 
He writes that " it was evidently crowned by a parapet ; 
but whether this was of the same thickness as the wall 
on which it stood, or whether it was thinner, leaving 
room for a walk behind it, on the same part of the wall, 
there is not sufficient evidence to determine."^ The 
consideration of the general plan of the fortress, with 
its outworks and the '* cyttiau" within its walls, will 
be dealt with at a later date, when the results of the 
survey are published. 

Before concluding this Report, we must briefly con- 

1 Arch. Cavib.j 4th Ser., vol. ii, p. (36. 
•^ Worlebury, pp. 21, 22. 

Digitized by VjOOQ IC 

AT TRB*R OEIBI IN 1906. 61 

aider any deductions that may arise from the result of 
the exploration. 

The " finds" of 1906 confirm, in the main, the con- 
clusions arrived at by Mr. Baring Gould and Mr. 
Burnard in 1903.^ 

In the Introductory Note by Professor Boyd Daw- 
kins, he considers we have suflScient evidence to assign 
Tre'r Ceiri to the prehistoric Iron Age. 

Whereas Mr. Baring Gould and Mr. Burnard con- 
sider there are no indications of a later occupation, 
Professor Boyd Dawkins is of opinion that it may have 
been, and probably was, used in later times by the 
Goidels of the district, whenever the country was being 
harried for purposes of defence. 

The conclusions arrived at by Mr. Baring Gould and 
Mr. Burnard are : — 

" 1. That the fortifications were probably erected and 
occupied by that people to whom the 'finds' appertained." 
" 2. That TreV Ceiri was only temporarily — and that 
for a short time — occupied in the summer season alone." 
** 3. That the race which erected the walls and con- 
structed the huts was Celtic, probably British ; and 
that the period to which they belonged was the first 
or second century of the Christian Era." 

"4. That the builders had not been influenced by 
the Roman art of wall building ; and this points to the 
erection of the fortress at an early period of the first 

In support of No. 1, the finds were all discovered 
on the true floors of the huts, and no object that can 
be assigned to an earlier period than the prehistoric 
Iron Age has been found. 

Mr. Baring Gould and Mr. Burnard base their con- 
clusions, with regard to No. 2, on (a) the small amount 
of charcoal found ; (b) that some of the " cyttiau" seem 
never to have been occupied ; (c) the exposed position 
in winter. The arguments (a) and (b) are confirmed 
by the result of the more recent excavations. We must 

1 Arch. Camb., 6th Ser., vol. iv, p. 14. 

Digitized by VjOOQ IC 


allow, however, that the absence of '* finds" ifti certain 
huts is no criterion that they were never occupied. We 
should especially take into consideration that the damp 
deposit within some of the huts seems ill adapted for 
the preservation of certain articles ; and, further, that 
nothing of distinct value to the inhabitants would 
wantonlv be left for the benefit of future ages. It is 
doubtful, moreover, with reference to point (c), if the 
severity of winter weather alone would be a sufficient 
obstacle in the way of a hardened race. 

With regard to conclusion No. 3, the following 
support is given by the results of the 1906 work, (a) 
The pottery is wheel-turned, and in the whole distinc- 
tively Celtic agreeing in this respect with that found 
in 1903, when in addition some fragments of Roman 
pottery were discovered. (b) The remains of iron 
found — some in conjunction with bronze — is sufficient 
to establish the claim of the Iron Age. The half of the 
iron loop was found close to the remains of the bronze 
" torque, or " armlet." (Hut 41) (c). The gold-plated 
bronze " torque" is, as Professor Boyd Dawkins points 
out, distinctively of the prehistoric iron Age. (d) The 
porcelain bead will go with those found in 1903. 

The result of the 1903 and 1906 combined excava- 
tions, in the number and importance of the ** finds," 
may, I venture to think, be considered highly satisfac- 
tory. The Cambrian Archaeological Association have 
to thank Mr. R H. Wood, the owner, for kindly allow- 
ing them to undertake the work ; and, by so doing, to 
throw so much light on the early history of our pre- 
historic fortified hill-villages. 

Additional Notes. — Colonel Morgan considers the true parapet 
on the ramparts at TreV Geiri indicates Roman influence. The 
second step, referred to by former writers, does not exist, but is 
only the wall of a hut placed against the rampart. Much of the 
pottery fonnd in 1906 is similar to that found in conjunction with 
Roman finds elsewhere. 

The stone ball, found in Hut 48, has been examined by Mr. 
A. B. Badger, who pronounces it to be of carboniferous limestone, 
probably from the shores of the Menai Straits or Anglesey. 

Digitized by VjOOQ IC 



Bt thb Rfv. GEO. EYRE EVANS. 

iJUad at LlamaifU, August, 1906.) 

When, a few weeks ago, I walked over most of the 
ground we are covering in this our sixtieth annual 
gathering, I little expected to be so fortunate as to 
find an early Christian inscribed stone which is not 
mentioned by Professor Westwood in his monumental 
Lapidarium Wallice ; and to which no reference what- 
ever has as yet been made by any previous writer in the 
authoritative pages of the Archceological Cambrensis ; 
where, by the way, there is but one single reference 
(iv, viii, 141) to Llansaint, and that only in connection 
with the larger and well-known " Vennisetli" stone. 

Llansaint is a small village set on a hill, part of the 
ecclesiastical parish of Llan Ishmael, and its Chapel, 
in which we are seated, is held with the mother 
Church, and served by its minister. The Chapel tower 
is conspicuous, and seen from afar. Its summit is 
reached by iron ladders inside, to gain admission to 
which you enter the tower through the doorless opening 
on the south side, by means of another and moveable 
ladder, which you mount for some three yards ere you 
plant your feet on the stone steps of the opening or 
doorway. There is one bell. 

Within living memory this tower was used as a 
•* gaol,*' or " lock-up " — both words are yet current in 
local allusion to it— for unruly parishioners. Now they 
proceed, but rarely, however, in a certain amount of 
state to Caermarthen, where the villagers say, " Ma 
Dai wedi myn'd i'r Casteir' ("he is gone to the 
Castle"). The county gaol is part of the once magnifi- 
cent fortress of Caermarthen, on the mound of which 
and on the " Castle Green" we stood this morning. 

Villagers also use the correct words of '* Chapel " to 

Digitized by VjOOQ IC 


designate this building, and " Meeting-house " for the 
nonconforming sanctuary. 

In leaving the village presently for Kidwelly, we 
shall travel in part over a road covered with cockle- 
shells, which frequent traffic pounds to a white powder 
so soon as a fresh supply of the disused houses of the 
toothsome bivalves is cast upon it. The latter portion 
of our road will be down the steep and ancient track- 
way, known as " Y portway," a road formerly used by 
farmers' pack-horses carrying lime or coal in panniers. 
The women of this village, in the season, earn as much, 
often-times more, than their industrious brick-making 
husbands, by "cockling " — i.e., cockle-gathering, on the 
Penbre sands. 

Ou the south wall of the nave is the outline of a 
door, walled up in 1862. On the south wall of the 
choir is a little stone tracery-work, the remains of a 
window. In the north wall of the Chapel boundary is 
inserted part of the head of what was probably the 
village cross, which was placed where it is for safety, 
in 1860-62, when the wall was first built. Prior to 
that time the burial ground was unenclosed. The 
chancel window was placed by Vicar Owen Jones to 
commemorate the repairing of the Chapel in 1862. 

A place-name near by is " Pare y CastelJ," and a 
house close to this field is known as " North-gat." 
" Cheeselands" is another field s name. 

In the report of the Commissioners appointed by 
Edward VI, in 1552, to take and make "a just 
viewe . . . within every parishe," we read : — " Saint 
Ismaells. It'm, a chalyce in Hawlkyng Churche, a 
chapel annexed to the same parishe." 

In a terrier of 1636, preserved in the parish register, 
we get : — 

Alken Church. 

Some say these are allusions to this Chapel; I incline, 
however, to the belief that they may refer to the 
sea and sand-buried hamlet of Hawton, which was 

Digitized by VjOOQ IC 

£ I 

< i 



o 3 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


< c 

5 ^■ 
-< r- 

as -5 
2 -< 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


demolished by a tidal wave, probably about 1630-40. 
Speede marks it on his map of 1610. Just now the 
shifting sands have laid bare a long line of low stone 
walls, with three or four stone uprights of doors or 
windows. These I have inspected under the guidance 
of the parish vicar, Mr. James, who keeps laudable 
watch for any further sand movement and disclosure 
of buildings. 

The previously undescribed stone, which we will 
now proceed to examine, was until recently covered 
with ivy. It is inserted in the wall upside down. 
With a bow to Professors Anwyl and John Khys, both 
of whom I see present, I suggest that the reading 
of the inscription on the stone is : — 


It measures some 1 ft. 6 in. long by 9 ins. broad. 

The other stone, 4 ft. 6 in. long, is given by Westwood 
as reading : — 


The only merit I claim for this paper is that the 
reading of it has occupied but seven minutes of your 
time, leaving us then with three minutes to the good 
of the ten allotted to me by our Committee. I thank 
our President, Sir John Williams, for taking the chair 
on this occasion. 

6th 8fiB., VOL. VII. 

Digitized by VjOOQ IC 



By Professor JOHN RHYS. M.A.. LL.D. 

Llansaint, Carmarthenshire. 

At the beginning of July, 1906, 1 received a letter from 
the Rev. M. H. Jones, Picton Terrace, Carmarthen, 
saying that he and others, including, I believe, Mr. Eyre 
Evans, had been going over the ground in order to 
prepare the way for the visit of the Cambrian Archaeo- 
logical Association ; and that they had come across an 
inscribed stone in the wall of Llansaint Chapel, which 
is not given in Westwood's " Lapidarium Walliae," or 
mentioned in my ** Lectures on Welsh Philology." This 
struck me as very strange, as I visited Llansaint in 
the Seventies, in order to examine a stone reading 
VENNISETL- FiLivs ERCAGN-. 1 saw no other inscription 
there ; but now another, only about two yards away 
from it, is suddenly announced. The explanation is 
that the latter was concealed by a thick growth of 
ivy, which was removed lately. The surface covered 
by the ivy is still eatsy to distinguish. The stone 
still unpublished has been horizontally built upside 
down into the south wall, about 4 ft. from the 
ground. It measures about 2 ft. by 6 or 7 ins. in 
the widest part. 

The lettering, or rather what is left of the lettering, 
is in two lines, reading a& follows : — 



Mr. Jones, in describing the stone, says : '* The 
letters are well cut, and it is easy to take a rubbing 
of them." But the final i of avicati, which is per- 
pendicular, is partly gone at the bottom ; enough, 
however, of that vowel remains intact to make its 

Digitized by VjOOQ IC 



identity certain. Here, on the right, began a splinter, 
which is gone with other letters, the tops of whicli 

■ m 




a o 





only are left. When my attention was drawn to them 
I tried to make them portions of the word fili, but 
that would not fit, so I conclude that what followed 
AViCATi was an epithet or surname which went 

Digitized by VjOOQ IC 


with that name, or else, more probably, iacit or 
Hic IACIT. One cannot stop there, as one wants fili 
somewhere, and I conclude that it followed cimesetl- 
in the first line. In other words, the end of the 
stone has been broken off since the inscription was 
made. What the length of the broken fragment may 
have been one cannot say; but as it consisted of 
an inscribed portion, together with a portion to be 
sunk into the ground, it is not improbable that 
altogether it was at least as long as the piece which 
remains. It is probably somewhere in the building, 
unless it is the third inscribed stone which is believed 
to be somewhere near the walled-up door in the same 
wall. The inscription should mean (" The place or 
monument) of Cimesetlas [son] of Avicatus : [here 
(he) liesj." I should perhaps explain that Avicattis 
was probably of the u-declension in Celtic, like the 
Latin fourth declension, as in the case of rtvagistratus, 
genitive magistrates; but nouns which made i in the 
genitive, like Cimesetli, had their nominative in os in 
Gaulish and early Brythonic, as in Latin also before 
it came to be changed into us^ as in dominv^, domini. 
So the Brythonic nominative corresponding to Cime- 
setli would be Cimesetlos, but I have treated it here 
as Goidelic Cimesetlas, because Goidelic shows -as in- 
stead of 'OS : compare Latin with -os, until it made 
way for -ibs, which helped the confusion with the fourth 

The first thing to call one's attention in the names 
is the common element setl-i that is setl-i in Cimesetli 
and Vennisetli. It is rather an unusual one, and its re- 
currence here makes it probable that the two bearers 
of these names belonged to the same family : let us say 
that they were, perhaps, cousins. Compare the series of 
men whose names began with (Bthel in the Saxon 
Chronicles, and in Welsh pedigrees with eddy such as 
Cadwallon, Cadwaladr, and the like. 

The word setl is in Welsh hoedlj *' life," represented 
by sel in the Irish compound gar-sele, in Welsh 6yr- 

Digitized by VjOOQ IC 


hoedledd,^^ shortness of life," so that Venni-setl- for 
Vendi'Setl' meant " him of white life," for the first 
element venni is one of a number of forms, such as 
vinne and vendu and vende, all meaning " white," in 
Gaulish uindoSj as in Penno-nindos, " white-headed." 
But in Celtic the word for " white," Irish Jinn, Welsh 
gwyn, feminine gwen, had secondary meanings, such as 
" blessed, or happy ; " and it is applied in Welsh to 
Heaven and the Almighty; also in terms like tad 
gwyn, " step-father" : compare the French use of heau, 
belle, in beau-pdre, belle-scBur. See also my paper on 
the " Celtic Inscriptions of Prance and Italy " (read 
to the British Academy on May 23rd, 1906, p. 12.) So 
much of VennisetldSj "him of the blessed, or happy life"; 
but what did Cimesetlas mean? I know of no Welsh 
word to throw any light on cime, but there was an Irish 
word, dmbj " silver," with which cime may be regularly 
connected on the supposition, that in it m stands for 
mm, representing earlier m6, just as nn in Vennisetli 
stands for earlier nd. The reason why we have cime, 
and not cimme, is probably the length of the combina- 
tion MM in writing, but instances of the simplifying 
of double consonants in our inscriptions are not rare. 
Take for example Cunoceni by the side of Cunocenni, 
and Vendubari by the side of BaiTivendi, also Cxino- 
gvM for Cunogussi, and others of the same kind. But 
what would " silver life" mean ? An answer was 
supplied by a member of the Association, when I was 
discussing this inscription at the evening meeting at 
Carmarthen, and it was to the effect that the name 
was synonymous with Vennisetl-i, meaning '* white 
life," that signification being suggested by the colour 
of silver. This is, however, hardly convincing, and I 
turn to Cormac's Irish Glossary, the earliest authority 
on cimb. There it is given as originally meaning 
" silver," but it is added that it was the word for the 
silver paid frequently and in large quantities as 
tribute to the Fomorian invaders. From that circum- 
stance it came, we are told, to be the name for every 

Digitized by VjOOQ IC 


kind of tribute. See Stokes's edition of O'Donovan's 
translation of Cormae's Glossary, pp. 32, 39, where the 
editor adds that cimh is perhaps " ransom-money," 
rather than "tribute," and cites the glosses cimhid 
" vinctus' and cimbidi " -custodias." Cimhid occurs 
in Cormac's Glossary , spelt also cimhith, '* a cap- 
tive ; so cimhid or cimhith may have meant a 
prisoner who was held to ransom. Stokes goes so far 
as to suggest that the Celtic word dmhri, explained by 
the ancients as latrones, is derived from cimh. Thus, I 
should be inclined to explain CimhH as meaning more 
precisely invaders who exacted tribute from their 
victims, raiders who made captives in order to obtain 
money as ransom, or thieves who were used to levy 
blackmail. In the case of the name Cimesetlas, 
whether cimh^ originally meant silver or not, I take it 
here to mean " ransom ; " that is, I should translate it 
" him of the ransomed life," and treat it as Christian 
referring to the Redemption. In fact, one may perhaps 
go so far as to suggest that the bearers of the related 
names Cimesetlas and Vennisetlas were the saints to 
which the name of the Church refers. The Church is 
called in Welsh Llan Saint, that is, " Ecclesia Sanc- 
torum ; " and the answer, or part of the answer, to the 
question who the saints were, naay be regarded as 
supplied by the two inscriptions. That seems to me 
more probable than the idea that it is a dedication to 
"All Saints." Vennisetli occurs in North Wales as 
Veiidesetli, which has in Welsh yielded the Saint's 
name Gwynoedyl, This last is reduced into Gvrynodl, 
and still further, in the name of the Church of 
Llan-gwnodl or Llan-gwnadl, in Lleyn : see the 
Myvyrian, vol. ii, p. 44. Whether the Llansaint 
Venyiesetly the Llannor Vendesetly and the Gwynodl 
of Llan-gwnadl are to be reckoned as three persons, 

1 What is one to make of Ponfc y Cim, **The Bridge of the 
Ctm," whioh is somewhere near Gljn Llifon, in Carnarvonshire? 
Cim may be for cym, a possible plaral of cww, a dingle or glen ; 
but I do not know whether the locality lends any countenance 
to snoh an interpretation. 

Digitized by VjOOQ IC 


two, or one, I cannot say : the saints of Wales moved 
a good deal frotn one place to another. In any case, 
the bearers of the names Vennisetlas and Ciraesetlas 
may be regarded as Goidels rather than Brythons. 
The Gwnnvvs inscription ^Cardiganshire) yields a 
purely Welsh name to ada to these two, namely, 
Hiroidil JUius CarotinUj where the d proves the t of 
setli to be standing here for ih = d. 

So far of the first name : as to the other there is very 
little to say. Holder, in his Altceltischer Sprachschatz, 
has a compound which we can compare with it, namely, 
AvicantuSj the name of a source god mentioned in an 
inscription at Nlmes : see the Berlin " Corpus Inscr. 
Latinarum," XII, 3077. The genitive Avicati is 
doubtless Latinised, and the Celtic would have probably 
been Avicatos of the i^-declension, the second element 
being catus^ " war or battle ; " Irish, caih ; Welsh, 
cat, cad, of the same meaning. The first element, avi, 
is supposed by Stokes to be of the same origin as 
Latin avere, avidus, and one may perhaps render it 
"fond;*' but whether in the subjective or objective 
sense it is not very easy to decide. Thus did Avican- 
tus mean a god who was fond of song, or a god of 
whom minstrels and musicians were fond. In the 
former case Avicatus would mean one who was fond of 
war and battle. The old Welsh for av-i was ou, 
whence colloquial ou (with ii or y) in South Wales, and 
euor au in North Wales, as in dau, '*two," (Demetian 
and Gwentian dou, day), and the plurals papou, 
•'popes," and loggou, "ships," in the Book of Llan 
Ddv (p. 120), become in later Welsh pabeu, pabau, and 
llongeu, llongau. This ou occurs also in old Welsh 
names like Oudocuy, in Latin OudoceniSyS^nd in Outigirn, 
Eutigim, which would seem to have meant '* fond of 
his t'eym or king, loyal to his prince." The represen- 
tative of aV'i having become successively ou and cu, 
homophonous as . it was with the pronominal genitive 
eu, " eorum, earum, their, theirs," came to be dropped 
in unaccented syllables, as for example in Oudoceius, 

Digitized by VjOOQ IC 


Oudogwy, whence Llan-dogo, in Monmouthshire. On 
the other hand, when it came in time under the accent, 
it has remained, as in Eudafj probably from an early 
Avitamos, *' most loving," or else '* most beloved." 

Since the foregoing notes were set up in type I 
received, in answer to queries of mine, the following 
letter from the Rev. R. J. James, vicar of St. Ishmael's, 
Ferryside. I had already most of the evidence of 
Rogers, the parishioner, cited by the vicar ; but he has 
kindly put more questions to him than occurred to me. 
He touches also on other points of interest, so I am 
glad, with the Editor's consent, to insert his letter in 
its entirety, especially as it may suggest to some of 
our members further enquiry regarding the Chapel of 
Llansaint and its surroundings : — 

Only at the very last, and quite incidentally, have I come 
across a man who worked at the restoration of Llansaint Chapel 
in 1862. His name is John Jenkins, a joiner, 84 years of age, 
and I give you his own words in reply to my letter asking what 
he knew about the two stones : — 

" Yes, I acted as foreman for my uncle, David Grower, at the 
restoration of Llansaint Chapel, and the architect told the Vicar, 
Mr. Jones, that he need not have a clerk of the works there as 
I was carrying out his plans to his satisfaction. The two stones 
were not removed ; they are now in the same place as they were 
when I was a boy, in the south wall, between the windows, only 
lower down in the wall, as far as I can remember. The stone 
near the ventilators (on a level with the ground) was not 
removed. The piece was chipped off it in its place so as it would 
be square with the ventilators, and lies in its original bed or 
place ; and so does the other stone.*' 

When I came to examine the stones carefully for myself I 
began to doubt very much as to whether they had been removed 
at all, and then I took Rogers to the spot and told him my 
opinion, and pointed out to him how that the mortar round the 
two stones was the same as in the old part of the wall, and that 
the mortjir used in 1862 was altogether different, being both 
lighter in colour and fresher looking. Rogers then hesitated 
(of course, you must allow for a man's memory forty-four years 
ago, and more especially as he was not working there, but was 
a mere casual passing observer, for he then lived two miles 
away), and admitted that he did not now remember to a 

Digitized by VjOOQ IC 


certainty that they were removed at all, and he doubted it very 
much upon examining the mortar closely ; and so, with John 
Jenkins's evidence, obtained since, you may be certain that 
they were not removed at all. 

I asked Sogers then, how about the remarks, made amidst 
laughter, when it was observed that the letters on the upper 
stone were upside down, or that that stone was placed upside 
down ; and he then thought that they must have observed it, as 
it was in its original place ; and that tlien it was sugorested to 
put it right, but that seeing they would have to take down 
some of the wall before getting at it, they, 'midst laughter, 
thought it as well to leave it as it was. 

Now, I asked John Jenkins in my second letter to him, 
seeing that he had not referred to that incident in his reply to 
my first letter (and in my first letter I purposely omitted 
referring to that incident, just to see whether he would mention 
it of his own accord), in these words : — " I should be obliged by 
your letting me know whether you heard any opinion expre^ssed 
by anyone at the time of the restoration as to why the letters 
on the stone — the one higher up and nearer the Tower — were 
upside down : that is, the stone itself being upside down, which 
it is?" 

He replied, evidently misunderstanding my query : — " I know 
no opinion expressed during the restoration what the letters on 
the stones were ; if that gentleman I mentioned in my former 
letter had written to me as he promised, very likely I would be 
able to say more." 

" In his former letter referred to, he said : — " I have seen some 
gentlemen there taking copy of the letters; one of them promised 
to let me know their meaning as soon as he could make out 
anything of them. However, he never did, so I am sorry to say 
I know nothing more about them." 

By that, Jenkins either knew nothing about, or has forgotten, 
what Rogers states as to the workmen having observed the 
letters to be upside down, and the ** laughing " incident. 

Now to take your letter of 21st September last, seriatim : — 
There is no record as to the stones. Their dimensions are : the 
one upside down, 2 ft. 4^ ins. by 7f ins. ; the other, the lower 
one, 4 ft. 9 ins. by lOJ ins. 

Jenkins in his letter terms them " bluish-grey,*' and says the 
general opinion was that they were quarried at the Treforris-fach 
Farm quarry. 

A joiner and builder at Llansaint who measured the 8U>nes 
for me in my presence said they were " Iwyd/' making a 
difference between Iwyd and 14s, but garreg l&s, I have heard 

Digitized by VjOOQ IC 


such coloured stones usually termed in the parish ; and there 
are some who think that they were quarried close by Ffynnon 
Saint, and there is an old disused quarry close by it. One on 
the Geological Survey spoke of that kind of stone as " green," 
and which, I suppose, would correspond to our las ; but the two 
stones seem certainly lighter in colour than the Treforris-fach 
kind, for there are some new houses built close by the church- 
yard of the latter, and which appear much darker than the two 
stones in the Chapel wall. They seem exactly alike in colour 
to the Ffynnon Saint quarry-stone. 

Both the stones are undoubtedly — so the builder above referred 
to thought, and others also — of the same kind. 

Did you notice the remnant of a cross fixed in the Llansaint 
churchyard wall (north) — that, I have heard, was put there at 
the time of the restoration in 1862 , but John Jenkins knows 
nothing of it having been removed and placed there. 

Jenkins says further : — " I was just thinking of the Chapel 
in my first remembrance about it — a high-pitched roof, as it has 
now, and an arch at the chancel, a stone seat round the chancel 
built in wall.** 

If there was a stone seat as described, surely it would never 
have been removed ; Jenkins must be mistaken, probably, but 
there is one such seat in the old parish church, St. Ishmael's. 
The stone is a white kind of fire-stone, like the Caen or Bath 

I wonder what the origin of that was ? Doubtless you k now 
Tliere is one like it, it seems, at Westminster Abbey. 

In Llansaint Chapel there is now, you may liave observed, in 
the east wall, a rough stone inserted and projecting. Could it 
have been used as a credence-table ? 

I fail to find anything in any way in connection with 
Ffynnon Saint 

I do not know whether the following will in any way help 
towards elucidating matters as to the stones and wall : can you 
tell me how Llansaint came to be called "Alkenchurch," or 
Alkenchurch Llansaint ? 

It seems that in the Report of the Commissioners appointed 
by Edward VI., 1552, " to take and make a just vie we in every 
parishe," occurs the following : — " St. Ismaelle's. It'm, a Chalyce 
in Hawlkyng Churche, a Chaple annexed to the same p'rishe." 

The terrier dated 1636, in the old parish registers, which I 
have, speaks of *^ the Villadge called Alkenchurch ;" a copy of the 
same terrier made in 1720, in the Diocesan Registry, has Alken- 
church in it as in the original ; and it seems strange that if it 
was known and spoken of as Llansaint in 1720, that the vicar 

Digitized by VjOOQ IC 


and churchwardens at the time, in copying, did not make some 
comment with regard to it. 

The first record I have of the village being called Llansaint at 
all is in an entry of burial, " — was buried at Llansaint." (I 
suppose John Harries was a bit of an alien) :" January 16th, 
1744," only twenty-four years after the copy was made; and a 
curious thing, I fail to find any trace of the word Alkenchurch 
from the oldest inhabitant ; and yet there are some now living 
who remember their mother telling them that she remembered 
the time when there was no wall round the Chapel, and that 
there was only one grave (though it is not of the same name as 
above, of January 16th, 1744) ; and yet to this day they speak of 
Carmarthen Gaol (built in the Old Castle) as *' y Castell,'* and 
the old Llansaint Chapel as " y Gapel" and the Nonconformist 
Chapel £is Meeting-house, or Ty Cwrdd, 

You must please excuse the length of my letter, but of all 
things I wish to be acctvrate ? 

Before leaving the neighbourhood of Llan Saint, I 
may mention that Sir John Williams took me to call 
on a farmer who was with him at the Swansea Normal 
College in their earlier days. This was Mr. John 
Lloyd Thomas, of Tan Lan farm, near Llan Ishmael, 
and my business was to ask him about the remains of 
the port of Aber Towy, to which I have referred in my 
Celtic Folklore J p. 513. His story is, unfortunately, 
very short : after the storm of 1896, Mr. Lloyd 
Thomas saw walls there, which were in some places a 
foot or two high. They formed rooms, and showed 
unmistakable fireplaces. He had some forty or fifty 
loads of the stones carted away to his farm. If he had 
not done it, he said that others would. The ruins 
extended, he thought, some 200 or 300 yards along the 
side exposed to the sea. He had no doubt that in 
front of the foundations, which he then saw, entire 
streets of houses had been swept away by the storms 
of previous ages. Let me add that, in the course of a 
previous day s rambles, the President took me tc see 
where exactly the T^f enters the Towy, for from the 
maps I never felt certain whether it entered the Towy at 
all ; but such is the case, and at low tide the Towy can 
be traced for miles beyond and below the mouth of the 

Digitized by VjOOQ IC 


TS.f. When I wrote before about Aber Towy, it was 
Apropos of the " Hunt of the Twrch Trwyth " ; and I 
now find that the Twrch, in order to make his way to 
the Loughor and Amman district, would naturally 
proceed down the right bank of the T^f until he came 
to Ginst Point, which ended then probably consider- 
ably in front of where it ends now, for the sea seems 
to have gained enormously about the mouth of the 
Towy. Even now, there would, at low water, be no 
great distance to swim between Ginst Point and the 
site of Aber Towy. Lastly, I should be glad to know 
what the local antiquaries make of Speed's Hawton^ : 
where exactly was it ; what is the history of the 
name ? and are there any traces of an old route from 
PeuHniog (or the district of Paulinus, about Llandy- 
silio) to Ginst Point and Aber Towy ? 

Llandawke, Carmarthenshire. 
On the Wednesday, during the Carmarthen Meeting, 
the President's party tried to meet the other members 
of the Association, but we found ourselves too long 
delayed at Laugharne to reach Eglwys Cymun in time ; 
so we met them at Llandawke, and betook ourselves 
to the examination of the ancient inscription kept in 
the church. This was my third time of visiting it : 
my first visit was early in the Seventies, when I 
failed to get the Ogmic portion of the legend right. 
The stone had had a piece broken off its length, and 
that piece had disappeared. The remainder, with most 
of the lettering, served as a threshold, and it had 
evidently done so for a pretty long time, as the edges 
were already considerably worn away by the tread of 
feet. This affected the Ogams more than the Latin 
letters, as it shortened the scores, where they reached 
originally up to the edge to which they belonged. My 
second visit took place in the company of my friend, Mr. 
Thomas, vicar of Laugharne, on Good Friday, 1898, 

^ Since writinef the above, it has occurred to me that Hawton and 
Hawkyn{g)y p. 74, above, meant the same name : which was the 
correct form, if either P 

Digitized by VjOOQ IC 



when we solved the chief difficulty of the Ogam, 
reading of the Latin is as follows : — 


-> 5« 


•^^ • 

►.^ <=> 

0*3 -a 

•S a « 
9 ® § 
^ S Q 

a §5 

^ as 

,-5 Q ® 

^ a -p 

^ 2 
§ o £ 




That is: "The place or monument of Barrivendas : 
the son of Vendubaras lies here." For I presume that 
the Hic lACiT though not on the front of the stone, 
is a part of one and the same inscription, with the 
wo lines on the front : it has sometimes been alto- 
tgether overlooked. The names mean " white-topped," 

Digitized by VjOOQ IC 


or ** white-headed," and are in Later Irish Barr- 
Fmn and Finn-barr ; in Welsh, Berwyn and Gwyn- 
far. As they have been so often the subject of 
comment, I pass on to give some notes on the 
lettering. It is generally pretty good ; the A is broad 
and angular in the first instance, but narrower in the 
second. The R is good, though open in the first two 
instances, but narrow and closed in the third, for the 
inscriber perceived that he was approaching the ground 
line, and Degan to crowd the latter part of the legend. 
The first v is peculiar in having its first arm nearly 
perpendicular : the others are not so. We have a 
horizontal i at the end of the first line. The F is pro- 
longed below the line, and has the short i, which 
follows, attached to its lower bar ; the s is reversed. 
The second D appears to have been punched for a B, 
when the inscriber found his error, and proceeded to 
make it into D ; but how far he carried his intention 
out one is prevented from judging by the lower portion 
of the letter having been worn away. The letters 
making Hic lACiT have their lower ends carried away 
by a flaking of the stone, and they were all so placed 
as just to avoid the scores of the Ogam. Thus the 
three last scores of the UJIL {qu) of maqui come right 
down to the top stroke of the T, and the second upright 
of the H is shortened so as not to join the subse- 
quent -j- (m) of the Ogam. Now, the Ogam portion 
of the legend is written as usual in the direction con- 
trary to the Latin, and is on the edges ab and CD. On 
the former edge (ab), the reading is — 

1 1 .../... ■ ...ill.. . . I . . . 


and on the latter (cd) the following : — 

/ . Mill .MM / I 
/I M M I / I 


Digitized by VjOOQ IC 


The first line began exactly opposite the v of Barri- 
vendiy and the other Une began almost exactly on the 
same level. Not only is the Ji (d) beginning the Ogam 
opposite the v, but the first score of it is to be seen in 
the opening of the v, while the other actually forms a 
part of the first arm of the v : possibly it accounts 
tor its being, as already mentioned, perpendicular; 
for the other vs have no perpendicular arm. The 
perpendicular is the normal direction in the case of the 
11. It is right, however, to say that the other u is 
not perpendicular, but slopes backwards : in fact, the 
Ogam scores here all slope, more or less, and especially 

These inscriptions raise various diflScult questions, 
and the first is, how much of the Ogam is missing ? 
Line a b is practically complete, I fancy : as it is, it 
ends abruptly with three scoies on the b side; but 
I have treated them as originally four, which would 
make 5, and finish a genitive Dumeledonas, That is 
probably all there ever was on that edge. The other 
edge has less on it, Maqui M, with the second m followed 
by one vowel notch, which might be a, making the 
commencement of another maqui ; but it is far more 
probable that the word is to be completed into mucoi. 
Maqui maqui would mean '* of the son of the son, fdii 
JUii, grandson's." It sometimes occurs, but a very much 
commoner formula is maqui mucoi, " filii generis," 
foUowed by the name of the non-Christian ancestor of 
the family. What that was in this case, I cannot say, 
unless perhaps it was Vendubari : this would imply a 
good length of edge to write on, but what there was 
originally of line cd, together with the top of the 
stone, may have possibly aflforded the length required. 
Following that out, we should have ** Maqui mucoi 
Vendubari, Dumeledonas," and construe thus — '* The 
burial-place of the son of the Kin of Vendubar, 
namely, Dumeledo." On the other hand, maqui mucoi 
may have been all that there was on that side : then 
we should have to read the two sides together, as 

Digitized by VjOOQ IC 

80 tiPlGRAPHlC NOtES, 

Maqui mucoi Dumeledonds " (Locus or Memoria) filii 
generis Dumeledonis." In that case, the departed's 
name is not given, unless we suppose the Ogam to be 
taken in connection with the genitive Bamvendi or 
Venduhari. This raises the question of the relation of 
the two scripts to one another : which was there first ? 
Barrivendi has the first score of the jj utilized, so to 
say, in making the v, while the other score hangs into 
the open space of that letter, and the arm of the Ogam 
for m penetrates into the semicircle of the second R. 
Now the inclination of the v, decided probably by the 
Ogam letter jj, would go to prove that the Ogam was 
there first. The Hic iacit, in its careful avoidance of 
the Ogam, distinctly shows also the priority of the 
Ogam on the other edge. Why, then, did the inscriber 
not keep the letters of Barrivendi clear of the Ogam, 
which he could easily have done by carving them a little 
further from the edge ? I can only suggest that he did 
not notice the Ogam as such on a b, but that it was so 
fresh on the edge c d that he could not avoid becoming 
aware of it. This all means that the two inscriptions 
had nothing to do with one another ; not to mention 
that they may be of different dates, the interval between 
them having perhaps been long enough for the con- 
nection of the stone with the grave of Dumeledo to be 
forgotten, and for the Latin inscriber to seize upon it 
for his own purposes. In that case also, one need not 
suppose the lost name as lengthy as Venduhari : let us 
substitute for it the Mini of the Treflys stone, to be 
mentioned presently, and then we should have — 

Maqui Mucoi Mini 

" Filii Generis Mini, 
i.e.. Dumeledonis." 

The name Dumeledo, genitive Dumeledonas, claims 
kinship with Dumelus of the stone at Llanddewi Brefi, 
reading DALLur DVMELur,^ Dumel-i from Gortatlea in 

^ This inscription is hardly to be recognised as represented either 
by Westwood or Hiibner. A good photograph or rubbing, or better, 
both, should be printed of it in this Journal. I have only seen it 
since the publication of Lewis Morris's Notes by Mr. Edward Owen 
in this Journal in 1896. 

Digitized by VjOOQ IC 


Kerry, and another genitive, Ddumileas, from Dunloe, 
in the same county : compare the Irish place-name 
Cluain Domail, ''D. s meadow," in "Gorman/ June 2nd; 
also a genitive Duimle in the Book of Leinster, fol. 368^ 
The meaning of the name Dumel-, eludes me ; and I 
have to make the same confession as to the ending 
-edu or -edo, genitive -edonas, -edona, later -edon. 
Take, for instance, the Ogams in the Isle of Man 
yielding the genitives Bivaidonas and Dovaidonay 
and the inscription in Inchaguile in Lough Corrib, 
which has Luguaedon^ while Aaamnan supplies Nemai- 
don (misprinted Nemanidon in Reeves's text). That 
is not all, for Holder has brought together nearly 
thirty instances, among which he includes CcUedu, 
which occurs on the Colchester bronze tablet as 
Caledo, meaning a Caledonian ; the singular, in fact, 
of CaJedones. Holder gives this and Ccdedones a 
long e, and so with the termination -edu, genitive edon- ; 
but in the words, Caledo, Caledones, at least he is 
probably mistaken as to the quantity, as proved by the 
Welsh Celyddon and the name of Dunkeld, which was 
Diin Chailden, " the dun of the Caledonesy 

Nbvbrn, Pembrokeshire. 

Note by the Editor. — Since there appears to be some doubt 
as to the exact circumstances under which the Ogam stone 
No. 2 at Nevem was found, it may be as well to state the facts 
of the case in a few words. Just as the members of the Asso- 
ciation were leaving Nevem, on the occasion of their visit to 
that place on August 17th, 1904, Archdeacon Thomas came up 
to me in the churchyard, and asked me whether I had noticed a 
piece of interlaced ornament on one of the lintel-stones of the 
narrow passage which gives access to the staircase leading to 
the so-called "priest's chamber." I replied that I had not 
observed it, and went back into the church to have a look at it. 
What I saw first is shown on Fig. 4. On examining the 
adjoining stone, my surprise and delight may be imagined when 
my eye caught the Ogams on the angle of the stone which had 
escaped the notice of the Archdeacon. However, I am quite 
willmg to share the honour of having made the discovery with 
liim, as I should never have found the Ogams if he had not 

CtH SEB., vol. VII. 6 

Digitized by VjOOQ IC 


sent me back in search of the stone with the interlaced 
ornament upon it. The reading of the Ogam inscription given 
in the Report of the Cardigan Meeting in the Arch. Uamb., 
6th Ser., vol. v, p. 167, viz., ... I cunan maqui . . . was mine 
and no one else's. I guessed the name to be cunan, and the only- 
letter I read wrongly was the final N TTTTT ^^ cunan, which 
should liave been s TTTT- I ^"®^ ^ ^^^ ^ Latin inscription 
on the under face of the lintel, but the stone was so dirty and 
the lighting of the passage so bad that I could not detect any 
letters. a1 far as I can remember, the only members of the 
Association who were present when the discovery was made 
were the Kev. G. Eyre Evans, Mrs. Thomas Allen, Mr. Herbert 
Allen and Mr. Edward Owen, Archdeacon Thomas had gone 
on with the rest of the party to the carriages. I might have 




* .^f^^^^^ 

<' J?^^"^* 



.;^f^r ■ 

Fig. 4. — Lintel-Stone with interlaced work at Nevem, discovered August 17th, 
1904, by the Yen. Archdeacon D. R. Thomas, F.S.A. 
Scale, \ linear. 
{Dravm by W. Q. 8. , from a nAbing by the Rev. O, Eyre Evans,) 

returned to Nevern again on a subsequent occasion, and 
endeavoured to get the stone removed, so that I could read 
the whole of the inscriptions correctly ; but I preferred to 
leave the task in the far abler hands of Professor John Rhys. 

We spent the end of the week at the hospitable 
home of Dr. Henry Owen, at Poyston, in Pembroke- 
shire, and on Saturday, August 18th, he drove me 
and Mr. Williams of Solva,^ proprietor of the Pembroke 
Cmmty Guardian — a newspaper which does good work 
for Demetian archaeology — ^to the pretty village of 
Nevern, to see the stone on which Mr. Romilly 

^ Alas ! I have jast heard the news of his death: lie was suffering 
wlien he was with ns. 

Digitized by VjOOQ IC 


Allen in 1904 read in Ogam . . . . i Cunan maqui : 
see the Arch. Camh. for 1905, p. 167. The stone 
forms a lintel of the door leading to a little staircase 
which brings one to what is called the "Priest's Room." 
Alongside of it is another stone, which may prove 
even more important than the first one. A certain 
quantity of ornamentation was visible on its lower 
mce, and it probably has on it an elaborate cross, 
accompanied very possibly with an inscription. But 
I must confine my remarks to the first stone and ite 
inscriptions. I put it in the plural, for before I could 
find the Ogam edge, I noticed a Latin inscription on 
the under face of the stone ; that is, the side of it over 
one's head. My first attempt, however, was to read 
Cunan in Ogam, and I was glad not to find there any- 
thing so late, but {cufiCLS, which showed me exactly 
where I was. Then we got a mason to clear away 
patiently the mortar and stones alongside in the direc- 
tion of the beginning, which in time he got clearly 
visible. To my joy, the name revealed itself as Magli- 
cunds, and opposite it, running in the contrary direction, 
I first made out ocuni, and after a while glocuni, a part 
of Maglocuni, the Latin genitive of the familiar name 
MaglocunuSy in Welsh Mailcun and Maelgwny which 
Welsh printers invariably wish to murder into Mael- 
gwyn. This was followed by fli Clut, and I felt 
certain that the whole of the last vocable would prove 
to be ClutorigL But I was quite wrong ; for, as the 
stone was long and had an ample grip of the wall, the 
mason punched away until he had the under surface 
of the lintel clear well past the last letter, and the 
name completed itself as Clutoriy beyond all doubt. I 
had, while the mason was clearing the Latin letters, 
been puzzling myself at what was left visible of the 
Offam, and ttiere also the scores for cl showed them- 
selves. Further punching revealed the notches for u. 
I could not see the writing any further, but I thought 
I could just feel the three scores for t. It was 
impracticable to make a hole further in that direction, 

Digitized by VjOOQ IC 



or in pursuit of the Ma of Maglocuni. The readings 
may be represented thus : — 

— AT- 


, .-.-JWt — 




3 -l 

I • 

^ 1 

ft- « 




o *2 





The G seems to be Q; the l inclines to be A, while 
the c approaches <, and the n is written M, but the 
latter perpendicular is lengthened, possibly to indicate 
the I , somewhat after the fashion of Roman inscrip- 
tions, but the I may be there as a very close parallel to 
the \A : I could not decide with the light from 
below. The F is much of the usual tjrpe, with its 
lower arm drooping a little. The second limb of 

Digitized by VjOOQ IC 


the V tends to end with a slight curvature, except the 
last V, which ends well under the roof of the T. The R 
is rather an elegant letter, but its perpendicular is 
prolonged somewhat beneath the line. The final i is 
of the usual horizontal kind. I had no time to try to 
take a rubbing of the stone, and even had I time, 
I do not think it could have been done to any advan- 
tage while the stone is in its present position. The 
letters ^lo are very faint, and some of the others are 
not much less faint ; so 1 gather that all that face had 
been a good deal exposed to the weather, or more 
likely, to the tread of feet, before the stone was placed 
where it is now. 

The name which yields the Latin genitive Chitori is 
new to me, and one of the principal disappointments 
connected with this fiLd is that we cannot as yet get 
at the end of the Ogam legend, so as to ascertain what 
the Goidelic genitive exactly may be, which stands in 
Ogam for the ClutoH of the Latin. This, and the 
probability of the other stone having on it both a 
cross and an inscription, make it highly desirable that 
both stones should be carefully extracted from the 

As to the other name given in Latin as Maglocuni, 
implying a nominative Moglocimus, I may say that 
one has usually taken for granted that the early 
Brythonic was Maglocunos, genitive Maglocuni, but 
the Goidelic genitive Maglicunas shows that we have 
been mistaken, for this last seems to imply a nominative 
Maglicu, and the Brythonic was probably Maglociiy 
genitive Maglocunos. That is to say, the second 
element was the word for '*dog/' the etymological 
equivalent of the English word hound, and the Greek 
KvtDVy genitive kwo^, nominative plural kvv^s. The early 
forms in insular Celtic may have been cu, possibly 
cud, or cuds, genitive ctinos, making in early Goidelic 
cu, genitive cUnas, modern Irish cti, genitive cdn, 
nominative plural cdin ; in early Brythonic en, cii, 
genitive cai\, modern Welsh cl, genitive own (preserved 

Digitized by VjOOQ IC 


in Mailcwn, Maelgvm), nominative plural cwm., "hounds, 
dogs." There is a difference here in the thematic 
vowel ofmaalos, as the Goidelic has i in Magli-cunas ; 
but the genitive occurs in a Latin list in the Book oj 
Leinster Tfol. 362') as Magla-coni. The variation is due 
to the inaistinct pronunciation of the thematic vowel, 
leading up to its ultimate elision ; while in Brythonic 
the thematic vowel of the o-declension remained, being 
protected by the stress-accent down probably to a 
comparatively late period. See my " Origin of the 
Welsh Englyn and Kindred Metres ( Y Cymmrodorj 
vol. xviii, pp. 6-10). It will have been noticed that 
the Welsh Maelgwn comes from the old genitive, or, 
more strictly speaking, it represents perhaps the stem 
of the cases, while the nominative is lost, except 
in so far as we have it in Greoffrey of Monmouth's 
Mailgo, genitive Malgonis. The Goidelic ci'i, genitive 
cwiiaSy meant, as already stated, " hound, dog," while 
the other element becomes in Welsh the personal 
name Mmjlos^ Mael. In Goidelic it should be maglas, 
and it makes in Irish mdl **a prince, a hero;" so 
the compound name in Irish should be Mal-chRy 
genitive McU-chon, but I have never met with either. 
However, in the list of the kings of the Picts, a Brude 
Mac Maelchon occurs more than once ; and I believe 
somebody has suggested that the father of the first 
Brude so described was no other than the Brythonic 
prince, Maelgwn Gwynedd. Be that as it may, the 
oldest form which we have of the Pictish king's name 
is that given by Bede in his Hist. Ecc, III, c. iv, in 
the well-known passage reading as follows in Plummer's 
text : ** Uenit autem Brittaniam Columba, regnante 
Pictis Bridio filio Meilochon, rege potentissimo, nono 
anno regni eius, gentemque illam uerbo et exemplo ad 
fidem Christi conuertit." In Meilochon the ch argues 
a touch of Goidelic spelling, while the name as a whole 
seems to have come from the Brythons at a time when 
the thematic o was still retained. Compare the Pictish 
Vipoig from Vepdgenos, and see ** The Englyn" 1. c. 

Digitized by VjOOQ IC 


The importance of this class of names makes it 
necessary to discuss them with more minuteness than 
is usual in our Journal ; and I wish to point out that 
the elements making up Maglicu, genitive Maglicunas, 
are practically the same also which we have in the 
compound Cuno-maglos, which occurs in the genitive as 
Cunamagli at Kirk Arbory, in the Isle of Man, and also 
in " Maglus Conomagli Filius/'^n the Bollandists' Vita 
S. Winwcdoei. The making of two compound names 
out of the same two elements is familiar in such cases 
as the Greek SeoStopo^ and ^^podeo^y "Imrapxo^ and 
''Apxt'inro^, and it can be matched in most other Aryan 
languages. There is a point to be noticed as to the 
connecting vowel in Cono-magli and Cuna-magli : the 
declension of the word for *' hound " is a consonantal 
one, and supplies no such vowel at all, but the analogy of 
the o-declension is followed, and the pronunciation is 
helped by recourse to the vowel o (changed in Goidelic 
mostly to a, sometimes to e or i). Compare again the 
analogy of Greek in such instances as /cvvo'/ceif>aXo^, 
tcvvo-irpoawTTo^, KVPO'fAopijH)^, and others. But the evidence 
of our Maglicunas means a catastrophe to the old 
explanation of names like Cunomaglos, Cunotamos, 
Cunomoros, Cunovalos, and many more beginning with 
cuno, cono, Irish con-, Breton ccm-, and Welsh con-y 
cin-y cyn- ; for that explanation postulated an adjective 
cunO'Sy which was supposed to mean "high." This, 
however, was never shown to have had any existence 
in any Celtic idiom, so far as I know,^ and now for 

^ Some of the Welsh words relied on to support the existence of 
the adjective cuno-a look rather Pnghean, such as civn " altitudOy' 
and cynu " turgere" The latter is Pughe's speHing of cynnu in pan 
gynnu, in the £ook of Taliemn (Skene, ii, 189), which, unfortu- 
nately, means "when it set," though from it he derives erchynu, 
"to rise," for which he gives no quotation. The nearest actual 
form is the Gwentian cumnu, " to rise, to raise," which is probably a 
contraction of cyehwynnu, " to start, to give a start to''; the third 
person singular present-future indicative is cychwyn^ and also the 
imperative singular ; so the corresponding parts of cumnu are cwyn, 
** rises, raises," and cu^, " do (thou) rise, or raise," which cannot 
be connected with cwnnu, except that be a ahoriening o£ cwyujiu from 

Digitized by VjOOQ IC 


"high" we have to substitute "hound." But what 
sense, the reader will probably ask, can there be 
assigned to the names in question. Irish literature 
enables one to answer, for with the ancient Irish the 
cu, "hound or dog," was the guardian, watchman, 
fighter, and protector par excellence. The name and 
story of the hero Cii-Chulainn, " Culann's Hound," 
together with the analysis of other cii names in Irish 
tales, amply, prove the term to have been at one time 
one of respect and regard ; nor is it wholly irrelevant 
to mention the fact that, according to Strabo,^ Britain 
exported, among other things, dogs fitted for the chase 
and for war as carried on by the Celts, by whom 
he meant the Celts of Gaul ; and that the same 
sort of exportation was continued in Ireland down 
to the time of St. Patrick (Bury's Patrick, pp. 31, 

c^wynnu = cyhwynnu = cychwynnu, than which the more asnal verbal 
noan is now cycJivoyn. The chwyn portion stands for squend, which 
is represented by the Irish verb scendim. " I spring.'* Led bj bad 
spelling, Dr Stokes thought that the Welsh forms pointed to an 
early squeiid rather than the squendd which the Irish forms postnlate; 
but the pronunciation is cychwy?maf, cychwyTiwol, etc., as the school 
of reformed Welsh spelling would write them — and as old authors 
did write them — with nn for early nd. It is useless also to invoke 
Welsh gogoned, "gloriosus," for the first two syllables of that vocable 
appear to equate with the name Gu-caun, Guo-caun, Go-gaun, Ga- 
gon,Gugan : compare Cat-gucann, Cat-gocaun, Cad-ugann, Cadwgan. 

^ See Meineke's edition, 199, 200 : the passage refers to the 
exports from Britain in the historian's time, and it reads to the 
following effect : — These products are exported from the island, and 
also hides, slaves, and dogs suitable for hunting ; the Celts employ 
dogs also in war, alike these British dogs and their native breed. 
My attention has also been called to Orosins v. 14, where Bituitos, 
king of the Arverni, who, when he had in the year 121 B.C. made 
immense preparations for fighting the Romans, met such a small 
army of them that he is said to have bragged that they would not 
suffice for a meal for the dogs which he had marching with him in 
his army. Some months later, Bituitos was seen on his silver 
chariot adorning at Rome the triumph of the Roman general, who 
had a medal struck on one side of which is to be seen a Roman 
soldier fighting with a big dog. Lastly, a friend has favoured me 
with a reference, possibly in point, to Gratius's Cyneyetica, lines 

Digitized by VjOOQ IC 


341) ; the Irish wolf-hound is not yet extinct, though 
doubtless he is rapidly becoming a sad mongrel. 

Everything, in fact, goes to show how important 
certain varieties of hound or dog were to the Celts, 
and in this light the proper names in question 
would seem to yield good sense ; thus Maglocu would 
mean a prince or hero who was a guardian or protector ; 
Cuno maglos, a guardian who was princely or heroic ; 
and Cuno-tamos {\x\ Modern Welsh Cyndaf)^ one who 
is in the highest degree a guardian or protector ; that 
is, if the word is to be treated as a superlative. How 
far the dog was regarded in the same light by other 
Aryan nations, I am not prepared to say ; but I notice 
that the Greeks had such personal names as l^vvayo^, 
KwovKKo^y ^i\oKva>u, Kvi/€a<;, and KwtWo?. Similarly, 
Forstemann gives a small number of Germanic names 
beginning with hund, " dog," such as Hundpaldy Hunt- 
prehty and Huntgar; and possibly some of those begin- 
ning with hun belong also here, such as Hunbert^ 
Hungar, Hunhilt, and Hunleib. But German scholars 
consider the first element, whether hund or hun, as of 
very uncertain origin and interpretation in this class 
of compounds. 

As the members of our Association are aware, 
Nevern has another bilingual inscription : that of 
Vitalianus, which I am inclined to regard as one of the 
oldest monuments of the kind in the Principality. In 
" The Englyn," p. 74, I have gone so far as to suggest 
that this stone commemorates the grandfather of 
Vortigern. In any case, the site of the village of 
Nevern, occupying a sheltered spot on a tidal creek, 
was probably one of the headquarters of the Irish 
D^ssi ; and this may prove the key to the early 
history of the Demetian district of Cemmes. The 
Welsh form of the name of Nevern is Nanhyfer, 
from an earlier Nant Nyfer^ which enables one to 
correct an entry in the Annales Camhrice into Cian 
nant nimer ohiit — " Cian of Nanhyfer died.'' The 
year appears to have been 865 : see PhilUmore s note 

Digitized by VjOOQ IC 



Kian a nd jn difieith on ad dray 

otnch pen bet alltad. 
Bet kindilic mab corknnd. 

in the Cymmrodor, ix, 165. Cian was a common Irish 
name at that time, and we have no means of ascertain- 
ing whether this bearer of it was a priest or a chief; 
but in the ** Englynion of the Graves," No. 41, a Cian is 
mentioned in a way that suggests a play on his name, 
as though it were derived from Welsh ci " dog," as 
follows, with cund corrected : — 

Cian howlH in the wolves' wilder- 

ness afar 
Over an alien's fjrrave — 
The grave of Ou-Duilich son of 


Corco-Nutan (Book of LeinsteVy fol. 350*, 365*") is not 
quite the equivalent of Cm^knudy but it is near enough, 
and Cian is here associated with two other men 
bearing distinctly Irish names, but we cannot locate 
him or the grave of his fellow Goidel. The lolo MSS., 
p. 78, give the name Ciariy there spelt Ceian, to a 
Goidel whom they represent invading Gower and 
Morgannwg. It is possible that the Cian of the 
Englynion and lolo's Ceian were one and the same 
man with the Cian of Nevem. Nevern is mentioned 
also in the '* Hunt of Twrch Trwyth " (Mabinogion, 
p. 138), where the place, instead of being called " the 
Dingle of Nyfer," is called Glynn Nyuer, " The Glen 
of Nyfer," and we read of Arthur's men stationing 
themselves on both sides of the Nyfer, whereupon the 
Twrch moves away to Cwm Cerwyn, where he fought 
fiercely and repeatedly against Arthur s men before he 
got to Peuliniog, and thence to Aber Towy and the 
Loughor district, as mentioned already. This story 
suggests that the Twrch was more or less at home in 
Nanhyfer, but it helps us to no date. My attention, 
however, has been called by Professor Anwyl to verses 
alluding to Nanhyfer in a poem by Meilir, who is sup- 
posed to have lived from 1120 to 1160. The subject 
was Trahaearn, king of Gwynedd, who was slain in 
the battle of Mynydd Carn, in 1079, when fighting 
against GrufFydd ab Cynan and Rhys ab Tewdwr, 

Digitized by VjOOQ IC 



aided by their Irish auxiliaries. The general sense of 
the passage is doubtful, but the words most in point 
are not hard to understand : see the MyvyriaUy I. 
192 :— 

Ny dot.jnt dros nor etwaeth 
Pobl anhjuaeth Nanhyaer 
Gwytyl dieuyl duon 
rsgodogion dynion lletfer. 

They are uot come across the sea 

yet — 
The illbred people of Nanhyfer, 
Goidels, demons black, 
Scottia's rabble, men weak-kneed. 

That a Welsh poet of the twelfth century should have 
given such a character to the people of Nanhyfer is 
very remarkable, and shows that it must have at least 
been well known that they were of Irish descent, and 
that they were in the habit probably of receiving and 
harbouring invaders from Ireland. How late this 
continued it is impossible to say, or to guess how long 
these men of Nevern retained the use of the Irish 
language, This last is a very important question, 
especially when one calls to mind the comparative 
lateness of the Trefgarn Fach inscription with Ogtenlo 
" Ogtiu's Grave," and the Llanvaughan one with 
Trenaccat-lOj " Tringad's Grave." Both are in Ogam, 
and with them may perhaps be chronologically asso- 
ciated some inscriptions which are in Latin alone, such 
as the Llanllyr one, in the Vale of Ayron. At all 
events, I gather that the Irish language persisted in 
parts of Dyfed considerably later than I have been used 
to think : possibly down to the days of Cian, late in 
the ninth century. 

The two inscriptions on the stone which have occupied 
us here are, it will have been noticed, exact counter- 
parts of one another, which is very seldom the case. 
But this Nevern find, which appears to have been 
made by Mr. Romilly Allen and Archdeacon 
Thomas, will not yield us all the lessons to be derived 
from it, until the two monuments are taken out of the 
wall to which the ignorance of a forgotten generation 
has recklessly consigned them. It is to be hoped that 
this will be seen to by the Vicar, Mr. J. O. Evans, who 

Digitized by VjOOQ IC 


kindly gave us all the assistance in his power on the 
day of our visit, and by the enlightened Squire who is 
the head of the great Welsh family of the Bowens of 

Trbflys/ Carnarvonshire. 

In October, 1904, 1 received a letter from Mr. Charles 
E. Breese, of Portmadoc, whose father, the late Edward 
Breese, was an ardent antiquary, conspicuous at our 
meetings in the Seventies. The son, I am delighted 
to say, has inherited his father's tastes, and his letter 
was about a find made near the old church of 

Fig. 6. -Inscribed Stone at Treflys, Carnarvonshire. 
{From a photograph by C. E. Breese.) 

Tretlys. I quote the following words from it : **A find of 
considerable interest has been made near the old church 
of Treflys, which, you may remember, stands about 
midway between here and Criccieth, in fairly close 
proximity to the sea. Some six weeks ago, some 
workmen were engaged in removing on the west of the 
church the boundary wall enclosing the graveyard on 
that side, in order to make room for an extension ; and at 
a depth of about 1 ft. 3 ins. or 1 ft. 8 ins. from the surface, 
and forming part of the foundation of the wall, they 
came across an inscribed stone. I send you a rubbing 

1 The inscribed stone at Treflys has been already described in the 
Arch. Camb. for Janoary 1905, and an account of the church by- 
Mr. Harold Hughes will be found in the Arch, Camb. for October, 
1906. This fact had wholly escaped me when I wrote these notes. 

Digitized by VjOOQ IC 


which I have taken of it, and also a very rude sketch- 
plan of the locale. The stone is perfect, and the 
incisions upon it in a very excellent state of preserva- 
tion. The stone is only very slightly chipped at the 
bottom end, and is of the blue-slate type peculiar to 
this country. The cross is interesting, and is very 
carefully incised." Since then Mr. Breese sent me a 
photograph of the stone, together with its dimensions, as 
follows : — 

Whole length of stone 4 ft. 6 ins. 

Whole length of inscription ... 2 ft. 6 ins. 

Depth of stone ... ... ... Sins. 

Width of stone ... ... ... 9 or 1 ins. 

Tapering at top end down to ... 7 ins. 

At Easter, 1905, 1 spent a week at Arianfryn on the 
Mawddach, and made an excursion to Portmadoc on 
April 22(id, when Mr. and Mrs. Breese took me to 
Gresel Gyfarch and Treflys. At the latter place the 
inscribed stone had been carefully placed inside the 
little church, and I found the reading exactly as de- 
scribed by Mr. Breese, as follows : — 

That is, lACONvr Fiuvr mini iacit, *' (Here) lies laco- 
nus, son of Min." This requires one or two notes : the 
sis of the tall gamma type; the f has the first i of filiics 
attached to its lower bar, while the Hits is represented 
by a sort of combination, which may be described 
as T, making an L upside down, to be read as li : then 
a sort of arm sloping upwards (from the middle of the 
perpendicular of the r) towards the right provides a v, 
while the arm and the lower half of the perpendicular 
yields an r. Thus I regard the word as meant to be 
read FiLivr, which is the case required. The cross in 
front of the first name is, perhaps, more correctly 
speaking, the monogram of XPI2T02, the part + being 
intended for X, and the prolongation into a curve to 
the right for P, the Greek rho, I cannot recall any 
exact parallel nearer than Cornwall : see Langdon and 
Allen's Catalogue of Inscribed Monuments in Cornwall, 
in this Journal for 1895, pp. 50-60. There No. 13 is 

Digitized by VjOOQ IC 



in point as to the form of the monogram, and Nos. 22 
and 35 also as to its position in a line with the 
legend in Latin. Before leaving the inscriber's handy- 

work, I wish to mention that I thought I detected 
five Ogam notches near the top of the stone on the 
right edge opposite the monogram : they are not so 
like notches, however, as little pits into which one's 

Digitized by VjOOQ IC 


finger-tips would fit. They formed the end of the geni- 
tive Ictconi, which would be — 

_ M M! I I I I I II 


A C O N I 

And looking further down, I seemed to detect the out- 
ward extremities of three of the four scores for c ; in 
other words, a flake had gone long ago, carrying on it 
the rest of the Ogam legend on the edge and to the 
right of it; but I should like to have the opinion 
(Jf others on this point. 

To come to the names, I have nothing to compare 
for certain with lACONVS, unless it be lacinipoi, in an 
Ogam inscription now at Donard, in co. Kildare ; and 
probably one must not associate with them Irish ice, 
" a healing, salvation," and Welsh iachy ** hale, sound, 
healthy," which seem to have bifurcated from a com- 
mon stem iecc. The Venedotian prince called lacoby son 
of Belly in the Harleian Genealogies ( Y Cymmrodory 
ix, 1 70), is called by Greoflfrey xii, 6, lago, in the accusa- 
tive Idgonerriy as if he had found a name laco^ accusative 
Idconerriy in a Latin document ; nor is it quite impossible 
that our laconus is a form of the Christian name 
lacobus : at any rate, it would be almost as near, with 
its n, to the original as the Italian Gridcomo, and the 
English James, with their m. If, for instance, 'laicw/So? 
was imported with its fi pronounced v, as has long 
been the case in Greek, there would be nothing strange 
in Idcovos being made by Goidelic Christians into 
Idconus; but, needless to say, this is all conjecture. 
As to the other name. Mini, it is hard to say whether 
the inscriber regarded this as the genitive of Minus or 
of a derivative MiniuSy as it would do for either. 
There is an Irish adjective min, which signifies '* soft, 
smooth, fine, small"; but whether our Mini has any- 
thing to do with it is hard to say. It is more, per- 
haps, to the point to mention the name of a family or 

Digitized by VjOOQ IC 

96 snaRAPHic hotbeu 

sept called Moeumin, in Adamnan's Life of Columha. 
It occurs in the names of two brothers, Lugbeus 
Moeumin and Lugneus Moeumin^ where the latter 
word means Mocu Miuj or the Kin of Min. Some- 
times Adamnan expands the name thus : Lugbeus, 
gente Moeumin, or Lugbeus of the gens called the Kin 
of Min. The references to Reeves's edition are i, 15 
(43), 24 (53), 28 ^56), ii, 18 (127), 27 (141): none of 
the passages enable one to find out where the Mocu- 
Min were settled. But until some evidence of the 
occurrence of a patronymic mac Min is found, I cannot 
help thinking that flius is here to be interpreted in a 
wide sense, such as was sometimes given to mac in 
Irish, and that JUhis Mini is to be treated as a loose 
translation into Latin of the Goidelic Mocu ^ Min, 
Mocu is neither declined nor translated by Adamnan, 
but in the Ogam inscriptions of Ireland it is almost 
always in the genitive, mocoi or mucoi. The latter 
occurs also in this country, at Bridell, in Pembroke- 
shire, and at Silchester, in Hampshire ; but nowhere, 
unfortunately, has this difficult term of Goidelic socio- 
logy been found rendered into Latin. 

I may add that I have had other letters from Mr. 
Breese, in which he alludes to various antiquarian 
remains at Treflys and in the immediate vicinity. They 
range from a spindle- whqrl to a cromlech. Subject to 
the Editor's approval, I should like to suggest to Mr. 
Breese that he should write for this Journal a paper 
on the antiquities of the district around Treflys. 

Llystyn Gwyn,* near Brynkir Station, 

On the 1st of July, 1902, I received a letter from 
Mr. R. Pritchard Evans, of Felin Llecheiddior, in- 
forming me of the discovery of an old inscribed stone 

' An illustration of the stone will be found in the Ai-ch, Camh, for 
1903, p. 288. It was first published in the Proc, Soc, Ant. Lond.y 
Ser. 2, vol. xix, p. 255. ... 

Digitized by VjOOQIC 


on the farm of Llystyn Gwyn, on the estate of Col. 
Lloyd Jones Evans, of Broom Hall^ near Pwllheli. In 
the course of the month, I went with Mr. Pritchard 
Evans to see the stone. As a result of my visit, I read 
an account of the stone to the Society of Antiquaries. 
It appeared in their Proceedings for 1903, pp. 255-271 ; 
and in it I stated^ among other things, that I had it 
from an expert that the j stone is one of fgranitic 
texture, which is to be found in situ in the Bethesda 
district. It measures parallel to the inscription about 
3 ft. 6 ins., by 3 ft. the other way ; and as to thickness, 
it varies from rather more than a foot at the edge just 
above the lettering to 6 ins. at the edge opposite, 
namely, the one furthest from the lettering. It is a 
peculiarity of this inscription that it is crowded into 
one corner of the surface : see Fig. 8. 

When I first saw the stone it rested on the edge b d, 
and I was not able to have it shifted, so I detected no 
more writing ; but as it stood in a very unsafe place, I 
urged the farmer to have it moved. In time he did 
so, and the photograph which Mr. Pritchard Evans 
procured for me represented it standing as above, as 
may be seen in the copy printed by the Society of 
Antiquaries. That photograph, though showing the 
edge B D clear of the ground, into which it pressed 
itself when I saw it, suggested to me no additional 
writing; but when in the course of the Portmadoc 
Meeting the Cambrians visited it, the Venerable Arch- 
deacon Thomas detected Ogam writing on the edge 
B D. Then as I could get no reading of the Ogam 
scores, I had to wait till I could find an opportunity of 
revisiting the stone myself This came during my stay 
at Penrhos early in September, 1904, when Sir William 
Preece drove me to Llystyn Gwyn. We found the 
stone by this time standing near the farmhouse, and 
we detected the Ogam scores at once on the rugged 
edge B D, for that is only a little less so than the broken 
edge c D. The top a b is so thick and rounded that it 
can hardly be said to offer an edge for the Ogam carver 

6th seb. vol. vu. 7 

Digitized by VjOOQ IC 



to practice on. There was left a c, which presents a 
fairly tractable edge, and in point of position that 


Fig. 8. — Inscribed Stone at LljHtyn Gw'jn, Caruarvou«hire. 
{Prom a photograph by J. Allen Jones, Ilvjh Street ^ Criccteth.) 

would have been the one where I should look for 
Ogams, and I believe I did so when I first saw the 
stone. The reason why ac had not been used was 

Digitized by VjOOQ IC 



Fig. 9. — Latin Inscription on the Llystyn Gwyn Stone. 

Fig. 10.— Ogam Inscription on the Llystyn Gwyn Stone. 


Digitized by VjOOQ IC 


probably the same as for crowding the Roman letters 
into the right top corner. That reason cannot, I think, 
be other than that more than one-half of the stone 
must have been covered by other heavy stones before 
either inscription was carved. What sort of burial 
that may exactly imply I am unable to say ; and it is 
a pity that the site should not be carefully excavated, 
for there is still one big stone there lying m situ^ 
perhaps more than one. The tenant can show the 
exact spot, and Col, Lloyd J. Evans would probably 
only be too glad to see the work done thoroughly : in 
any case, there would not be much to do. 

One of Sir William Preece's party took a photograph 
of the stone, but as the day was not favourable the 
result did not come up to our expectations. I read the 
Roman letters as before : icoRi filiv f/potenti/ni, that 
is Icori filiu[s] f(ili) Potentini — "I. son of the Son 
of Potentinus. ' The Ogam reading up at right angles 
to it is as follows : — 

IMM MM ■ I ///// I I I 1 I // ■ 

"///// ""•// ' MM 


Some of the vowel notches are indistinct, but there is 
no serious difficulty as to the reading. It is right to 
say that the two inscriptions come in one another's 
way, for where the second F comes to the very edge of 
the stone, its two bars interlock with the arms of the 
Ogam J-l^ ; but they are, I thought, just kept from 
touching one another. 

Now, as to the names, one sees at once that Icorigas 
is the genitive of the name which is given as ICORI in 
the Latin, and that it is impossible to regard ic as 
standing here for the Latin hie, as I wrote before seeing 
the O^am. Unfortunately, there are two ways of 
regardmg the declension of the word : (1) the form 
ICORI stands for an earlier Icori-s of the z-declension, 
making in the genitive Icoriy-dSy written Icorig-dS, to 
be compared with Avittorig-es, the genitive in es of the 

Digitized by VjOOQ IC 


name written in Latin Avitoria in the nominative. 
(2) Or else it may be treated as standing for Ico-rlx, a 
compound making its genitive Ico-rlg-as, involving, in 
fact, the word for king, Old Irish H, genitive rig, 
Welsh rhi. Holder supplies two forms in point, and 
on the whole they may be said to favour the second 
view. They are Icongiumy a place-name from the 
neighbourhood of Treves, formed most likely from 
Icorix. The other is Icovellauna, the name of a 
goddess in inscriptions from the vicinity of Treves and of 
Metz. Vellauna is the feminine of Vellaunos, meaning 
probably one who rules or reigns, a prince : compare 
CdssivellaiinoSy Catuvellaunos, and the like. But all 
this does not enable one to fix the meaning of the first 
element ico in the composition of names such as Ico-rlx 
and IcO'Vellauna. 

As to the other name, PotentinuSy I have given it 
as mv opinion that Filius Fill Potentini may be a 
literal translation from Goidelic ; and, as I have got 
no '* forwarder" since, I cannot do better than repro- 
duce it in the form in which it occurs in the paper 
read to the Society of Antiquaries, as follows : — " The 
name Potentinus occurs in one of the Roman inscrip- 
tions at Caerleon, and we have Potenina, which looks 
like a reduced form ofPotentina, on a post- Roman stone 
found at Tregaron, in the neighbourhood of the Roman 
site of Llanio, in Cardiganshire. Potentinus^ as a 
derivative from potens, 'powerful, strong/ has its 
parallel in Irish in the name Ceithemach, which comes 
from mediaeval Irish cethern or ceithem, in Welsh 
cadam, ^ potens^ strong, able-bodied,' literally * fit for 
war/ cognate with cad, ' battle,' Irish cath of the same 
meaning; but the Irish word cethern has only come 
down in the sense of ' soldiers,' or rather perhaps a 
* band of soldiers,' as it is used in the singular with a 
plural meaning, and it has been borrowed into Welsh as 
such, while in English it became hern and oateran (see 
O'Dono van's Battle of Magh Rath, p. 140, and Stokes's 
Saltair na Rann^ line 3538). The kern seem to have 

Digitized by VjOOQ IC 


earned at an early date a very bad reputation, and 
y gethem is usually connected in Welsh with heU, and 
means the rabble of demons associated with it. To 
return to the inscription : Filius JUi Potentini may be 
treated as the equivalent of some such a mediaeval 
Irish designation as Mac Meic Ceitherrmich, or 
Mc Ceithemaigh's son. In fact, it is perhaps needless 
to look for any other, as Mac Ceithernaigh occurs as a 
proper name in Irish annals, for instance, in those of 
Ulster, A.D. 1382 : in the translation of the Four 
Masters it is anglicized as * Mac Keherny,' and it was 
borne by one of the chieftains of Connaught." 

Curiously enough, the meaning given to the feminine 
collective cethern in Welsh, recalls the strong words in 
which Meilir indulged in reference to the Groidels 
of Nanhyfer : it testifies to a racial animosity which 
has taken centuries to die out in the Principality. 

Note. — With regard to the illastrations of this paper, it should 
be mentioned that Mr. Worthington Smith has done his best with 
the materials placed at his disposal, which consisted chiefly of 
rubbings. These are occasionallj misleading, if not corrected bj 
moans of photographs and sketches taken from the stones them- 
selves. The most satisfactory results can only be obtained by photo- 
graphs of casts of the stones, and these are not as yet available. The 
rubbings of the Nevern No. 2 stone were taken by the Rev. G. Eyre 
Evans in the depth of winter, at considerable inconvenience to 
himself.— Ed. 

Digitized by VjOOQ IC 


Cambrian ^trdbaeologtcal 2[0£(ociatuin. 




On MONDAY, AUGUST 13th, 1906, 






JjOccU Committee. 
Oaiman.— ALAN STEPNEY-GULSTON, Esq., Dkrwydd. 

The Mayor of Carmarthen (H. E. Blagdon- Richards, Esq.). 

Sir James HiUs-Johnes, V.C, G.C.B., Dolaucothi, Llanwrda. 

Mrs. Johnes, Dolaucothi, Llanwrda. 

Sir Lewis Morris, Penbryn. 

The Venble. Archdeacon Evans, Carmarthen. 

Rev. T. R. Walters, Carmarthen. 

Rev. W. Davies. Llanfihangel Abercowin. 

Rev. J. Thomais Laughame. 

Rev. D. D. Evans, Llangunnor. 

Rev. George Eyre Evans, Aberystwy th. 

Rev. W. W. Poole Hughes, Warden, Llandovery College. 

Rev. J. Marsden, LlanDwch. 

Rev. Alexander Williams, Llangatheu. 

Colonel Gwynne Hughes, Glancothi. 

Meuric Lloyd, Esq., Delfryn. 

G. G. T. Treherne, Esq., 28, Bedford Row, London. 

D. Lleufer Thomas, Esq., Hendre, Swansea. 

T. W. Barker, Esq., Oaklands, Carmarthen. 

Digitized by VjOOQ IC 


Local OomnUttee. — Continued, 
T. E. Brigstoeke, Esq., King Street, Carmarthen. 
J. D. Jones, Elsq., Bronydre, Carmarthen. 
H. C. Tierney, Esq., Editor, Wdshman, Carmarthen. 

A. LI. I>avies, Esq., Brynderw, Carmarthen. 
J. B. Morgan, Esq., Architect, Llanelly. 
Mrs. D. Pugh Evans, Parade, Carmarthen. 
Miss Spurrell, King Street, Carmarthen. 

H. S. Holmes, Esq., Training College, Carmarthen. 
R. M. Thomas, Esq., Town Clerk, Carmarthen. 

E. V. Collier. Esq., Architect, Carmarthen. 

Arthur R Davies. Esq., 5, Quildhall Square, Carmarthen. 
John Lewis, Esq., Arybryn, Carmarthen. 

B. A. Lewis, Elsq., Morfa House, Carmarthen. 

F. W. Childs, Esq., Architect, Carmarthen. 

Hon, Local Treasurer, 
P. J. Wheldon, Esq., Nat. Prov. Bank, Carmarthen. 

Hon, Local Seeretwrtu, 
Rev. M H. Jones, 22, Picton Terrace, Carmarthen. 
Walter Spurrell, Esq., King Street, Carmarthen. 

General Secretaries to the Association, 

Rev. Canon R. Trevor Owen, M.A., F.S.A., Bodelwyddan Vicarage, 

Rhuddlan R.S.O., North Wales. 
Rev. C. Chidlow, M.A,, Lawhaden Vicarage, Narberth. 

Digitized by VjOOQ IC 



MONDAY, AUGUST 13th, 1906. 

A CoNVBBSAZiONE aod reception of the members of the Assooiation 
by the Mayor and Mayoress of Carmarthen, Mr. and Mrs. H. E. 
Blagdon-Riohards, was held in the Assembly Rooms at 8 p.m. 

The Mayor, in offering a welcome to the members of the 
Association, said it was a very carious fact that on the first visit of 
the Association to Carmarthen it had jnst terminated thirty years of 
its existence, and now at the time of its second visit, it had jast 
completed the second thirty years. He took it that in the usual 
course of events the third visit of the Association would be in the 
year 1936. He and all of them would very much like to put that 
date a little closer to the present year. He felt highly gratified — 
not to say honoured — that it fell to his lot to tender to each 
distinguished member of the Assooiation the warmest and heartiest 
welcome into their midst that it was possible for a man to offer. 
He did so on behalf of himself and the Mayoress, and on behalf of 
the Corporation of this ancient borough, with the assurance also of 
the fullest appreciation of his fellow-townspeople. He should like 
to emphasise that assurance, because he wished them to believe him 
it was not a mere idle sentiment prompted by ordinary courtesy. 
He knew it to be based upon a very lively feeling of satisfaction 
that existed in the town over this present visit of the Association. 
They in Carmarthen took a very great pride in the history and 
historical status of the town, and especially of being St. Peter's 
boys, of which he was ona Of course, it was very natural and 
pardonable for anyone to eulogise his own town, but it was more 
than usually justified in their case. He supposed ho might assume 
everyone present had read every historical document relating to 
the ancient charter of this old town. It was simply teeming with 
eulogistic references to the town. One in particular read very 
nicely, and he had made a copy of it. In a certificate made out in 
the year 1548, it stated: ''The town of Carmarthen is a fair 
market town, having a fair haven and the fairest town iu the whole 
of Sonth Wales, and of the most civility.'' He really did not 
consider that by any means an exaggerated way of putting it, 
because, in his experience, which was a very long one, he had 
known this statement to have been made scores and scores of 
times ; in fact, whenever visitors came to Carmarthen there were 
four things they did. First of all, they visited the Market-place, 
especially if they could do so on a Saturday. Then they went down 
to visit the Carmarthen docks ; then they admired the beauties of 
the town from its various aspects, inside and around ; and lastly. 

Digitized by VjOOQ IC . 


bj no meftns the leasfc, thej fallj appreciated the oiTilitj shown 
them by the Carmarthen people. He held that there was no single 
spot in the Principality that contained, in proportion to its size, 
landmarks more calcnlated to aronse the dreamy interest of the 
antiquary than were contained in this small area. He said this 
advisedly, becanse those who had made themselves familiar with 
the town and its history by means of stndy and previous visits would 
bear him out in that statement; and with regard to those who 
had not had the inestimable privilege of visiting this town before, 
they would, during the course of the week, come to realise the truth 
of every statement he uttered there that night. He could give them 
instances beyond nnmber, but would content himself with drawing 
attention to just one spot of interest and great antiquity in the 
town — that was the Old Oak in Priory Street, the site of the Old 
Priory. It had been said — and there were many believers in the 
faith to-day, who maintained the legend — that when the Old Oak 
falls Carmarthen will be swallowed up by the ravages of the iiea. 
And this belief was perfectly real and clear, because it was shown 
by the amount of attention given to this old oak. It was alive at 
the first visit of the Association, but since then had suffered 
premature decay ; and when they came to visit it sometime that 
week, they would be astounded to know that all that rested between 
Carmarthen and utter destruction was the little support given fo 
the old tree by means of mortar and bricks and iron bands. 

The Mayor then vacated the Chair in favour of Archdeacon Thomas, 
St. Asaph, who thanked the Mayor for his warm welcome on their 
second visit to this ancient and historic town. He could hardly 
support the statement that it was thirty years since their last visit ; 
it was thirty-one years, and he was afraid that would add one year 
more to their happiness. However, he hoped it would not make 
ranch difference when the time came round again. There wore 
some there now who were present at the first meeting. Some 
very active members and workers in the Association — ^their 
Secretary for South Wales, and also their Editor, joined on that 
occasion. Those present would remember the admirable address 
with which the then President, Bishop Basil Jones, opened the 
meeting; they would think what great giants they were in 
those days, and he was afraid they would look down perhaps 
a little on their followers of the present day. Work went on 
though the workers changed ; and when they came here this 
time they had an advantage which those members of 1875 did 
not possess. They found, in their itinerary course from North 
to South, that their endeavour to stir up interest in archsdology 
did bear some considerable amount of fruit. A good many 
societies had arisen in the kingdom that were doing admirable work 
in the country; and here in Carmarthen they had the great 
advantage on coming amongst them, of finding that not only would 
they have interpreters on the spot, but that the excellent society 
that had been established here had evidently taken root and done 

Digitized by VjOOQ IC 


good work. Referring to the antiquarian records, he said, that 
looking orei* the index of parks published^ be saw a verj great 
variety and extensive meaning of place-names. That showed that 
one or other had looked np the facts and the history of those varions 
places, and there was very valaable material for their fntnre 
historian. He had been to Carmarthen many times sinoe 1875, and 
there were two places he made a point of seeing — one, the 
Parish Charch, with its very interesting monuments, which was in 
many ways an object of great interest as well as care, lie wished 
he could say the same of the other object which he had come from 
time to time to look at, and it was one of the moat interesting 
things — the remains of the ancient Castle. When he came to stand 
opposite that fine gateway, and saw how little of it was visible, he 
had from time to time a spirit of sadness that it was blocked np 
as it is by the surrounding buildings. He thought as it was their 
object to stir up interest, and try to bring about a better arch»ological 
condition of things. It would be a matter of great credit to the town 
if they were to start — and indeed it would be extremely gratifying 
to the Association if their visit led to a removal of those unsightly 
buildings, and to disclose to the public that fine gateway. He 
threw this out for, from what the Mayor had said he evidently took 
a deep interest in archaeological matters. They had their local 
Society, and they had as representative of the Association their 
good President, and he hoped that when 1936 or '37 came round they 
would see the gateway of the Castle, to say nothing of changes 
inside. He did not suppose he would bo here in 1936, but no doubt 
members of the Association would then appreciate the kindness and 
welcome that was extended to them, and would rejoice more than 
those present could now rejoice, at the completion of what he had 
thrown out as a suggestion. In vacating the chair in favour of Sir 
John Williams — who, he said, wns exceedingly competent, and had 
a special claim to fill it— he observed that in Sir John they had one 
who had shown by his energy and skill what a Welshman could 
do. By his talents and merits he had risen to a position they 
very much envied. Now he had reached the top of the ladder ho 
had come to live among them in Carmarthenshire, and give full 
vent to that love for Wales and its literature which they knew he 

Sir John Williams then took the Chair, and said he joined in the 
welcome given by the Mayor to the company to the ancient capital 
of the Principality of Wales. A year ago the antiquaries of the 
town joined the Society and the Cambrian ArchaBological 
Association to place on record the historical and architectural 
buildings of the town. It was now a lusty and thriving infant, 
and had unearthed at Cwmbrwyn remaimi of Roman works that 
had remained hidden from sight and lost to memory for one thousand 
years. He expressed regret that the Rev. M. H. Jones, one of the 
founders of the Society, the Editor of its records, and one of the 
Secretaries, was about to leave Carmarthen for another sphere, where 

Digitized by VjOOQ IC 


he had been appointed to a post of great importance and far- 
reaching inflaence in the religions world. He took this opportnnitj 
of wishing him Gk>d-speed in his new undertaking. The r resident 
then proceeded to deliver his address. 

Thb Presidbnt's Address. 

It is with much pleasure that I join his Worship the Mayor of Car- 
marthen in welcoming yon to the ancient capital of the Principality 
of South Wales. An important station during the Roman occupa- 
tion, its after-history is intimately connected with the story of the 
legendary Myrddin, of the Princes of South Wales, and Rhys ap 
Thomas, epoch-makers in the history, not only of Wales but of 
Oreat Britain. 

A year ago, the antiquaries of the town and county joined to 
form a society — The Carmarthen Antiquarian Society — the object 
of which is to place on record and to preserve that which is left of 
the historical and architectural antiquities of the county. It is a 
lusty and thriving infant, of great promise, and on Wednesday you 
will have an opportunity of seeing one of its early achievements, in 
the excavations made at Cwmbrwyu to reveal a Roman building 
which has been hidden from sight and lost to memory for more 
than a thousand years. Thursday morning will be devoted to visit 
places of interest in the town — some of them Roman, others of a 
later period. To-morrow morning the Society will visit Llanstephan, 
where a castle, a church, and two holy wells await inspection by the 

The story of the Castle is but little known, and little has been 
written of it That which was known at the time Mr. Waters 
wrote is recorded in his valuable essay on The History of Llanstephcm^ 
Past and Present, The exact date of its foundation has not been 
ascertained, but it must have been built during the very last years 
of the eleventh century or the early part of the twelfth ; for in the 
year 1137 it was destroyed by Owen and Cadwaladr, the sons of 
Gruffndd ap Cynan, Prince of North Wales. The earlier Welsh 
castles which were burnt by the Welsh princes were not the massive 
stone buildings which have come down to us. 

The date of this, the first destruction of the Castle known to us, 
is usually given as 1136, and Mr. Waters adopts it in his History of 
Llanstephan^ but there are reasons for believing 1137 to be the 
correct date. The Armales Cambrice, and Caradoc in his Historic of 
Cambria^ before ih^i HistoHe was *' greatly improved and enlarged" 
by Wynne, give the date as 1137. Moreover, the story of the 
BnU^ which is the only authority adduced in favour of the opinion 
that 1136 is the correct date, under the year 1136,. reads thus : — 
" Yn y vlwyddyn rac wyneb," which is translated " in the year 
ensuing," and apparently means the year 1137. ^' In the ensuing 
year " it reads '^ Grnffudd, son of Rhys, died. ... In the same 

Digitized by VjOOQ IC 


year Groffudd, son of Cynan, died. ... In that year also the sons 
of Grnffadd, son of Cynan, came the third time into Ceredigion, and 
homed the Castle of Ystrad Mearig, the Castle of Llanstephan, 
the Castle of Hamfrey and Carmarthen.*' 

Now in what year died G-raffadd ap Rhys and Groffadd ap 
Cynan ? The writer of the AnncUes states definitely that their deaths 
took place in the year 1137. Caradoo is equally definite. His 
words are : *' The year 1137 died Graffyth ap Bees ap Theodor. . . . 
Also, toward the end of the same year died Graffyth ap Conan, King 
or Prince of North Wales.*' The three anthorities agree that the 
destruction of the Castle was effected in the same year in which the 
Prince died, and Caradoc and the Aivnales state definitely that they 
died in 1137. We may therefore conclude that the burning of 
the Castle by the sons of the Prince of North Wales took place in 
the year 1137. 

Twice in the previous year, 1136, had the same sons of Grnffudd 
ap Cynan raided Ceredigion, once with the aid of Gruffudd ap 
Rhys ; and it was on the third occasion of raiding that district that 
they marched south and destroyed the Castle of Llanstephan. 

Had anything more than love of country and hatred of the 
Norman adventurers been wanting to infuse them with courage, to 
give strength to their arms and keenness to their vision, it was 
furnished by the cruel treatment of the Princess Gwenllian, who 
was the daughter of Gruffydd ap Cynan and the wife of Gruffudd ap 
Rhys, by Maurice de Londres. Gwenllian was taken prisoner in 
the battle of Cydweli. The date of this battle is generally given 
as 1135. 

The Brut and Annates make no mention of Gwenllian or the battle, 
bat Giraldus in his Itinerary states : *^ In this district, after the 
death of King Henry, whilst Gruffydd, son of Rhys, Prince of 
South Wales, was engaged in soliciting assistance from North Wales, 
his wife Gwenllian (like the Queen of the Amazons and a second 
Penthesilia) led an army into these parts ; but she was defeated by 
Maurice de Londres, Lord of that country, and Geoffrey, Constable 
to the Bishop." 

Henry died on December Ist (December 3rd, according to the 
Brut), 1135, so that the battle of Cydweli could not have been 
fought until the very last days of 1135, and may not have been 
fought until early in 1136. The treatment of Gwenllian by Maurice 
may account for the three raids made into the Marches of South 
Wales by the sons of Gruffudd ap Cynan, in such rapid succession 
in the year following the death of Gwenllian. 

We know of no attacks upon the Castle from this time until the 
year 1146 (1147, according to the Annates), 

When Gruffudd ap Cynan died, there was some prospect of a union 
of the Welsh Princes, but in the year 1143 Cadwaladr killed 
Anarawd, the son of Gruffudd ap Rhys. This led to disunion among 
the Welsh. Meanwhile, the Lord Marchers who were united became 

Digitized by VjOOQ IC 


so aggressive that the sons of Graffadd ap Bhjs in South Wales 
with difficnltj held their own. Under these ci reams tanoes, Hy wel 
and Owen, the sons of Owen Gwynedd, came to their aid with a 
large army. They besieged the Castle of Carmarthen, and after five 
days oaptnred it. Afterwards, Cadell, Rhys, and Maredudd, the 
sons of Gmffndd ap Rhys, marched on Llanstephan, and conquered 
the Castle. 

Henry II, who was at war with Philip of France, died at Chinon 
in 1189, and in this year Llewelyn ap Grnffndd took possession of 
the Castles of St. Clare, Langhame, and Llanstephan. 

The year 1215 will remain ever memorable in the story of Britain, 
for in that year the English barons, in arms, appeared before King 
John to demand the charter of English liberty. Llewelyn allied 
himself with the barons, and together with the Welsh princes 
^thered a large force, marched upon Carmarthen, captured the 
Castle and rased it to the ground. They then demolished the 
Castles of Llanstephan, Talacharn, and St. Clare, and subdued the 
whole of South Wales except Pembroke and Glamorgan. 

For forty years after this event, it does not appear that the 
Castle was taken or attacked by the Welsh ; but in the year 1255 
the Lord Marchers became so aggressive and tyrannical that the 
Welsh nobles *' came to Llewelyn, having been robbed and made 
captive, and complainingly declared to him that they would rather 
be killed in war for their liberty than suffer themselves to be trodden 
down in bondage." Llewelyn, together with Maredudd ap Rhys 
Grug, invaded the midland country of Perfeddwlad, and subdued it 
before the end of a week. He gave Bnilth to Maredudd, and banished 
the owner, Rhys Vychan. 

This led to another battle in the following year ; for " Rhys 
Yychan, meaning to recover his lands, obtained of the King a large 
army, whereof one Stephen Bacon (or Banson) was captain." They 
came to Carmarthen, and having devastated portions of the district, 
marched upon Dynevor. Here they met the Welsh in force, and 
having suffered a disastrous defeat, fled, having lost, it is said, two 
or three thousand soldiers. Afterwards, the Welsh army went to 
Dyfed, and burned the county and destroyed the Castles of Aber- 
cowan (Laugharne), Llanstephan, Maenclochog, and Narberth. 

After this we find no record of fighting at Llanstephan, until the 
year 1403, when Owen Glyndwr captured John Penres, the keeper 
of the Castle. 

The first Lord Marcher of whom we find mention is Geoffrey 
Marmion, who was lord about the middle of the twelfth century. He 
may even have been the first Lord Marcher of Llanstephan. He 
granted the church of Llanstephan, with some glebe land and 
other privileges to a certain Master of the Slebech Comraanderj of 
St John of Jerusalem, as is mentioned in Owen's Pembrokeshire ; 
where is also to be found an inventory of the gifts made to the 
Slebech house. 

In making this gift, however, Geoffrey planted a seed which 

Digitized by VjOOQ IC 


after some years bore fniifc in the form of a lawsuit. The year of 
his death is not known. He had a daughter named Albreda, or 
Albrea Marmion, who was his heir. She was married to William de 
Camville, who was son of Richard de Gamville, one of the leaders 
and constables of the fleet of Richard I, and who died at the siege 
of Acre, 1191. 

They had a son named Geoffrey de Camville. In the year 
1200 King John granted a charter to Geoffrey, confirming to him 
the Castle and town of Llanstephan, as William de Camville held 
them, on the day he gave them to Geoffrey. 

Geoffrey was to hold the Cattle by the service of one knight's fee, 
to be performed in Sooth Wales for all services, as the charters of 
William his ^Either and Albreda his mother *' reasonably testify." 
This Lady Marcher, Albreda Marmion, who was the heir of 
Geoffrey Marmion, appears to have reserved some rights and 
claim over lands in Llanstephan, when she gave her hand 
in wedlock to William de Camville ; for the confirming charter 
refers to the charters of William and Albreda: and in the 
year 1228 we find that Albreda Marmion appeared before the 
King at Gloucester, and " quit-claimed to William de Camville all 
right and claim she had in the land of Llanstephan.'' Now her son 
Geoffrey had a son named William, and I am of opinion that 
Albreda Marmion quit-claimed all her rights in Llanstephan lands 
to her grandson and not to her husband. 

This lucky William, the pet of his grandmother, succeeded to 
the lordship when ho was still a minor ; and whilst he was a minor 
in the custody of the King and the Earl of Salisbury, a certain 
Conan Howell, a Welshman, came and occupied during the custody 
of the King. 

William had a son named Geoffrey. This is Geoffrey number three. 
The Normans seem to have been as chary as the Welsh of adding 
to the Christian names in the family. This third Geoffrey is he, 
during whose tenure of the lordship the gift of the Church of 
Llanstephan to the Slebech Commandery fructified and bore a 
lawsuit He instituted proceedings against William de Hamleye, 
Prior of the Hospital of St John of .Jerusalem in England, and 
Gilbert de St Augustine, Master of Slebech, to recover the 
advowson of Llanstephan Church, deforced from the Lord of 
Llanstephan, by the Prior of St. John and the Master of 

There was some hard swearing in the course of the litigation, 
but the Lord Marcher won. He won, however, to find himself out- 
manoeuvred by Thomas, Bishop of St David's, who wrote : " To 
the Venerable and discreet man, Lord Robert de Tybetot, justiciar. 
. . . whereas according to ecclesiastical laws . . . churches ai*e to 
be vacant only for times defined by law, and the Church of Llan- 
stephan, owing to the plea moved between Geoffrey de Camville 
and the Master of the House of Slebech, hds now been vacant for 
a long time, and beyond the term of the statute, on which account 

Digitized by VjOOQ IC 


the collation has fallen upon as, loci of the same diocese and 
bishopric, and it is onr will to provide properly for the indemnity 
of sonls, lest the rapacious wolves should destroy the Lord's flock, 
destitute of a pastor, the said Church of Llanstephan being vacant 
beyond time. . . . Grant to our beloved in Christ, Thomas de 
Goedeli, enjoining him, etc." 

Geoffrey, the successful suitor, had to solace himself with an 
earthly reward — the damages of the valor of the church for two 
years, to wit, 120 marks. 

The suit interests us chiefly because the evidence given in the 
course of it enables us to complete the list of the Lord Marchers of 
Llanstephan, from Geoffrey Marmion, who may have been the first, 
to the year 1338. 

The litigious Geoffrey de Camville died in 1308. 

He had a son named William. This is William the third. He 
was a Knight of Paine de Chaworth, Lord of Cydweli, in tbe war 
with Llewelyn in 1282-83. This, the third William, and the last of 
the Lords of Llanstephan who bore the name of Camville, ^i€»d in 
1338, and lefb five daughters — two of whom, Matilda and Eleanor^ 
are of interest to us in this connection, because their father was 
granted and given licence to enfeoff them with the Manor of Llan- 
stephan in 1337. Eleanor became the wife of Richard de Penres. 

Twenty-nine years passed, and in the year 1377 the lordship fell 
into the hands of the Crown by the forfeiture of Robert Penres, 
because that he, on Sunday next after the Invention of the Holy 
Cross, 1370, feloniously killed Joan, the daughter of William Ap LI', 
at Llanstephan, of which felony *' he was convicted on Monday 
after Holy Trinity, 1377." The wheels of justice moved somewhat 
slowly. This felonious Robert Penres, was evidently the son of 
Robert Penres, who immediately preceded him as Lord of Llan- 
stephan, who again appears to have been the son of Richard Penres, 
who married Eleanor de Camville, and through his marriage 
succeeded to the lordship. 

Upon the forfeiture of Robert Penres, Richard, Prince of Wales, 
granted the Castle and lordship to Simon de Burley. In the 
course of the following year, 1378, two other charters were granted 
by Richard to Simon, each giving him greater power and more 
extensive privileges. 

Simon, however, was not allowed to enjoy his possessions un- 
disturbed, for in 1379 — two years after the first gi*ant to him — he 
was subjected to proceedings in a writ quo loarranto, respecting his 
liberties in Llanstephan. In view of the charters I have already 
mentioned, Simon had a strong case — a complete answer to the 
enquiry. But the King, to remove all doubts respecting the rights 
of his favourite, granted new and sufficient letters- patent, confirm- 
ing the gift, in which are enumerated the reasons for the gift in 
very touching words. " In consideration,'' it is stated, " of the good 
service which the aforesaid Simon has done us, and the important 
position he held for us, from our tender age up to this, in all the 

Digitized by VjOOQ IC 


estates which we have held, to wit, before we received military 
orders, and likewise when we assumed that order, also afterwards 
before we were Prince of Wales, and daring our time as Prince, 
and further until made King bj Divine Grace, and as an envoy 
about our marriage, returning with the Queen-Consort from her 
land to this country, wherein we are informed that he had to sell 
all his lands in Suffolk," eta 

Richard was faithful to his favourites ; but, alas ! Simon Burley, 
with eight others, were impeached in 1388, and the ^'Merciless 
Parliament passed judgment against them. Simon Barley was 
condemned to be hanged, drawn, and quartered. Richard could 
not save him, but he did what he could; and the sentence was 
changed to one less disgraceful and more surely instantaneously 
fatal. Simon was beheaded. 

By Simon de Burley's forfeiture the lordship came into and 
remained in the hands of the Crown for three years ; and in the 
year. 1391, in consideration of the payment of 500 marks, the Castle 
and manor were demised to Robert de Penres, knight, son of 
the Robert de Penres by whose forfeiture they fell to the Crown in 
1377. Ten years after the forfeiture of Simon de Burley, that is 
in the year 1398, Roger, the son of Sir John de Barley and nephew 
of Simon de Burley, endeavoured to obtain possession of the Castle 
and lordship. The King addressed the Sheriff of Hereford in the 
following terms : *' We desire the Castle of Llanstephan ... to be 
restored to Roger de Burley, in accordance with the tenour and 
effect of the consideration, judgment, and statute made in our last 
parliament . . . and we command yon to cause Nicholas Clerk, 
Philip ap Cradock . . . tenants of the aforesaid Castle and lord- 
ship, as it is said, to appear before us in our Chancery . . ." upon 
which the Sheriff returned into the Chancery *' that . . . Nicholas 
Philip, Philip Cradock . . . still held by the feoffment of Robert 
Penres, the Castle and lordship aforesaid, enjoining with Eynon ap 
Jevan yet surviving, in the same writ not named." The cup of 
Richard's follies was now full, and he was deposed in the following 
year (1399), and the suit of Roger de Barley failed ; for we find in 
June, 1403, that John Penres, keeper of the Castle, was captured 
and detained by Owen Glyndwr, and that the custody of the Castle 
and lordship was granted to one David ap Howell, Armiger, to guard 
it and the adjacent county by placing in it ten men-at-arms and 
twenty bowmen. This was a liberal allowance of men to garrison 
the Castle, for by an ordinance made by the Prince's (Richard II) 
Council in 1369, the garrison of each of his castles in Wales and 
Chester consisted of one constable and twelve archers only. 

Letters of protection were granted to David ap Howell, and John 
Swetappul was appointed to provide food for the towns and Castles 
of Carmarthen, Kidwelly, and Llanstephan, and sustenance for the 
soldiers, *' and our liege people there;" so that Llanstephan Castle 
was garrisoned in the early part of the fifleenth century. 

John Penres, who in 14f03 became the prisoner of Owen Glyndwr, 
6tu ske., vol. vu. 8 

Digitized by VjOOQ IC 


appears to have obtained his liberty, for half the lordship of 
Llanstephan was granted to him in the year 1408, by reason of the 
forfeiture of Henry Owyn, and he was again keeper of the Castle at 
the time of his death in 1411. It was then seized 'tby Thomas 
Bede, who held it in his demesne as of fee of onr Sovereign lord 
the Prince." 

I know not what became of Thomas Rede, but the Castle appears 
to have been soon granted to one William Owyn and his son Henry, 
for in the year 1416 it came again into the hands of the Crown '' by 
the rebellion and forfeiture of William Owyn, fietther, a^ well as by 
the forfeiture of Henry Owyn, his son, who was slain at Aginoourt 
in the ranks of our adversaries." It was then granted to Humphrey, 
Duke of Olouoester. 

The Duke of Oloucester was childless, and in the year 1443 the 
reversion of the Castle and lordship was granted by Henry VI to 
William de la Pole, Earl of Suffolk, and Alice his wife, with the 
style and title of the Earl of Pembroke. 

Seven years later (1450), they came again into the hands of the 
Crown by virtue of an Act of Parliament. They were farmed by 
' — Nicholas, armiger. 

In the year 1453 the King became insane, and Parliament made 
a grant of *^ dower lands" to Queen Margaret, including among 
others the *' Castle, lordship, and town of Llanstephan, and all its 
appurtenances, forsooth, the lordship of Penryn and la verye, eta" 

During the wars between the Houses of York and Lancaster, Sir 
William Herbert, son of William ap Thomas of Baglan, who 
fought in France under Henry Y., and was made a knight-banneret, 
had proved himself a devoted adherent of the House of Tork ; and 
Edward lY., in the first year of his reign (1462), raised him to the 
rank of a baron for his good services against Henry Yl., Henry, 
Duke of Exeter, Jasper Tudor, and other rebels ; and granted to 
him the Castle, lordship, and town of Llanstephan, the lordship and 
manor of Penrhyn and la veire, etc. 

William Herbert died in 1469, and during the minority of his son, 
John Donne, one of the King's bodyguard, was appointed to tlie 
ofiice of seneschal of the Castle and lordship of Llanstephan, as 
well as to several other such offices in South Wales. 

In the year 1482, through an exchange of lands between the 
Prince of Wales (Edward Y.) and the Earl of Huntingdon, the Castle, 
lordship, and town of Llanstephan, the manor of Penrhyn, and la 
Yerye, and other properties were made parcel of the Duchy of 
Cornwall (Act 22 Edward lY). 

In the year 1484 Richard Williams, '* one of the keepers of our 
chamber," was appointed seneschal of our Lordship of Llanstephan, 
as well as of several others in South Wales, by Richard III. How 
long he held it is not recorded. 

The Act 22 Edward lY., making Llanstephan parcel of the 
Duchy of Cornwall, was annullecl in 1 495 ; and the lordship, together 
with others, reverted to Jasper, Duke of Bedford. 

Digitized by VjOOQ IC 

carmaHThen meeting. — fetepont. 115 

Law was unknown in March land. The Welsh laws oonld not be 
enforced, and the King of England's writ ran not there. The only 
principle of action recognised was " Trecha reisied gwana gwiohed." 
Might was the only right. Acquisition of territory bj the Lord 
Marcher remained unnoticed, or was encouraged of the King, 
nntil it had attained a magnitude such as to be a possible source of 
danger to the Crown. Robbery of the Welsh was a virtue, until it 
becam^ a danger to the English ruler. The King winked at the 
deedsoftheadventurer— or rather freebooter — Norman; while he at 
the same time kept a watchful eye on his steady and often rapid 
increase of power, and endeavoured to solve the question : *^ HoW 
shall I cnrb him ? " He favoured the Lord Marchers to fight the 
Welsh, annex their land, and wear out their resistance ; at the same 
time he feared their growing power, and endeavoured to control it. 
In the same way a custom grew up — the custom of the March. This 
custom depended upon two principles. One was, might is right. 
This was the principle upon which both king and barons were agreed 
in their treatment of the Welsh people. The Marchers held the same 
principle in their dealings with .the King ; but the King would 
none of it. Mr. Morris, in his work on the Welsh Wars of 
Edward I., defines the custom of the March in these words : ** The 
custom to fight and annex without restnction from the Crown of 
England, and to allow no appeal from the sub-tenants of fhe March 
to the King of England as overlord." In fact, the Lord Marcher 
was absolute lord of his March. His will was law in it. He held 
his lordship by his sword, and not by chai*ter of the King. Mr. 
Morris adds : ** One right was always enjoyed by the Crown. If a 
Lord Marcher lost his lands by a successful Welsh rising, and if 
the aid of the Royal forces was called in to reconquer it, the land 
thus reconquered reverted to the Crown.'* A continuous struggle 
was being carried on between the King and the Barons ; sometimes 
qaietly, and then it amounted to watchfulness and passive 
resistance; sometimes violently, then to aggression and conflict. 
These conflicts generally ended, not in victory for either party, bat 
in a compromise, and a compromise meant a restriction of the 
irresponsible power of both King and barons ; thus verifying an 
old adage, for by the conflicts between King and barons, the people 
had their rights enlarged and the bounds of freedom were 

The chief means by which the King was enabled to restrict the 
powers of the Lord Marchers was the successful raids made by the 
Welsh Princes into the Marchers' territory ; for the King coming to 
the aid of the Lord Marcher to recover his lost land, became 
possessed of the land by conquest, and the Marcher became the 
King's tenant. It appears, however, that there must have been 
other ways in which the King was able to acquire control over the 
Lord Marchers and their lands ; for we find that in the latter half of 
the twelfth century, Llanstephan Castle was held by charter from 
the King. This is about forty years only after the earliest 


Digitized by VjOOQ IC 


notice we have of the Castle ; and it does not appear from the 
records that the Welsh, when they had taken or homed the Castle, 
held it for any time. On the contrary, the Castle was bni*ned in 
the coarse of a snccessfal raid, and the Welsh, after haying 
completed the work of destruction, returned to their homes. It is 
not likely, therefore, that the Castle fell into the hands of the 
Crown by re-conquest, when it had been lost to the Lord Marcher ; 
and yet we find it held by charter at a very early period of its 

Professor Tout is of opinion that '^as early as 1256, Edward I., 
set up a rudimentary county organisation, in those southern and 
detached parts of the Principality where the power of Llewelyn ap 
Gruffudd was weak, and the traditions of the March recent. 
Carmarthen, which was in his hands, was the national seat of the 
county and the new offices." Mr. Morris states : '* Then he 
(Edward I.) pushed his influence southwards, and his design was to 
create two counties under Royal auspices in Cardigan and 
Carmarthenshire. For this purpose he partly strained the rights 
of the Crown to overlordship over South Wales, and partly he 
seemed to have claimed the right of conquest. Also at 
Carmarthen, Edward instituted a County Court or comttatus, 
to which neighbouring Lord Marchers wer^ constrained to do suit 
and service." The evidence for the view of Professor Tout and 
that of Mr. Morris is the fact that the lords of Llanstephan, 
Laugharne, and St. Clare did suit and service in Carmarthen for 
their holdings. Pain de Chaworth, Lord of Cydweli, was ordered to 
do likewise, but the order was afterwards withdrawn; while 
William de Braose, Lord of Gower (for Oower as well aa 
Llanstephan formed part of the county of Carmarthen at that early 
period) did not obey. 

It may reasonably be inferred from these fieicts, that Edward 
strained his overlordship over the Lord Marchers of South Wales, 
and that the smaller submitted to his wise and powerful tyranny, while 
the greater and stronger resisted successfully. I think it probable 
that the Kings before Edward practised the same tactics ; and the 
fact that a small March, such as Llanstephan, far away from the 
English border, was held by charter and service at a very early 
period inclines me to draw such an inference ; so that the 
organisation of counties by Edward is but another step in the 
process of curbing the irresponsible powers of the Lord Marchers ; 
a process which had been going on for nearly a century, for 
Geoffrey de Camville, in 1200, held Llanstephan by charter and 
service, and William, his father, and Albreda, his mother before 
him, had held it on the same terms. From this time on, control 
over the Lord Marchers seems to have steadily increased, for we find 
that — 

In 1276 Geoffrey de Camville and his bailiffs were ordered to 
prohibit his tenants to furnish provisions or supplies to the Welsh 

Digitized by VjOOQ IC 


In 1277 and 1282, he is gammoned to %lit in the war with 
Llewelyn. He obeys, and takes his qnota with him, which consisted 
of two knights and twelve lances : in all fifteen lances. In 1287 
Oeofirey is enjoined to reside on his own demesne and lordship 
until the rebellion of Rhys ap Meredith is pat down. In 1316 the 
Prior of Carmarthen, his men and tenants, are commanded to 
receive their measures, scales and weights from the King's minister 
of the new town of Carmarthen, the keeper of the King's measares, 
scales and weights, just as the Barons of Llanstephan, Talacharn, 
and St Clare do. This shows us that the bearer of the sword is 
being made more and more subject to the jurisdiction of the county, 
and that the wearer of the mitre will have in this respect to keep 
him company. 

In the year 1324, Roger Mortimer escaped to France, and 
together with Isabella — Edward's (II) Qaeen — fomented war against 
England. This appears to have produced a state resembling panic 
in the English Court, if we may judge from the following orders, 
issaed to the Lords and Bailiffs of Llanstephan and others. 

In 1324, an order to cause all ships capable of carrying forty tons 
and upwards to be arrested and equipped for the King's service. 

In April, 1325, and again in May, a proclamation in favour of the 
men of Flanders. 

In December, 1325, an order to cause all ships entering Llan- 
stephan and other ports, or wishing to leave the same for parts 
beyond the sea, to be searched, and to arrest any whom they shall 
find with letters prejudicial to the King, etc. 

In the month of January, 1326, a similar order, but extending 
to horses, arms, gold and silver taken out of the country by any 
except merchants. 

In August of the same year, an order to cause all owners of ships 
of the burthen of fifty tons and upwards to repair to Portsmouth, 
on Sunday after the Decollation of St. John Baptist next, with their 
ships found with arms, victuals, and other necessaries, with double 
equipment, to set out in the King's service against the French. A 
farther order to all owners of ships of smaller burthen than fifty 
tons, not to leave port for any purpose, under pain of being taken 
and imprisoned. 

In 1328, an order enjoining all owners and masters of ships to 
cause all their ships of less than forty tons burthen that are out- 
side their port, to be brought back to the port, lest the malefactors 
from Normandy and Poitou take them. They are to certify to the 
King as soon as possible of the number of the ships and their mem- 
bers, and the harden of their ships. 

(Verily, the Lord Marchers have become submissive !) 

Lastly, in 1361, an order not to admit any earls, barons, knights, 
or men-at-arms, to cross to parts beyond the sea, or to take horse 
or arms without the King's special license. 

This order was issued not without reason, for some years after- 
wards, Thomas and John Fort, of Llanstephan, assumed the Eoyal 

Digitized by VjOOQ IC 


jurisdicfcion, and granted a safecondnct to one John de Ispania, a 
subject of the King of Castile, and an enemy of the King of 
England, and showed him the secrets of eleven castles in South 
Wales — are pardoned. 

One year later, John Fort was again pardoned for scaling the 
walls of Langharne Castle, and robbing Gay de Brian of £25 in 
gold and silver. The Forts clearly had friends at Coart 

Such is a short sketch of the story of Llanstephan Castle, up to 
the time that Henry Tudor became king. Then it was an imposing 
and frowning fortress, overlooking the Bay of Carmarthen as far as 
Tenby, Gower, and the coast of Devon — the home of soldiers, the 
scene of armed conflicts ; now a picturesque ruin, the haunt of 
toarists, holiday-makers, and lovers. 

At the conclusion of the Address a vote of thanks was proposed 
by General Sir James Hills-Johnes, seconded by Mr. A. Stepney- 
Gulstou, and briefly acknowledged by the President. 

TUESDAY, AUGUST 14th, 1906. 

A Public Meeting was held in the Assembly Rooms at 8.30 p.m. 

In the absence of Professor J. E. Lloyd, of Bangor, his Paper ou 

** Carmarthen in Olden Times " was read by the Rev. C. Chidlow, 
General Secretary for South Wales. 

Archdeacon Thomas remarked that the Paper helped very largely 
to elucidate the early history of this town. 

Mr. Lleufer Thomas being called upon, said he did not feel com- 
petent to offer any criticism of Professor Lloyd's exhaustive history 
of that period. He was particalarly gratified with what was to 
him a new solution of the difficalty with reference to Llanteilyddog, 
his identification of it, and the manner he had worked it out. 

They must congi*<itulate themselves upon having that new con- 
tribution to the history of Carmarthen. 

The Rev. Griffith Thomas asked whether the burgesses of Car- 
marthen paid homage to William I. It was a moot point, because 
Wales did not submit to the Crown of England for some two 
hundred years after. 

Professor Anwyl said he was very much in the dark as regarded 
this period, but so far as he could follow Professor Lloyd's Paper, 
he rather gathered the borough of Carmarthen, as a borough, did 
not exist so early as that, so the burgesses could not have sworn 
fealty as burgesses at all. What he understood from the Paper was 
that there was an ecclesiastical establishment of the Welsh kind, 
the old British kind, a close, as it was called, in connection with 
the Church dedicated to St. Teilyddog ; and in course of time the 
Castle came to be built of the stockaded type, and at that time the 
Castle was at Rhydygors, and not in the present Carmarthen. 

Archdeacon Thomas thought the existence of Rhydygors was of 
a very short daratioi), and it was afterwards that the military centra 

Digitized by VjOOQ IC 


was transferred to the new Carmarthen. That would hardly admit 
of the burgesses having sworn allegiance to William I. 

Rev. Griffith Thomas : I think he mentioned it was not given a 
charter till the reign of King John, but it is shown there were 
privileges to the borough previous to the reign of King John. 

Professor Anwyl said there might have been privileges enjoyed 
there without a charter, which were such that settlers gathered 
at the place. 

Mr. T. E. Brigstocke said apparently the Roman sefctlemenfc was 
identified with the old city of Carmarthen — the ecclesiastical city — 
and there was some difficulty in his mind : how did he identify the 
fortified situation of the present Castle and the Roman settlement ? 
All the discoveries they had come across, and the coins, had all been 
found apparently in the old city of Carmarthen, lying eastward to 
the walled town of the new Carmarthen — the old Carmarthen site 
of the Romans — and the Castle formed by the Normans was a 
departure from the old settlement of the Romans. 

Archdeacon Thomas said GKraldus Cambrensis wrote *' that 
ancient city is beautifully enclosed with walls of red brick.'' He 
evidently looked upon Carmarthen itself as having evidence still 
existing of Roman antiquity. 

The Rev. W. Done Bushell, of Caldey, was then called upon to 
read his Paper on '* Neolithic Monuments.'' 

Archdeacon Thomas said the subject was extremely abstruse, and 
perhaps to the young members of the audience one of great difficulty. 
But still, one could not listen to the able and clear, argumentative, 
and eloquent address of Mr. Done Bushell without feeling that a 
new light had been thrown upon an ancient monument close at 
hand. He had read now and then of theories about Stonehengc 
and other places, but he had never listened to anything, or read 
anything, that appeared to him so clear and so satisfactory as the 
address they had had the pleasure of listening to. 

This closed the evening's proceedings. 

WEDNESDAY, AUGUST 15th, 1906. 
There was no Evening Meeting on this day. 

THURSDAY, AUGUST 16th, 1906. 

The Annual Business Meeting of the Association was held in the 
Assembly Rooms, at 8.30 p.m. 

The President, Sir John Williams, Bart, took the Chair ; and 
after the Minutes of the previous meeting had been read and con- 
firmed, he requested the Rev. Canon R Trevor Owen, Senior 
General Secretary, to read 

The Annual Report op the Association. 
The Journal. — The following Papers have been published in the 
Archaologia Cambrensis between July, 1905, and July, 1906 : — 

Digitized by VjOOQ IC 


Prehistoric Period, 
"Oq the Discovery of Prehistoric Hearths in South Wales." By T. C. 

Cantrill and 0. T. Jones. 
" The Early Settlers of Cardigan." By Professor E. Auwyl. 
"The Exploration of Pen-y-Qaer above Llanbedr-y-Cenin. " By Harold 

" On the Defences of Pen-y-Qaer." By Willoughby Gardiner. 
" Peny-Gorddyn, or Y Qorddyn Fawr." By Harold Hughes. 

Late-Celtic Period. 
No Papers. 

RomoAM-Brttisk Period, 
" Roman Remains : Pen-y- Darren Park, Merthyr Tydfil." By F. T. James. 
" The Town of Holt, in the County of Denbigh.** By A. N. Palmer. 
**The Ordovices and Ancient Powys." Hy Archdeacon D. R. Thomas. 

Early Christian Period, 
" The Llandecwyn Inscribed Stone." By Professor E. Anwyl. 

Mediaeval Period, 

" A History of the Old Parish of Gresford, in the Counties of Denbigh and 
Flint." By A. N. Palmer. 

" Some Notes on Mediaeval Eifionydd." By Professor J. E. Lloyd. 

'• The Vairdre Book." By Dr. Henry Owen. 

" Allen's Pembrokeshh-e." By E. Laws. 

** Welsh Wooden Spoons, with Ornamental Carving and liOve-Symbols. " By 
J. R. Allen. 

"On Some Sacramental Vessels of Earthenware and of Wood." By Arch- 
deacon D. R. Thomas. 

" The House of Scotsborough, near Tenby." By E. Laws. 

" The Religious and Social Life of Former Days in the Vale of Clwyd." By 
the Rev. J. Fisher. 

" Old Radnor Church." By Ernest Hartland. 

The following books have been received for review : — 

" Edward II in Glamorgan." By the Rev. John Griffith. (Cardiff : Western 

Mail, Limited. ) 
" Lampeter." By the Rev. George Eyre Evans. (Aberystwyth : William 

Jones. ) 
" Owen's Pembrokeshire," Part III. By Dr. Henry Owen. (Bedford Frees.) 
" A Digest of the Parish Registers of Llandaff." By the Committee of the 

Llaudaff Diocesan Conference. (Cardiff: Wm. Lewis.) 
" Welsh Abbeys." By John A Randolph. (Carmarthen : William Spurrell 

and Son.) 
" Castell Morgraig." By John Ward, John W. Rodger and John Stuart 

Corbett. (Cardiff : Wm. Lewis.) 

The Special Illastration Fand has been made ase of by the 
editor to obtain photographs of the old house at Scotsborough, near 
Tenby, for Mr. E. Law's paper on the subject of the Norman 
Fonts at Lamphey and at Bed berth in Pembrokeshire. 

The thanks of the Association are due to those authors who have 
supplied drawings and photographs to illustrate their papers in the 
Journal. Amongst these are Mr. A. N. Palmer, Mr. Harold 
Hughes, Mr. Willoughby Gardner, Mr. F. T. James, and Archdeacon 
D. R. Thon^as. The photographs of the sculptural panels on the 

Digitized by VjOOQ IC 


Norman lavatory at Wenlock Priory were taken by Mr. H. E. 
Forrest at the request of Miss Anden. 

The Official Set of the Archcedogia Cambrenna, — This is now 
complete with the exception of the following three volumes : — 

Sen 1, Vol. 4, 1849. 

Ser. 3, Vol. 1, 1855. 

Vol. 2, 1866. 

Progress of Welsh Archaeology in 1905-6. — The Committee begs to 
call the attention of the members to the following matters of 
importance, which have come under the notice of the officers of the 
Association during the past year. 

No specially interesting accidental finds of antiquities appear to 
have been made in Wales during the last twelve months, or if they 
have been made they have not been reported by the Local Secretaries. 

Some of the recently- formed local archoeological societies have 
done good service in excavating prehistoric defensive earthworks, 
and sending the results to the editor of the Journal for publication. 
As instances of this may be mentioned, the explorations made at 
Pen-y-Qaer (above Llanbedr-y-Cenin) and at Pen-y-Gorddyn by the 
Nant Conway Society and described by Mr. Harold Hughes in the 
July number of the Archceologia Cambrensis. Much good might be 
done in the future by friendly co-operation between the Cambrian 
Arche&ological Association and the local Antiquarian societies and 
field clubs. The Association might grant funds for the examination 
by the spade of promising ancient sites on the understanding that 
the work shall be scientifically carried out by the local societies 
under proper supervision. The results could afterwards be described 
and illustrated in the Archceologia Cambrensis^ and such reprints as 
might be required could be supplied to the members of the local 

It is gratifying to find that after a long interval the Romano- 
British occupation of Wales is again attracting the attention of 
contributors to the Journal^ as is shown by the valuable papers in 
the July number on the Roman Remains at Merthjr Tydvil by Mr. 
F. T. James, and at Holt, near Wrexham, by Mr. A. N. Palmer. 
Nevertheless, it is greatly to be regretted that it is still necessary 
to appeal to England for an expert opinion on the antiquities of the 
Romano-British period. Surely there should be some classical 
scholar of one of the Welsh Universities who could train himself to 
occupy the same position as an expert in this period in Wales as 
Dr. F. Haverfield, F.S.A., or Mr. G. B. Fox, F.S.A., now take in 

Further discoveries of importance have been made by special 
explorations, at the expense of the Association at TreV Ceiri, 
Carnarvonshire and Y-(Jaer, near Colbren, Glamorganshire, of 
which reports are laid before the members. 

Digitized by VjOOQ IC 



Election of Officers and New Members, — The Committee propose 
that the Right Reverend the Lord Bishop of Llandaff and the 
Right Hon. Lord Harlech be enrolled amon^ the Pati*ODS of the 
Association; and that Robert Cochrane, Esq., F.S.A., I.S.O., 
Honorary Secretary of the Royal Society of Antiquaries of Ireland, 
and Colonel Morgan, R.E., be elected Vice-Presidents; also that 
the thanks of the Association be presented to the Venerable 
Archdeacon Thomas for his services to the Association as its 
President during the past year. The retiring members of the 
Committee are : — 

T. M. Franklen, Esq., 

The Rev. John Fisher, B.D., 

The Rev. E. J. Newell, M.A., 

and the Committee recommend their re-election. 

They farther propose the following members as Local Secretaries 
for their respective counties : — 

The Rev. C. F. Roberts, M.A., Rectory, Uanddulas, for Denbighshire. 
R. Jones Morris, Enq., Tycerrig, Talsarnau, for Merionethshire. 
J. Bancroft Willans, Esq., Dolforgan, Kerry, for Montgomeryshire. 

The following are proposed for membership : — 

Mrs. Gordon, 9, St. German's Blaoklieath 
Mr. Price, 48, Pall Mall, S.W. 
Lieut. -Col. Q. Tucker Thomas, I. M.S., The Bush, 

North Wales. 

Miss Hampton Lewis, Henllys, Beaumaris 


Mr. WiUoughby Gardner, F.L.S,. F.R.G.S. 


Mr. Dodd .... 


Mr. S. H. Harrison, F.R.G.S., F.R.S.A. 


Mr. W. I. P. Story, Rhyl . 

The Marches. 

The Right Hon. Lord Harlecli 

Proposed by 
Mr. Pepyat Evans. 

Mr. Edward Owen. 

The Rev. J. G. Swainson 

. Mr. J. E. Griffith. 

. Canon Trevor Owen. 

Canon Trevor Owen. 

. Yen. the Archdeacon of 

. L. L Roberto, H.M.LS. 

H.M. Lieutenant of Meri- 
onethshire, W. R M. 
Wynne, Eaij., Peniarth. 
. Canon Trevor Owen. 

South Walks, 

Davies, Mr. John, Bridge Street, Lampeter . Rev. G. E. Evans. 

Ellis, Thomas, Esq., Glascoed, Aberystwyth . Rev. G. E. Evans. 

I^wes, Miss Evelyn, Tyglyn-Aeron, Ciliau Aeron Rev. G. E. Evans. 

Phillips, Rev. Thomas, Rectory, Aberporth . Rev. D. D. Evans. 

Recs, Rev. R. J., M.A., Rhos, Aberystwyth . Mr. D. Samuel. 

Digitized by VjOOQ IC 



South Walks. 

Proposed by 
Rev. C. Chidlow. 

Mr. W. Spurrell. 
Rev. C. Chidlow. 
Mr. J. F. Hughes. 
Rev. C. Chidlow. 
Mr. C. H. Glascodine. 
Profe«i8or Anwyl. 
Mr. T. E. Morris. 

Mr. Edmund Jones. 

Rev. C. Chidlow. 
Mr. J. E. Kichards. 
Mr. H. W. WUUams. 
Mr. Edmund Jones. 
Mr. Pepyat W. Evans. 
Rev. C. Chidlow. 
Mr. J. E. Richards. 
Mr. Pepyat W. Evans. 

Professor Rhys. 

Rev. C. Chidlow. 
Rev. C. Chidlow. 

Ca rmarthenshire. 

Collier, Ernest, Esq., M.S. A, Carmarthen 
Da vies, A. Llewelyn, Esq., Brynderw, Carmar- 
then .... 
Gabriel, J. R., M. A, Technical College, Swindon . 
James, Daniel, Esq., Vrondeg, Llandeilo 
Jones, Arnallt, Elsq , M. D. , Carmarthen 
Ludford, T. li., Esq., Llanelly 
Thomas, Mrs. R. M., Llanddowror 
Thomas, Rev. 0. J., Llandyssilio Vicarage 
Williams, Mrs. W. J., 91, Picton Terrace, Car- 
martheu .... 


Llandaff, The Lord Bishop of. The Palace, Llan- 

daff .... 

Davies, D. Jones, Esq., Rugby Road, Neath 
George, Isaac, Esq., The Grove, Mountain Ash . 
Gibbins, F. W., Esq., Garthmor, Neath 
Gordon, Mrs., Nottage Court, Porthcawl 
Jenkinb, Mrs., Gtellystone, Llandaff 
Jones, Rhys, Esq., Godrecoed, Neath . 
Phillips, Rev. T. C, Vicarage. Skewen 
Williams, Arthur J., Esq., Plfts Coed-y-mwstwr, 

Bridgend .... 


Lewis, Rev. J., Lampeter Velfrey Rectory 
Phillips, Rev. John, Uzmaston Rectory 

Tre*r Ceibl 

At the meeting of the Committee of the Cambrian Arohsdological 
Association, held at Shrewsbury, on August 14th, 1905, it was pro- 
posed by Mr. T. E. Morris, seconded by Mr. A E. Bowen, and 
carried, "That Professor Boyd Dawkins be asked if he would 
kindly consent to the excavations at Tre'r Ceiri being carried out 
under his direction, with the assistance of Colonel Morgan and Mr. 
Harold Hughes." 

Professor Boyd Dawkins kindly consented to undertake the work. 
Through the assistance of Mr. D. R. Daniel, of Fonrcrosses, eight 
labourers were obtained and work commenced on June 5th, 1906, 
and continued till June 16th. Unfortunately, Professor Boyd 
Dawkins was called to London before the excavations were com- 
pleted, and Colonel Morgan was nnable to be present during the 
whole fortnight. On the other hand, very valuable assistance was 
given by Mr. C. B. Breese, who devoted several days to the work. 
I was present throughout the whole time occupied by the works of 
excavation. A full and detailed report of the result of the fort- 
night's work is in course of preparation, and will be laid before the 
Cambrian Archeeological Association. 

The following, however, is a brief summary of the work, together 
with Professor Boyd Dawkins*s view as to the position Tre*r Ceiri 
occupies in relation to the history of Wales. 

Digitized by VjOOQ IC 


In an introdaobory noto to the report, Professor Boyd Dawkins 
writes : — " It is one of many similar villages, oconpying a com- 
manding position for parposes of defence, in the neighbonrhood . . . 
containing the remains of rude stone huts, called by the inhabitants 
of the district * cy ttian gwyddelod ' . . . the huts of the Goidels. 
This popular attribution of the Goidels, the conquerors of the 
original Iberic Welsh, who in their turn had to submit to the 
mastery of the Brythons, is in my opinion true. They are pro- 
bably the dwellings of the Welsh Prehistoric Goidels, and have no 
necessary connection with the Irish Goidels, who were undoubtedly 
in close touch with this as well as with other districts in Wales in 
the historic period." . . . "This class of fort is proved, by the 
remains found in various places, to have been occupied at vnrious 
periods, ranging from the Bronze Age to the Prehistoric Iron Age, 
and well into the Historic period. The bronze sickle found in Dun 
Aengus proves that it was used in the Bronze Age ; while bronze 
pins with ornamentation of the Prehistoric Iron Age indicate that 
it was occupied at that time, and a bronze ring with cable decoration 
that it way not without inhabitants in the fifth century after Christ." 
After referring to the cashel on Inismurray, Professor Boyd Dawkins 
proceeds : — *' This class of fort in England is clearly proved, by the 
results of the exploration of Worlebnry, to belong to the Prehistoric 
Iron Age. Here the inhabitants belonged to the aboriginal Iberic 
stock, the ancestors of the Silures of the north side of the Bristol 
Channel. Equally good evidence is presented by the brooch, found 
in the excavations of 1903 at Tre'r Ceiri, that it also belongs to the 
Prehistoric Iron Age. It may, however, have been — and probably 
was — used in later times by the Goidels of the district, whenever 
the country was being harried for purposes of defence.'' 

Between June t5th and 16th, thirty-five sites were excavated. 
The finds include : — 

Three portions of a bronze gold-plated beaded ornament (torque 

or armlet). 
A bronze pin, with indications of gold plating. 
Fragment of a thin plate of copper or bronze. 
A pewter object, bearing in shape a resemblance to the pommel 

of a sword-hilt. 
Remains of a lump of lead. 
A blue porcelain bead. 

An iron object (the shape of a ladle in outline). 
Remains of iron (spear- head P). 
Many small fragments of iron. 
Iron object, shape of heel-plate (? modern). 
Remains of large earthenware vessel, interior finished with coarse 

gravel (quartz), probably for grinding (Mortarium). 
Many fragments of black pottery. 
Fragments of red pottery. 
Hone stone. 

Digitized by VjOOQ IC 

CARMA&THEN meeting. — REPORT. 125 

Pounders and mbbers. 

Bebblee (sling-stones and pot-boilers). 

Witb reference to the most important finds, Professor Boyd 
Dawkins writes: — "The bronze object (beaded) is distinctly of 
Prehistoric Iron Age, and is probably a torque or armlet as you 
suggest. The glass or porcelain bead — some of these found at 
Glastonbury —in the Prehistoric Iron Lake Village. The (pommel ?) 
is not very far removed from one discovered in Prehistoric Iron Age 
fort at Hod.'' 

The black pottery and the iron ladle-shaped object, he write?, 
would belong to the same period. 

Place of Meeting for 1907.— -The Committee recommended that 
Llangefni, Anglesey, should be chosen as the place of meeting for 

The adoption of the Report was proposed, seconded, and carried 

FRIDAY, AUGUST 17th, 1906. 

A Public Meeting was held at the Assembly Elooms, at 8.30 p.m. 
The Chair being taken by the President, Sir John Williams, Bart., 
he called upon Professor E. Anwyl, of Aberystwyth, to read his 
paper on *' Early Settlers of Carmarthenshire.'* 

Sir John Williams said he felt deeply indebted to Professor 
Anwyl for the excellent paper they had just heard. It was a most 
interesting and valuable description of their ancestors, ages ago that 
they could not count, and of their companions, gentle and ungentle, 
tame and savage, and of the manner of their lives when Carmarthen 
Bay was not yet. 

Mr. E. Laws said, as he was the oldest cave-digger present, he 
bad to thank their friend for his excellent paper. He thought they 
must remember the oldest of their cave-diggers in this part of the 
world, Gilbert Smith, rector of Gumfreyston, who began to dig there 
before the value of cave-digging was actually recognised. He was 
a clergyman, and as he dug he was actually afraid of what he was 
doing. It partially shattei'ed his faith, and still the plucky old fellow 
went on at it ; he dug in fear and trembling, and he (Mr. Laws), was 
his pupil. In those days Mr. Dawkins was down there a good deal, 
and what they found had been pretty well explained. He thought 
wolves very scarce at Hoyle, but they found them in considerable 
numbers on Caldey. The commonest beast was the horse, and most 
puzzling, the hippopotamus. They must remember if he got caught 
in the winter he was done. He lived at the bottom of the river, 
and could not exist in freezing rivers ; and as the rivers here were 
freezing, he must have come only for a short time. He did come 
because he (Mr. Laws) had dug him up himself. Palasoiithic Man 

Digitized by VjOOQIC 


was about tho scarcest beast. He (Mr. Laws) never found him in 
Hoyle, and he did not think anybody else had, but they did find sh'ght 
traces of him on Caldey. As a matter of religion, Neolithic Man pat 
his dead away with what they wanted for daily life, and he seemed to 
have believed the dead man continued to live in the place where his 
body was put. The Brass Man had totally different notions, some- 
thing like onr own. He thought there was another world, somehow 
or other. He bamed his dead, and made a ghost of him. He broke 
his axe and everything he had, and threw it all away, and so made a 
ghost of everything that they should go with the dead man. He 
believed there was a totally different idea between the Stone Ag^ 
Man and the Brass Man. Mr. Laws concluded by a description of 
the cave near Penally. 

Mr. Stepney-Gulston, Chairman of the Local Committee, thanked 
Professor Anwyl for the very kind remarks he had made with 
regard to the newly- born Association, the Carmarthenshire 
Antiquarian Society. They were all beginners, and they were more 
than pleased and gratified to be taken in hand and encouraged by 
those who had worked in the direction of archeology for. many 
years. They came forward to give them encouragement and 
enthusiasm. While regretting they were losing the energetic 
Secretary, Rev. M. H. Jones, be hoped in his new sphere he would 
be of the great value he had been to them. 

Mr. W. Spurrell then read a paper by Professor Sayce upon 
"What can be done for Archaeology in Wales." The writer 
referred to the method of excavation in Egypt, and stated only 
properly qualified persons should be allowed to excavate. In this 
country it was left to chance and hazard. Two objects should 
be aimed at : a thoroughly working survey of Wales and 
border counties, and the training of as many as possible of the 
younger members of the community in habits of careful and 
accurate observation, and in methods of modem archaeological 
science. The soil of these islands was full of relics. ExcavatioTi by 
untrained amateurs did more harm than good, and he suggested 
that members of the Archaeological Association should start 
investigating in a small area round where they lived. 

Sir John Williams said the paper was full of suggestions, and made 
an appeal to the Cambrian Archaeological Association to carry out this 
work. That was the meaning of the paper, and two things were 
required : men to do the work, and money to help them to do it. 
That was the thing they were apt to forget. He asked what was 
being done in regard to Pembrokeshire P 

Mr. E. Laws — The Archaeological Survey is not going to stop ; 
it is earned on by one man. 

Sir John Williams said he was sorry and he was glad, but he 
thought it was stopped because he had never been asked for his 
promised subscription. 

Digitized by VjOOQ IC 

CARMARtHilN MBEtlNG. — ttBPOJlT. 12? 

Mr. £. Laws — I gave up snryeyin^ because I felt I was pfetting so 
old, and was afraid I could not carry it od. My friend, Dr. Henry 
Owen, has taken it up, and is carrying it on at his own cost, and is 
going to finish it at his own cost. 

Professor Anwyl proposed a vote of thanks to the Chairman and 
Local Committee, remarking that they owed a deep debt of gratitude 
to them for organising these meetings so well. « 

Mr. E. Laws seconded. He thought it had never struck many of 
them, when they came there and found carriages ready, and 
luncheons and everything, what a lot of trouble it had given to 
somebody. He had gone through the mill himself, and knew, but 
some of them had no conception of what a bother it was. When 
done, there was apparently no difficulty about it, but he could 
assure them it was a troublesome thing. Thanks were due to the 
Chairman and Committee who had carried out the programme so 

The motion was heartily carried. 

Mr. Stepney-Gulston, in responding, said the Secretaries^ Rev. M. 
H. Jones and Mr. Spurrell, and the Committee connected with him, 
had worked like horses, and the difficulties which had been referred 
to had been a matter of pleasure to them. Everyone had been 
proud to have an opportunity of putting his shoulder to the wheel 
as representing their new-bom Society, and also as representing 
the town and county. It had been a great pleasure to them to 
welcome the Cambrian Archaeological Association there. 

Mr. Meuric Lloyd proposed a vote of thanks to all the proprietors 
of grounds and curios, who had allowed them such ready access upon 
their various expeditions to view them. They had been rather an 
invasion, he fancied. Their numbers had been far greater than 
personally he had seen on these occasions, and he thought perhaps 
they might have done a certain amount of damage. Possibly it 
might be some satisfaction to those who had allowed them to go to 
think they had afforded the members a good deal of pleasure. 

Mr. T. E. Morris seconded the vote of thanks, which was carried. 

Canon Rupert Morris proposed a vote of thanks to the Local Secre- 
taries, who had done such excellent work in preparing for those 
meetings. He had had some experience in managing and preparing 
for the annual meeting thirty-one years ago, when he was one of the 
Local Secretaries with Captain Philipps for Carmarthen; and he knew 
what a long time they took in going over every part of the ground, 
and in writing letters and making the various arrangements, espe- 
cially about the carriages. The work had been done very well indeed 
by Mr. Jones and Mr. Walter Spurrell. Mr. Jones bad the true 
archasological spirit, and in Mr. Spurrell they had the son of Mr. 
Spnrrell, the former publisher of the Haul, who had written an 

Digitized by VjOOQ IC 


excellent history of Carmarthen. He had, with his oolleagae, con- 
ducted the work extremely well, with considerable firmness, bat 
with proper sympathy and tact thronghont. With respect to the 
Goygan Cave, he (Canon Morris) explored some of it with Dr. 
Hcarder, the late Superintendent of the Asylam, and they fonnd in 
addition to what had been mentioned, the woolly mammoth. He 
had some large teeth at home, deer bones, etc. They lost them- 
selves in the cave, and there was some diffionlty in getting out of it. 
The next time they took some thread to enable them to find their 

Canon Trevor Owen seconded, observing he knew how very im- 
portant it was to have good local secretaries, because a great deal 
depends npon them whether the meeting was a success or not. 

The motion having been carried with applause, Mr. W. Spurrell 
responded, stating it really had meant a great deal of work, hot 
considering the number of members this time he must say on 
the whole they behaved very well indeed. They had a heavy pro- 
gramme, and it had been a g^reat gratification to him that it had 
not been found necessary to leave out any part except the very 
last item that day. 

The proceedings then terminated. 

Note. — In the compilation of this report we have largely made 
use of the reports of the meetings given in The Welshman. 

Digitized by VjOOQ IC 

APRIL, 1907. 


By Colonel W. LL. MORGAN, R.E. 

At the meeting of the Cambrian Archaeological Associa- 
tion, held at Shrewsbury in the autumn of 1904, I was 
asked to conduct some excavations at the camp at 
Coelbren, to ascertain, if possible, the approximate 
date of its construction. 

This camp is a conspicuous object from the Swansea 
and Brecon line, about half a mile to the right, imme- 
diately after leaving Coelbren Station. It was well 
known to Mr. Glascodine and myself, and on our 
frequent walks on the Sam Helen we had traversed 
that road throughout its whole length. Some doubtful 
places we had visited three or four times, thereby 
rectifying several errors on the Ordnance Survey map 
of the road. 

The history of the Sarn Helen is so inseparably 
connected with that of its camp, that I have commenced 
with a description of the former. 

The road known as Sarn Helen is undoubtedly of 
Roman construction. It connected the Nidum of the 
twelfth Iter Antoninus (which is supposed to be Neath) 
with Bannuum, afterwards known as Caervan, and now 
as the Gaer, near Brecon, on its course to Chester; but 
whether this poition was made on the line of an earlier 

6th skb., vol. vn. 9 

Digitized by VjOOQ IC 


British trackway is more than doubtful. It certainly 
does not run as straight as Roman roads in general, 
and it is not an unreasonable idea th^t in pre-Roman 
times an original British track from the centre of Brit<iin 
to the sea coast might have taken this line ; but though 
older authorities incline to this theory, in view of the 
nature of the country I consider it to have been 
entirely the work of the Romans. It was probably 
much used up to Norman times ; then, for several cen- 
turies it fell into disuse, for the reason that there was 
little intercourse between the Normans of Glamorgan 
and those of Breconshire : the latter being connected 
with their kinsmen in Gower, and their line of com- 
munication came down the opposite side of the Swansea 
valley by Llanguicke Church. In later times, and up 
to the present day, different stretches of the road have 
been in use to connect the several farms, but very little 
of it has become a main road. 

The site of the Roman station of Nidum is doubtful. 
From the similarity of names, it has been taken for 
granted that it must be Neath, which probably, then as 
now, was at the head of the tidal water of the river. 
Three Roman roads seem to converge on this site, but 
they cannot be traced within two miles of the present 
town of Neath, nor have any Roman remains been 
found within this area. 

Though there are no traces of the road now existing, 
probably (as mentioned by Jones) after leaving Neath 
it was carried across the marsh, and then kept to the 
high ground to the back of Ynisgeryn, where it ascen- 
ded the hill. The line as given on the Ordnance Map 
is conjectural until it reaches Llettyrafel, where the old 
ro^d can be seen for a short distance, where it is again 
lost in a ploughed field. It then runs on the line of 
the present road to the Downs, and though much altered 
in modem times, it still retains the unmistakable 
appearance of Roman origin. The pitched paving can 
be at once distinguished on ascending the Downs ; 
though in many places it is covered with accumulations 

Digitized by VjOOQ IC 


. MsM^ Maooc 


w WAy^^^y^^yi> 

1% y^ys-y^cw^wYN 

Fig. 1. — Map showing position of Camp at Coelbren. 

Digitized by VjOOQ IC 


of soil, it shows for many miles at intervals of a few 

In places the curbs are still to be found. The 
roadway is generally slightly raised above the ordinary 
ground level, and in one spot there is a cutting in the 
hill. The roadway then follows the crest of the hill 
(the Hirfynydd), and leaves Waunglynnyd on the right. 
The road then makes a bend to the right, after which 
it runs straight over the Downs, until opposite to 
Blaennant, where it takes up the line of the stone wall. 
Half a mile further the paving is in remarkable preser- 
vation, though the general character has been destroyed 
by scratching the ground for the stones. In many 
places the wall is actually built on the line of the road. 

Cam CornaP (a fairly large cam) stands half a mile 
to the left. The road runs alongside the wall (in some 
places well defined, in others it is obliterated) before it 
makes a sharp turn to the right, near the plantation of 
fir trees. 

The wall is now thirty yards to the right of the 
road, which runs along a cutting, which I think is part 
of the original scheme. It again turns to the right, 
round the shoulder of the hill, but whether it be 
rejoined by the wall is uncertain, as no decisive trace 
can be seen until another turn to the left at Clawdd y 
Fan wen, where both foundation and pitching are once 
more apparent. Its course is now broken up by old 
coal levels, through here and there the line can yet be 
traced, till it descends the hill opposite to Tynyrheol. 
The pitching there is in good preservation, and the 
curbs are in sight in various places. The line then runs 
opposite the old tramway until the railway crossing at 
Tafarnybenwen Common, where it has been covered by 
the present road, which deviates again at Llwynpica, 
from whence the pitching of the old road can be 
traced to the southern entrance of the camp, commonly 

^ The Boman sfone removed to the Gnoll, Neath, stood somewhere 
near here. It is mentioned in Q-ough*8 Camden^ p. 473, vol. ii ; also 
in Arch, Camb., 3rd Ser., vol. xi, p. 59, and 5th Ser., vol. xi, p. 338. 

Digitized by VjOOQ IC 


called the Gaer, from whence it emerges on the eastern 
side, and is traceable down to the river, where it is lost 
for about twenty yards, beyond which it is again visible 
near the hedge. It crosses an old lane, and can be 
followed across a field. Further on, a wall has been 
built on the line of the road, which has obliterated all 
trace along several fields. We find it again in thfe open 
ground beyond, the pavement and euro both in good 
condition. Ton-y-ffildre, or as it is locally called, Ton- 
y-ynwl-y-dref (the town of the soldiers), or the " flat- 
land at the edge of the town," is to the left of the road, 
which here is 21 ft. between the earthworks, and 14 ft. 
between the curbs. 

The line across the morass is very distinct, and traces 
of pavement appear wherever the water has washed 
away the accumulation of soil. It is visible in the 
hollow where it passes Cefngwenynawg, and alongside 
the hedge, though there it has been mended and made 
up with modern materials. Crossing a small brook 
(Nanthir), it can be followed over the mountain in a 
well-defined line, until crossing another brook it 
descends the hill to Gwaunymaerdy, to avoid a steep 
ravine which intersects the direct line. It now turns 
sharp to the left, through a well-marked cutting (which, 
like the former one, I think is original). On the side 
of the hill, the roadway in many places has been scooped 
away by the rain, leaving portions of the pavement 
3 ft. or 4 ft. up the side of the bank. It descends again 
over a small moor, and remains well-defined until 
opposite to Cefnucheldref. Here, according to the 
Ordnance Map, it turned to the left, round the hUl, and 
after crossing the River Nedd re-ascended the hill to 
the spot where there are undoubted traces of the road, 
but this is incorrect. 

From Cefnucheldref the line of the road ran straight 
on into the adjoining field. The traces near the hedge 
may be illusory, but in the further field at the head 
of the ravine the line is distinctly marked as far as the 
hedge. After this, all trace of the road itself is ob- 

Digitized by VjOOQ IC 


literated, but the heaps of stones in the fields correspond 
exactly to similar ones across the river. For the length 
of three fields the hedg^e probably stands on the site of 
the road, which must have passed the River Nedd 
somewhere about the present ford. On the other side 
of the river we again find distinct signs of the road in 
the fields above Coedygarig. Here the stones of the 
pavement were taken up within the last sixty years, 
and still stand in heaps along the line of the road. 
Over the wall there is one of the most perfect pieces of 
pitching on the whole length of the road, and it is 
incomprehensible how it could have been overlooked by 
the Ordnance Surveyor. The road then falls ixito a 
modern trackway to the Maen Madoc.^ 

Another stone has been found near to the same site 
on the mountain to the left. It has been called the 
Penymynydd Stone. It is now at Ty-mawr, Aberdare. 
It was visited by the Cambrian Arcnseological Associa- 
tion in 1900. The road descends the hill, and is plainly 
discernible until obliterated by the modern turnpike 
road to Brecon. Beyond the cross-roads, in 1896, the 
pitching could be recognised in many places at the side 
of the road, but in 1902 these traces were by no means 
so evident. 

Leaving the Maen Llia to the right, the old road 
continues until it reaches the steep scarp of the Llethr, 
overlooking the Senni valley. From this point to 
Fedwenunig, two miles distant, its course is purely a 
matter of conjecture. On the side of the hill some 
remains of an old road are seen alongside the present 
road, which possibly may indicate the site of the Roman 
road, but that, too, is lost in the cultivated ground 
below. The idea of some modern writers that it went 
down the Senni valley to Rhydybren is untenable. 
The more likely route is, as suggested by Jones, that it 
passed above Blaen Senni, and there met the lane from 

^ Archoeologia, iv, Plate 1 ; Gough's Camden, 11, Plate 14 ; We^t- 
wood, p. 64 ; Ardi, Camb., 4th Ser., vol. v, p. 332. 

Digitized by VjOOQ IC 


Fedwen at the turning to Gelliauisaf, crossing the 
dingle of the Cwmddu at the easiest point. It is not 
unlikely that frotn this point to Fedwen the lane runs 
on the line of the old road, though there are no traces 
lea of it. 

A little above Fedwen, we come on the road, which 
is lost in a gorse plantation, but reappears in a field on 
the further side, until lost in another large plantation, 
and again to be found in a field beyond. It is now 
lost for some distance, to be seen further on, where it 
runs the whole length of the field next the cross-roads. 
Beyond the cross-road, it is traceable in another field 
for some little distance, but is lost until it ascends the 
hill above the plantation. In the low ground beyond 
all trace again is lost, and where it crossed the Llestyr 
is uncertain. In the field across the brook a broad line 
of scattered stones mark the Roman paving turned up 
by the plough year after year. The crossing of a brook 
and the line up the hill to the common is plainly 
discernible, but it is lost on the other side, and the 
crossing the Camlais is uncertain. It ascended the 
Mynydd Iltyd on the same line as the modern trackway, 
which has obliterated the old road, except for some 
detached pieces of pavement on the ascending slope. 
Where it leaves the common (with Blaengwrthyd on 
the right), the road and pavement are again visible 
(this length is marked on the Ordnance Map as ** Roman 
Road "). It is seen in the field after passing the gate 
and skirting the hedge in the next one. It is then lost 
amid the ruins of a farmhouse, to reappear on the other 
side above the fishpond, when it is finally lost. The 
course from hence to The Gaer is only theoretical ; when 
lost, the line was pointing in the direction of Penpont 
Ford (which is the best ford on the Usk for many 
miles), and local tradition has always asserted that it 
went through the park at Penpont, and joined the Julia 
Maritima at the foot of the hill. 

Digitized by VjOOQ IC 

136 report on the excavations 

The Gaer at Coelbren. 

The camp is situated on the crest df a rounded hill, 
730 ft. above the sea-level. There is a fall of 30 ft. to 
40 ft. in every direction, from the ramparts to the 
general level of the country around. 

The hill is of a stiff boulder-clay, overlaying the out- 
crop of the coal measures, and is now covered with short 
hill-grass, with patches of heather. The sides of the 
camp are almost an exact square of 160 yards, with the 
usual rounded angles. The highest point is about 60 ft. 
inward from the western entrance, to which is a fall of 
1 ft., of 5 ft. to the north-west angle, 7 ft to the south- 
west, 10 ft. to 15 ft. to the western side. The ram- 
part all round the enclosure has been partially levelled. 
It now has the appearance of a broad platform, 30 ft. 
to 40 ft. across, in some places having a slight fall to 
the interior at the inner edge, but generally it is 
levelled off to meet the rise of the ground. 

At some period the whole interior appears to have 
been ploughed up. The marks of the plough are more 
evident on the platform than elsewhere. A short 
distance to the north and under the hill is a large bog 
— Gors Llwyn — difficult to cross at any season, but 
impassable in winter. To the west the country is open, 
to the south runs the Camlais brook, and to the east 
the Nantybryn, though neither could now be considered 
as a military defence. Altogether, it was a well-chosen, 
easily-defensible site. With a moderate surface drainage 
it could at any time be made dry and habitable. On 
the west are two ditches, 4 ft. to 6 ft. below the present 
ramparts, the outer ditch being rather lower than the 
inner one (Fig. 2). On the north front the inner ditch is 
not so apparent, but the outer one is well marked. On the 
north-east front both ditches have almost disappeared, 
though the rampart is well marked. On the south- 
west the ditches have entirely disappeared, and the 
rampart is spoiled by a modern hedge ; the same to the 
south, though here and there the ditch can be traced. 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


Fio. 2. Plak of C^ 

Sceil^ ^^ 

Digitized by VjOOQ IC 




Digitized by VjOOQ IC 

Digitized by VjOOQIC 



The entrance on the west front is represented by a 
causeway across the ditch. That to the north may 
have been where a modern footpath has made a breach 
in the rampart, bat on the south and east, though the 
roads are clearly to be traced up to the encei7ite, all 
trace of the entrances have been destroyed. In the 
interior are two heaps of stones, evidently surface 
stones, dislodged by the plough and collected together. 
There are also three large boulders, doubtless placed 
in position by the hand of man, though for what 
purpose is yet to be seen. The adjoining farm is called 
Tonycastell, and the next TonyflSldre. This might 
mean '* Town of the Soldiers," but the local people call 
it ** Ton-y-ynwl-y-dref," or the **flatland at the edge 
of the town." There is a local tradition of a town about 
here in former times, but they consider that it is to be 
found under the Gors Llwyn, not at the camp. The 
remains of ancient scratchings for iron are to be seen in 
the immediate neighbourhood, some in close proximity 
to the camp. 

Some two years ago, about twenty large buttons 
with holes in them were said to have been found. 
With that exception, no finds have been made in the 
interior of the camp. I could find no trace of these 
so-called buttons, and I consider that this find is iden- 
tical with the discovery of bronze harness in 1903 
(Arch. Camb., Sixth Sen, vol. v, p. 127), and that they 
were not found in this camp. The bronze celts now in 
the possession of Mr. Price (described in Arch. Camb., 
Sixth Ser., vol. i, p. 163) were found near here (Pen- 
wyllt), and a bronze celt was found on the adjoining 
hill to the south, about half a mile from the Roman 
road. A Roman coin was also found near the camp, 
just off the road, but I could not trace it. 

The Excavations. 

The Ramparts. — Sections were made across the ram- 
parts in several places, to ascertain, if possible, its 
original form and relation to the ditch. Some of these 

Digitized by VjOOQ IC 



^Oi///9 ^9^^/ 

^o^/p ^^^o 








(the earlier) were excavated under 
such adverse circumstances, aiused 
by the bad weather, that the results 
might perhaps have been considered 
hardly commensurate with the labour; 
but the structure proved to be of such 
a unique character as to amply justify 
the time and money expended. In 
some places the base of the rampart 
rested on a layer of logs, laid at right 
angles to the length. These were on 
the undisturbed boulder clay, about 
3 ins. to 6 ins. below the present sur- 
face ; in other places on large boards 
of oak, or rough stones ; whilst in some 
parts little more than a layer of de- 
composed brushwood could be detected 
above the clay. The pavement was of 
such interest that it was thoroughly 
and carefully investigated (Fig. 3). 
Section A is made at the south-west 
angle, where the log pavement was first 
detected. The details of this section 
are in a great measure applicable to 
those at the other angles. The logs 
were 17 ft. long, 8 ins. to 9 ins. in 
diameter, at right angles to the length 
of the rampart, with a slight fall to 
the outer side. The butts were on 
the inner side, and were accurately 
laid. Towards the exterior the line 
varied by some inches, according to 
the length of the logs. The marks of 
the axe were plainly to be seen in the 
butts. In some cases the larger logs 
had been split in two. The bark had 
not been removed. The wood was in 
various stages of decomposition. Some 
logs had become a haid "bog oak," 

Digitized by VjOOQ IC 

AT COfiLBREN. 139 

others were almost charcoal, and otliers again had de- 
cayed into a soft, spongy condition, easily squeezed in 
the hand, and soon crumbling away ; but when first ex- 
posed they had retained their form better than any 

Above these logs was a layer 1 ft. thick of black soil, 
which certainly contained much decomposed vegetable 
matter ; then another layer of logs, smaller than those 
of the lower stratum, seldom exceeding 6 ins. in dia- 
meter. These were laid irregularly, never close together; 
in some cases 1 ft. apart. They had more the appear- 
ance of branches than hewn logs, and the layer was 
more in evidence towards the inner side of the rampart. 
In eveiy section it was laid at the same height — 
1 ft. 2 ins. above the lower logs. This upper layer was 
mainly composed of birch, which had decomposed into 
a black band of vegetable matter. A small number of 
branches alone remained intact, though here and there 
the bark of the silver birch was easily discernible. 

The natural clay contains a good deal of iron in 
solution, which, acting on the tannin in the oak, tended 
to preserve it ; but it would not affect the birch, which 
fell into a black mass.^ 

Above this came various layers of yellow clay, with 
intervening bands of vegetable matter, together about 
3 ins. in thickness, showing that this part of the 
rampart had been constructed of turves and branches 
of trees. Longitudinal sections showed these layers 
almost parallel to each other, but transversely they had 
been squeezed by the superincumbent weight of the 
ramparts into various curves of contortion. Over the 
inner edge of the logs, and for 3 ft. towards the centre 
of the rampart, these bands had almost amalgamated 
into one continuous black mass, from the centre to the 

^ The further action of the iron on the vegetable matter, tbrongh- 
oat the ramparts and elsewhere, formed a sort of black ink, which 
had stained the adjoining clay to sach an extent that it g^ve a first 
impression of a far larger quantity of vegetable matter than was 
actually there. 

Digitized by VjOOQIC 


outer edge. These bands decreased until they were 
scarcely traceable. * This decrease was particularly 
marked at the top, excepting in one spot, there they 
could be traced to the top of the outer edge, giving 
the impression that the scarp face of the rampart 
was originally composed entirely of white clay. At the 
bottom of the outer edge, and 6 ft. to 8 ft. outwards, 
these layers were invariably found extending over the 

It cannot be said with certainty whether the scarp of 
the rampart extended beyond the logs, or whether the 
superincumbent bands had been squeezed outwards by 
the weight of the rampart. The latter would be the 
more reasonable conclusion, as the decaying brushwood 
would in course of time readily allow water to soak 
into the centre of the rampart, and the clay would have 
been kept in a plastic condition. Beyond the inner 
edge this black mass extended some 9 ft. towards the 
interior area, gradually tailing off to nothing. Here 
and at other sections there was some evidence of a 
distinct vertical face over the interior ends of the logs, 
but it was not enough to be conclusive. 

The log footing now lies about 3 ins. to 6 ins. below 
the present surface, which represents a fairly level 
platform, 35 ft. wide, extending over both edges of the 
footings. At the centre of the rampart there is only 
4 ins. of arable soil over the yellow clay, but over both 
inner and outer edge the depth increases to 18 ins., 
which gives the peculiar rounded contour apparent in 
all the sections. The greater part of this increased 
depth of soil consisted of mixed clay and vegetable 
matter, evidently the result of the levelling of the 
rampart. It extended towards the interior 1 ft. over 
and beyond the black mass before noticed ; like that, 
it tailed off to nothing, while towards the exterior it 
extended over the whole width of the berm, and nearly 
filled up the inner or main ditch. The berm was 16 ft. 
in width to the edge of the ditch, which was 18 ins. 
below the level of the footings. 

Digitized by VjOOQ IC 


Longitudinal sections, about 10 ft. long, were dug at 
both the inner and outer edge of the logs, to test their 
length and their direction at the curve of the angle of 
the camp. They were found to be laid at right angles 
to the tangent of the curve. The butt-ends were 
always in juxtaposition on the inner edge, and were 
accurately lined, the top-ends spreading out in the 
form of a fan. The intervening spaces were apparently 
Qot regularly filled in, though a few stones were found 
here and there. 

Section B, Fig. 4, at the other end of this curve gave 
a perfect exposition of the logs, which here were 17 ft. 
long and 9 ins. to 12 ins. thick. The 3 ft. band of 
black soil over the inner edge was very apparent, and 
the vertical face was more in evidence here than in any 
other spot. The bands on the interior edge of the 
rampart were much contorted, and gradually diminished 
towards the exterior. The rounded aspect of the section 
was particularly marked. 

Section C. — The logs were laid in two rows of equal 
lengths, the inner row at right angles to the southern 
face, the outer row following the curve of the angle of 
the camp. 

Section D. — The logs here were again in one length, 
set at right angles to the course of the rampart, and 
were both longer and broader than those elsewhere, 
the largest 1 ft. 3 ins. across (though from the afberwork 
in other sections it is probable that this was rather a 
board than a log ; at the time this was not noticed). 
There was nothing remarkable in this section, which 
was purposely made, where there was a sensible bulge 
on the inner face of the rampart, that gave the impres- 
sion that some foundation might be found. However, 
there was nothing more than an extra amount of soil, 
deposited there at the time of the levelling of the 

The Section E, Fig. 5, on the north-west angle differed 
slightly A from on the south-west. The logs were 

Digitized by VjOOQ IC 











Digitized by 


AT OOELBRErr. 143 

shorter, the greater number not more than 11 ft. 11 ins., 
and were so laid that in some cases the outer thrust 
of the weight of the rampart was taken by roughly- 
squared oak piles, 1 ft. across (Fig. 6). Where one log was 
too short for the position, it was wedged up to the pile 
by a partly-dressed stone. In another spot, a log some 
2 ft. longer than the others projected over the berm. 
and terminated with a pile. Another pile stood some 
distance outside, but tnere was no indication of what 
it had supported. The weather was bad, and the 
trenches here so waterlogged that it was impossible to 
proceed. This angle would be an interesting spot for 
further excavations. This section in general resembled 
the others, but that the bands had been less contorted. 

Fig. 6.— Camp at Coelbren : Logs at North-east Angle. 

and towards the inner side the black mass of vegetable 
matter was wider, and the colour more intense. This 
mass extended 20 ft. beyond the end of the logs, and 
the signs of decomposed wood were more apparent here 
than elsewhere ; whereas, in the other sections, there 
was room for doubt whether the decomposed mass now 
behind the ramparts might not be the accumulations of 
after-years. In this spot there could be no mistake 
that it was a regularly-laid filling (probably turf and 
branches of trees) ; and it points to the conclusion that 
the rampart, not here only but elsewhere, extended 
beyond the footing of logs. At the north angle the 
section was dug merely to ascertain the continuance of 
the foundation of logs. When they were duly found 
in the expected place, the section was not continued 

Digitized by VjOOQ IC 



Section F, S.-E. angle, Fig. 7. — The logs were here 
laid in two lengths, with an intervening space between 
them. The inner row was not more than 5 ft. long, a 
4-ft. interval, and the outer row 9 ft. long. They were 



^llzzriaZ ^ot^l- c^o 

TTvwnbcr' do 

Fig. 7. — Camp at Coelbren : Section at Sonth-ea«t Angle V. 
Scale, 10 ft. = l in. 




Fig. 8. — Cnmp at Coelbren : Logs at South-east Angle. 

12 ins. to 15 ins. wide, and 3 ins. thick ; in fact, rather 
boards than logs. Both extremities of the logs com- 
posing the inner row were regularly lined, as also those 
on the inner edge of the outer row ; but the outer 
edge was irregular. One board rested against a pile 
9 ins. across, the others projecting beyond it from 1 ft. to 

Digitized by VjOOQ IC 



3 ft. (Fijr. 8). Here the rampart was constiiicted with a 
second layer of logs, separated from the upper by an 
intervening space of 1 ins. to 4 ins. of very black soil, 
in which the remains of heather was still discernible, 
The smell of accumulated marsh gas was here most 

Fig. 9.— Camp at Coelbren : South-east Am?le. 
Scale, 20 ft. = 1 in. 


Fig. 10.— Camp at Coelbren : Logs at F. 
Scale, 10 ft. =1 in. 

oflFensive. The inner edges of both outer layers were 
in line, but the outer edge of the lower projected some 
6 ft. beyond the upper layer. The outer edge just 
projected over the base of an existing hedge. It is 
possible that the boards may have been shortened when 
that was made^ There was no visible sign of the ditches; 



Digitized by VjOOQ IC 


doubtless they had been filled in and the ground 

Fig. 9. — Some 19 ft. behind these logs was found, 3 ft. 
below the surface, a transverae log F, 9 ft. long, 6 ins. 
across, and well squared, laid at right angles to the 
diagonal of the camp. It was kept in position at its 
northern end by a large stone, 12 in. by 8 in., and by a 
small pile on the outside (Fig. 10). Its end was secured 
to the next log, 6 ft. inwards, by a cross-transverse, 
which was further secured by a stone, 7 ins. across, and 
a small pile. As these transverse logs appeared to extend 
inwards for some distance, the section was enlarged, to 
ascertain their number and position, as it was possible 
that they might be the foundation of a ramp leading to 
the angle of the camp. A third log was found 6 ft. 
behind the last, and a fourth 4 ft. distant from the 
third. Their outer ends were connected by cross-pieces, 
but they did not represent any definite line, the lengths 
of the logs being unequal. The appearance favoured 
the foundations of a house rather than a ramp. See 
account of " Interior." 

In order to test the further direction of these trans- 
verse logs, a trench (Section G) was cut towards the 
outer bank, but they aid not appear again (Fig. 9). At 
12 ft. some stones were found, which possibly might have 
formed a step, and at 28 ft., stones, which had some 
appearance of a wall, but further digging did not 
confirm this theory. They were found 2 ft. above the 
wooden foundation. Here, as on the other side, the 
logs were in two lengths, with a space between. On 
the inner row the logs were 6 ft. long, a space of 4 ft., 
and the outer row about 8 ft. ; the exact dimensions 
uncertain, owing to an underground drain. They were 
decidedly more boards than logs. The usual upper 
layer, in the same place, was here very pronounced, and 
composed of really fine logs. Within a short distance 
of the outer layer, and 1 ft. above them, a wall was 
found, which curved in the direction of the curve of 
the angle of the camp, which was eagerly followed up. 

Digitized by VjOOQ IC 


but it proved to be a modern conduit from a spring to 
a well in the hedge outside. 

A trench, H, was then dug along the base of the 
Sector formed by Sections F and o, to ascertain how 
the log foundations were carried round the curve. 
Ten ft. from the pile x was another, projecting upwards 
1 in. to 3 ins. All the intervening space was close- 
boarded with wide boards. One, 1 ft. 3 ins. across, had 
had a large piece cut out of one side by a saw, 
probably, needed for some other purpose, which shows 
that any odd pieces — provided they were large enough 
— were used for this pavement. The outer edge of 
these boards was irregular ; the first three being too 
short, had the length made out by pieces about 1 ft. 
long, jammed against a stone 2 ft. 4 ins. by I ft. 2 ins. 
by 1 ft. 3 ins. thick. Further on, the boards overlapped 
the stone by many inches. At the pile y they pro- 
jected 3 ft. ; taking the line of piles as a datum, the 
boards might be considered to overlap at last 3 ft. 

Beyond the pile the character of the paving changed : 
the boards lay further apart, and logs reappeared in 
some places 1 ft apart. The lower layer throughout 
was formed of wide boards, placed close together. It is 
evident that the first — or lower — layer had been found 
an insufficient foundation, and that a second layer had 
been laid above, to secure a better result. 

However faulty the method of construction, the 
ultimate result was good ; as, though the angle was 
the weakest point of the whole front — ^and doubtless 
the spring of water was as much hindrance to the 
Romans as it was to us — ^yet the foundation has not 
shifted in the slightest degree. No contortions of the 
layers of peat and clay are here seen in the rampart. 
In point of fact, when cleaning the soil in order to 
extract one of the boards, a space of 2 ft. by 3 ft. was 
pared away, in alternate horizontal layers of black soil, 
grey soil, and perfectly white clay. 

As the wooden pavement had been proved at all the 
angles, and along a considerable stretch of the southern 

10 « 

Digitized by VjOOQ IC 


front, there is little doubt than on that front it con- 
tinued the whole way ; but on Section I on the 
northern front, where the outer edge of the log pavement 
was expected to be touched, there were small pieces of 
wood 1 ft. 6 ins. by 9 ins. in diameter. Beyond these 
there was no further trace of wood. The log pavement 
was absent ; in its place was a layer of black vegetable 
matter overlaying the natural clay. The bands in the 
rampart were fewer in number, and in parts the clay 
was perfectly clean. The black mass in the rear of the 
rampart was, however, wider here than elsewhere. 

Section J was likewise bare of interest. It was cut 
transversely across the road, and continued longitu- 
dinally along the rampart, to test the road at the 
entrance ; and to ascertain if any wall or fblmdation 
showed the division between the road and the rampart, 
the excavation was carried 2 ft. down in the alluvial 
soil. At the end of the road, yellow clay of the 
rampart with the intervening bands appeared. 

Finding no trace of either wall or of logs, the trench 
was now cut diagonally across. After cutting 6 ft., the 
depth of alluvial soil had diminished to 10 ins; the 
original rampart increased to 1 ft. 9 ins., but the black 
bands had amalgamated into a mass. After 15 ft. this 
mass gave place to the mixed soil, the produce of the 
rampart. It was conclusive that on this side there was 
no log pavement. The base of the rampart may have 
rested on stones, as was found to be the case on the 
western side. 

Section K was made across the next front to test the 
ditches, and was carried into the rampart. The soil 
was hard clay and gravel — a good foundation for any 
weight. There was no trace of log paving, and the 
cutting was not carried through ; but by a hole dug ia 
the rampart on the opposite side, it was ascertained 
that the base rested on a layer of stones, about 9 ins. 
across, set close together. 

This concluded the investigation of the ramparts ; 
and the conclusion deduced therefrom is, that the log 

Digitized by VjOOQ IC 

AT COfil.BRttK. 149 

pavement was laid to enable the rampart to stand on 
the natural soil without slipping. Where the founda- 
tion was most treacherous, wider boards, and even a 
double layer, had to be used, but that no connection 
could be traced between the log pavement and any 
wooden superstructure which might have been erected 
on the top of the rampart. 

That tne log pavement designed to withstand the 
outward thrust of an excess of weight on the rampart 
is found under the four angles, and is generally absent 
from the sides, would lead to the conclusion that engines 
for missiles were placed only on the angles, as they 
alone would have necessitated this unusual foundation. 
That on the other three sides a layer of stones, or a 
thick layer of brushwood, was found sufficient base to 
secure the rampart from slipping, and that the log 
pavement was laid along the southern front, on account 
of a defective foundation, or of springs on the treacher- 
ous clay. 

The Ditches. 

Section A, south-west angle. — The width of the berm 
was 16 ft ; the edge of the ditch 1 ft. 6 ins. below the level 
of the log footing. The ditch was 9 ft. wide at the top. 
The sides having been cut in the stiflF yellow clay, were 
well preserved. The ditch itself was filled up with 
black decayed vegetable matter. At the depth of 2 ft. a 
piece of common red Roman pottery was found. At 
6 ft. down the character of the filling changed to a 
mixture of silt, gravel, and decayed vegetable matter. 
Around this were found a large number of oak stakes, 
9 ins. to 12 ins. long, pointed at one end (possibly 
charred), with a curious notch below the pointing, 
giving them the appearance of modern tent-pegs. They 
were lying flat, and not in situ. Also some pieces of 
cut oak, 6 ins. long by 4 ins. wide. It may be con- 
jectured that they were portions of obstacles for the 
defence of the ditch (fig. 11 on next page). 

At the depth of 6 ft. 6 ins. was found a leg bone of 

Digitized by VjOOQ IC 











*> a 




an animal, either cow or deer, and a rib and a vertebral 
bone of some young animal. The ditch was further 

Digitized by VjOOQ IC 


opened out, but further investigation was frustrated 
by the rain. The sides of the ditch fell in, the clay 
became unworkable, and it was impossible to decide 
where the ground had been previously worked. The 
section of the ditch was certainly triangular, and when 
the digging was abandoned, at a depth of 6 ft. 6 ins., 
the apex was apparently about 3 ft. below. 

A bank of natural stiff yellow clay, 7ft. to 8 ft. wide, 
divided the inner from the outer ditch, which was more 
distinctly marked on the surface than the inner ditch. 
It proved to be about 7 ft. wide, triangular in section, 
and seemed not more than 6 ft. deep. It was di£5cult 
to distinguish between the filling and the undisturbed 
ground, as after the first foot of peaty ground had been 
dug, there came a mass of silted clay, which under the 
constant rain became perfectly unworkable. Nearly at 
the bottom of this outer ditch was found an oak stake, 
3 ft. long by 9 ins. across. It was lying flat in the 
ditch, and was much decayed. In the hope of finding 
more, the length of the ditch was dug for 15 ft., but no 
more were forthcoming. This also had to be abandoned 
on account of the wet. 

A Section (M) of the glacis was made rather higher 
up, to ascertain where the soil from the ditch had 
been deposited. It was found that a quantity about 
equal to the size of the outer ditch had been placed 
on the crest of the glacis, varying in depth from 
9 ins. to 12 ins., tapering off to nothing after a length 
of 30 ft. It was clean clay, and the original surface of 
the turf beneath could still be discerned. 

A Section (K) of the ditches of the west front was 
•made, rather to the north of the entrance. They were 
more distinctly marked than elsewhere ; the section 
differed slightly from those on the other side. The 
•inner ditch was 1 1 ft. wide at the top, and triangular 
in section. Beneath 5 ft, of alluvial soil there was 
2 ft. 6 ins. of silted clay ; then it became impossible to 
determine whether the disturbed ground went down 
further. Two oak pegs, similar to those before described, 

Digitized by VjOOQ IC 


were found immediately at the bottom of the black 
soil. The natural bank between the ditches was about 
15 ft. wide, of yellow clay mixed with gravel. The 
outer ditch was 8 ft. across at the top. The level of 
the present ditch is 1 ft. below the surface of the 
ground, und the filling of black matter 4 ft. Below 
that the fine silt was met with : apparently, the ditch 
was not much deeper. 

The natural soil under the rampart was a hard gravel, 
mixed with clay ; yet, judging from the result of the 
digging in the ditches, some few feet below the stratum, 
there would be a vein of finely-silted clay, saturated 
with water, almost in a running condition. 

At Section E on the south-west angle the ditches 
are fairly well marked on the surface, but below they 
are very distinct. They are here filled up with intensely 
black soil, and the yellow clay of the nattiYal soil, as 
underneath the rampart, which forms the bank between 
the two ditches, is very light coloured, 'i'he ditches 
seem practically the same size us those at the south- 
west angle, but there was a great accumulation of 
water, and difficulty in draining it off, so that digging 
had to cease after about 3 ft. had been excavated. An 
attempt was made to run a drift up the outer ditch, so 
as to drain the bottom some yards higher up ; butas it 
appeared to act as a drain for the whole field it was 
abandoned. The only way to excavate these ditches 
would be to run a drift up one of the ditches, and allo\y 
it to drain the ground before commencing the work. 

Section L across the centre of the northern front 
proved the most fruitless of all. The outer ditch, well 
defined on the ground, proved to be full of black earth. 
The mound of natural clay between the two was very 
distinct. The inner ditch, nearly obliterated on the 
surface, proved to be full of black earth like the other. 
The outer ditch had been excavated down about 3 ft., 
and the filling was beginning to assume a more settled 
character, when the accumulation of water rendered 
further work impossible. It was then utilised as a 

Digitized by VjOOQ IC 


reservoir for water baled out from the drift towards 
the ramparts. In the inner ditch the usual black earth 
soon gave place to half-silted clay, rough gravel (re- 
sembhng the soil found in the outer ditch on the south- 
west angle). This ditch appeared about 7 ft. wide, but 
it was a matter of conjecture where the filling ended 
and the scarp begun, or whether it was silting or a 
of a natural vein of the boulder clay. About 7 ft. undier 
the supposed berm, traces of wood seemed to be present. 
If this were correct, the ditch along this front must 
have been both wider and deeper, and the berm pro- 
portionately narrower. 

The Entrances. 

There are supposed to have been four entrances to 
the camp. 

The position of the south entrance can be approxi- 
luately fixed, as the pitching of the Rpman road leading 
to Clawdd-y-Banwen is visible on the outside. The 
modern pathway probably follows the course of that 
road. As this path is much used by coUiers going to- 
and-fro after dark, it was not advisable to make a 
trench across it. 

The position of the north entrance is not so well de- 
fined, as the existing break in the rampart might have 
been made for the footpath which now crosses the area 
,of the camp. If it be the entrance, it would divide the 
northern front into two unequal parts. The trench 
dug showed more stones in the soil under this^ footpath 
than on either side of it. As this trench also had to 
be closed before nightfall, no conclusive results were 
obtained. The eastern entrance should be more easily 
located, for the Roman road from the east can be traced 
as a ruined causeway in the field outside. A trench 
was cut behind the hedge, 8 ft. from the corner. For 
a length of 6 ft. there was a depth of 1 ft. 9 ins. of 
.alluvial soil, for the next 7 ft., 1 ft. of alluvial soil, and 
1 ft. 6 ins. of mixed clay and stones. No signs of 

Digitized by VjOOQ IC 


pitching, though there were significant traces of a 
destroyed roadway ; after which the tailing of the end 
of the rampart was apparent (see Ramparts). A long 
trench (N) was dug 30 ft. behind trench J, nearly in 
line with the corners of the hedge. The alluvial soil 
was found 1 ft. 9 ins. deep; at 15 ft. a good deal of 
iron was found. On cutting across the prolongation of 
the road, stones were found about 9 ins. below the 
surface, but no positive indication of a road. The 
thickness of the natural soil was 1 ft. 3 ins. The stones 
continued to the end of the trench, but at a greater 
depth, the natural soil running to 1 ft. 9 ins. deep. 
Some traces of the continuation of this road were found 
in the diagonal trench T. A trench (P) across the 
caubeway of the western entrance, in line with the 
inner ditch, showed that the so-called causeway had 
never been excavated. Upon the old natural surface, 
represented by a black band, 2 ins. thick of decayed 
vegetable matter, had been deposited a layer of yellow 
clay 16 ft. wide. This was laid 9 ins. deep in the 
centre, tailing oflF to nothing at either end. Over this 
^lay the alluvial soil was 8 ins. deep in the centre and 
1 ft. at the ends, giving the causeway a slightly rounded 

In the centre of the causeway the top of a large 
stone almost protruded through the yellow clay coating. 
For what reason it was so placed could not be deter- 
mined. Further trenches, 13 ft. in length, were ex- 
tended on either side of the causeway, ending in a 
ditch. The inner and outer ditches on the south side 
were here joined by a ditch, running parallel to the 
causeway, 7ft. wide, 5 ft. deep, and triangular in sec- 
tion. On the north side the ditch was 6 ft. 6 ins. wide, 
and not more than 3 ft. deep. The ditches here con- 

In another section (Q), made 21 ft. behind the former, 
traces of made ground were apparent on the spot where 
the road should have been, but no pitching. The 
trench was lengthened to determine if there were a 

Digitized by VjOOQ IC 

At COELfenEN. I56 

junction with the rampart, but the result was indefinite. 
A hole dug 20 ft further on the same line proved that 
the rampart was here laid on a layer of large stones. 
A third trench (R) was cut 30 ft. behind the second. 
The ground was very stony, but no decided sign of a 
road could be detected. Some 30 yards further, on the 
conjectured line of the road, stands a considerable heap 
of stones. From their size and appearance they might 
have been taken from the pitching of the road. These 
stones could not have been found in the boulder clay : 
they must have been brought hither for a special pur- 
pose, and it is evident that that purpose was for the 
pitching of the road. If any portion of the road can 
be found undisturbed, it is probably under that very 
heap of stones. 

The pitching found in Section W, and for a distance 
of 30 ft. beyond, was to some extent a prolongation of 
the line of this road. It ran fairly parallel to the north 
side of the camp, and would point not very wide of the 
northern gateway, though the line cannot be said to be 
direct. The stones were upon the original ground, 
without any intervening gravel. It is quite possible 
that they may have been those dispersed by the 
plough, and that the actual site of the road had 
not Deen hit off. 

The pitching in Section X, towards the southern side, 
is more regularly laid, and has not been disturbed. It 
might have been a portion of a good road, but it 
appeared to be that it is more likely to have been the 
pitching of a yard. From experience of the pitchings 
of a similar character on the Rhondda Hills, I should 
say that it was the work of iron-smelters in post-Roman 

The ridge of gravel which runs across the eastern 
side of the camp is probably the site of a road. The 

f ravel had been carried there for a special purpose, 
he layer of decomposed peat is placed between the 
made ground and the original clay. It is 1 ft. 6 ins. 
in the deeper part. The width varies considerably. 

Digitized by VjOOQ IC 


No trace of the pitching reraaine. The plough has here 
done its work of destruction most effectually. 

The Interior. 

Septions were dug across the four diagonals of the 
carap, to endeavour to locate any pitching of road or 
foundations of houses ; if possible, to reconstruct the 
plan of the camp. Wherever any stones lay in apparent 
order, the digging was carried round to ascertain their 
size and direction of the area. 

Section Z, a continuation of F, from south-west 
angle diagonally across to within 30 ft. of the footpath, 
running across the centre of the field. The portion 
through the rampart and the transverse baulks has been 
described under ** Ramparts." 

A little above the level of the transverse baulks came 
much burnt earth and clay, with ashes (or, at least a 
black band) underneath. These remains were not level, 
but had more the appearance of low mounds with a 
hollow between. Above these was a layer of blue clay, 
with 1 ft. 6 ins. of arable soil on top of it. Much 
glass, mainly inj;an un worked stage, and pottery was 
here found. The position of these finds was various, 
though the pottery was mostly immediately over the 
brick ashes, while the glass was mainly immediately 
over the alluvial soil. About 2 ft. below the surface 
were found several stones. Thirty feet further up, 
2 ft. deep, was an area 3 ft. by 3 ins. of stone, laid in 
some order. It was neither the foundation of a wall 
nor the pitching of a road. A good deal of red earth, 
with black matter below, here extended about 20 ft. up 
the trench ; it might have been a floor. Pottery, both 
red and grey, and much glass, was also found. The 
depth of the natural ground was about 2 ft. At 30 ft. 
the trench crossed the layer of gravel 1 ft. 6 ins. thick, 
25 ft. wide, under a layer of soil ; and below the gravel 
a black band of decomposed peat overlying the natural 
ground. This gravel must have been brought there by 

Digitized by VjOOQ IC 


man. Four feet beyond this gravel, at the depth 
of 2 ft., there was a good deal of ironstone. At 80 ft. 
from the wall, pottery, glass, and a good many rubbed 
stones. The brick earth was here again conspicuous, 
and had the appearance of a floor, though irregular in 
thickness, varying from 1 ft. to 2 ft. below the surface. 
Beyond this the brick earth disappeared ; the layer of 
black matter also, and the last 20 ft. of the trench 
showed no sign of man's habitation, the soil being but 
1 ft. 6 ins. deep. 

Trench S, 30 ft. northward, parallel with A. Natural 
soil, 2 ft. from surface. The first 30 ft. showed a layer 
of red ash 1 ft. 6 ins. from surface, with black clay and 
decomposed matter below. Bead No. 2 was here found, 
just above the red ash, also glass and pottery. Fifty 
feet, a stone for grinding, glazed marble stone, and 
small fragments of ghxss and pottery. Sixty -one feet, a 
large boulder stone, 1 ft. below the surface, planted in the 
natural soil, its use unknown. A few feet further was 
a large piece of lead ; a space 6 ft. by 4 ft. was worked 
round, but besides glass and pottery nothing more was 
found. At 95 ft., crossed the gravel track, about 
28 ft. wide, 1 ft. surface-soil, 1 ft. gravel, black band 
1 in. thick, of decomposed peat immediately below ; then 
followed a long length of red ash, 4 ins. to 5 ins. thick, 
the usual 1 - in. black band underneath ; beyond, no 
further trace of habitation. 

Trench S (a). — A small trench, connecting ends of 
Z and S, primarily for draining purposes. It was dug 
through 2 ft. of alluvial soil, and was rich in pottery 
and glass. Here, at a depth of 2 ft. were found the 
piece of pottery of the peculiar yellowish-red glaze, and 
the glass of brownish tint. The alluvial nature of the 
soil laid both finds open to suspicion. 

Trench S (b), a second cross -trench, from 20 ft. up 
S to the supposed wall in Z. About half-way the red 
ash was crossed, and then an area of stones : a possible 
pavement. This pitching was 2 ft. 6 ins. wide at the 
trench, and ran for 12 ft. parallel to A and B ; it wa4 

Digitized by VjOOQ IC 


then widened to 4 ft. At first the stones were fairly 
regular, then they were raore disturbed. 

Section T. — From north-east angle of camp, diagonally 
across to the centre. The presence of the log layers 
(upper and lower) was ascertained. The lower layer 
lay 4 ft. 6 ins. below the surface ; 4 ft. behind the logs 
were several large stones, 9 ins. to 12 ins. across, in no 
regular order ; 16 ft. along, the trench cut into a 3-ft. 
area which might have been a pitching, but no extension 
could be traced on any side. The next 25 ft. was 
rich in pottery ; stones also, but in no definite order. 
Natural soil, only 1 ft. deep. At 39 ft., a 12 ft. length 
of brick earth, about 2 ins. thick, 1 ft. 6 ins. deep, not 
laid level ; but wavy, as had been noted in other places, 
though it had more definite appearance of a floor than 
the similar deposits found elsewhere. Some of the 
pieces were certainly fragments of bricks and not brick 
earth. The accurate area could not be settled. 

At 127 ft. stones were laid in some order at a depth 
of 9 ins. The stony area was worked round with no 
results. Some kind of pitching it certainly was, with 
many of the stones removed, and others disturbed by 
the plough. Many showed the mark of the plough- 
share. This pitching lay about the centre of the gravel 
ridge, which was not so clearly defined here as in other 
places. From hence a trench was run along this gravel 
ridge to Section S, which proved the presence of the 
gravel layer very near the surface, for the whole length, 
but no finds were made. 

Section T was continued from the spot where it 
crossed the prolongation of the road ; for a stretch 
of 7 ft. it had somewhat the semblance of a road, 
though there was no regular pitching, and the stones 
were little more than 6 ins. below the surface. Behind 
both these stretches of supposed road, at a distance 
of 20 ft., trenches were run to ascertain if such traces 
ran further, but nothing more was found. 

Section W, diagonally across the field, from the foot- 
path to the north-west angle, only a few pieces of 

Digitized by VjOOQ IC 


pottery were found, under 1 ft. 6 ins. of accumulated 
8oil, until the top of the field was reached, there, the 
trench crossed remains of much disturbed pitching, of 
of which a length of 30 ft. by 15 ft. was opened out. 
It ran fairly parallel to the north side, and if prolonged 
would have passed not far from the E. gate. Whetner 
it were pitching or road was difficult to decide. Half- 
way to the end of the Section, 1 ft. 6 ins. below the 
surface, was a pocket of charcoal 2 ft. in diameter, 1 ft. 
thick, of which 6 ins. were sunk in the natural clay. 
Towards the end of the Section some red pottery was 

Section X, diagonally across from centre to south- 
west angle. — The accumulated soil was seldom more 
than 1 ft. thick, and but few signs of man's hand 
appeared. Towards the end the trench crossed a 
pitching of large stones, laid in fairly regular order, in 
an oblong area, 13 ft. by 15 ft., I ft. below the surface. 
The stones were from 9 ins. to 12 ins. across, laid on 
the natural soil. The appearance was more that of 
pitching than of a road. A heap of large stones near 
this spot, on the edge of the western rampart, was 
carefully examined. It proved to be the fragments of 
a large boulder that had been blasted into three pieces 
(and other smaller portions which had been removed), 
with an accumulation of field stones thrown in between 


The timber foundations of the ramparts — the spikes 
and other obstacles found in the ditches already de- 
scribed (Fig. 11) — ^are most interesting, and to a certain 
extent are unique. It is to be regretted that the finds 
in the interior of the camp are not merely commonplace 
but poor. They can, however, be identified as Roman, 
though from the utter absence of any coins it is difficult 
to assign any nearer and more definite date. The 
potsherds are the most easily identified as of a parti- 
cular period, and from their characteristics a definite 

Digitized by VjOOQ IC 


opinion may perhaps be looked for. These potsherds 
divide themselves into several groups. 

Fragments of very rude Yellowish Ware, which vary considerably 
Iq texture, some of comparatively fine grain, others very coarse, 
hardly to be distinguished from sandstone, the usual paste used 
for Amphorae and other large vessels. The authorities of the 
British Museum have dated them as about the third century. But 
few of these pieces exceeded 6 ins. across, and varied fi-om ^ in. 
to 1 in. in thickness. Fig. 12, the handle of an Amphora. There 
are seven rims of various vessels, probably also Amphorae. Of 

Fig. 12. — Pottery found at Coelbren. 

one, the colour is more decidedly red. It must have formed 
part of a vessel 7 ins. in diameter across the mouth. Some of 
the others might have been part of larger vessels. 

Fragments of Ware. — Yellowish drab, or fawn-coloured, almost 
black in the interior, with a fine, even surface. Several frag- 
ments were found, but the one — the lip of a mortarium — is the 
most interesting. Anotlier lip of a similar vessel is a more 
damaged specimen. 

Black Ware. — These potslierds vary in hue from deep black 
to slaty grey. They are generally fine in grain, though some 
few are coarse, and are mostly considered to be no older than 
the second or third century. A considerable number of frag- 
ments of finely-moulded rims, pertaining to articles of domestic 

Digitized by VjOOQIC 



Fig. 13.— Pottery found at Coelbren. 

Fig. 14. — Pottery found at Coelbren. 

Fig. 15. — Pottery found at Coelbren. 

use, vases and pots (Fig. 13), and a lesser number of bases 
(Fig. 14) ; but the latter, as a rule, seem to have belonged to 
vessels of a smaller size. The sides have been reduced to very 
small pieces, some ojf which are decorated. 
Fig. 15 has bands of irregular concentric curves, resembling 
6th 8KB., vol. vn. U 

Digitized by VjOOQ IC 



the marks made by a large thumb, similar to those figured in 
the Gellygaer record. Such pieces were found widely apart over 
the ground, showing that it was a common form of decoration. 

Fig. 16. — An ornamental band, which might possibly be a 
series of curves, though more like chevrons. 

Fig. 17, besides the band, has prominent projections on the 
outer surface; the purport of which is inexplicable. As two 

Fig. 16. 

Fig. 17 

Fig. 18. 
Pottery found at Coelbren. 

specimens were found, it was not an accidental defect, but is part 
of some design. 

Another fragment of the black ware had several globular 
projections, whether accidental or by design is open to con- 
jecture. Some pieces have a series of etchings, i in. long, in 
two bands. All these decorated fragments are of thinner and 
harder material than those undecorated, for which reason they 
are in better preservation. 

Digitized by VjOOQ IC 


Bed Ware. — Fragments of ordinary red ware ; also some rims 
and bases of small vessels. Fig. 18 is one of two decorated 
pieces found ; all others are perfectly plain. They are generally 
in a soft, rotten condition. There are also several pieces of a 
better style of ware, with a perceptible glaze on the inside, 
having the appearance of a polish. Several frj^ments of one 
piece of rich salmon-coloured ware ; but possibly this difference 
in colour is due to the particular environments : some of a 
coarse-grained paste, intermediate between the ordinary red and 
the first-mentioned coarse yellow ware. 

Beddiah-Brown Ware. — A few specimens of a fine reddish- 
brown glazed ware, varying in colour from deep red to shades of 
brown. These were mixed with the old Eoman remains, and 
some have been pronounced by the British Museum to be as 
late 8is the seventeenth or eighteenth century. Those were 
found in two spots only. 

There is also the handle of a pitcher, with a greenish glaze, 
probably of the same late date, though it certainly has a look of 
mediaeval age. 

Several other specimens with a yellow glaze may also be of 
the same age. The modern specimens were never more than 
12 ins. below the surface. For that reason, those found deeper 
may be considered as older. 

Bricks. — One specimen, 6 ins. long, of a coarse grain, though 
much broken. It is decidedly a brick. A great quantity of 
burnt clay, which might have been bricks, though in some cases 
the pieces might be the harder portions of brick earth. 

TUee. — ^No tiles, either of brick or stone. 

Bones. — In addition to the bones found in the western ditch, 
small fragments of bones were found over the camp. Nothing 
could be deduced from their presence ; they might belong to 
any animal at any date. 

Coal. — Numerous specimens of coal, all of the seam which 
outcrops half a mile distant They are too numerous to have 
been accidentally brought on the ground by manure carts, which 
might be a feasible theory to account for isolated pieces; con- 
sequently, the coal must have been designedly conveyed there, 
probably for smelting purposes. 

Charcoal — Charcoal was found in various places, but on only 
one spot was there a considerable amount. This pocket was 
not far from a pitching, which might have been i hearth. Here 
Uie bits of charcoal, though small, were distinct and well 
preserved. In other places it was difficult to decide whether it 

Digitized by VjOOQ IC 


had been charcoal or wood, now turned into bog oak ; though the 
remains of coal and charcoal taken all together were of con- 
siderable amount, yet there wsts not enough to represent smelting 
on a large scale. 

Glass, — The bits of glass have more interest than the potsherds ; 
though stray fragments were scattered all over the area, the 
greater number were found on the south-eastern side. They 
have oxidized in an irregular manner; the smooth faces are 
little affected, but the rough unmoulded piece has a dull patina 
of oxide. At first glance there is little difference to be discerned 
between the old Roman glass and the modern derelict ; but when 
seen in juxta-position, the peculiar greenish-blue shade of the 
former is easily distinguishable. There are four bits of bases of 
bottles of the square Roman type (Fig. 19) ; two good examples of 

Fig. 19. — Glass found at Coelbren. 

the ribbed side of a bowl or cup, and the riiris of several vessels ; 
besides many examples of unmoulded glass, either direct from 
the furnace or melted into their present shape as the result of a 
conflagration in the camp ; and of the two classes together, rims 
and sides joined to unmoulded glass. 

At first, the condition of these remains would give rise to the 
theory that they were damaged articles from a manufactory on 
the spot, for which the silica sand could have been procured from 
the Penwylt Mountain, about four miles distant. 

But the idea of a general fire in the camp is the more feasible, 
for the edges of many of the moulded fragments have been 
blunted by heat subsequent to their original burning, and no 
remains of either moulds, hearths, or crucibles were found. 
A great many pieces of glass, varying in thickness, the colour 
ranging from blue to light green, apparently portions of vases 
or vessels of some sort; some few approximate closely to 

Digitized by VjOOQ IC 


window-glass. These were collected all over the area, at such a 
depth that it is improbable that they could be modern products, 
in contradistinction to the modern fragments which were scat- 
tered over the surface of the soil. Some are open to suspicion, 
but, on the whole, the sorting has been correct 

One bit of very thin yellow-brown glass, which has many ot 
the characteristics of a modern hock -glass— except that it has 
an S curve of surface — was found alongside undoubted Roman 
remains. It is impossible that it could have worked down from 
the surface, and the ground has not been ploughed for eighty or 
one hundred years. At that time, and for some generations 
previous, there was a considerable population in the vicinity 
engaged in working the old Banwen iron- works, the ruins of 
which are a conspicuous object in the distance. 

Fig. 20. — Iron Spur found at Coelbren. 

And a small piece of beautiful blue glass ; this likewise does 
not look very mediaeval, yet it may be genuinely old. 

Iron. — The few articles of wrought iron had nothing definite 
to mark their age, except two spurs, considerably oxidized, 
which have been pronounced no earlier than mediaeval days. 
One (Fig. 20) has a shank 3 ins. long, the rowels still apparent, 
as well as the bar for grasping the boot ; the other is more oxi- 
dized. Several nails 4 ins. long, lumps of oxidized iron of 
uncertain use, and several scraps of iron tubing about the thick- 
ness of tobacco-pipes. 

Various deposits of scoria and slag, as also iron-stone for 
smelting, and limestone for flux, some of the latter glazed by 
the action of tire. It is possible that we did not come across the 

Digitized by VjOOQ IC 


main refuse tips ; but, from what was discovered, no considerable 
smelting could have been carried on within the camp. 

Lead, — A lump of melted lead, of irregular shape, about 1 lb. 
in weight, near a pitching of stone, which may have been a 
hearth. Small pieces of worked lead, in some cases resembling 
modern window-work, were found in other parts of the camp. 

Stones, — A large number of sharpeners of various sizes and 
shapes, some more or less square, in section, with rounded 
edges (others almost round), generally about 1 in. across by 
6 ins. to 8 ins. long, all more or less broken ; smaller ones also in 
considerable numbers. They were formed from sandstone or 
slate, which must have been brought from long distances. A 
stone ball 5 ins. across, which looks like millstone grit or sand. 

Fig. 21. — "Melon" Bead found at Coelbren. 

stone conglomerate ; either a hammer-stone or a projectile for a 

Several stones (sandstone) with hollows 3 ins. to 4 ins. deep, 
1 ft. to 1 ft. 6 ins. across, all broken in two pieces. No mark of 
grinding could be detected ; it is difficult to judge if the hollows 
be natural or artificial. The stones are so rough, that they 
could never have served as mortars or for grinding ; but the 
hollows do not look as if made by Nature. If they had been 
used as moulds, for which purpose they would have served, the 
surface of the hollow would have been glazed by the molten 
glass or iron. They are not from the boulder clay ; they have 
been procured from the bed of a river if natural ; if artificial, it 
is difficult to divine their use. 

Sundry stones, much worn by use, 6 ins. by 12 ins. by \ in., 
of hard sandstone. The flat side had been used for polisliiug, 

Digitized by VjOOQ IC 


and one end either for polishing or moulding the inner rim of a 
pot On one stone there appeared to be an excrescence on the 
flat side — of burnt material of some sort — iron or what not. 

Discs of sandstone 2^ ins. across, J in. thick ; fairly round ; 
probably used as covers for small vessels. 

Flints. — Only three flints were found. One was outside the 
ramparts on the berm ; one of the others is much burnt. Flints 
were used by the Romans usually for agricultural purposes, 
therefore their presence in this camp would hardly be expected. 

Oem and Beads. — A very small amethyst, amid other re- 
mains. It has a slight groove cut in the back, as if to attach it 
to the setting. It is possible it may not be a genuine " find '* ; 
but as the men were warned that no reward would be paid for 
** finds," they had no inducement to introduce any extraneous 
objects. Five terra-cotta beads in graduated sizes, "melon" 
shaped, of a greenish- blue colour ; one 1^ in. across (Fig. 21), 
one rather larger, fragments of two about the same size, and 
one very much smaller. 


The result of the treQches, etc., is fairly conclusive 
that Coelbren was not a walled town, and that if there 
were houses in the interior, they were not constructed 
of stone ; though the structures (of whatever material) 
were without doubt arranged in the usual regular order 
of Roman camps. There are traoes of the roads, or 
rather of their foundations ; the pitching is plainly- 
discernible close to the exterior on the southern side of 
the camp, yet in the interior all trace is lost ; which 
indicates that it has been purposely destroyed, probably 
when the interior area was first broken up by the 
plough. About one hundred years ago, the farmers 
in the neighbourhood had a perfect craze for collecting 
the field stones (the course of the Roman road to Neath 
in parts can be followed by the heaps of stones); though 
a large amount was used to construct stone walls on 
the neighbouring farms, many heaps still lie unutilised. 
If the roads which, without doubt, existed have been 
so completely demolished, it is easy to comprehend how 
all traces of human habitation have been obliterated. 

Digitized by VjOOQ IC 


In several spots there were signs that brick floors 
had been laid on wooden foundations ; it was impossible 
to trace any definite outline, for such floors had gene- 
rally fallen into red brick earth ; but in every case an 
intervening layer of black earth lay between the brick 
earth and the natural surface of the ground. This 
black earth might represent either decomposed timber 
or the original peat. 

In many cases the brick earth had the irregular ap- 
pearance of cinder-tips, but the thickness was generally 
uniform, the layer of black earth underneath being more 
or less present. The remains conjectured to be Roman 
were generally found within a few inches of this black 
layer, or, when it was almost absent, on the original 
surface of the ground ; but the specimens of glass were 
found either on the surface of the brick earth, or but 
slightly embedded in it. 

With the exception of two specimens (noted after), 
no remains are to be traced either in the rampart — which 
was constructed of clean materials, showing no sign of 
previous occupation of the adjoining land — or on the 
levelled terrace. The latter is remarkable, as the top 
of the rampart is the spot where potsherds, etc., would 
have been deposited ; and such remains would have 
been found in the levelling of the top. Yet this surface 
is as clean as the portion of the rampart left intact. 
This can only be accounted for on the supposition that 
during the time of occupation the top of the rampart 
had been protected, as was the case with the defences 
constructed by Caesar in Gaul. There strong towers at 
the angles and the entrances, defended by ballista and 
other engines, were connected by wooden galleries and 
bridges ; a wide berm in front was covered with wooden 
spikes, and beyond were ditches filled with obstacles. 
The whole outer defences could be swept by missiles 
from the towers and ramparts. 

The theory of wooden galleries on the ramparts of 
Coelbren may be considered as inconclusive, out the 
other points seem to be irrefragably proved. The 

Digitized by VjOOQ IC 

AT CO^LBR^. 16& 

wooden footings to the ramparts and angles were quite 
unneeded to take the weight of the rampart itself, and 
must have been laid with a view to the strain of the 
ballustrae and other heavy engines of war. The width 
of the berm and its protecting spikes are there to be 
seen ; while the insignificance of the ditches show that 
they were designed for the protection of obstacles, not 
as defences in themselves ; and their relation to the 
ramparts show how these ditches were defended. 

This being so, the wooden galleries on the ramparts 
would be a part of the design, and the defences would 
have been the counterpart of those described by Caesar 
in the construction of a camp during a campaign against 
the Bellovaci (St. Pierre de Ch^tres), at the conclusion 
of the Grallic war. Such was the system of fortification 
brought by the Romans to England. It is probable 
that all their first stations were constructed on these 
lines, in after-years to be either remodelled in stone or 
levelled to the ground, in after-times ; and it is only in 
such out-of- the- world places as Coelbren that there is 
any chance of finding the original type. 

This type is not uncommon in Scotland, for there the 
period of Koman occupation was so limited that there 
was no opportunity of converting them into walled 
towns ; they were deserted while still in good preserva- 
tion, and there was no inducement (as there was at 
Coelbren) to subsequently destroy them. This same 
type of fortress was adopted later by the Romano- 
British in their struggles against the Saxons ; and, so 
well did they copy these Roman models, that often it 
is impossible to say positively whether they were the 
work of the masters or the pupils ; and many so-called 
British camps may actually be the work of the Romans, 
though they differ as much from the preconceived idea 
of a Roman camp as does the camp of Pierre de 

The point for consideration is. Whether the rampart, 
after the disintegrations of many centuries, was levelled 
for the purpose of working the plough through it, 

Digitized by VjOOQ IC 


about one hundred years ago, or whether it was done 
by the Romans themselves to destroy it as a military 
work ? The inferences seem to favour the latter 
theory. The appearance of the rampart itself gives no 
clue : provided the work were done by the spade — 
either in the year 200, for the purpose of destroying it, 
or in 1800, for the purpose of utilizing the soil — the 
result would be much the same. The solution must be 
looked for elsewhere. 

If the ditches had been filling up with sittings and 
decayed vegetation for 1200 years, and then the soil of 
the ramparts had been shovelled down into them, the 
blackest soil would be at the bottom, the lighter above 
it. The contrary is the case: the soil from the ramparts 
appears at the bottom of the ditch, and the decayed 
vegetation appears to have accumulated afterwards. If 
the oak spikes had remained on the berm, or if they 
had fallen into the open ditch, they would have rotted 
away ; but if they had been thrown into the ditch, and 
the levelling of the rampart had been cast on top of 
them, they would have been preserved, as we found 

The absence of potsherds and the dSbris in the filling 
up of the ditches shows that during the occupation 
the soil had remained perfectly clean. The 2-ft. block 
of vitrified earth at the top of the south-east angle, 
outstanding below the alluvial soil, shows that the fire 
was made after the levelling of the rampart, and before 
the alluvial soil had accumulated. There is no positive 
testimony to the date of this fire ; but the inference is 
that it was coeval with the other remains of smelting, 
which would carry the date of the levelling of the 
rampart back to an early date. The small bit of 
Roman glass found under similar conditions at the 
north-east angle would suggest that it had worked into 
this position long before the plough had come over the 

The entrances to the camp, as they now stand, were 
totally unprotected; in fact, must have been the weakest 

Digitized by VjOOQ IC 

At COfiLiREK 171 

part of the defences. This could not have been the 
case ; if their towers and guard-rooms had been left to 
perish by decay, some traces would still remain. On 
the contrary, every vestige has disappeared. The total 
destruction must have been done designedly, and such 
is applicable only to the Roman occupation of England. 
All circumstances point to the conclusion that the de- 
fences at Coelbren were of a semi-permanent character, 
and were intended only for a limited time of occupation. 

At first glance, this theory would seem to support 
the supposition that Coelbren was one of the forts con- 
structed by Ostorius about 50 a.d., after his successful 
campaign againt Caradoc, as after a limited occupation 
the Roman forts were taken, and the invaders were 
forced to retire ; in which case the victorious natives 
would have burnt and devastated the camp, levelled 
the ramparts, filled in the ditches, and rendered it 
useless for any subsequent reoccupation. But there is 
no reason to conclude that Ostorius penetrated so far 
into the heart of Wales ; and from the accounts as given 
by Tacitus and the Welsh historians, the fieldworks in 
the Margam mountains represent the extreme limits of 
his conquest, and the scene of his disastrous defeat. 

The period of the invasion under Julius Fron., 70 a.d., 
or perhaps that of Julius Agricola, some years later, 
seems to fit in better : for it is stated that he so far 
subdued the Silures that, to consolidate his conquest, he 
made the two highways — the Julia Maritima and the 
Julia Montana, running parallel to each other for a long 
distance, and uniting at Maridunum. These were con- 
nected by the Gelligaer cross-road, running from Cardiff 
to the Gaer, at Brecon (Banium) ; and it is possible 
and probable that the Sarn Helen from Nidum to 
Banium was constructed at the same time and for the 
same purpose. This theory fits with the situation of 
the camp as a protection for a road running through a 
hostile country, as the road makes a wide detour, as if 
to secure a good strategical site for the camp. Our 
knowledge of the history of those times is imperfect, 

Digitized by VjOOQ IC 


yet it seems not improbable that this road with its camp 
should have been constructed at the same period as the 
highway, as a part of the one general scheme. 

After the country had settled down under the Roman 
rule, and it had become feasible to reduce the standing 
forts and garrisons, the stations along the Julia 
Maritima (which was selected on account of its superior 
strategical position) were converted into walled towns ; 
and all these stations can be identified at the present 
day. Westward of Cardiff, on the Julia Maritima — 
sites of the stations Bovium, Nidum, and Leucarum — 
(which were not converted into walled towns) are lost. 
How long these stations were held it is impossible to 

Mr. Ward considers that the walled town of Gelly- 
gaer was abandoned as early as 90 a.d. This is rather 
an early date. How could the roads have been made, 
the temporary camps converted into walled towns, and 
the country so settled that the garrisons could have 
been withdrawn, in such a short interval of vears? If 
he be correct, the occupation of Ooelbren (which, being 
un walled, would have been vacated before the excava- 
tion of Gelly-gaer) would be less than fifteen years. 
There is one disturbing factor to my theory : that is, 
that the remains found, to a large extent, are typical of 
the second or third century. If that be proved, then 
the camp and the Sarn Helen must have been con- 
structed at that date, as it is impossible that the camp 
could have been occupied from the first to the third 
century without showing signs of successive occupation. 
In that case, the extension of the Sarn Helen must have 
been an afterthought, long subsequent to the general 
conquest of the country, and the camp constructed to 
defend this road was occupied but a short time, when 
the general withdrawal of the garrisons caused its 
evacuation, when it was demolished, the ramparts 
levelled, the ditches filled in, the buildings burnt, that 
there should be no occupation by the natives. The 
only tokens of habitations left were fragments of glass 

Digitized by VjOOQ IC 


welded together by the conflagration ; molten iron, 
which had formed a slag with the surface clay, lead 
run into a natural mould, broken scraps of pottery, 
charcoal, and burnt wood.^ This destruction of the 
camp in no way impaired the usefulness of the road; 
and, as stone is scarce in the neighbourhood, the pitch- 
ing inside the camp might have been taken up to mend 
the road along the Sarn Helen. 

The number of memorial stones of Romano-British 
time found along the line of the road show that the 
road was still a highway ; while the remains of ancient 
hearths in the neighbourhood prove that the iron in- 
dustry was still in work. Possibly, the remains of iron- 
smelting found in the camp may be referred to this later 
date. This would be another factor in the obliteration 
of marks of Roman occupation. These later occupiers 
have left no further sign of their presence. It may be 
that these rude smel ting-hearths appertain to a far 
later date, when Royalists or Roundheads attempted 
to repair the damages after a fight, at a solitary forge in 

^ The final destruction of the camp by fire is, however, problema- 
tical, the evidences being conflicting. A portion of the lower band 
(having all the appearance of the debris of a great fire), consisting 
of barnt clay, bones, charcoal, and broken glass, was submitted to 
Mr. Seiller, the borough analyst, who is himself a keen antiquary. 
He kindly subjected this debris to a chemical analysis. He reports 
that the upper portion was in places white in colour, and proved to 
be chiefly calcium phosphate, with some iron and aluminium phos- 
phate. He considers that it is bone, from which the organic matter 
had been removed by decay, and the lime partly replaced by iron 
and alumina. He gives the analysis of the clay, which he describes 
as a siliceous clay, containing 1.80 per cent, of tatanium oxide, and 
apparently had not been exposed to fire. 

The glass was soda glass, containing no lead. From the sharp 
edges of the fragments sent to him, he does not consider that such 
portions had been subjected to the action of fire. 

He had not completed the analysis of the charcoal, but is of 
opinion that it is bog oak. This would point to the conclusion that 
the apparent result of fire is merely the natural blackening eflect of 
tatanium on the iron of the natural clay. On the other hand, the 
blunted edges of other pieces of glass, and the welding together of 
various portions, would indicate the destruction of the camp by fire. 

Digitized by VjOOQ IC 


the old camp, leaving a few bits of broken crockery as 
token of their presence. The whole countryside was 
then almost deserted, and was little better than the 
Great Forest of Brecon, which it practically adjoins. 

The final work of destruction was carried out by the 
considerable population drawn to the neighbourhood 
during the working of the old Banwen Ironworks. 
Advantage was taken of the amount of basic slag and 
decomposed vegetable matter in the ground ; the site 
was converted into agricultural land, and the plough 
made short work of ramparts and Roman remains. 

I cannot conclude without expressing my appreciation 
of the kindness received from everyone while carrying 
out these investigations : To Mr. Ward, of Cardiff 
Museum, for valuable hints how to commence the 
work ; to Mr. Morgan Williams, of St. Donat's, the 
owner ; to Mr. Miers, the Lord of the Manor, and to 
the tenant, who assisted in every way ; to Sir Griffith 
Thomas, without whose assistance I could not have 
procured men ; to Mr. Cunnington, who devoted much 
time to superintend the work ; to Mr. Seiller, whose 
chemical knowledge was of first importance ; to the 
Station-master of Coelbren for all his kindly, willing 
help ; and to Mr. Lloyd, Mining Engineer, for his care 
in the superintendence and the drawing of the plans, 
without whose hearty cooperation I could not have 
attempted the work. 

Digitized by VjOOQ IC 




By T. C. CANTRILL, B. Sc, LoND. 

CwMBRWYN is a prettily-situated Carmarthenshire 
farmhouse, about 10^ miles west-south-west of the 
county town as the crow flies, in the parish of Laug- 
harne, and three miles west-by-north of its church. It 
is on the northern fringe of a stretch of uplands, which 
extends some fifteen miles west of Laugharne, and is 
bounded on the north by the valley of the T&f and its 
tributaries, and has Carmarthen Bay to its south. The 
house overlooks a wooded dell, or " cwm," down which 
splashes a clear stream, which ultimately debouches 
into the T§,f at Llandowror, and it lies near the south- 
east side of the road between St. Clears and Marros. 
The district is sparsely inhabited, and abounds in 
ancient remains. 

The remains which are the subject of this Paper are 
on the western and highest side of a field, known as 
DwrbwIl-fsUjh, to distinguish it from the adjoining field 
to the west, Dwrbwll-fawr, which contains a pool, 
whence the names. The ground here gently ascends to 
the west, and beyond the second field makes a rapid 
drop into a second ** cwm," the stream of which joins 
that referred to above, in the vicinity of Llandowror. 
The site we are considering is 388 fl. above the 
Ordnance datum, but it does not occupy the highest 
point, as the field behind continues to gently rise 
(Fig. !)• The view from here is very fine and extended. 

Before the recent exploration, the site presented an 
irregular oval space, enclosed by a low rampart, with 
the faint outer hollow of an external ditch, both inter- 

Digitized by VjOOQ IC 


rupted by the shallow opening of an entrance on the 
east side, the whole being grass-grown like the rest of 
the field. The shape would be more exactly described 
as between an oval and an irregular polygon, with 
rounded angles, as will be seen from our plan of the 
enclosure (Fig. 2). The exact line of the western ram- 
part was not easy to determine from the surface 
indications, as it is occupied by the hedge which 
divides the two fields. But this hedge here makes a 
slight outward or westward swing, and it is more 
highly banked than elsewhere ; and in the field on 
its further side are some faint traces of the hollow 
of the ditch. From these indications, it was tolerably 
clear that advantage was taken of the rampart on 
that side by those who made the hedge, and that 
the two lines approximately coincided: an inference 
amply corroborated in the subsequent exploration. The 
height of the rampart scarcely exceeds 2 ft., but is 
naturally more obvious on the outer side, the hollow of 
the ditch increasing the apparent elevation some 3 ft. 
The width may be set down as approximately 30 ft., 
but it is difficult to demark its gentle slope from the 
normal surface. The hollow of the ditch is less deter- 
minable, but it is somewhat narrower. The length of 
the area within these envelopes is about 130 ft., and 
width, 110 ft. ; or 280 ft. and 200 ft., respectively, 
including them. The ditch is not continued across the 
front of the entrance, but ends on either side of it, 
leaving a causeway about 20 ft. or more in width. 

These earthworks are plainly visible in the field, and 
even more so from the road ; and it is curious that 
they should have escaped the observation of the 
Ordnance Survey officials, as also of antiquaries, until a 
circumstance in 1890 forced them upon their notice. 
Yet it would be incorrect to say that the nature of the 
site was wholly unknown, as the ground had on several 
previous occasions been disturbed for the sake of the 
material for building purposes it yielded. It is said 
that, many years ago, a former Lord Kensington, who 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


lived at Caatell Lloyd, about two miles to the south, 
removed much stone from it for various works he had 
in hand ; and a fine slab at Castell T6ch is pointed out 
as from the same source. The circumstance referred to 
above was one of these delvings for stone. Mr. Bowen, 
the owner and occupier of the farm, resorted to this 
expedient for materials wherewith to construct a cul- 
vert. It was the first time during his twenty years' 
residence at Cwmbrwyn that he had broken into the 
ground, and all he expected to find was stone ; but he 
found something more, and this excited his curiosity, 
and soon attracted the attention of others interested in 
the local archaeology. One early visitor to the site was 
Mr. H. C. Tierney, of Carmarthen, the Editor of the 
Wdshman, and this was followed by a long and detailed 
account of the discovery in the issue of that paper of 
June 27th, 1890, under the nam de plume of '* Peter 
Numskull." A few weeks later, this was followed by a 
visit from Mr. Edward Laws, F.S.A., of Tenby, at the 
request of the Editor of Archceologia Camhrensis, who 
communicated the results of his enquiry to that publi- 
cation (5th Ser., vol. vii (1890), p. 334.) The two 
accounts do not agree in every particular, but this is 
due to the writers having to mainly depend upon 
hearsay, their visits being after the demolition of the 
remains thus brought to light. Under these circum- 
stances, it will be best to describe the remains exhumed 
during the recent exploration fii-st, and then to cor- 
relate the statements of these communications with 
them. The little hollow left by Mr. Bowen's diggings 
is shown in the north-west corner of the plan (Fig. 2). 

The writer's first visit to the site was on October 2nd, 
1905, at the instance of, and accompanied by Mr. G. 
G. T. Treherne, who was then President of the newly- 
formed Carmarthenshire Antiquarian Society and Field 
Club ; and Mr. T. C. Cantrill, of the Geological Survey, 
and Mr. William Clarke, of LlandaflF, were also present. 
On that occasion several labourers were engaged ; and 
with Mr. Bowen's kind permission, two diagonal trenches 

6th 8BB., YOU VII. 12 

Digitized by VjOOQ IC 


within the enclosure, and a smaller one on the site of 
the gateway, were cut. This preliminary investigation 
bj^ught to light the remains of a wall and undoubted 
e wlences of Roman occupation, in the form of roofing- 
slates, pottery, and other objects, and it convinced the 
party that the site was worthy of a more systematic 
exploration. On the evening of that day, some of the 
chief finds were exhibited at a meeting of the Com- 
mittee of the Society at Carmarthen, and Mr. Treherne 
urged that the Society should take up the work. Early 
in the following year (1906) this was decided upon, and 
a fortnight's digging was arranged to begin on Whit- 
Monday, June 4th, the present writer being asked to 
direct the operations. The work commenced on the 
day arranged, under the superintendence of%fr. Tre- 
herne, Mr. Clarke, and the writer, for the first week, 
and under that of Mr. Cantrill for^the second. The 
members of the Carmarthenshire Society paid a visit to 
the excavations on the 7th. 

At the conclusion of this fortnight s work, it was still 
evident that further digging would be necessary ; and 
in anticipation of the visit of the Cambrian Archaeo- 
logical Association to Carmarthenshire in August, the 
Committee decided upon another fortnight's work. 
This second work began on August 2nd, under the 
superintendence of the writer, Messrs. Treherne, 
Clarke, and Cantrill being unavoidably absent. The 
visit of the Association to the site took place upon the 
15th following. The exploration ceased on the same 
day ; and as there was little prospect that further 
digging would materially add to the information 
already gained, arrangements were made for the early 
filling-in of the trenches — a work which Mr. Bowen 
kindly undertook to superintend. 

The Exploration op the Outer Works. 
Several trenches were cut through the rampart, 
A — H, Fig. 3, and one on the east side, A, was extended 
across the ditch. A section of the rampart and ditch 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 








Fio. 2. CwMBRWYN : Pjlan 

aoft. toi 

Digitized by VjOOQ IC 

!■■■■ t ':,,;«^^ 


^RK Excavation. 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


as seen in this trench is given on Fig, 4. The ditch 
here was V-shaped, but irregularly so. The bottom, 
which for a depth of 2 ft. 6 ins. was cut into the rock, 
was reached at about 8 ft. 9 ins. below the present 
surface. It is probable that the original form was less 
irregular than we found it, and that the irregularities 
were largely due to the dislodgment of soil from the 
upper parts of the sides. On the bottom and tailing 
off up the sides, was observed a layer of yellowish 
loamy soil with few stones, closely resembling the 
undisturbed soil of the site, but softer, and having 
the appeafance of "wash-down" from the sides. A 
• similar '* wash -down " was observed by the writer 
in the ditch of the Roman fort of Gellygaer, but 
it was more easily distinguished from the hard clayey 
soil there. Immediately above this deposit was a 
darkish soil, also with few stones, about 18 ins. thick 
at the bottom and thinning off up the sides. This 
indicates a gradual silting-up of the ditch to that 
extent, and the darkness was undoubtedly due to 
vegetation. Above this, the ditch was filled with a 
jumbled mass of normal soil, with an abundance of 
stones, which tended to lie parallel to the slopes of the 
sides, but about the middle they formed a pSle-mSle 
accumulation. This certainly was mostly derived from 
the rampart, and it suggested an intentional fiUing-in 
of the ditch in order to lessen the inequalities of 
the surface, and so render the field more fitted for 

The usual form of a Roman ditch is angulated with 
straight sides, or strictly V-shaped ; but examples are 
known in the North with narrow flat bottoms, convex 
sides, and rounded brinks. It is difficult to say for 
certain what was the exact original form of our ditch : 
probably, however, it was of the ordinary type, and the 
observed irregularities were due, partly, as stated 
above, to the dislodgment of soil from its sides, and 
partly to our inadvertent removal of portions of the 
natural soil. Assuming the angulated form, the width 

12 2 

Digitized by VjOOQ IC 


would be about 17 ft. or 18 ft., and the depth from the 
Roman surface^ 8 ft. 

The section of the rampart as disclosed in this trench 
(a) had a well-defined underlying layer (indicated by 
small crosses on the plate) of clayey consistence, redder 
than the normal soil of the site, and almost devoid of 
stones. It had a tolerably uniform thickness of 8 ins., 
and width of about 14 ft. 6 ins., the ends being abrupt • 
It was near — if, indeed, it did not actually rest upon — 
the old natural surface, and was set back from the ditch 
about 6 ft., thus leaving an intervening shelf or berm 
of that width, no doubt to insure the stability of the 
rampart. Somewhat behind the middle, this layer was 
slightly dished ; and immediately above was an accu- 
mulation of stones, the weight of which may have 
caused the hollow. The soil above this layer was 
faintly bedded, the beds dipping towards the back, 
that is, towards the interior of the site ; while in the 
opposite or upward direction they became confused 
and lost. The uppermost, which directly overlaid the 
clayey layer for several feet towards the back, was of 
dark earth. A similar clayey layer was observed in a 
trench across the south rampart at G (Fig. 3), but it 
was not so well-defined, and its width was about 
15 ft. In a trench, which was cut into the west 
rampart at D, and tunneled for a short distance under 
the hedge — representing in all, perhaps, two-thirds the 
width of the rampart — a somewhat thicker clayey seam 
was observed to rise, following, in so doing, the natural 
rise of the ground here ; and it rested upon a thin bed 
of sandy loam, below which was the normal soil. This 
seam may have been of natural formation, as in several 
other places the undisturbed soil was observed to have 
a surfacing of finer soil. In the remaining trenches 
cut through or into the rampart, the artificially-placed 
soil was found to be more clayey below than above, 
but it was not distinguishable as a separate layer. In 
none of the sections except that of A, and in less 
degree that of d, did the upper portion of the rampart 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


Pio. 3. Cwmbewyn: Pijm 
30 ft. to 1 i 

Digitized by VjOOQ IC 



fWR Excavation. 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


exhibit definite stratification ; but^ in all, the proper 
soil of this structure was more or less distinguishable 
from that which covered its sides and smoothed off the 
contour, and which was undoubtedly derived from the 
original summit. 

The presence of a more or less well-defined bottoming 
of clayey soil in all these trenches can hardly have been 
accidental. It rather indicates that the first step in 
the construction of the rampart consisted in the laying 
down of such a layer, 15 Roman feet wide, and repre- 
senting the width of the intended bank. It appears, 
as a rule, to have been laid directly upon the natural 
surface, but it is probable that its irregularities were 
first filled in. The object of the bottoming may not 
be clear, but we know that the Komans were often at 
considerable pains in preparing the sites of their earth- 
works. The Antonine Wall and the ramparts of 
Birrens in Dumfriesshire, were raised upon a spread 
of stones of the requisite width ; while those of 
Camelon, near Falkirk, rest upon a foundation of clay 
and brushwood between marginal strips of rough 
stones. Split timbers and branches have also been 
observed in a similar position. The source of this 
clayey soil at Cwmbrwyn presents no diflSiculty, as it 
occurs close at hand ; the pond in the field behind, for 
instance, is sunk in clay. 

The inclined stratification on the upper part of the 
first cutting seems to indicate that the rampart was 
piled up towards the front of the foundation-layer, and 
that the materials trailed down by a gentle slope 
towards the back. This would necessitate a revet- 
ment of some kind along the front, unless the rampart 
was very low. But no indication of a revetment of 
stone, turf, or timber was found, although a special 
trench was cut along its presumed line. The many 
large stones in the tilling of the ditch may seem to 
point to one of stone, but it is hard to understand how 
its foundations should have entirely disappeared ; 
equally, if of timber, why its post-holes should not 

Digitized by VjOOQ IC 


have been apparent. The most reasonable conjecture 
is a turf wall, which, pushed forward by the weight of 
soil behind and the turves mingling with the fallen 
materials, would leave few, if any, traces of its former 
existence. The rampart would have a flat summit 
sufficiently wide to provide a walk for the defenders ; 
and there must, of course, have been a parapet, which 
may have been of turves or timber. If all die soil 
from the ditch was used for the mmpart, it must have 
been of considerable height ; but the Romans some- 
times, if not usually, raised an cKternal low mound 
corresponding with the modern glacis, to accentuate 
the height of the counterscarp. No trace of such a 
mound, however, was observed in our first trench ; but 
there is a slight rise on the outer side of the ditch at 
the north-east which may be artificial. Any moiind in 
this position would be necessarily low, so as not to 
interfere with the " command " of the rampart, and it 
would probably take the form of a mere spread of soil 
rather than an actual mound. 

In its present condition, the rampart shows as a low 
and gentle mound of greater than its original width, 
the inner slope encroaching on the interior of the enclo- 
sure, and the outer covering the berm and encroaching 
on the ditch. From the data obtained from the various 
trenches referred to above, it is comparatively easy to 
determine the exact limits of the Roman rampart and 
ditch (which are shown in Fi^. 3), except along the 
west side, where visible indications of the outer line of 
the rampart and of the ditch are well-nigh obliterated.^ 

The excavation of the gateway brought to light the 
remains of the side walls, which were 13 fl. apart, and 
between them a spread of rough stones on the level of 

1 Mr. David C. Evans, of St. Clears, writes : " It appears to me 
that one may be fairly certain that the ditch went round the back. 
There are depressionn in DwrbwU-fawr corresponding exactly with 
that running roand the rampart& The vegetation is ranker there, 
and Mr. Bo wen's son tells me that there is a marked difference in 
the soil." 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 



— -t»l TC H t H G 

R A M P A R T- — 



T lO t^ 





Fio. 4. CwMBRWYN : Plan op Buildik 
Plan— 20 ft. to 1 in. 

Digitized by LjOOQ IC 

"!Ns' F \ ^: 

■ "Z'l! -1: _!'- — I." -ii^ -hiL"-: 

■ ■ ' ■ ■ ■ ■ 




-D I T C H 


J L 

ttD Sections of Rampakt aud Ditch. 
Seetiona— 8 ft to 1 in* 


Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

W Jereini/.] [Photo 


Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


the old surface. Little of either wall was left, hut that 
to the north was the better preserved. Of this wall 
only two huge stones remained in position, and these 
were rough as quarried, with no sign of dressing of any 
sort (Fig. 5). Together they extended 5 ft. 6 ins., and 
rested about midway upon a foundation of smaller 
stones, about 10 ft. long. Whether this foundation re- 
presented the original length of the wall is uncertain. To 
the west it ended abruptly with the line of the back of 
the rampart, and this probably represented its original 
termination in that direction, but to the east it was 
indefinite. Probably the wall originally extended to 
the front of the rampart. Of the south wall, only a 
few rough stones of the foundation remained, and a 
single one of the actual wall. The spread of stones 
between these walls was too roughly laid to be regarded 
as either pitching or paving ; it seemed rather to be 
the foundation of a gravelled road, as the soil above 
contained much small broken stone. 

The scanty remains of this gateway were exceedingly 
rude for Roman construction. No dressed stones were 
found about the site. What was left of the side walls 
suggested masonry of Cyclopean type — large irregular 
stones with their gaps filled in with small stones.^ In 
a preliminary cutting made on October 2nd, 1905, at 
the south-west angle of the site of the gateway, an 
interesting and distinctively Roman object was found 
— ^the iron sheath, or shoe, which lined the socket for 
a door-pivot. When found, it presented an irregular 
mass of rust-cemented fragments of stone and earth ; 
but after chipping off these extraneous matters it 
proved to be a short cylinder, with sides and bottom 
lormed of thin iron, about -^ in. thick, the internal 
dimension being 1 J in. deep and 3^ ins. wide (Fig. 6). 

A short digression upon the gateways of Roman 
cities and forts, of which many examples have been laid 
bare, will be helpful here. They may be broadly classed 
as single and double. The Roman north gateway of 

» Ct Fig. 10. 

Digitized by VjOOQ IC 



Cardiff Castle and the two exposed at .Caerwerft wef e 
of the former type, that is, they consisted of a single 
opening or span each ; while those of Gellygaer were of 
the latter type. With the former we must also class 
that of Cwmbrwyn, for not only is it too narrow for 
subdivision, but no trace of an intervening wall or 
spina was found. In other respects, the planning of 
Koman gateways was, with few exceptions, remarkably 
uniform. The side walls of the opening of the single 
type, or of the single passage in the double type, had 
pilaster-like projections or returns in front, which con- 
tracted the opening, and they carried the arch, and 

Fig. 6. — Cwmbrwyn : Iron Shoe of Gate-socket. (J. ) 

often these were repeated at the back. Within the 
angles of the front projections were the sockets in 
which the door-pivots turned, each door consisting of 
two leaves which swung back against the side walls 
when the gateway was open. The sockets were sunk 
in blocks of stone firmly embedded in the roadway, as 
at Cardiff, or in the ends of a stone threshold, as at 
Gellygaer, or of a timber one, as at Silchester. Occa- 
sionally, the remains of the iron linings of the sockets 
have been found, as at Cardiff, where the bottom plate 
of one was still in position. Usually, the threshold had 
on its outer side a luised lip, or curb, which sheathed 
the foot of the doois when glosed. This at Gellygaer 

Digitized by VjOOQ IC 


was coDstractdd of flagstones, set on edge in the 
ground ; but at Cardiff there was a central stone stop- 
post instead. 

From these data it is possible to complete the plan 
of the Cwmbrwyn gateway with a considerable degree 
of probability. The iron shoe was loose amongst the 
dSbriSy and in a position where the socket-stone might 
be expected. This stone, however, was not found ; 
nor were the corresponding shoe and stone on the 
opposite side of the passage. It is, of course, possible 
that /these shoes were let into a wooden sleeper, which 
had entirely disappeared by natural decay ; but this is 
hardly likely, as suitable stone for the purpose is 
abundant in the district. It is more likely that the 
whole of the front portion of the gateway has been 
rooted up for building material, and that this accounts 
for the absence of the socket-stones. The side walls 
would certainly extend in a forward direction to the 
face of the rampart, as in Fig. 7, in which the remain- 
ing stones of the walls are shown black, and those of 
the foundations in outline, while the probable original 
planning is indicated in close diagonal shading, A A, be- 
ing the area covered by the rampart, and b b, the ditch. 
The jambs may have taken the form of inward returns 
at their extremities, as at 6, but more probably they 
were set back some distance, as at a, as Roman gate- 
ways were often recessed, and the spot where the iron 
socket-shoe was found is more consistent with this view 
than with the former. The jambs would reduce the 
actual portal perhaps to 10 ft. or less. It is reasonable 
to think that, following the usual custom, the gate 
was arched, but there was no evidence for this. If the 
portal was recessed, the arch would be set back with 
the jambs. It is also reasonable to think that the side 
walls would be returned for a short distance at their 
front ends, as indicated at c. The exploration supplied 
no hint whether the stop took the form of a curb (as 
shown on the plan) or a central post. 

A trench was cut about 12 ft. in front — that is, east 

Digitized by VjOOQ IC 



— of the gateway, and this brought to h'ght the spread 
of rough stones of the roadway, about 1 ft. below the 
present surface. The direction of this road beyond is 
unknown, as no indication was to be seen in the field, 
but it is possible that a faint difference in the colour of 
the herbage may reveal the course in a very dry 


/ B 


Fig. 7. — Cwmbrwyn : Conjectural Plan of Gateway. 

The Exploration of the Interior. 

A.t an early stage of the exploration there was 
evidence that the back or western side of the enclosed 
space had been occupied by a long narrow building 
with a slated roof, shown on a large scale on Fig. 4. 
Little more remained of its walls than their founda- 

^ There is a faint ridge mnnizig oonoentrioallj with the earth- 
works aboat 100 ft. to the soath and south-east, bat it appears to 
represent the ontcrop of the hard sandstone met with at the bottom 
of the ditch and of the pit g^ and thus to have no archsBological 

Digitized by VjOOQ IC 

CAftkAR^HENSHlRlL 18? 

tions, and considerable lengths of these had been 
wholly removed. The main structure was oblong, 
97^ ft. by 25 ft., with a southern adjunct or exten- 
sion, A, of a shape to accommodate it to the restricted 
space within the rampart. This was of slighter con- 
struction than the main building ; and as its walls 
were not bonded into the latter it appeared to be 
an addition, but not necessarily of a later date. 

The foundations of the main building were about 
3 ft. wide, and consisted of rough stones deposited in a 
trench of the same width. Here and tnere, these 
stones were laid more or less on end, like rough 
pitching. On the west side they formed a single 
course ; but on the east, where the natural ground 
is lower, there were two courses, evidently with a view 
to bring the summit of the foundations to a common 
level. Of the actual walls only short lengths of the 
lowest course remained — a broken length of about 
26 ft. on the western side, 3 ft. on the eastern, about 
9 ft at the north end, and a single corner-stone at the 
south-west angle, all indicated in black on the plan. 
These fragments of walling were 2 ft. thick, and were 
carefully constructed of slightly hammer-dressed stones. 
The mortar was reducea to an earthy consistence 
through the dissolving out of the lime. Very little of 
the foundations of the northern third of the building 
remained, except those of the northern end. It was 
across this portion of the building that Mr. Bowen's 
diggings took place in 1890, their approximate area 
being indicated by the dot-and-dash line. As he 
followed up the foundations for the sake of the stone 
they yielded, it may be presumed that their removal 
beyond the limits of this area was then accomplished. 
The highly interesting remains he met with will be 
considered presently. The east wall of the building 
was continued 9 ft. to the north, where it ended in the 
inner slope of the rampart, thus enclosing between the 
north wall and the rampart a small open triangular 
space^ B» which was entered by a doorway in the pro- 

Digitized by VjOOQ IC 



longed east wall, at K. Thus far, the general plan and 
construction of this building. 

The exploration of the portion south of Mr. Bowen's 
diggings, consisting of rather more than two-thirds of 
the main structure, yielded no trace of cross- walls; 
but there may, of course, have been timber partitions, 
which had disappet\red by decay. Wherever the ex- 
cavations (which are indicated by thin broken lines) 
were made, its floor was revealed, consisting of the 
gravelly soil of the site, well compacted by beating ; 
and probably it was originally mixed with lime, which, 

Fig. 8.— Cwmbrwyn : Roofing Slates. (J.) 

as usual, has disappeared. Portions of a similar floor 
were also revealed at the north end. This floor was on 
a common level throughout, and is shown as a dotted 
ground on Fig. 4. No trace of a floor was discernible 
in the southern adjunct. Lying upon the floor, and 
on the soil immediately above, over the whole of the 
building were innumerable fragments of roofing-slates, 
and many whole ones. They were of the usual Koman 
shape, about 11 ins. wide, with parallel sides and rec- 
tangularly pointed below ; while in the rough upper 
end was a nail-hole, which occasionally retained the 
rusted head of the iron nail ; but a few of these slates 
had square lower ends, whidi, with little doubt, were 

Digitized by VjOOQ IC 



used for the eaves (Fig. 8). These slates would pro- 
duce a pleasing lozengy pattern in combination on the 
roof, as indicated in Fig. 9. A roof thus covered had 
usually a stone ridge, but no fragments of ridge-stones 
were found. Several small pieces of Roman ted roofing- 
tiles were turned up about the area of Mr. Bowens 
diggings, but not in sufficient numbers to warrant the 
belief that they were used as roofing material : the 

Fig. 9.— Cwmbrwyn : Restoration of Roofing. 

Romans were wont to use these tiles for a variety of 
other purposes. That the building was of a single 
story may be inferred from the thinness of the external 
walls ; and these had too much disappeared to provide 
a clue as to where it was entered. Several fragments 
of window glass were found about the site, indicating 
the former presence of windows or skylights, and many 
fragments of square flue-tiles, of which more anon. 

Attention must now be directed to the northern 
part of the building, the seen© of Mr, Bo wen's diggings. 

Digitized by VjOOQ IC 


It Will be observed on the plans that a drain (l) extends 
from the east wall, pointing in a north-eastern direction. 
At a distance of about 80 ft. it was found to pass under 
the north rampart, and with little doubt it emptied 
itself into the ditch. It was of very simple construc- 
tion, consisting of a steep V-shaped trench, about 3 ft. 
deep, with a rounded bottom, which had been covered 
with flagstones, of which one was found in situ^ the 
rest having been pulled up. In its passage through 
the rampart, the upper part of the trench had been 
filled with broken stone instead of soil. We knew that 
the various structures found by Mr. Bowen within the 
building had been removed, but we cut a wide longi- 
tudinal trench from the north end for about 25 ft., in 
the hope of finding some indications of what had been. 
The normal floor, as described above, extended 10 ft. 
from the north wall, at the end of which a sudden drop 
of nearly 2 ft. brought us to a new level c, consisting 
of the natural soil, flat and somewhat hard. This 
continued about 8 ft., when a transverse ridge of 
natural soil, D, about 3 ft. wide at the base, and with 
sloping sides, was reached ; and the north side of this 
had been puddled with white clay, patches of which, 
about an inch or more thick, remained. On the south 
side of this ridge the ground fell to the former level, e, 
as far as the trench extended, 5 ft. We now excavated 
to the east, and almost immediately met with a vertical 
face of natural soil, 7 ft. from the east side of the 
building, which was capped with the gravelled floor. It 
was clear, then, that within the area dug by Mr. Bowen 
there was a sunk space, crossed by a ridge running east 
and west. The filling of this space consisted of soil and 
debris, mixed with a large amount of charcoal, which 
gave it a dark colour. 

What Mr. Bowen found hereabouts was related by 
Mr. Tierney in the Welshman, and by Mr. Laws in 
ArchcBologia Cambrensis, as stated above; and the 
former gentleman has kindly forwarded further parti- 
culars from memory, also the gist of an account which 

Digitized by VjOOQ IC 


appeared at the time in a Tenby paper. These accounts 
do not precisely agree, but this much is tolerably clear ; 
Mr. Bowen unearthed three parallel walls, running 
east and west, and about 5 ft. apart. Of these, the two 
outer walls were well built, and laid in mortar, while 
the middle was of " dry" masonry, with the interstices 
filled up with common clay. The first wall discovered 
— that to the north — was about 4 ft. high, and was 
traced for 24 ft., when it came to an end. This, it will 
be observed, represents the width of the building, so 
we may conclude that it extended across its full width. 
The south wall is described as a *' half- wall," that is, 
it was properly faced on one side (the north side), and 
was rough, as though built against the soil on the 
other. The face of this wall was plastered, according 
to Mr. Laws, with brick and lime concrete ; and he 
states also that the south face of the north wall was 
similarly treated. The length of this south wall is not 
stated, nor that of the middle one. 

Between the north and the middle walls was a 
curious construction, which is variously described as a 
masonry bench, or block, or concrete floor. Mr. Tierney 
thus describes it : ** Mr. Bowen found something which 
at first sight resembled a cist without the usual flag- 
stone covering. A number of thin flat stones of some- 
what irregular shapes and size, were set on their 
edges, so as to enclose a space 6 ft. lon^ by 2 ft. 9 ins. 
wide. This space was filled to a depth of 3 or 4 ins. 
with a bright red clayey, or rather gravelly, substance, 
which would almost appear to have been artificially 
coloured. At first it was nearly of vermilion shade, 
but on exposure to the air for some time its colour 
became a good deal duller. The composition of this 
red powder resembled that of the cement, of which 
I shall speak presently. The stones resting on their 
edges and the red materials were then removed, 
and directly underneath it, was found a bed of the 
same dimensions of beautiful concrete or cement, 4 ins. 
deep, and reddish in colour. It was not nearly so 

Digitized by VjOOQ IC 


bright a hue as the' red powder just described, but it 
appeared to contain a good deal of the same ingredients. 
Among the rest, quarry and other stone, not found, I 
believe, in the neighbourhood, appeared to enter in a 
granular form into its composition. The bed of con- 
crete was removed with a pickaxe, and disclosed a 
layer of flagstones, some of them over S2 ft square, 
and altogether making up the same area as the cement. 
When the flags were raised, and the layer of concrete 
of the same thickness as before, but this time a light 
grey colour, came to view. The men who made this 
grouting, concrete, or whatever it should be called, 
were masters of their trade, for even now it is hard to 
break it, even with a pickaxe. A large portion of it 
came off like a big flagstone, and it is almost as hard 
and solid as the stone of the locality. However, Mr. 
Bowen succeeded in getting it all away, and once more 
he encountered flags like those above referred to. He 
determined not to give in just then, and went on to 
raise the flags. It seemed to be labour lost, for no 
sooner was the second bed of flags out of the way, 
than, lo ! here comes the cement again — this time of a 
dark slaty colour, and coarse in structure. What lies 
beneath the dark cement, if anything, is not yet 
known/' The residue of this structure was, however, 
subsequently removed, so the writer is informed by 
Mr. Bowen, from whose description it appears to have 
been merely a foundation spread of rough stones. 

According to Mr. Laws, this " bench " was '* about 
2 ft. high, very strongly built of alternate courses of 
mortar and flags, on the top of which was a sort of 
tray made with flags, containing clay burnt, ground 
fine, and mixed with quartz. This bench was 7 ft. 
long and 8 ft. (3 ft. ?) wide." 

In each of the passages or spaces between these 
walls were many fragments of flue-tiles, and, according 
to Mr. Laws, there was " a considerable quantity of 
wood ashes ** in the space to the south. We also met 
with both in abundance in this part of the building. 

Digitized by VjOOQ IC 



No perfect flue-tile was found ; but a comparison of 
the fragments showed that they were of the ordinary 
Roman form — square tubes about 10 ins. long, and 
from 5 ins. to 6^ ins. wide on each side. They were, as 
usual, scored on the front and back, in order to make 
mortar adhere to them. These scorings were made 
with a three- toothed instrument or ** scratch.'' On 
some of the tiles they simply crossed the face diago- 
nally, saltire-wise, from corner to corner ; on others 
they formed a more elaborate design, consisting of two 

Fig. 10.— Cwmbrwyn: Flue-Tile, (i.) 

intei-secting semicircles, with wavy lines in the inter- 
spaces. Some, at least, if not all, had lateral openings 
(Fig. 10). Vertical flues formed of these tiles were let 
into the walls of a heated apartment, and by this 
means the heat of the hypocaust radiated from the 
walls, as well as from the floor. In the sudatories ot 
baths the walls were sometimes wholly lined with 
them, the lateral openings, just referred to, allowing 
the heated gases to freely circulate from flue to flue. 

The writer's first difficulty was to correlate these 
remains with the building as a whole ; but Mr. D. C. 

6th seb., vol. VII. 


Digitized by VjOOQ IC 


Evans, of St. dears, recently interviewed Mr. Bowen, 
who pointed out the spot where the ** bench '' was 
found. This was about 18 ft. from the north wall of 
the building, and about 5 ft. from its east wall (f in 
the large plan, Fig. 4). From this, it is clear that all 
Mr. Bowen's walls were internal, and corresponded 
with the sunk area we found. His north wall was 
evidently huilt against the north side of this area, 
where we observed the sudden drop from the gravelly 
floor on the normal level. This would account for the 
height of the wall, 4 ft., reckoned from the bottom of 
the depression ; if reckoned from the normal level, it 
would have protruded above the present surface. It 
also accounts for the south face only being plastered. 
The ridge of natural soil observed about 8 ft. to the 
south may have been connected with the dry wall, 
which, allowing for the thickness of the former wall, 
would leave an interval of about 5 ft. or 6 ft. The 
south wall was evidently also a retaining-waW^ as its 
rough back indicated ; and it, presumably, formed the 
southern limit of the sunk area somewhat to the south 
of our excavation. If so, the second depressed space 
could not have reached the east wall of the building, 
as we found here about 6 ft. of the normal floor. The 
** bench " evidently lay to the south of our trench 
at F, and its vicinity to the end of the drain should 
be noted. 

The "bench" is puzzling; but it so exactly cor- 
responds in construction with the usual substructure 
of a Roman tank,^ that the writer is inclined to regard 
it as the bottom of one. The thin slabs surrounding it 
would be the lower portions of its sides. The red 
stucco of brick and lime was the usual lining of built 
receptacles for water, and the bright red gravelly 
substance which overspread the ** bench" was probably 

^ At Gelljgaer the lower part of a similarly constructed tank was 
fonnd, only each stratum consisted of stones packed together on 
end, instead of laid horizontally as at Cwmbrwyn. Roman Fort of 
Gellygaer, p. 69. 

Digitized by VjOOQ IC 


the decayed stucco of the sides. The wood ashes and 
flue-tiles are suggestive that the sunk spaces between 
the three walls formed a hypocaust. We have here, 
then, all the elements of a small Roman bath, con- 
sisting apparently of two heated rooms (a tepidarium 
and a calidarium), and we may reasonably suppose that 
the space between these and the north end of the 
building also formed one of the suite of bathing 
chambers. As this space had no hypocaust, it would 
be the combined cooling and dressing-room (frigida- 
rium and apodyterium). A careful study of the plan 
(Fig. 4) will convince that the stokehole of the hypo- 
causts could hardly have been otherwise than on their 
west side ; but the external wall here was reduced to 
patches of foundation rubble, too vague to indicate any 
traces of a passage through it ; but just outside its line 
was found a rough structure of several large stones at 
G, which may have been one of the cheeks of the fur- 
nace. The tank described above had a solid bottom, 
so was incapable of being heated ; we must therefore 
regard it as the cold-water plunge of the frigidarium. 
There should be a hot-water alveus, and this was 
normally constructed over the hypocaust and close to 
the furnace. It would therefore be in the second 
heated chamber {i.e., the chamber to the south), and at 
its west end, with the flue of the furnace passing 
under its bottom, which would probably be formed of 
a large flagstone.^ Nothing, however, answering to 
this seems to have been found by Mr. Bowen, nor any 
remains of the suspended floors and the piles which 
supported them, of the heated chambers. Nor did he 
apparently find any of the flue-tiles in position — it is 
very evident that he was not the first to resort to this 
part of the site for building materials ! A bath attached 
to a Roman house rarely, if ever, communicated directly 
with it; and if our conjecture that the space to the 
north was the frigidarium, it is here that we must 

^ Such as the slab at Castell T6ch, p. 177 ; but this is traditionally 
said to have come from the gateway. 

Digitized by VjOOQ IC 


look for the entrance to the suite of rooms. Now the 
intact piece of walling forming the eastern half of its 
north side had a tolerably well-formed square end at H, 
suggestive of the side of a doorway from the little 
yard, B, which, as already intimated, was reached by a 
doorway in its east wall. The frigidaiium entered, its 
opposite side would probably present a wide recess to 
the left, containing a cold-water plunge, and to the 
right a narrow doorway into the little tepidarium. 
Altogether these curious remains — slight as they are — 
are consistent with the hypothesis of a bath of 
thoroughly normal planning. 

We must consider the remains in the enclosed space 

Pig. 11.— Cwmbrwyn : Retaining Wall. 

external to the building just described. The building 
occupied, as already stated, the west side of the enclo- 
sure. The southern adjunct, in spite of the slope of its 
west end to accommodate it to the curved sweep of the 
rampart, encroached upon the inner slope of the latter. 
This necessitated the cutting away of part of this 
slope, and the insertion of a concave retaining-wall, 
18 ft. long (i on Fig. 4), to support the remaining 
portion of the earthwork. This walling was construc- 
ted of large rough stones, with their irregular interspaces 
filled with smaller st/ones, all being bedded in red clay 
instead of mortar (Fig. 11). It still remained to the 
height of nearly 4 ft., and owed its preservation to its 
utilisation for the field-fence. There appears to have 

Digitized by VjOOQ IC 


been a similar revetment towards the opposite end of the 
building at J, where a broken line of large stones was 
brought to light at the foot of the hedge, about 24 ft. 
in length. This was at a suflScient distance from the 
west wall and north-west corner of the building, to 
allow of a passage from the little yard at the north end 
to our presumed stokehole. Between these two re- 
taining walls (i and j) the rampart sloped down to the 
foot of the building. 

Attention is now directed to the rest of the interior 
east of the building. A series of diagonal and other 
trenches proved that the whole central portion was 
devoid of buildings and other structures ; that it was, 
in fact, an open space. The road through the gateway 
continued across it, or rather projected into it, for its 
traces became obscure as the building was approached. 
The track consisted of a spread of rough stones, about 
25 ft. in width, which with little doubt was originally 
gravelled. The old surface on either side disclosed 
patches of finely-broken stone, which suggested that 
the space generally was also thinly gravelled. In the 
more central area, the surface, as a rule, was clean and 
free from finds ; but in the vicinity of the gateway, 
and especially of the building, the soil was discoloured, 
and fragments of slate and pottery were frequent. This 
open space, or yard, extended unimpeded to the por- 
tion of the east rampart south of the gateway and to 
the south-east corner, and hereabouts the old surface 
appears to have been left in its natural condition. 
Elsewhere, as the trenches approached the rampart, a 
different condition of things was observed, which will 
now be described. 

Along the north side, the east side to within a few 
yards of the gateway, and especially within the north- 
east corner, patches of rude paving and dark soil were 
met with between the rampart and the broken line a a. 
Fig. 3. 

Extending eastwards from the northern prolongation 
of the east wall of the building were the remains of a 

Digitized by VjOOQ IC 


slight retaining-wall 6, at the foot of the rampart, 
definitely built at the west end, and becoming a 
tumbled line of stones towards the east. Near the 
north-east curve of the rampart, a line of tumbled 
stones, c, was again met with in this position, and 
between the two, many stones were found which 
may have belonged to this wall. Southward of this 
curve, and also along the inner foot of the rampart, 
was a row of several shallow holes, rf, about 1 ft 
in diameter, and surrounded with stones. Four of 
these holes were observed, three of them 7 ft. apart, 
and the northernmost 14 ft. distant, with some vague 
indications of an intermediate one. They were evidently 
post-holes, and suggestive of a building or shed, of 
which the rough paving in front was the floor. No 
trace of the opposite side of this structure, either in the 
form of post-holes or otherwise, was found ; the only 
indication of its width being the paving, which ceased 
at a distance of about 15 ft. Immediately north of this 
row of holes, and extending into the rampart, was a large 
hole e, filled with large stones, charcoal, and dark earth, 
but it is impossible to say what it was intended for. 
From these slender data it would seem that the yard 
was bordered on the north and on the east to nearly as 
far as the gateway, with a range of timber buildings, 
or sheds, constructed against the rampart. 

On the south side of the yard we again met with a 
line of stones, y, at the foot of the rampart, which 
appeared to relate to a slight retaining-wall. At the 
west end of this was an oval pit, g, 8 ft. 6 ins. deep, 
with its bottom sunk about 2 ft. into the rock. It 
measured across the top 5 ft. 8 ins. east and west, and 
4 ft. north and south, but its sloping sides reduced the 
bottom to 3 ft. by 2 ft. 6 ins. The sides above the 
rock were roughly lined with stone, like a well. The 
filling consisted of soil, dark from the presence of 
charcoal, and many stones. What this pit was used 
for it is impossible to say. It certainly was not a well ; 
and as no drain opened into it, it could hardly have 

Digitized by VjOOQ IC 



been a cesspit. At a short distance northwards of the 
eastern end of this ruined retaining- wall were found the 
remains of a trough-like furnace, or, to be raore precise, 
the flue of a furnace, h, also Fig. 12. It was sunk 
about 9 ins. into the old surface, and the sides and 
west end were built of two courses of stone bedded in 
clay, the east end opening into a shallow excavation 
about 3 ft. across. The internal dimensions of the 
flue were 2 ft. 3 ins. in length, about 1 ft. in width, and 
9 ins. in depth. The floor was of earth, much burnt, 
and the clay in which the stones were bedded, was 
reduced to a crumbly brick-like consistence by the 

Fig. l2.^Cwmbrwyn : Plan of Furnace. {^.) 

action of fire. Much charcoal was also found about the 
site. This flue in its original condition would be covered 
with a structure of stone, surrounding a cauldron or boiler 
of some kind, and it would be stoked from the depres- 
sion. Similar flues have been found at Silchester, Caer- 
went, and elsewhere. 

In the open space in front of the long building, and 
near its south end, was a roughly-pitched area, about 
10 ft. square, i (Fig. 3). It was covered and sur- 
rounded with black earth containing much charcoal; 
and the fact that several pieces of iron cinder or clinker 
were found on the site, renders it probable that it 
was the floor of a small smithy. No trace of an en- 

Digitized by VjOOQ IC 


closing wall was noticed, so that it is probable that 
it was a timber building. On its north side was 
a trail of stones, suggestive of a fallen structure of 
some rough description, About 15 ft. to the east of 
the pitched area was a shallow depression, /, contain- 
ing dark earth, but nothing was observed to indicate 
its use. 

The Finds. 

The fragments of pottery, ornaments, and other 
objects of a more or less portable nature found during 
the exploration were comparatively few, and, with only 
three or four exceptions, of no special interest ; they 
were, however, thoroughly representative of the usual 
finds on Roman sites. They mostly occurred on and 
about the site of the building, and in the fiilling of the 
pit, g, Fig. 3. 

As usual, fragments of pottery predominated, and, 
with one exception, related to the commoner types of 
Roman vessels. They included several pieces, all 
plain, of the lustrous red so-called Samian ware, 
of which one was the bottom ot a shallow patera 
with a ** pushed-up " centre and faint indications of a 
potter s mark, the rest apparently belonging to small 
bowls or cups. The majority of the potsherds were of 
the common grey and black wares, evidently of different 
makes, as their texture and finish differed considerably. 
Several related to the familiar globular jars with out- 
curved lips, almost invariably found on Roman sites ; 
others to shallow dishes, and to bowls with straight 
tapering sides and simple or moulded flanged rims. 
The most interesting potsherd was a fragment of the 
upper part of a cup or cup-like jar, with an eyelet 
handle, as shown in Fig. 13, which also indicates the 
probable form of the vessel. It was of coarse black 
ware, and the eyelet could only have been used for 
suspension, as its aperture was less than ^ in. in 
diameter. These vessels are rather rare, and, if we 
mistake not, are the survival of a British Iron-Age 

Digitized by VjOOQ IC 


form. Several of similar type were found by the late 
General Pitt-Rivers at Rushmore, and are figured on 
Plate XXXIX of his Excavations in Cranhorne Chase 
near Rushmore^ vol. i. 

Several buflF-coloured potsherds were found, five or 
six belonging to mortaria, all with the usual broad 
roll-and-elevated-bead rim. Of a much coarser variety 
of this ware were several pieces of large amphorcB. A 
few pieces of red pottery, resembling that of a modern 
flower-pot, were also met with, and one of these related 














Fig. 13. — CwmbrwyD : Eyelet Handle of Vessel. (J.) 

to a shallow hemispherical bowl of fine texture, with 
an external moulding l^ ins. below its lip. It retained 
some indications of a well-smoothed surfacing. 

Of stone objects the most notable was a portion of 
the upper stone of a quern, of convex form, with a 
beaded shoulder and slightly dished summit. Fig. 14. 
It was carefully shaped out of hard gritstone, 15 ins. in 
diameter, with an **eye" tapering from 2^ ins. across 
the top to 1^ ins. at the bottom. The grinding surface 
was concave, and the handle-hole, in the side. A 
portion of the nether stone of another quern of 
rude workmanship, with a flat grinding surface, was 

Digitized by VjOOQ IC 



also found. In the filling of the pit was a curious 
and very roughly shaped hemispherical stone, about 

Fig. H.—Cwmbrwyn : Portion of Upper Stone of Quern. (J.) 

Fig. 1.5.— Cwmbrwyn : Slate Disc. (J.) 

9 ins. in diameter, with a shallow circular hole or 
socket in its summit. If ins. across. Several slate 

Digitized by VjOOQ IC 



disc3, ranging from 1^ (Fig. 15) to S^ ins. in diameter, 
rudely chipped out of slate, were met with. Similar 
discs, but of pennant grit, have been found on Roman 
sites at Llantwit-Major, Gelljgaer, Ely Racecourse 
(CardiflF), and Caerwent, and they have been regarded 
as covers for vessels or as objects used in some game. 
Whether we can class with these the remarkable disc 
of the local sandstone found in the pit, and shown in 
Fig. 16, is uncertain. It will be observed that its 

Fig. 16.— Cwmbrwyn : Incised Stone 
Disc, (i.) 

Figs. 17 and 18. 

Cwmbrwyn : Spindle-whorls of 

Stone and Slate, (i.) 

upper surface is incised with a wheel-like device, the 
'* spokes " of which radiate from a neatly-formed hemi- 
spherical depression. It has been suggested that it is 
an unfinished spindle-whorl, the central hole of which 
was never completed ; but the depression seems to 
have been purposely made as it is, and the object, 
which is 2f ins. in diameter, is unduly large for a 
spindle-whorl. Three undoubted spindle-whorls were 
found, one of sandstone. Fig. 17, and the other two of 
slate, one of which is shown in Fig. 18. Several whet- 

Digitized by VjOOQ IC 



stones and stones with flattened surfaces, which may 
have been used as rauUers, were also found. 

The only undoubted fragment of a Roman glass 
vessel was a small piece of the upper part of a thin 
colourless beaker, or goblet, about 3 ins. in diameter. 
The inner surface was smooth, but the outer was 
slightly rough, the lip gently curved outwards, and 
the sides were ornamented with shallow horizontal 
grooves about ^ in. wide. Fragments of precisely 
similar vessels have been found at Gelligaer and Oaer- 

Only three bronze objects were yielded by the ex- 
ploration. One of these was a small coin of Carausius 
(a.d. 287-293). It is of a common type (Fig. 19), 

Fig. 19. — Cwmbrwyn : Coin of Carausiua. (|.) 

having the Emperor's bust with a rayed crown to the 
right — IMP caravsivs pp avg, and on the reverse, 
Peace standing to the left, and holding in the right 
hand an olive branch, and in the left a staff or hasta 
pura — pax avg. The coin is patinated, and the figures 
and lettering are singularly sharp, exhibiting little 
sign, if any at all, of wear. It is evident that when 
it was lost it had not long been in circulation. The 
little object, Fig. 20, which was found in the drain, is 
very carefully and neatly shaped. The shank of the 
disc-like stud is hollowed and polished on its upper side 
(as shown in the figure) by wear, and the upper end of 
the curved bar has a fractured surface, showing that 
what remains is only a portion of the original object 
What that object may have been puzzled the writer at 
first, but eventually he concluded that it was the 

Digitized by VjOOQ IC 



handle of a small bronze bucket, or situla, such as have 
been found at Pompeii, but not in Britain, so far as he 
is aware. Mr. Treherne submitted it to Mr. Reginald 
Smith, of the British Museum, who was of the same 
opinion, and considered it to be of early character, 
certainly very early in the first century, a.d., if not 
earlier ; of British rather than Roman workmanship ; 
and made of bronze containing much tin. Another sug- 
gestion is that it is part of a spur, but it seems to 
too slender for that purpose. The third object, a pair 
of tweezers, which was exhibited on the occasion of the 
visit of the Cambrian Archaeological Association, pro- 

Pig. 20.— Cwmbrwyn : Bronze Fragment, (i.) 

bably also came from the drain, for it was picked up 
from the spoil in its vicinity. It is formed of a thin 
band or ribbon of bronze, with embossed decoration on 
the outer side, and, if the writer remembers aright, had 
some traces of gilt. 

The only iron objects which could be identified were a 
few nails, all others being reduced to shapeless masses 
of rust. There were several fragments of much cor- 
roded and oxidized lead, the largest of which was 
a^ thin piece about 3J ins. long, with a longitudinal 
ridge. It was almost certainly a piece of pipe, made 
of sheet lead, the ridge being the soldered edges. It 
was found at the foot of the south rampart. 

Digitized by VjOOQ IC 


A few fragments of animals' bones were met with, 
but as a rule they were too decayed for identification. 
Several belonged to some large animal — perhaps the 
ox or the horse. A few oyster shells were also met 
with. Charcoal was found in comparative abundance 
in many places, as already stated. It occurred freely 
among the stones in the north-east corner of the 
enclosure, suggesting that the timber structures there- 
abouts had been destroyed by fire. Several pieces of 
anthracite were noticed in the filling of the pit, g 
(Fig. 3), and elsewhere. 

Mr. Bowen's diggings in 1890 brought to light three 
supposed inscribed stones, which were briefly described 
by Mr. Tierney in the Welshman. One of these was a 
block, 12 ins. by 7 ins., which was built into one of the 
walls then found. It bore a number of incised markings, 
which this gentleman states in a recent letter to the 
writer had a remote resemblance to Greek characters. 
The late Mr. Alcwyn Evans suggested they were Runes. 
The other two stones were smaller, and were probably 
fragments of a larger block, and they also bore similar 
enigmatical markings. One of these stones — presuma- 
bly one of the latter — was seen by Mr. Laws, who, 
however, regarded the incisions as probably masons' 
marks. It is possible that those of the first stone, at 
least, were rough cursive Roman characters, such as 
one of the inmates of the building may have made in 
an idle hour, but unfortunately these stones are now 
lost. Several stones with incised markings were found 
during the recent excavations, which throw some light 
upon the question. All of these were of fine-grained 
sandstone, and several of the grooves were certainly 
produced by the sharpening of pointed implements of 
some kind, as arrow- or spear-points. Others, however, 
could not have been so produced. On a piece of broken 
stone, for instance, were some markings which appeared 
to be part of a rude cruciform device, with the limbs 
terminating in, or rather intersecting, semicirclea It 
was such a device as any one in any period, attracted 

Digitized by VjOOQ IC 


by the smooth surface of a stone in a wall, might have 
made to while away the time. 

The exploration naturally suggests the question, 
what are these remains at Cwmbrwyn ? That they 
are wholly Roman seems beyond all doubt. Nothing 
was disclosed to indicate that the site was British, and 
was afterwards adopted and modified by the Romans ; 
nor that it continued to be occupied after the Roman 
period. But the nature of this occupancy is by no 
means clear. The strong defensive works, the planning 
of the building, and the crudeness of the structures, are 
not consistent with the hypothesis of a " villa ; " they 
are, on the contrary, suggestive of a fortified post of 
some sort, in spite of the remarkable divergence of the 
form of the site from the normal rectangularity of a 
Roman fort. But the numerous Roman forts which 
have been more or less completely explored in this 
country — leaving out of the question the large 
legionary stations like Caerleon, Chester, and Lin- 
coln — were on a larger scale than Cwmbrwyn, and 
were designed to hold a cohort, whether small or 
large, whether consisting of six centuries, as at Gelly- 
gaer, or nominally a thousand strong, as at House- 
steads. In these, each century was housed in a 
narrow building from 120 to 140 ft. or more long; 
but at Cwmbrwyn we have only one building com- 
parable with these. May we infer from this that it 
was a fortlet designed to hold a small detachment 
consisting of a century ? But the fortlets comparable 
in size with it, such as the few described in Watkin's 
Roman Lancashire^ or even the smaller mile-castles on 
the Wall of Hadrian, are square, which makes the 
abnormal form of Cwmbrwyn all the more remarkable. 
Is it possible that it was a mansio or a mutatio ? 
Unfortunately, we know nothing of the planning of 
these posting stations, but presumably they were 
fortified. The open yard in front of the long building 
would be convenient to receive vehicles and baggage, 

Digitized by VjOOQ IC 


and the timber structures on the north and east sides 
would provide accomraodation for relays of horses. But 
against this hypothesis must be urged the apparent 
absence of any important Roman road in the vicinity. 
Perhaps it was the outpost of a Roman fort, the 
remains of which occupy a typical position on fhe 
banks of the Tk£ two miles to the N.N.E., and are to 
be seen on the left side of the road between Llandowror 
and St. Clears Station. Cwmbrwyn is admirably 
placed for a detachment of cavalry to thwart a hostile 
landing from the estuary of that river. 

Did we but know the exact age of the remains, it 
might help to a clue. The fresh condition of the coin 
of Carausius shows that the place was in occupa- 
tion during— or at least shortly after — that emperor's 
time, but it does not prove that it had not already 
been long occupied. On the other hand, the absence 
of any signs of rebuilding, and the general paucity of 
the ^' finds," militates against a long occupation. All 
we can say is, that the site was in use in comparatively 
late Roman times ; and we know that in late times 
changed military conditions had Mnrought changes in 
fortifications. Traditional forms and arrangements 
were no longer strictly adhered to. 

It will be evident to the reader that Cwmbrwyn is, 
— and is likely to become more so — a point of great 
interest in Romano-British archaeology, and that all 
archaeologists will congratulate the new Carmarthen- 
shire Antiquarian Society on this, their first important 
work of investigation. Little is known of Roman Car- 
marthenshire, and this county offers a fertile field ot 
research, to which this Society will do well to give 
special attention. In conclusion, both Society and 
writer are indebted to several others besides the gen- • 
tlemen whose assistance has been referred to in the 
foregoing pages. A special word of acknowledgment 
is due to Mr. Bowen and his family for their patriotism 
and kind hospitality, and to Mr. W. Jeremy, of 
Laugharne, for a series of photographs, which provide 

Digitized by VjOOQ IC 


a record of the work at various stages ; and the hearty 
co-operation of the diggers and their foreman, Mr. E. 
Evans, of Greenbridge, all of whom laboured with 
intelligence and care, both facilitated the work and 
materially helped to bring it to a successful issue. 


By T. C. CANTRILL, RScLond. 

Oeographical. — The remains described above lie on the 
northern slopes of a range of hills which here form the coast of 
Carmarthenshire. Prom Laugharne the range extends for some 
miles westwards, through Llansadurnen, Eglwys-Cymmyn, and 
Tavemspite towards Narberth ; the plateau which forms its 
top attains an elevation of some 500 ft. or 600 ft. above sea- 
level. On the south it presents an almost unbroken front, 
where it rises sharply from the coastal alluvium of Laugharne 
Marsh; on the north it is trenched by several deep "cwms," 
between which the plateau descends in stages towards the 
valley of the T&f The Koman remains stand at an altitude 
of 390 ft on one of these minor plateaux, between the dingle 
of Cwm-bn^n on the east and that of Las-f4ch on the 
west Southwards the site is dominated by the higher ground 
of Castle-tdch ; northwards the surface falls away towards Llan- 

These physical features do not, however, appear to throw 
any light on the position selected; we must suppose a road 
of some sort passed near the place, yet any main line of 
road following the coast might have been expected to traverse 
the district along the crest of the hills, as does the present 
road from Laugharne past Three Lords' Bushes towards Mar- 
ros. Such a road would have avoided crossing the deep 
dingles which cut into the north side of the plateau. It is 
probable, therefore, that the building was approached by a 
minor road, the course of which has still to be traced. 

The Site, — The country-rock is the Eed Marl subdivision of 
the Lower Old Eed Sandstone, and consists of a great thick- 
ness of red and chocolate-brown marl, with occasional beds 
of sandstone of divers characters. Some of the marls contain 
twigs and pellets of grey calcareous matter ("race") of in- 

6th 8KB., VOL. VII. 14 

Digitized by VjOOQ IC 


organic origin ; not infrequently these have been dissolved 
out, leaving a loose residue of brown earthy matter. The 
occasional bands of sandstone vary much in character : one 
type is micaceous, soft, fine-grained, and flaggy, and readily 
splits along the bedding-planes. These micaceous sandstones 
are generally purplish-red in colour, but are occasionally pale 
green. Another type is coarse, open-grained, gritty, and often 
pebbly ; this is generally brown or greenish-grey, and not 

The prevailing dip of the strata is a little west of south 
at an angle diminishing from 50 deg. a few hundred yards 
north of Cwm-bn^yn to 5 deg. a mile to the south. Under the 
Roman site the dip is about 10 deg. 

The bed-rock immediately underlying the site is red marl 
of the ordinary type, with some wlcareous bands. The rock 
was reached in the exploratory trench cut across the vallum 
and fosse north of the gateway, showing that the fosse had 
been at this point cut down through the subsoil and several 
feet into the solid rock. Also, the excavations showed rock 
under about 7 ft. of natural rubble in a pit at the south- 
west part of the site. 

The subsoil covering the site is the direct product of the 
weathering of the underlying rocks ; these break down into 
angular lumps which, when traced upwards, are seen to grow 
smaller and smaller, and to be embedded in an increasing 
amount of fine loamy material till the actual surface-soil is 
reached. This is often paler in colour than the subsoil, owing 
to the bleaching action of vegetable acids. In the case under 
description, the subsoil consists of a red loamy material, full of 
small angular pieces of marl and sandstone, chiefly red ; about 
the middle of the area the subsoil contains much ddbris of green, 
fine-grained sandstone, a thin band of which no doubt crops 
out at that position. Fragments of a similar green sand- 
stone are present in the concrete found on the site. There 
is an entire absence of any drifted materials — boulder-clay, 
gravel, or sand ; the subsoil and soil have been derived imme- 
diately from the subjacent rocks. 

The Building - Stones, — These consist of rough undressed 
blocks and slabs of coarse grit, medium-grained sandstone, 
flaggy sandstone, and marl, all of which could be obtained 
from the Old Red Sandstone of the neighbourhood. The 
blocks were undoubtedly derived from different beds, and not 
from a quarry in any one bed, though some may have come 

Digitized by VjOOQ IC 


from the bottom of the fosse ; in fact, it would seem that 
the builders explored the whole neighbourhood for a mile or 
so around, and gathered from the surface whatever lumps of 
rock met their notice, or could be extracted from the soil 
with little trouble. This would account also for the weathered 
appearance and absence of sharp angles which characterise 
most of the blocks. One lump of vein-quartz was noticed ; 
also a piece — 6 ins. long — of under-burnt limestone, in the wall 
at the south-west angle of the building. Limestone could 
be obtained from the Carboniferous Limestone of Coygan, 
or Pendine, each about 2| miles from Cwm-br>^n. 
The blocks and slabs range up to 3 ft. in length. 

Other Stones. — Two stones of exceptional character — possibly 
used in some grinding or pulverising operations — were exposed 
within the area. One is a water- worn pebble, roughly tri- 
angular in form, 5 ins. thick, with sides about 2 ft. long ; it 
consists of fine-grained, hard brown sandstone. The other 
is similarly water-worn, is somewhat almond-shaped, 1 ft. 
IJ ins. long, 6 J ins. across, and 2 J ins. thick, and is of rock 
similar to the other. Probably both are beach-pebbles from 
the coast west of Pendine, along which various hard sand- 
stones of such a character crop out in the Millstone Grit and 
Coal Measures. Several small rounded pebbles of sandstone, 
ranging up to 4 ins. in length, were noticed ; these were cer- 
tainly brought to the site for some definite purpose. 

The Boofing'SLaies, — The roofing slates are of two varieties : 
a pale greenish-grey slate of granular texture, and a blue- 
black slate of finer material and somewhat silky lustre. There 
is little doubt that the first has been obtained from a bed 
of volcanic ash interbedded with the Didymograptits-hijidus 
Beds of the Arenig Series, a subdivision of the local Ordovician 
System of rocks. The second variety is probably attributable 
to the Uidymograptus-bijidtcs shales themselves. The beds 
which would yield such slates crop out over a large area 
along the southern slopes of the Preseley range of hills in 
North Pembrokeshire, and some 10 miles north-west of Cwm- 
br>^yn. In the present poverty of our knowledge of the 
geology of that region, it is not possible to specify more 
precisely the actual localities from which the slates could 
have been derived. The district in question is traversed by 
the Via Julia on its way from Carmarthen to St. David's, so 
that a knowledge of the existence of such materials was pro- 
bably soon acquired by the Eoman settlers. 


Digitized by VjOOQ IC 



Coal, — Numerous fragments of anthracite, both burnt and 
unburnt, were found associated with that part of the building 
enclosed in the dot-and-dash line in the plan on Plate II. 
The nearest points where such coal crops out are the Gwen- 
draeth Valley district, near Kidwelly, 10 miles to the east, and 
the Amroth district, about 6 miles to the south-west. The 
latter is the more accessible, and was probably the source of 
this fuel 

Digitized by VjOOQ IC 


eambrtan larrljaeological lasdoctatton. 


{Continued from page 128.) 



Boute. — The members assembled in the Guildhall Square at 
9 A.M., and were conveyed by carriage along the road on the west 
side of the River Towy, 8 miles in a south-west direction to 

The return journey was through Llanybri to Trefenty (5 miles 
north-west of Llanstephan) ; thence 2^ miles north to the new 
church of Llanfihangel Abercowin ; and then back along the St. 
Clears road to Carmarthen, a distance of 8 miles east. 

The members were entertained to luncheon at Plas, Llanstephan, 
by the President, Sir John Williams, Bart., and to tea at Trefenty 
by the Rev. W. Davies, Vicar of Llanfihangel. 

Ystrad House. — The first stop was at Ystrad House, the residence 
of Major and Mrs. Evans. Here the sculptured base of a font, 
supposed to have belonged to the demolished Church of St. John, 
Carmarthen, was inspected, photographed, and sketched. It is 
3 ft. 2 ins. high, with four sides 9^ ins. square, each side bearing a 
carved figure under a canopy. These represent respectively a 
mitred bishop, St. Peter with the keys (P), the builder with a model 
of the church in his hands, and a worn-out figure of a person with 
hands uplifted to bless. The other stone on the lawn was supposed 
by local archaeologists to be a public Roman altar, but the general 
opinion of the experts was that it was the capital of a column. It 
was dug up between St. Peter's Church and Priory Street, and 
removed by Mr. John Jones, M.P., to Ystrad about 1830. 

Llanstephan Church. — Llanstephan was reached soon after eleven 
o'clock, and in the Church the Rev. J. M. James read an exceed- 
ingly interesting Paper regarding that edifice. At its conclusion, 
Archdeacon Thomas, in proposing a vote of thanks, said they were 

Digitized by VjOOQ IC 


reminded that they were very near Pembrokeshire by seeing the arches 
cnt out of the walls, a featare very common in that coanty. There 
was also a sqnint in the north wall, by which those who sat in the 
transept won Id be able to see the elevation of the Host. 

Llanstephan Castle. — From the Church to the Castle the walk 

was interrupted by a heavy shower of rain, bnt reaching the rnin 

the party entered an ancient room, where Colonel W. LI. Morgan 

gave particulars as to the architecture of the building. They had 

heard, he said, that the Castle was last destroyed in 1256. He 

divided the Castle into two parts — the old and the new — and 1256 

was the dividing date between the two. The new Castle was built 

after 1256, and the question was how much of the old Castle 

remained to the present day. They had heard the previous night of 

the number of times the Castle was destroyed. They would see 

outside there was undoubtedly part of the older Castle in existence. 

The lower part of the walls was undoubtedly different from any of 

the newer part. It was impossible to tell from the masonry whether 

it was earlier or not than 1256 ; but as there was work which 

undoubtedly was done after that particular date, he thought they 

would all agree that earlier than that would belong to the old 

Castlo that was so often burnt and destroyed. The wall of what 

had been called — erroneously he thought — the "keep'* was the old 

Castle of Llanstephan. From outside they would see the outline ot 

the wall most distinctly, far better than on the insida There might 

be some question whether the square tower was old or not. In 

those early days they did baild towers, but all he had known had 

been very dififerent from this. He thought it represented the old 

peel tower of the Castle. The greater part of the wall had been 

destroyed, but they could see where it went into the ground, and 

the area given, although small, would still represent the Castle of 

those days. Inside the Castle there were alterations made when it 

was restored, and most peculiar and distinctive work they found 

there. When the Castle was restored, the place in which they were 

standing was the main gateway. They could see the remains of 

the portcullis, and outside it was defended by a shoot extremely 

well preserved, and the lancet loops were very characteristic 

of the reign of Henry III, or at all events early in that of 

Edward I. Lancet loops were quite as characteristic as tracery in 

the windows. It would be 1270 most likely when the Castle was 

restored. There were several points of great interest about this 

that he should like to argue out. The room up above was a rather 

fine room, but insignificant, and had got a very large fireplace in it, 

and therefore, he thought, the kitchen. The top room of all was a 

magnificent room, and was the State hall. They would find a very 

good fireplace indeed, witli beautiful carvings of the Early- Englisli 

date. There were two windows ; the tracery of one was well 

preserved, and the other had been destroyed. It was characteristic 

of between 1270 and 1300. The rooms each side of where they 

Digitized by VjOOQ IC 


were standing were guard-rooms, and that comprised the main 

At some other time either the garrison had been reduced, or tbej 
did not want to maintain the number of men necessary to keep the 
gateway going, so they started a new gateway on the other side 
that could be worked and guarded by a very much less number of 
men. When that was done was quite problematical. He certainly 
thought from the arrangement of the gateway it was not very long 
after the Castle was actually restored — probably within 100 years. 
After the original owners had died out, and it had got into the 
hands of the Crown, the place would be left in the hands of a 
Gt)yernor, who turned that main gateway into his own apartments. 
He blocked it up by two walls. It was totally incompatible with 
even tbeir Welsh taste that that could have been done when a good 
owner was in the Castle. It seemed a very flimsy way of doing it. 
The next tower had been called the chapel tower ; the upper room had 
got good windows in it, and it had all the appearance of a chapel, 
except that it had got a fireplace in it. There was no other place 
that could have been a chapel of the Castle, and there must have 
been a chapel, therefore he thought that might have been it. The 
room underneath was the priest's room. The tower on the right 
was the guard-room. It was a very remarkable thing how very 
well this Castle was laid out, because the ramparts and also the 
roof could be sentried and guarded by men from the guard-room 
without any connection whatever with the state-rooms. Those were 
apparently kept quite distinct. The sentry walk was quite distinct 
from the walk to the chapel tower, and that was why he thought it 
must be the chapel, because they could go from the state-rooms to 
the chapel without going into the inner court. The further tower 
had no roof to it, but was put there t*o guard the angle of the 
chapel. Several points of that wall were worth going to see. The 
Castle was surrounded by a very good ditch, and on the outside to 
the west there were some very fine earthworks. The date of them 
was rather problematical. They would naturally think they were 
put up in the time of the Civil War, but they found so many of 
them in these Welsh castles that played no part in the Civil War 
that they must be condemned for that purpose. He was not sure 
that they were not of the time of Owen Glyndwr, but they seem to 
be intimately connected with the building of the Castle. If they 
looked at Buck's print, 1740, it showed a very different state to 
now. He showed a fine wall outside that would really turn that 
part of the Castle into a concentric castle, fie thought he was 
right there, and that these outworks were really coeval with the 
date of the restoration of the Castle. The only thing he could not 
understand was where they got their drinking water. It seemed to 
him utterly impossible that they could have had water at all. 
There might have been a tank underneath that building; but if 
there was the only place was in one of the gpmrd-rooms, which 
showed some approach to a tank. 

Digitized by VjOOQ IC 


The Pilgrims' Church at Llanfihangel Abercowin. — The mined 
oharch is situated a quarter of a mile south-east of Trefentj, in the 
angle formed by the junction of the Afon Oywyn with the fiiver 
T4f. On arrival at the church, the Rev. W. Davies, vicar, read the 
following Paper : — 

The time at my disposal here this afternoon is limited, though I 
have much ground to cover. However, my Paper will not be 
long, and my remarks brief and concise. I believe this is the first 
time the Cambrian Arch»ological Association, or any other body of 
experts, have ever coma to thb interesting old Church of Llanfi- 
hangel Abercowin, which is, as you see, in ruins. So we feel our- 
ourselves greatly honoured by having a visit, in a very out-of-the- 
way place, from such a distinguished company as we have here this 

The church is known by another name, the Pilgrim's Church, in 
consequence of the legend connected with the pilgrims who are 
supposed to lie buried here. As the name of the parish signifies, 
the church is dedicated to St. Michael. The Welsh prefix " Llan" 
means an enclosure. In Welsh place-names it generally signifies a 
church, probably including the churchyard. The name Abercowin 
is evidently added, from its position at the mouth, or estuary, of the 
River Cowin. Aber means the confluence of a smaller river into a 
larger one, or any river entering the sea. According to Place-Names 
in Wales, the River Cowin, or Cy wyn, flows into the River T&f at 
the place, hence the name. The popular word Cumu — rising — comes 
from the verb Cywynu — to rise, to mount up. The water at the 
month of the river rises twice a day by means of the tide from the 
sea. Churches dedicated to St Michael were often celebrated places 
for pilgrimages ; hence, perhaps, the great attraction for pilgrimages 
to Llanfihangel Abercowin, to implore the aid of angels in times of 
persecution, and also of destitution. 

The fact of the church being dedicated to Mihangel, or St. 
Michael, is one presumption of its great antiquity. Probably it 
takes us back to a period immediately succeeding the year 700. 
Churches and parishes dedicated to St. Michael represent the later 
Christianising of districts which lay out of the beaten track, in 
places inaccessible by reason of their mountainous or marshy char- 
acter. Even the ruin in which we are gathered hardly represents 
the first Christian building. In those early days the churches were 
made of wattle, or wicker-work, covered with mud. So, when and 
by whom the church was founded is enveloped in the mist of the 
far and distant ages. It is not improbable that the foundation leads 
back to one of the early centuries, when Christianity and Roman 
occupation marched together amongst the early Britons. 

I am not going to describe the architecture of the building, nor 
call your attention to the difierent historical features that are in it, 
and that for two reasons. 

In the first place, time will not allow me, and in the second place, 
Mr. D. C. Evans, F. O. S., of St. Clears, has kindly done so in an 

Digitized by VjOOQ IC 


able and interesting Paper, which I find is distributed among the 
members here this afternoon. I personally feel mach indebted to 
Mr. Evans for his kindness in taking off a good share of the work 
that would naturally have devolved upon myself. 

The great attraction to this church — which causes so many visits 
to it — is the three pilgrims* tombs in the churchyard ; hence the 
name ** Pilgrims' Church'* given to it 

In Black's Guide to Wales the following reference is made to the 
graves : — " Tradition relates that three holy palmers, meeting here 
in great destitution, prepared three graves, agreeing that two should 
be put to death, and that the third, after burying them, should lie 
down in the remaining grave, and pull over him a large stone : and 
this was done. As far as I am aware, there is no historical evi- 
dence for the story ; it is tradition, but tradition often carries 
some truth with it" 

For the reasons already given, I am not going to explain the 
monumental slabs, with their effigies, and the carved symbols of the 
respective trade-guilds to which probably they may have belonged. 
Besides, I hope light will be thrown upon those points by some who 
are present this afternoon, and well versed in such subjects. There 
are more graves than these three, in which pilgrims are said to have 
been buried. The reader of this Paper discovered two more twenty 
years ago, having practically been covered under the open surface 
of the soil, which, at that time, was also overgrown with weeds and 

Besides the five recumbent monuments lying in the churchyard, 
there is a much smaller one, which was also found by the writer, in 
the hedge on the east side of the church. This is now lying within 
the ruined walls of the church. 

There is another curious local tradition prevalent, to the effect 
that if the pilgrims' graves were disturbed or neglected, that the 
peninsula on which the church and churchyard are situated would 
become infested with venomous reptiles. There seems to be some 
truth in this tradition, for it is a well-known fact in the parish that 
during the dark age in the history of the place, venomous reptiles 
were so numerous in the churchyard that they were a living plagae 
and a terror to any who might come near. But we are in a position 
to know that now, since the peaceful repose of the deed has been 
asserted, and the graves of the holy palmers restored and properly 
looked after, these venomous reptiles have almost, if not entirely, 
disappeared. So no one present need get alarmed by hearing the 
story of the snakes. 

Apart from its architecture and legendary associations, however, 
the ancient edifice possesses an abiding interest for many. The 
Rev. Thomas Charles, of Bala, was born in the parish at a farm- 
house called Pantdwfn, about a mile distant, between here and St. 
Clears. At the old font, which used to be here, he was baptised on 
October 26th, 1755, when twelve days old. From the pulpit, the 
base of which can be seen on the south side, close to the arch leading 

Digitized by VjOOQ IC 


into the chancel (and on which I stand at present) the same Thomas 
Charles preached his first recorded sermon on Sunday, August 16th, 
1778, after his ordination on the previous Trinity Sunday. The 
English and Welsh Bibles that were on the pulpit on the occasion, 
together with the English and Welsh Prayer Books that were in 
use at the time, are still in my possession. The English Bible bears 
the date of 1680, and so it is 226 years old ; the Welsh Bible, 1690 ; 
the English Prayer Book, 1768; and the Welsh Prayer Book, 1770. 

There is an old chalice still in existence which bears this inscrip- 
tion : — ** Poculum eclessie de Llanfihangel Abercowin.'' No date is 
given on it; but, according to the opinion of two gentlemen who 
have lately seen it, and who are well versed in church plate, it is a 
1574 one, and so 332 years old. 

With regard to the old font to which I have just referred, it is a 
Norman one, of much beauty, as well as of great antiquity. It was 
removed from here in 1848, and is now placed in the new church, 
and is still in use. Those who may wish to inspect it shall have an 
opportunity of doing so on their way home this evening. It is 
supposed to be 800 years old. The old books, the old chalice, and 
other church plate may also be seen at the new church. 

Some here may naturally ask, Why has this old church been 
allowed to get into and remain in this ruinous state ? I will try and 
explain. In the year 1848, Mr. Richard Richards, of Trecadwgan, 
in this parish, built at his sole expense a new church some three 
miles distant, in a more populous and a more central part of the 
pai-ish, for the convenience of the people. As this new church was 
Rubstituted for the old one, and became in every sense the parish 
church, the old building was neglected, and at last abandoned, as 
you see it now. It has been crumbling gradually, and falling into 
decay for the last fifty-eight years, and so far no attempt has been 
made for its restoration, inasmuch as its services are no longer 
required since the building of the new church. The new church 
was consecrated on October 3rd, 1848, by the late Bishop Thirlwall, 
and no services have been held here since : with the exception of the 
memorial service, which is held annually in the open air, either on 
the last Sunday or the last but one in July. The first of these 
services was started in 1882, and this year marked the completion of 
the first quarter-century of its existence. This is a very popular 
institution, and people for miles around look forward to it every 
year. Hundreds of people gather together from a wide area, and 
the scene in and around the roofless edifice is most impressive and 
unique in character. And the reverent conduct of the people during 
these services is a feature to be greatly commended, and it is to be 
hoped will be continued for future generations. But something is 
intended to be done before very long to prevent the falling of the 
old building into further state of decay. Mr. Weir, from the Society 
for the Preservation of Ancient Buildings, visited the place two 
years ago, and reported thereon. Mr. Weir's report, we are pleased 
to state, meets with the Society's entire approval. It is not intended 

Digitized by VjOOQ IC 


to restore the old church: only prevent the walls and the tower from 
suffering further from the destructive ravages of the elements. The 
approximate cost of the necessary repairs to the fabric, pointing the 
walls with cement, and protecting top surface, would probably 
amount to the sum of ^100. So far, only a few pounds are now in 
hand, being the profits from the sale of some views connected with 
the old place. And we cannot expect much help (in the way of 
money) for the repairs, only from those who take interest in such old 
relics of the past, and are desirous of preserving them from total 
oblivion. Early next year Mr. Clark, of Llandaff, intends to make 
casts of the stones which are on the pilgrims* graves, with the 
object of placing them in the Welsh National Museum at Cardiff. 
Also, we have been advised by experts that the stones should after- 
wards be fixed in their place in concrete, so as to avoid the possibility 
of their being lost. They (the stones) are unique throughout Wales, 
and have a national value. 

The field on the east side of the church is called Paro-y- Parsonage, 
in which may be seen to-day raised embankments running in dif- 
ferent directions. There is a tradition that there was once a village 
here, and the raised embankments may bear the traces of the 
buildings and the garden enclosures. It is quite possible that there 
was also a parsonage-house somewhere in the field, hence the name 
Parc-y- Parsonaga 

Two fields distant, on the north-west side, there is an ancient and 
historic farmhouse, called Trefenty. In its very centre may be seen 
to-day the old passage through which it is said the parishioners had 
to pass on their way to church, and there record their names. 
Probably this was a device to preserve private interests The 
people were asked to go through it only on special occasions, and 
once a year. There was no right-of-way ; and to prevent that 
being created, the people had to be occasionally subject to the yoke of 
the passage. The funeral processions passed through for the same 
reason, and the offertory was taken there at one time at least, but 
not during the latter half of the eighteenth century. 

For carrying the dead to the churchyard, instead of the bier or 
hearse of to-day, the parish possessed a horse-bier, horse-litter, or 
corpse-litter — in Welsh elorfeirch — which occurs in the Bible in the 
last chapter of Isaiah. A description of this horse-litter is given in 
the Arc/i, Camb, for last April, p. 136: "It had long arms, or 
shafts, behind and before, into which the horses were put, one in 
each shaft, and secured by specially-made gear.'* Some people who 
lived in the parish fifty or sixty years ago remember speaking to 
old people who had seen this horse-litter in use, though for many years 
previous it had gone out of use. I believe it was peculiar to this 
parish only in this part of the country. It was generally used in 
some district in North Wales. This kind of bier was very necessary, 
not only because of the long distances (for some funerals came 
many miles outside the parish), but also because of the badness of 
the roads of those days. 

Digitized by VjOOQ IC 


Trefenty farmhouse is sapposed to ocoapj the site of a moaasterj. 
(Curtis, 253.) 

I find the time is going, and I shall only touch very briefly on a 
few more points of interest 

Opposite the front of the house, i.e., Trefenty, there is a site of an 
old encampment, or earthwork. Here, tradition says, a great 
battle was once fought. In the same field, within living memory, 
there was to be seen an entrance to a subterranean passage, or 
a fine arched cave, which was supposed to pass to Laugharne 
others say to Llangunnock, and even so far as Abergwili. The 
mouth of this cave was closed some sixty-five years ago, being a 
constant danger to animals to fall into. I might have dwelt on the 
beautiful well that was once on the north side, inside the churchyard, 
and on the tradition attached to the same, (it was closed in my 
time, some twenty years ago) ; on the state of the fences here 
twenty-five years ago, and the large sums of money that have from 
time to time been spent on the same, in order to keep out the 
animals, which were once allowed to graze here ; on the briars, 
thorns, and nettles that grew in wild luxuriance ; on clearing and 
levelling the ground, and planting it with ornamental trees and 
shrubs ; on the number of headstones that were found in the 
hedges and ditches and in the surrounding farmhouses, but which 
now have been replaced in God*6 Acre. 

I had better not dwell any longer on this matter — it is such a sad 
history. The church and churchyard pi*esented a picture of great 
neglect and desolation not easily imagined. However, I think I 
ought to call your attention to the church of St. Teilo, Llandeilo- 
Abercowin, which stands on the opposite side of the river, and 
which some have visited this afternoon, or at least intended to do 
so. It is a plain building, but is supposed to be very old — older 
than this one. The renewed GriflBth Jones held the rectory, to^ 
gether with that of Llandowror, for nearly fifty years. There is an 
ancient building near the church which bears marks of great age. 
The lower part of the building is now used for a dairy. 

In conclusion, I wish to say that what we see left of this old 
building can only be described as a fragment, yet a fragment 
that testifies to the grandeur of the building in ages past. History, 
written and unwritten, bears ample testimony to the fact; the 
church and churchyard have truly shared in the joys and sorrows of 
the parishioners for centuries, and afibrded them, regardless of 
wordly rank and station, a peaceful resting-place from their various 
labours — and " May they rest in peace.*' 

Archdeacon Thomas said the church would originally be an 
oratory on a pilgrim-route. They were close to the Laugharne 
river ; and although he did not know the geography of the place, 
he had very little doubt there was a pilgrim-road running from 
there probably to St. David's, and taking Whitland on the way. 

Rev. J. Thomas, Laugharne, said there was a *' H^n Ff ordd " 
leading down to the river. 

Digitized by VjOOQ IC 


Archdeacon Thomas went, on to say that the stones belonged to 
pre-Norman days, and he took it that the font to which their 
attention had been drawn was earlier than Norman. It had, of 
course, the ronnd circles and the subdivision into the acnte arch, 
bat there were many features about it that belonged to the earlier 
period, and were more characteristic of the ornamentation they 
found in Anglo-Saxon drawings and carvings. He thought their 
attention was not drawn to the two brackets on the east wall 
under the window, where probably there might have been altar- 
lights, or possibly images ; but on both sides they noticed there was 
a high recess, and that on the north side, he fancied, and that on the 
south, formerly contained shrines. The remark made about the 
field on the right-hand side being called Parc-y- Parsonage rather 
implied that in the early days, when Palmers passed that way they 
must have had lodgings, and if they could only dig below the 
surface they might find the foundations of the Palmer's houses. 

The question was asked whether there was any evidence that the 
Palmers were bound for Whitland or St. David's, and Archdeacon 
Thomas said St. David's would be the ultimate destination. In 
North Wales they had pilgrim roads in many parts towards the 
Island of the Blest — Bardsey. Witli regard to the horse-bier, there 
were two in existence : one between Towyn and Barmouth, in 
a church now disused ; and another in a church on the banks of 
Bala Lake, looking exactly as described in the Paper. 

Mr. E. Laws remarked that the headstones they had examined 
were, to his mind, the most valuable seen in Wales. Those on the 
top had beasts on them, something resembling the ones at Penally. 
Those at the bottom had got mounted men. He believed if they 
looked at them they would see they were things which ought to be 
very carefully copied for their Joui^nal, 

Rev. W. Da vies said there would be casts taken of them by next 

Mrs. Allen said she was there fifty years ago, when the roof .was 
on the buildiTig, and the gravestones were not in the place where 
they are now. There were three gravestones to the west of the 
tower, then in an upright position, and not as they are now in 
divisions. She thought she had a drawing somewhere of the church, 
with the roof on, that she made. She rode there on horseback 
to see the pilgrims' graves, and it made an impression on her. 

Archdeacon Thomas supposed the pilgrim stones, as they were 
now, represented a good many more than three. 

Mrs. Allen : There were three upright then. 

Bev. W. Da vies said some people in the parish remembered sixty 
years ago, and he never heard of the upright stones from them. 

Colonel G Wynne Hughes : Is there any idea of the approximate 
age of the stones ? 

Archdeacon Thomas : If I ventured to guess, I would say about 
the year 800. 

The party then proceeded to farther inspect the interesting stones, 

Digitized by VjOOQ IC 


and subsequently left Trefenty, where they had had tea, by the kind 
inritation of the Vicar and Mrs. Davies, for the new church of 
Llanfihangel Abercowin, where the Norman font of the old church 
was seen, and rubbings taken of ifc; also the ancient Bibles and 
Prayer Books referred to in the Vicar's Paper, as well as the church 
plate. The drive home was on the St. Clears road, passing at a 
distance Bwl-y-Seiri (a British camp), Castell-y-Qtier, Derllys Court, 
and Llanllwch. 

With regard to the fund started for the preservation of the ruins 
of the Pilgrims' Church, we may state that there was collected on 
the spot, £5 158. ; and donations have since come to hand of 
£3 6s. 9(£., making the total £9 Is. 9d. Included among the donors 
were Rev. J. G. Swainson, M.A., £1 lOs, ; Rector of Wistanstow, 
£1 1$. ; Mrs. Pnghe Evans, £1 ; "Antiquary," £1 ; Mr. Stepney- 
Gulston, 10s ; and Mr. Foulkes Roberts, Denbigh^ 10«. Besides 
these, £2 78. 8d. was realised by the sale of views connected with 
the old church, sold on the day, making a grand total of £11 14«. 5d, : 
an excellent start of a fund for a very deserving object. Sab- 
scriptions may be sent to Mr. P. J. Wheldon, National Provincial 
Bank, Carmarthen (Treasurer) ; or to the Rev. W. Davies, Vicar of 
LlanBhangel- Abercowin, St. Clears. 

The following drawings by Mr. D. C. Evans, F.G.S., of St. Clears, 
were exhibited in the Temporary Museum formed during the 
meeting : — 
Sketches of " Pilgrim Stones," Llandowror : 

a. Grave i. Face. b. Grave. Reverse. c. Grave ii. 

[These two stones had been taken out of the ground for this 
visit; the sketches show the entire stones.] 
Sketches of " Pilgrim Stones,*' Llanfihangel-aber-Cywyn : 

a. Grave i. 

[E£5gy with crossed arms; headstone with two concentric 
circles; plain footstone, modern.] 

b. Grave ii. 

[Effigy with crossed arms ; on either side of head, an animal, 
left, a stag (?) or goat (?) ; right, a stag-hound ; right 
hand grasps a javelin or spear; headstone with cross- 
formed in raised circle ; footstone weathered.] 

c. Grave iii. 

[Coped slab ; along ridge lies main beam of a calvary ; head- 
stone has a wheel cross ; part of footstone missing.] 

d. Grave iv. 

[Slab almost identically the same as iii ; headstone has plain 
face, edge ornamented with lines in chevron pattern ; foot- 
stone very dilapidated.] 
Grave v. 

[Seems to be incorrect copy of No. i, much broken, portion 
carrying head is missing; head- and foot-stones each bear 
figure of man or woman on horseback ; edges ornamented 
with cable pattern.] 

Digitized by VjOOQ IC 



Grave yi. 

[Small stone, now loose in nave of rained charoh ; figare of 
a child in long robe, bar across the hips. All these stones 
are fully described by the draughtsman ^Mr. D. C. E7ans 
— and these sketches are reproduced, with the articles, in 
Transactions of C. A. S., vol. ii.] 

Fig. 1. — Norman Font from the Old Church of Llanfihangel Abercowin, 
now removed to the New Church. 

The New Church of Llanfihangel Abercowin. — This is situated 
3 miles north of the old church, on the high road from St. Clears 
to Carmarthen, at the point where the branch road from Trefenty 

Digitized by VjOOQ IC 


joins it. The new church was built in 1848, in the revived Gothic 
stjle of that period. The only object of interest here is the arcaded 
Norman font (Fig. 1) removed from the old church. 


Boute. — The members assembled in the Guildhall Square at 
8.45 A.M., and were conveyed by carriage to St. Clears (10 miles 
west) ; thence through Llandow^or to Bglwys Cymmyn (5 miles 
south-west of St. Clears) ; and on through Llandawke to Laugharne 
(5 miles east of Eglwys Cymmyn), 

The return journey was made through St. Clears (A^ miles north 
of Laugharne) without any stops. 

The members were entertained to luncheon at Cwmbrwyn by 
invitation of Mr. and Mrs. R. M. Thomas, and to tea at Laugharne 
by Mr. and Mrs. J. M. Power. 

St. Clears. — On reaching the gates of St. Clears Church, where 
the fine Norman arch was to be inspected, there was a torrential 
downpour of rain, which caused those in the rear vehicles to dart 
into the doorways of houses and shops, to await a cessation before 
they could proceed into the church. Here, in the absence of the 
Vicar (Rev. C. F. Owen, M.A.), Archdeacon Thomas read an inter- 
esting Paper prepared by him, dealing with the history of the 
church and priory, and also the meaning of the name St. Clears. 

Commenting upon it, the Archdeacon said they were very glad to 
have this compendious summary of the history of the church, in 
which there was a very good Norman arch, but of such a very 
depressed type as was unusual. He did not remember seeing one 
in that form before. On the capitals was some curious carving. 

It being too wet to go to Banc-y-Beili, Mr. D. 0. Evans, P.G.S., 
St. Clears, gave a description, inside the church, of this old " motte 
and bailey" Castle. He said there had beeu some misapprehension 
as to the site of the old Castle of St. Clears. Of course, most 
writers referred to the mound there as having probably formed a 
part. However, a few writers, some years ago, made out that the 
Castle was situated up the town, at that part where the *' Blue 
Boar*' was. But he thought it was not only probable, but quite 
certain, the Castle was situated down there where they saw the 
mound. It was probably a ** motte and bailey" Castle," that was, 
an earthwork surrounded by a structure of timber. The " keep " 
was placed on top of the mound. If they went to the top of the 
mound and looked across the field, they would see permanent 
railings, and he had measured them, and found them enclosing a 
space of 56 yards long and 44 yards wide. That was probably the 
inner "bailey." If they looked round to the Co win, they would 

Digitized by VjOOQ IC 


find there was a ridge, which was probably the onter rampart 
enclosing the onter wall. On the left there was a smaller monnd, 
which probably carried a small turret to defend the gateway between 
the small mound and the bigger wall. This he called the Water-gate, 
for just at that point the River Co win came up close, so that this 
gate could be entered either by water or by land. The rampart on 
the eastern side might be traced for some distance straightforward 
towards the junction of the Co win and the T&f, but the southern 
part of it had been made use of as a backing for limekilns, which 
have now disappeared. At the corner of the field there was a rise, 
which suggested there was a smaller mound there, probably defend- 
ing another corner of the outer wall. Immediately north of the 
mound there wore traces of other ramparts, showing ramparts 
coming round to the *' keep," and approaching very nearly to the 
smaller of the two mounds. It was probable there was a little 
turret there as well, to cover the gate he had mentioned Other 
onter ramparts had existed, but had disappeared. They had probably 
been levelled, and now tho site was occupied by gardens and cottages. 
He examined, some time ago, the structure of the ground about 
there, and he found there was a small hill there previous to 
these works being carried out, and this small hill was a very 
convenient place for the Norman to erect his castle. These mounds 
and ramparts had been constructed from material immediately 
at hand, at a minimum of cost and labour. The first mention of 
the Castle was by Giraldus Cambrensis, who passed there in 1187 
with Archbishop Baldwin. The Castle had probably changed hands 
several times during the short time of its existence. It had been 
suggested to him it was improbable the Castle was there, because of 
the higher ground, from which it could easily be taken. It was, in 
fact, taken very frequently, and burnt very frequently, which seemed 
to indicate very clearly the Castle was buUt of something very com- 

Llandowror Church. — Leaving St. Clears the party journeyed to 
Llandowror, where the church was interesting, because of its asso- 
ciation with the Rev. Griffith Jones, "the 'morning star' of the Welsh 
Reformation, and the founder of the itinerary schools." Here a 
most instructive Paper was read by Mr. D. C. Evans, who referred 
to the two so-called pilgrim tombstones to be seen in the field close 
to the churchyard. He said there were some genuine pilgrim 
monuments to be met with occasionally, but here there was nothing 
but tradition to uphold the theory. The first slab was of local 
stone, which appeared to have been exposed to the weather, if not 
water- worn befere being made use of as a gravestone. There was no 
trace of tooling, except the sculpture of a cross — one on the back 
and one on the face. The number of crosses, as well as the character 
of the workmanship, seemed to suggest they were not of the same 
age, and that they indicated three separate burials. The second 
slab had no inscription of any kind either, and it was quite evident 
6ru sBtt., VOL. VII. II 

Digitized by VjOOQ IC 


from the plaoe of the cross these were intended to be upright, and 
not recnmbent. A woald be idle to estimate as to the age, but thej 
were older than any portion of the churoh as it now stood, and 
marked the burial-place of some of the earlj sons of that district. 
This was not the church in which GriflBth Jones was wont to 
officiate. He concluded with some particulars of the Rev. Griffith 
Jones's connection with the district. 

Archdeacon Thomas wished he could hare told them somethiug 
more about the old church, but they were under the influence of 
Griffith Jones, and Wales owed a great debt of gratitude to his 
noble serrice, and to the great Society which enabled him to do so 
much good work — the Society for the Promotion of Christian 
Knowledge. He should like to know what special pilgrim marks 
Mr. Evans referred to, the absence of which made him doubtful 
as to the age of those stones. 

Mr. Evans replied they were usually marked with a wallet or* 
scrip, a staff, scallop-shells, and branches of palm, and various other 
things indicating the fact that they were pilgrims. 

Archdeacon Thomas : I do not think those marks belong to that 
period. These stones must have been of earlier date. I have no 
doubt of that. There is nothing to show they are pilgrim stones, 
but they are very ancient stones. 

On an examination of the stones, Mr. A. Stepney- Gulston said 
they were found very nearly where they now stood, and that field 
was probably part of the churchyard. There used to be three of 
them, but one had been broken or lost, but it was hoped it would be 

A good deal of interest was taken in the chair that belonged to 
Madam Bevan, a financial helper of Griffith Jones, which was on 
exhibition outside the Old Tavern. It was incidentally mentioned 
that Madam Bevan was a sister of Mr. Stepney Gnlston's grand- 
mother in the fifth degree. The party then preceded to Cwmbrwyn. 

Gwmbrwyn. — Continuing the journey, a short drive brought the 
party to a spot where a lane led off from the road, and was believed 
to be a short cut to Cwmbrwyn. It turned out to be a muddy, 
winding path, and those who kept to the road had the best of the 
walk. Arriving at the farm owned by Mr. Bo wen, they made an 
inspection of finds at thesite of a Roman settlement then being 
excavated by the Carmarthenshire Antiquarian Society, with the 
assistance of Mr. John Ward, Curator of the Welsh National 
Museam, Cardiff. These included fragments of stone querns or 
handmills, portions of upper stones of common Roman form; a 
bronze coin of Carausius, a.d. 286-93 ; stone spindle-whorl ; frag- 
ments of Samian pottery, fine and coarse buff and reddish ware, 
with coarse black and grey waves ; window glass ; fragments of 
red roofing-tiles, consisting of fragments of flat tegulce and half- 
round imbrices ; part of bronze handle of backet-like vessel or 
tittUa, similar to those found at Pompeii : fragments of flue- tiles, 

Digitized by VjOOQ IC 


and roof-riates of common Roman form. With these Owmbrwyn 
finds, Mr. Ward exhibited, with the kind permission of the owner, 
Mr. HugheS'Garbett, of Bristol, the Roman bronze sancepan-like 
patella and the strainer which were found at Kyngadle, near 
Laugharne, some time ante 1889. They represent one of the chief 
Roman finds in Wales, and were fnlly described and illustrated by 
Mr. J. Romilly Allen, F.S.A., in ArchcBologia Cambrmsis about six 
years ago. The patella was first described as a " sacrificial censer" 
in the Gentleman's Magazine of 1839. When found, the patella con- 
tained many coins, mostly of Carausius (ld. 286 to 293), which have 
long been lost. 

On the excavated site of the Roman settlement Mr. Ward gave an 
extremely lucid account of the discoveries made there. He said he 
remembered, some few years ago, showing a lady friend a field like 
* that : a larger site, with magnificent excavated trenches and pitfalls, 
• and afterwards asking her, '* What do you think of it all ?" ** Well," 
she replied, ** I think it is a good field spoiled." He was afraid 
their good friend, Mr. Bowen, would have in his own mind pretty 
much the same feeling. He had been very good, not only in giving 
permission, but rendering every assistance, and taking the greatest 
interest in the work. The site was about 240 ft. long, north to south, 
in exterior length, and about 140 ft. internal length. It was ur - 
rounded by a rampart, which was formed of the stuff thrown up 
from the ditch, and the ditch was of the usual Roman V-shaped 
form, about 17 ft. wide, and 8 ft. 6 ins. deep, cutting down into the 
rock at the bottom. Between the ditoh and the rampart, which was 
originally 15 ft. wide^ was a space of 6 ft. or 7 ft. It was impossible 
to say how the rampart was constructed beyond, that it consisted of 
the stone and clay, etc., out of the ditoh ; they could find no sign of 
a retaining or external wall, but on the inner edge they found here 
and there tumbled-down stones, which might be the remains of an 
inner retaining wall to support the foot of the rampart ; or, possibly, 
i^e rampart itself might have been surmounted by a wall, and 
that might account for the large amount of stones they found roll- 
ing down the sides of the ditch, and also the stones on the inner side 
of the rampart. But it was quite impossible to say. They had 
looked for any kind of a base or support, to support the earth of the 
rampart. As they saw the rampart now it was spread, and not 
much more than 2 ft. high anywhere, and it had partly filled the 
ditch — the ditch and rampart showing a slight hollow and a slight 
rise 30 ft. wide. On the west side was the single entrance, with side 
walls, of which they had the foundations still left, with an opening 
about 11 ft. or 12 ft. wide. Through the opening came the track- 
way or road which extended across. They would see some remains 
of the paving or foundation of the road. On each side of that they 
had a yard, and the yard was gravelled. About that corner they 
saw a tumbled amount of stones, which suggested buildings or sheds 
of some kind. Along the back was a long building about 110 ft. 

Digitized by VjOOQ IC 


by 28 ft wide. That extended the full length of the back, and 
behind that was the back rampart and the back ditch. 

On the other side they had a cnrions bit, very ronghly laid, of 
stonework, which they thonght was the well, but they found out after- 
wards that it was cut into the rock, and there it stopped. It was 
very likely that was a cesspool to catch drainage. Below that was a 
small flue — two little walls about 9 ins. high, with a space between 
of 1 ft, and an open space in front where the fire was kindled. 
Such a little flue had been found in Silchester and Gaerwent, and 
had been used to heat coppers and other things. At the back 
was a very rough patch of pitching, covered with earth and cinders, 
where there was a little wooden building — a smithy, or something 
of the'^kind. The long building had a little building at the end, 
which appeared to have been tacked on, probably at a more recent 
date. The main building was a parallelogram, of which the founda* 
tions remain, bat not entirely, and the floor of the building where- 
ever exposed was formed of the natural soil, mixed with gravel, and 
rammed down very hard, and no doubt mixed with lime. Many 
years ago, Mr. Bo wen made some cattings there for the sake 
of the stone ; and as far as one could see there might have been a 
hypooaust, or some arrangement for heating the place. They went 
down a depth of 2 ft or 3 ft. from the main floor to a hard sarface, 
and nearly all those plain tiles came from that part. As to what 
the long building was, he really could not tell them. There 
appeared to be no cross walls, but all the heating was done from one 
end. The fortified enclosure was not square — and Roman forti- 
fications were almost invariably square — ^yet there was some sem- 
blance of squareness along the back and side. He thought at first 
it might be a villa, but they did not find villas with fortifications as 
large as some of the largest schemes. It was obvious nobody would 
go to the trouble of fortifying a house to this extent. (R«v. 
J. Thomas : In Wales ?). Well, perhaps the wild Welshmen were 
very troublesome. It was not a Roman fort, becaase it was al- 
together too small ; but then he thought it might be a Roman 
mansio. According to classical writers, along the lines of Roman 
main roads there were stations and mansiones. These mansiones 
were to all intents and purposes imperial posting-stations, where 
relays of horses were kept, and where, in a partly-subjected country, 
they might keep a detachment or " century" of soldiers to the neigh- 
bouring station. The only difficulty was, there was no indication of 
a main Roman road. His attention had since been called to the 
Roman camp on that side of St. Clears, and it was just possible this 
little work might be really a sort of redoubt connected with that 
fort or camp — ^assuming it to be one — about two miles away. They 
got on Hadrian's Wall a succession of stations and small fortlets, to 
which daily or weekly detachments could be sent from the main 
fort. The long building tended to confirm that view. In all Roman 
forts of which they had plans, they would find barracks consisting 
of a long building, 120 ft. to 150 ft, to accommodate a '* century' 

Digitized by VjOOQ IC 


— eightj men, with a centurion and nnder-offioers. It was, of 
conrse, much smaller than that at Gelligaer, which was a cohort 
camp, where they found eight of these buildings. It struck him as 
being connected with some larger camp, to which they might send 
a ** century " of men to occupy it periodically. In such a case the 
open space in front would be very useful for massing a body of troops. 
All the 6nd8 were de6nitcly Roman, and this was no doubt a 
Roman site. But towards the end of the building, where the 
ground was very much disturbed, were things which might be very 
much later, of medisBval date : an ale-pot of 1 700, or earlier, and 
rather deep down what appeared to be an ordinary wine-bottle of 
100 or 150 years ago. Mr. Bowen said he believed there used to be 
a small cottage built out of the ruins where he found the ground 
disturbed. There was no evidence of the site being pre- Roman, and 
they had slight evidence it might have formed just a little domain — 
a small cottage — but one find threw an interesting light upon 
the whole thing : that was a single small bronze coin of the 
Imperator Carausius, who seized the sovereignty of Britain in 286, 
and came to the end of his tether in 298. That was in a very fresh 
condition, and very sharp, so it could not have been long in circula- 
tion. That rather suggested, at any rate, that about the close of the 
fourth century this site was in occupation. It might have been 
200 or 300 years earlier. He daresay they saw also the very inter- 
esting bronze saucepan, or patella, with a little colander, or sieve, 
which came from Kingaddle, that, according to the discoverer, 
about a century ago contained coins of Carausius. 

Rev. J. Thomas : In the south transept of Laugharne Church an 
urn was found with a great number of coins of Carausius. 

Mr. E. Laws : Coins of Carausius are common in Pembrokeshire. 

Mr. Ward said in the time of Carausius our shores were very 
seriously raided by successive attacks on the east, by the Irish on 
the west, and by the Scotch on the north. It was during that 
period that we got our coast forts, like Cardiff Castle, Richborough, 
and others. It was just possible this might be a late Roman fort- 
let, having been connected with keeping off pirates from the 
sea. He believed there was some evidence that Laugharue itself 
was a Roman station ; and if that was so, this might very easily be 
a sort of outpost of Laugharne. Shortly there would be a full report 
published, so they would have an opportunity of following out his 
stat-ement in detail. 

Archdeacon Thomas proposed a vote of thanks to Mr. Ward for 
his interesting and instructive address. They would not only be 
able to look around them with intelligence, but look forward with 
great interest for the fuller account he had promised. He took this 
opportunity of congratulating the Carmarthenshire Antiquarian 
Society on being so vigorous and so active, and taking a work of 
this kind in hand. He also congratulated them upon having as 
their exponent such an expert aa Mr. Ward. He had further to 
thank them for giving the Cambrian Association that opportunity 

Digitized by VjOOQ IC 


of seeing what they were doing, and learning about Cwmbrwyn. 
Reference had been made to the work that was formerly done there, 
and he thoaght Mr. Laws knew something of that, and perhaps he 
would tell them a little about what was done then. Then of 
Colbren, which was not excavated; he believed Colonel Morgan 
.could tell them something, and they had also among them a very 
able expert upon ancient camps ; and he hoped Mr. Willoughby 
Gardner would have a word on the subject 

Mr. E. Laws snid he heard of the digging up there some years 
ago, and came and spent a few hours there. They dug at one end, 
and they saw and left in sight a little heap of hypocanst tiles. That 
was all he saw, and all he could tell them about it 

Colonel Morgan, referring to Colbren, in Brecon shire, on the road 
between Neath and Brecon, said they had discovered there certain 
things that had not been found at any other Roman station before. 
In all other stations the foundations for the rampart had been found 
to be either stone walls or stone paving. In this case he found the 
most beautiful log pavements, sometimes 1 ft 3 ins. in diameter, 
extending under the wall of the rampart, and running the whole 
length of it. It was in a beautiful state of preservation — some very 
nearly turned into bog oak ; but they could see perfectly plainly the 
marks of the axe. On the outside was a very wide brim, and it 
showed the outer entrenchments extremely accurately. Beyond 
that they found two trenches — not of great importance as far as 
size was concerned, but showing a particular stage of Roman forti- 
fication not very much studied in England. .They found obstacle 
trenches, and they found obstacles in the shape of oak spikes as 
perfect as the day they were put in, sharply pointed, only not in 9itu 
because they had tumbled down. They found a very large number 
of them in that ditch, and any amount of them could be found there 
now. Sucb works were only occupied a short time — about thirty 
years — and they had not time to replace them by stone walls. In 
England all the early Roman entrenchments were afterwards con- 
verted into stone-walled camps, and obstacle trenches were done 
away with. He hoped, if they had a little better weather this 
autumn, he might be able to make a fuller report of what he 
thought would turn out one of the most interesting stations in 

Mr. Willoughby Gardner, speaking of the Cwmbrwyn discovery, 
said he never saw anything quite the same as this before. It was 
unique of its kind. All he could do was to congratulate Mr. Ward 
and the members of the Association on what they had done. It was 
very remarkable and interesting. 

Eglwys Gymmyn. — Continuing the journey, the party, after a 
pleasant drive, arrived at Eglwys Cymmyn Church, where it was 
disappointing to learn that Mr. G. G. T. Treherne, of London, who 
has made the antiquities of what he calls " Laaghameshire'* his 
special study, was unable to be present owing to indisposition. 

Digitized by VjOOQ IC 

CAtlMA&THBN M^firmO. — teXOXTRStONS. 231 

Unfortanafcely, too, the Papers which Mr. Treherne had promised to 
read, thoagh thej had been posted, had gone astray in transit ; bat 
the diflScolty was got over by some observations on Eglwys Cyramyn, 
Parc-y-Cerig Sanctaidd, and Landawke, by Archdeacon Thomas 
and other speakers. With regard to the ancient Oh arch of Eglwys 
Cymmyn, Archdeacon Thomas said it was evident this was a very 
early church, becaase of the extremely sharp-pointod vanlted roof, 
and the chancel arch was very rude, being simply cat out of the 
wall. There was no timber or pillar, or anything of the kind. 
It was of the Pembrokeshire type, and like what they had seen in one 
or two places already. The porch was also vaalted, and it was 
very carions to see another door so close to the original one. He 
did not know how that was to be accoanted for. In the side of the 
chancel was a piscina with a ledge, and under the window they 
wonld see in a glass case a small vessel of glassware that was 
found close to the church, when some parts of it were being repaired 
or rebuilt. Then there was an Ogam stone at the west end of the 
church inside a box. There were also stained windows of St. 
Margaret of Antioch, St. Margaret of Scotland, and St. Margaret of 
Marros, the daughter of Guy de Brian. 

The Bev. Geo. Eyre Evans said that thirty years ago he was 
in the church, and they would hardly know it was restored, so well 
had the work been carried out. They had there an example of 
what could be done, using restoration in the right sensa He had 
remembrance as a child of seeing what was now missing — a silver 
chalice. That chalice had disappeared. There had been rumours 
of its whereabouts, and possibly it might be recovered. But it was one 
of a very beautiful series, of which they had so many in Carmarthen- 
shire, dated 1574. No one could tell how much that church owed 
to Mr. Treherne — how much and how wisely he had worked there. 
They had got a master-hand who saw how things should be rightly 
and properly done; and the glass case in the chancel preserving a 
relic was an object-lesson to all of them as to what they might do in 
some way. In Cardiganshire, Bishop Morgan's Bible was put in a 
glass case, in a dry part of the church, where all could see it. They 
found evidence all round of great taste in restoring the building. 
It was intended to put a window on the site of the old door, but it 
would be put in so that people would know a door was there. 

Archdeacon Thomas said the church stands in an old earthwork, 

The Rev. J. Thomas : We are now standing within the boundary 
of the fort. This was an old British track which led by Tavern 
Spite on to Menapia. He therefore thought the Roman fort would 
be on the burrow. 

Mr. Egerton Phillimore said this cair would have nothing to do 
with the boundary of the land of Llandowror. The name cair was 
the name of a brook. 

Professor Anwyl proceeded to explain the inscription on the 
Ogham stone. He saw the Ogham was fairly plain, and the Latin 

Digitized by VjOOQ IC 



was a translatdon of the Ogham. The words of the Ogham are 
Avitoriges (the g probably not being pronounced) inigina Gunigniy 
and the Latin read Avitoria filia Cunigni, that was to say, 
" Avitoria, the daughter ot Cynin" — Cunignos at that time. Avi- 
toriges soemed to be a compound word, meaning the granddaughter 

Fig. 2. — Inscribed Stone at Eglwys Cymmyn, Carmarthenshire. 
[Prom a Photograph by T. Mansd Pranlderty Esq.) 

of Toros. Further, the name Cunignos was the same as Cynin, 
which they got in Eglwys Cymmyn. 

Parc-y-Cerig Sanctaidd. — Leaving Eglwys Cymmyn for Lang- 
hame, a halt was made at Parcy-Gerig Sanctaidd, where some 

Digitized by VjOOQ IC 


disonssion took place in regard to the " holy stones." Mrs. Allen 
thoaght the round stone was the ba£(e of a cross. She woald 
like to know whether there had been a battle anywhere there, 
because sometimes they erected a cross on the '* field of sorrows*' 
after a fight. There was one in North Wales. 

Rev. J. Thomas said in a field near there was a large tumulus 
ploughed down, and they were surrounded by very ancient habita- 
tions. The tradition of the locality was that when funerals came 
along the road to Llansadwrnen they used to go round there to 
rest the coffin, and use that hollow stone as a stoup for holy water. 

Mr. E. Laws pointed out that if they looked under the stone with 
a cross they would find a lot of small white stones. In any sepul- 
chral place he had opened he had found them. 

The Rev. J. Thomas said they were commonly called " cursing 

Mr. Ward said, so far as he understood, these stones did not 
occupy quite the same position as they did formerly. According to 
the late Miss Curtis, in her gossipy little book on Laugharne, she 
mentioned these stones as resting- stones. When coffins passed 
there to church, they rested upon these stones while they repeated 
the Lord's Prayer, or something. The parish boundary went along 
there ; and Mr. Treherne and himself went very carefully over the 
ground, and had reason to think a direct lane to the parish church 
passed over there, so that the present road was probably modern. 
That was the highest point of the road, and what more natural than 
that there should be there a wayside cross — these crosses were 
common thi*oughout the country in pre- Reformation times — and it 
was preceded by a pre-Norman cross for the very same purpose ; or 
it may have marked the tomb of somebody, and placed near the 
wayside. According to that, the four stones were simple founda- 
tions. The cross was a thirteenth- or fourteenth-century cross.^ 

Mr. E. Laws thought it was sepulchral. 

Llandawke. — Proceeding to Llandawke, the party inspected the 
church and an Ogam stone kept therein, Professor Rhys giving a 
description of the stone. He said it was peculiar in several respects. 
Generally, when they had an inscription in two languages on stones 
in this part of the country, one was more or less a translation of the 
other, but in this case it was not. The Latin was ^' Barrivendi filius 
Vendubari," and there was '' Hie jacet " on the edge of the stone. 
The man must have thought he had not room to write " Hie jacet," 
and in the early copy of the inscription that was not seen. He first 
saw the stone as a threshold, and the end had been smashed. A 
big piece had been splintered away, which ought to be found yet, 
and the surface was a good deal polished, and many of the strokes 
worn by the feet of the parishioners. It commemorated the son of 
the son of somebody, and was one of the earliest inscriptions. He 
should say it went iMok to the fifth century. 

* It appears to us to be much earlier. — Ed. 

Digitized by VjOOQ IC 



Laughame. — The day's work was broufifht to an and at Laag- 
hame, where the Castle was gone over, and Mr. Power gave some 
explanatory notes as to its constraction and history. The earliest 
part is the round tower, and it was the military base to cover the 

Fig. 3.— Pre-Norman Cross at Laugharne, Carmarthenshire. 
{Prom a Photograph by T. Mantd PranJdeny Etq.) 

constrnction of the main castle against attack. The portion called 
Sir John Perrott's gateway was added to the Castle by him in 1560, 
and no doubt represented the best apartments. It was besieged by 
Cromwell for somewhere about a month, and eventually lost, largely 
from the same cause that reduced Pembroke— the cutting-off of the 

Digitized by VjOOQ IC 


water-supply. The Governor of the Castle, — General Langhame — 
wan first a ParliamentariaD, and aabseqnentlj held the Castle for the 
King. The garrison made a sortie from the gateway, and oat to the 
lane to recover the water-supply, bat were defeated. The attackers 
rashed the gate, and then Cromwell ordered the place to be dis- 
mantled. Like every other castle, it had served as a qaarry for the 
town, and that had done more harm than anything. 

In the Town Hall the Recorder of Laugharne, Mr. Jeremy, gave 
some particulars of the government of the town by the ancient Cor- 
poration, stating the same method as two hundred years ago was 
still carried on. He mentioned that in 1731 a bai'gess would not 
take the oath of allegiance, so was not elected Portreeve. Ancient 
deeds were produced and inspected, as well as an old Winchester 
measure and some tally-sticks. The property of the Corporation 
consisted of some cottages and 725 acres of land, some portions of 
which were divided among seventy-six senior burghers and held by 
them for life. There was an annual beating of the boundaries, with 
certain halting-places for refreshments, and for hoisting and whip- 
ping the boys. The Portreeve of 1864 was the last to act as 

The Church was also visited, and here the Rev. J. Thomas read a 
Paper, and a large quantity of silver plate, some dating from 1600, 
was on view, and including a silver flagon and paten presented to 
the church two years ago. After a long day, the party reached 
Carmarthen at ten o'clock p.m. 


iRoute. — The members assembled at St. Peter's Church, and spent 
the morning in inspecting the antiquities of the town of Carmarthen. 

After luncheon the members assembled at the Great Western 
Railway Station at 1 p.m., and were conveyed by train to Ferry side 
(eight miles), and thence by carriage through Llansaint to Kidwelly 
(four miles south-east of Ferryside). 

The return journey was made by carriage through Llandefeilog. 

The members were entertained to tea at Kidwelly by the Mayor 
and Mayoress (Mr. and Mrs. H. E. Smart). 

Carmartheit — St. Peter's Church, the old Castle, and other 
historic remains in Carmarthen, occupied the attention of the mem- 
bers of the Cambrian Archaeological Society on the morning of 
Thursday, August 16th. The assembly was at the Church, and 
here Mr. T. E. Brigstocke, whose knowledge of the subject is pro- 
fuse and sound, read a Paper on the ancient edifice. He pointed 
out the features of. interest, including the tomb of Sir Rhys ap 
Thomas and Dame Eva, his second wife, which was originally in 

Digitized by VjOOQ IC 


the monastery of tbe Grey Friars, in Lammas Street^ and was 
removed therefrom some 350 years ago to the chancel of thechnrch, 
and afterwards, on the erection of the organ, placed in its present 
position. He raised an interesting point in regard to the plainness 
of the architectnre of a chnrch of sach dimensions and importance, 
suggesting that possibly it included the remains of a still earlier 

Archdeacon Thomas referred to a picture he had seen of the 
church, in which a very curious feature was shown by the south 
door, namely, a little corner walled off which was called the 
" charnel-house," or Golgotha, where in former times the bones 
collected in the churchyard were placed. Those who had been in 
Brittany would remember what a great feature the Golgotha was 
there. The iiesh turned to dust, but the bones were collected 
together and placed in these, while the skulls were put on raised 
shelves in the cemeteries. 

In the absence of Mrs. Dawson, daughter of Archdeacon Bevan, 
her Paper on St. Tewdrig was read by the Rev. Charles Chidlow. 

Archdeacon Thomas said the life of Tewdrig belonged more to 
Tiutern than to Carmarthen, It certainly did not fit in with the 
conclusions brought before them by Professor Lloyd, in his Paper 
on Monday evening, that it was the Church of Teilyddog. He did 
not think St. Tewdrig had anything to do with Carmarthen. 

Passing through the churchyard to the Vicarage, the ladies of 
Dolaucotbi were exceedingly amused by the specimen of colloquial 
** Welsh" used by a native of Carmarthen, in response to a question 
as to what they were doing to the paths. " Ob," he replied, " we 
buildo yr wall all round." 

In the Vicarage garden the members inspected a Roman domestic 
altar, another carved stone with a boss, and a third built into a 
wall inscribed " R. P. Nato," which were briefly described by Mr. 
Walter Spurrell. 

Proceeding to the Castle, the party inspected the outer wall, and 
then being admitted through the prison gates, ascended to the top 
of the old mount, where Mr. W. Spurrell read a Paper by Mrs. 
Armitage, who, he said was particularly interested in the early Nor- 
man castles, and, as many of them knew, had rather strong views as 
to the age of these fortifications. She did him the honour of calling 
upon him when visiting the town ; and when he asked Mr. Holmes 
to prepare a Paper on the Castle, he thought he would write to Mrs. 
Armitage, and ask her what conclusion she had arrived at as to the 
Castle mound on which they were standing. Mr. Holmes's Paper 
took the form of a criticism of Mrs. Armitage*s Paper and other 
authorities. In her Paper Mrs. Armitage said she had considerable 
doubts about the stone keep, as to whether it was an abutment of 
the upper portion of the *' motte," or whether it stood on the original 
top, and hfikd been filled with soil. Outside there was a considerable 
rise of tower, but from the inside none. After the introduction 
of artillery, it became usual to construct a wall and fill up with earth 

Digitized by VjOOQ IC 


to resist shot. She would be sfratefnl to know whether there was 
evideDoe of an old entrance. Basement entrances were rare before 
the thirteenth centnry. The best " motte" she saw in the neigh- 
bonrhood of Carmarthen was at Wistou, in Pembrokeshire. It was 
a wall apfainst which lean-to buildings in wood were supposed to be 
erected. In 1096 Ehydy^ors was abandoned to the Welsh, but 
restored, and afterwards they never heard of it again ; but in 1113 
they heard for the first time of the Castle of Carmarthen. Where 
was the castle of Rhydygors ? It had been sought for and placed 
at a farmhouse called Rhydgors, near Carmarthen, opposite to 
which there are earthworks : some think the embankments of the 
river. The Castle of Carmarthen was for many centuries a royal 
castle, and it was extremely probable that it was built by order of 
William Rufus. Undoubtedly the mound was the " motte" of the 
typical early Norman castles. These castles were not of stone, but 
of earth, with wooden superstructures. More than 90 per cent, of 
the castles built by the Normans were of this description. The date 
of the castle stone keep was difficult to determine, as both keep and 
** motte" are so travestied by modern arrangements, that it would 
be difficult to plan them out correctly. Carmarthen Castle was 
razed to the ground in 1215, and it was possible this keep repre- 
sented the rebuilding which followed that event. The gate-house 
was of the Perpendicular period, and probably there was no masonry 
older than Henry III ; it was not unlikely there was no masonry 
building there until the fourteenth century. 

The Paper by Mr. H. S. Holmes, B.Sc., B.A., Vice-Principal 
of the Training College, Carmarthen, tended to show that the 
structure was much older than Mrs. Armitage put it. Qiraldns, in 
1204, described the town as an ancient city with walls. If the town 
was enclosed by strong walls, the castle would be more than an 
earth-and-timber block house, built by the Normans to keep in check 
the Welsh guerillas. In 1273 the walls were stated to be in a 
mined condition. The town was sacked in 1244, and again in 1246 ; 
but the castle was apparently too strong ; or, on the other hand, the 
attacks on the town were merely raids. Therd was no record of 
remains having been found within the area of these walls, but con- 
siderable Roman remains of different kinds had been found east 
of the town wall. 

Colonel Morgan being called upon to give his opinion, said that 
was undoubtedly a stone revetment against a moated mound, and 
he did not put the revetment anterior to Charles I. He thought 
there were innumerable signs this was of the date of the Civil War. 
Then the face was made, and he could see no break in the other 
part; so, though there might have been a shell- keep that was 
older, the greater part of the stone revetment was not earlier than 
Charles I. 

Rev. J. M. Phillips asked if there was any reason to suppose that 
the Castle built at Carmarthen could be called the Castle of Rhydy- 

Digitized by VjOOQIC 


Archdeacon Thomas : I do not think it is an nmianal thing when a 
castle is replaced by another one close by, or within a short distance 
of it, for the second one to take the name of the former. It supple- 
ments the first. 

Mr. J. M. Phillips : But if there was no castle down at Rhydygors 
originally, why should the name be attached to the castle here, as 
Mrs. Armitage suggested ? 

The question was asked, how far they were from the ford ; and 
the Rev. J. M. Phillips replied, half a mile. 

Mr. A. LI. Davies : There is a ford just down here. 

Mr. Egerton Phillimore remarked that these names did get mis- 
placed, and he mentioned a place on the Severn, the name of which 
was taken by a farm a mile and a-half from the Severn now. 

Colonel Morgan : A short distance from here you have got the 
remains of outworks of Charles Fs time, quite unique, not only in 
England, but in the whole of Europe. They are beautiful specimens 
of the bastion traces of Charles I's time, which have been destroyed 
everywhere with the exception of at Carmarthen. 

The party then went round the southern wall of the Castle, 
taking note of a portion which served as an outer wall to a modern 
dwelling-house, and inspecting the rooms above the fine old gateway. 
Afterwards they proceeded to the Diocesan Registry Qjice, where 
they viewed underneath the offices an extensive crypt and vault, 
which are supposed to have formed a portion of an early church — 
Prince Edward's Chapel. Mr. T. W. Barker had laid out in his 
rooms a portion of a stone cross and column, lent by Mr. Victor 
Jones, which came from the old Priory of St. John. Here also were 
to be seen the early manuscripts of episcopal acts, including the 
earliest book, which was lent by permission of the Record Office, 
and dated from 1399 ; also other books and articles of value con- 
nected with the diocese. Visits of inspection were subsequently 
paid to Bishop Ferrar's tablet in Nott Square, the vault beneath the 
Sheaf Inn, the town walls in Quay Street and Blue Street, the 
sculptured stone at the rear of the Town Hall, and the remains of 
the Grey Friars' Monastery in Lammas Street. They then went to 
the Dyke and Ditch behind Christ Church, upon which Colonel 
Morgan dilated at some length. He considered this to be one of 
the most valuable possessions of the borough. At the time of the 
Civil War, he said, it was decided to fortify the town of Carmarthen, 
and the ramparts they then saw were the remains of works that 
originally went all round the town. The last remaining of the 
works were destroyed about ten or fifteen years ago, when Francis 
Terrace was mada These works were quite the finest specimens of 
the bastion traces as executed in the time of the Civil War in 
England, and probably even on the Continent. The system of 
fortification here introduced had been first started by an engineer 
of the name of Erard, who published his works in 1594. They 
seem to have been adopted up to the time of the end of the Civil 
War, when they were supplanted by a style of Count d© Pargon, 

Digitized by VjOOQ IC 


-who published his works in 1665 ; so that these particnlar ramparts 
must have been executed some time between 1594 and 1665. The 
characteristic feature of the Erard system was that the flank was 
placed at right angles to the curtain, while de Pargon had it at 
right angles to the face of the bastion. He showed how the defen- 
ders could sweep down this earthwork to clear the attacking partj. 
Nothing of this style remained on the Continent. 

Llansaint. — In the afternoon the members proceeded by train to 
Ferryside, where brakes were in readiness to convey them to Llan 
Ishmael. This was one of the most interesting churches visited 
during the tour, and a Paper was read by the Kev. G-eorge Eyre 
Evans, dealing with its chief characteristics. 

The drive was continued to Llansaint, in the same parish, passing 
on the road the buried village of Hawton, demolished by a tidal 
wave about 1639, and marked on Speed's Map of 1610. The party 
entered the chapel, where there was one of the largest gatherings of 
the Association, under the presidency of Sir John Williams. Here 
a Paper was read by the Rev. George Eyre Evans on the newly- 
discovered inscribed stone, which he read — 


and another and larger stone, 4 ft. 6 ins. in length, which is given 
by Westwood as reading — 


Much interest was taken in the newly-discovered inscribed stone, 
and Professor Rhys complimented Mr. Evans upon his find. He 
said he was not going then to say very much about the stone, as 
he hoped to have an occasion to say something more at length that 
evening. He had examined the stone, and had some trouble in 
copying the inscription, which was upside down. The other stone 
was all right ; but he got definite evidence from a man in the 
village— whom the young fellows called an old man — he was only 
69, and he (Professor Rhys) did not agree with them — that both 
stones were taken out of the walls of the previous chui-ch, forty-five 
or forty-six years ago. That was a fact that, he supposed, could be 
easily ascertained when the restoration took place. He remembered 
distinctly that stone being put like that, upside down, and then 2 ft. 
built on it, when there was a great fuss and controversy about the stone 
having been put upside down. They wanted to get it out, but the 
contractor — a certain Wm. Matthias — would not undo the wall, as 
that, he (Professor Rhys) supposed, would cost money. That was 
definite evidence that the stones had been moved from the walls of 
the earlier church, but he could not ascertain whether they were 
inside the church or outside. 

The Vicar said the man told him they were outside. The wall 
was taken down with a view of throwing out a vestry there, and 
the stones were in the original wall of the building. When the 

Digitized by VjOOQ IC 


vestry was thrown out they were removed, and placed where they 
are now. 

Professor Rhys : Rogers did not know, according to his deliberate 
statement to me, whether they were inside or ontside. 

At the Evening Meeting on that day, Professor Rhys gave a 
learned address of over an hoar's duration, apon the various inscribed 
stones visited during the meetings of the Association in South 
Wales. Referring to the recent discovery made by the Rev. George 
Eyre Evans, Aberystwith, of the inscribed stone inserted upside 
down in the wall of Llansaint Church, which had hitherto been 
covered with ivy, he said that the lettering, "Cimesetli Avicat," 
seemed to imply a place or monument to " a man of ransomed life " 
— son of Avi Caton (that is, " one admired as a warrior"). Dealing 
with the name " Llansaint," Professor Rhys suggested that it was 
dedicated to two relatives whose names ended in *' Setli," who were 
probably Irish saints. 

Kidwelly. — Proceeding further towards Kidwelly there were to 
be seen the ivy-clad ruin of Penalit Priory, and Clomendy, a well- 
preserved old pigeon-house, which probably belonged to the Priory. 

Arriving at the corporate borough, the party were welcomed at 
the Castle by the Mayor and Mayoress (Mr. and Mrs. H. B. Smart). 
An interesting history and description of the architecture of the 
ruin was given by Colonel Morgan. 

Some photographs of the group, which have resulted in excellent 
pictures, were taken by the ex-Mayor (Mr. A. Stephens, Broomhill). 

An adjournment was made to the Town Hall, where, at the 
invitation of the Mayor and Mayoress, tea was partaken of ; after 

Canon Morris proposed a vote of thanks to them for their hos- 
pitality. It was very important to their Society, and archoeology 
generally, that they should have the support of the authorities in 
the various towns ; and it was very gratifying to know in every 
place they had been to they found this hearty welcome, and this 
readiness to help them in preserving monuments and other interest- 
ing things in connection with the past history of the country. 

The Mayor said his wife and he were extremely obliged to 
them for the vote of thanks for the little they had done. He 
could assure the Association they were exceedingly pleased to have 
had the opportunity of entertaining them. He was sorry that the 
Corporation had not much to show. The old charters had been 
lost, and they had tried many times to find them. Although they 
had not succeeded, they had not given up the search, and they 
hoped in time the charters, which had somehow or other been lost 
or mislaid, would be found. They had the two silver maces of the 
borough, copies of one of the old charters (but the wording of it was 
not interesting), and the borough seal. Also a piece of old cloth 
dated 1759, and having on it the name of Griffith Jones, Mayor, 
which used to be on the magistrates' bench. 

Digitized by VjOOQ IC 


The parish church was then risited, and here a Paper on the 
stmofcare was read by the Rev. D. D. Jones, Vicar, and listened to 
with much interest. 

Canon Owen proposed a vote of thanks, and Mr. E. Laws secon- 
ded, remarking he had seen a great many charches, bnt he did not 
know anyone that had such interesting features in it as this one. He 
conid not lielp thinking there was Early Engh'sh work in the arch. 
Sir Gilbert Scott notwithstanding. More than that, they had two 
tombs, one to the Lady Ysoude, and the other a civilian holding his 
glove in his hand. He thons^ht they ran into Early English times, 
too, and that rather backed him np in his heresy. One of the most 
significant things was the enormous number of staircases. There 
was a staircase to the rood-loft, anotlier going to the room over the 
sacristy, and there was a very interesting little wall- window above, 
for the priest to look down apon the altar, in order to ring the bell 
on the elevation of the Host. There was a staircase, which the 
Vicar thought went to a stone pulpit, which had disappeared, and 
still another staircase ; also a beautiful piscina and sedilia, and a 
charming little window decorated. 

After spending an interesting half honr inspecting the chnrch, in 
the vestry being seen an elaborate alabaster figure of the Blessed 
Virgin and Child, which was for years buried in the churchyard, the 
party drove home vid Llandefoilog, where the Rev. Peter Williams, 
theedit or of the first Welsh annotated Bible, printed by John Ross, 
in 1770, is buried. 


Route. — Members assembled at the Great Western Railway 
Station at 9.50 a.m., and were conveyed by train to Whitland 
(18 miles west), and thence by carriage to Parcau, Gwarmacwydd, 
and Llandyssilio (seven miles north-west of Whitland). 

The return journey was made by carriage to Clynderwen Railway 
Station (two miles south of Llandyssilio), and from there back to 
Carmarthen by train. 

The members were entertained to luncheon at Gwarmacwydd by 
invitation of Mrs. Bowen Jones, and to tea at Llandyssilio by the 
Vicar, the Rev. 0. Jones Thomas. 

Whitland. — On Friday morning, August 17th, members travelled 
by train to Whitland, where they were met by conveyances, and 
driven to the Abbey (Ty-Gwyn-ar-Daf), where an able Paper was 
read by Mr. E. Ltiws, Tenby, and an interesting discussion took 
place regarding the name, which Archdeacon Thomas said was a 
curious study in philology. The old name, Ty-Gwyn-ar-Daf, meant 
the white house on the h^nks of the Taf River. That got translated 
6th ber., vol. vii. 16 

Digitized by VjOOQ IC 



first of all with the omission of the definite article, and they Bad 
Tygwyn-landaf. Then they got Alba Landa ; the house was left 
ont, and they had white and landa, which became misleading — 
it was white on the banks of the Taf. Then came another stage, 
the English of Alba Landa — Whiteland, the Taf was lost altogether. 
Tlien came another change, the dropping of the ** e" in the white 

Fig. 4. — Inscribed Stone at Parcau, Carmarthenshire. 

— Whitland, and when they got to Whitland they got to the reverse 
of Lantwit, from which Paulinas was said to have come. 

The Rev George Eyre Evans asked if anything was known of the 
small circular gold vessel dug up twenty-three years ago in the 
garden in which they were standing ? He had spoken to the man 
who saw it dng up. 

Mr. D. 0. Evans said he had made enquiries, but could not find 
it The name had led people astray, and placed this spot in the 
Isle of Wight, 

Digitized by VjOOQ IC 

Fig. 6. Inscbibed Stone from Castbll Dwyban, now at Gwarmacwydd, 


{From a Photograph by J. E. Omoer, 7, Laminas Street, Carmarthen.) 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


Mr. Egerton Pbillimore said G-iraldns Cambrensis made it too. 
There was not the slightest gronnd for belieWng that Panlinas had 
e^er anything to do with Whitland. He had looked at all the 
manascripts he oonld get at Oxford on the *' Life of St. David ;" 
and in the latest version by Usher, the word he had to copy was so 
difficult and so unintelligible, that he said an " insula quadem" — a 
certain island. In the twelfth century it was written, '^ insula 
inwincdi lantquendi ;" some of the later manuscripts say, ** insula 
whitlandi ;" others say, " insula whit." Professor Rhys thought 
** lantquendi" did not stand for Whitland. He (Mr. Phillimore) did 
not think it did before the time of HywelDda. He could not believe 
'*alba landa" had anything to do with the Taf ; the place had been 
put in Whitorn, in Galloway. The latest manuscripts say " insula 
whit." In some of the later manuscripts — in some of the Irish ones 
— these names were corrupted into " delanda bendi." 

Mr. D. C. Evans remarked that there was a hill close by called 
Castle Hill, which to some extent had been quarried away. When 
digging the ground twenty-five skeletons were come across, showing 
that the graveyard went that way. 

A short time was spent in examining the surroundings, where 
remains of ironworks and earthworks are to be seen ; and consider- 
able interest was taken in a beautiful piece of white marble, with 
designs in relief, which had been found. The coat-of-arms over the 
hall-door of the present residence also received attention, showing 
a Tudor rose, portcullis, dog and griffin, fleur-de-lis, and lions, 

Parcau and Owarmacwydd. — Leaving Whitland, the party pro- 
ceeded to Parcau, where the inscribed stone " Quenvendani fili 
Barcuni" was seen (Fig. 4), and on to Gwarmacwydd. Here, encircled 
by a wooden fence, was the Ogam stone with Latin inscription, 
*' Memoria Voteporigis protictoris " (Fig. 5) ; and several members 
busied themselves in taking a rubbing of the Ogam. This stone was 
removed from Castelldwyrau churchyard. 

Llandyssilio. — Afterwai*ds the party resumed the journey to 
Llaudyssilio, where the parish church was visited, and a great deal 
of interest was taken in the three inscribed stones to be found in 
the outer south wall of the edifice (Figs. 6, 7, and 8). Inside the 
church a silver cup, dated 1636, was inspected, and another of 
Early Elizabethan period, also a register dated from 1720 to 1814. 

Archdeacon Thomas said the present church was modern ; but in 
the old church there was what they did not often see, the font 
brought up near the chancel, whereas its proper position was at the 
entrance to the church. It was symbolic, as so many parts of 
the church were, to the gradual growth of the Christian life, and 
the means of grace, leading up from the entrance by the font to the 
Lord's table. There was on the south side of the chancel a small 
piscina. It was very simple, but was large enough for the purpose 

16 « 

Digitized by VjOOQ IC 



for which it was used — the rinsing-ont of the sacred vessels. The 
ohanoel and the chancel arch had not been touched at all ; it was a 
plain rude pointed arch, cut out of the wall apparently, and on the 
north side there are two depressed arches separating the chancel 
from what was now the vestry-room, bat which he had no doubt 
was at one time a chantry chapel ; very plain and rude, and cut out 
of the wall, as in the neighbouring churches of Pembrokeshire. 


Fig. 6.— luscribed Stone No. 1, at LlandyMUio, Pembrokeshire. 
Scale, T^y linear. 

There was a very curious book there : a small diary kept by a former 
curate, who was the Vicar of Llan-y-cefn, John Griffiths. No doubt 
he had his full duties to do on the Sunday, and he (the Archdeacon) 
was quite sure he had a great deal more than he ought to have 
undertaken or laid upon him daring week days : because in the little 
diary book, which he seemed to have carried about with him, and 
filled in day by day, he jotted down marriages, birtlis, and deaths 
in the different parishes during the week day, representing his 
secular duties. He was sorry to say that the aggregate of those 

Digitized by VjOOQ IC 



pan'sliefl in which be had to do duty during the week day was no 
fewer than twenty- six. A thing of that kind was a great abuse. 
He had heard of three or four parishes worked together, but twenty- 

Fig. 7. — Inscribed Stone No. 2, at Llandysdlio, Pembrokeshire. 
Scale, ^ linear. 

Fig. 8.— Inscribed Stone No. 8, at Llandyssilio, Pembrokeshire, 
discoyered by Mrs. Thomas Allen. Scale, i linear. 

six was almost incredible. That was what they were told in the 
diary, if he understood it aright. The Communion plate dated 
from 1651, and it was rare to find Communion plate of that date; 
it was during the interregnum of the Commonwealth. The then 

Digitized by VjOOQ IC 


Vicar remained in charge, he anppoBed ; but not having registers 
going back so far, they could not say for certain. 

Bev. J. M. Phillips : I do not think he did. 

Archdeacon Thomas : Yon believe he was deprived ? 

Bev. J. M. Phillips : I believe he was. 

Archdeacon Thomas said the chalice or Communion cup of Egre- 
mont was also there for inspection^ and that was a much earlier one 
than the Llandyssilio cup. It was an Elizabethan cup, but was not 
dated. It had the form and the characteristic band which was 
invariably found on Elizabethan cups. 

Mr. Stepney- G-ulston said the date of the cup belonging to that 
church was 1651, but the hall mark was 1632 or 1684 probibly. It 
was interesting to show that the date upon it was subsequent to the 
time of its original making. 

Mr. Egerton Phillimore made some very interesting remarks upon 
the inscribed stones. 

Mr. T. E. Morris referred to the small book in the church, which 
he said was highly interesting in these days of public libraries : 
inasmuch as it said that, as early as the year 1761 there was 
in that parish a circulating libi*ary, and it gave the names of 
all the books in the library. He found that in the year 1761 
there were in the possession of the parish as many as 800 books, 
among others some of the best and well-known Welsh books — Bardd 
Ctosg and Drych y Prif OeB-oedd, There was also a catalogue giving 
the price of some of the books, and he saw there was paid the sum 
of \8, for Bardd Ctrng. It also gave the names of the persons to 
whom the books were lent from time to time. There were also a 
number of very interesting sentiments and toasts — he was not going 
to give any extracts — but anyone who was curious would be amply 
rewarded if he looked through that book. 

Archdeacon Thomas thought in a large number of parishes they 
found Dr. Bray's libraries — possibly in each deanery. 

Owing to the time having expired, Egremont Church, where 
there is an inscribed stone, had to be left out of the programme, and 
the members returned by train from Clynderwen. 

ADOITST 17th. 


Route. — The members assembled in Guildhall Square at 9 A.M., 
and were conveyed by carriage up the Valley of the Gwili to Conwil 
Elvet (seven miles north), and then four miles further north along 
the road to Llandyssyl, which follows the course of the Afon Duad^ 
to the earthwork called Clawdd Mawr, opposite Nant-yr-hyddod. 

Digitized by VjOOQ IC 


Fig. 9. Inscribed Stone No. 1 at Traws Mawr, Carmarthenshire, 

Removed from Newchurch, aud now u.sed at* the Pedental for a Sundial. 

{From a Photograph hy T. Maimd FranUen, Es<j.) 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 


The return journey was made through Conwil Elvet and New- 
church (four miles north of Carmarthen). 

Luncheon was provided for the members at Conwil Elvet. 

Glawdd Mawr. — The party proceeded through the Vale of Gwili 
to Conwil, and thence to Clawdd Mawr, an ancient earthwork 
composed of a dyke and ditch, about a mile and a-half long. 
Opposite Nant-yr-hyddod Farm a longitudinal section of the earth- 

Fig. 10. — Inscribed Stone No. 2, at Traws Mawr, Carmarthenshire. 

work was examined by Mr. Walter Spurrell, Mr. Glascodine, and 
Colonel Morgan. About 10 ft. above the base of the section a 
horizontal layer of peaty material, about 3 ins. in thickness, was 
observed, affording evidence that this was at one time the surface of 
the ground. 

Traws Mawr. — After lunch at Conwil, and a visit to the neigh- 
bouring church, the company proceeded to Traws Mawr to view two 
inscribed stones, and a stone with an incised cross, standing 
upright in the private grounds of the mansion. On one of the 

Digitized by VjOOQIC 


upright stones was the ioscription, " Severini iili Severi*' (Fig. 9), 
and on the other "Cunegni" (Fig. 10). The inscription on the 
" Severini " stone has every appearance of having been re-cat and 
otherwise tampered with. 

From Traws Mawr the party visited *• Castell-y-Gaer," an ancient 
earthwork, with a saucer-shaped hollow on top, about 90 ft in 
diameter. Bound this mound there runs a wide ditch, but not 
filled with water in the manner of an ordinary castle moat. 

Fig. 11. — Rude Pillar Stone, with Incised Cross, at Traws Mawr, 

Owing to the heavy rain, it was found impossible to visit Gum 
Fawr and the Caturus stone in the Church of Merthyr Monach. 
The party returned through Trevaughan, reaching Carmarthen 
soon after 5 p.m., having spent on the whole a delightful day. 

Note. — This report has been compiled chiefly from the account 
given in the Welshman. 

Digitized by VjOOQ IC 



Carmarthen Mbbtino, August, 1906. 

Subtcriptiofu to Local Fund, 

Sir John Williams, Bart., President 

Earl Cawdor 

Alan Stepnej-GulstoD, Esq,, Chairman 

T. W. Barker, Esq. ... 

Henry Owen, Esq., D.C.L. 

R. H. Wood, Esq. 

Colonel W. L. Moi^gan... 

R. E. Jennings, Esq. ... 

Sir James ffills-Johncs, V.C., G.C.B. 

Lady Hills-Jolmes 

Mrs. Johnes 

Rey. J. Thomas 

Rev. W. Davies 

H. Meuric Lloyd, Esq. 

Rev. T. R. Walters ... 

Rev. W. W. Poole Hughes 

Rev. W. Done Bushell 

H. S. Holmes, Esq. ... 

Mrs. D. Pugh Evans ... 

Miss C. M. C. Stepney 

Dr. W. W. Leigh 

Charles Lloyd, Esq. ... 

Rev. D. D. Evans 

Miss Evelyn Lewis 

Dr. Charles Spurrell ... 

Colonel H. Davies-Evans 

Miss Thursby Pelham ... 

R. E.Williams, Esq. ... 

D. Moi-gan, Esq. 

F. W. Gibbins, Esq. ... 

John fVanoii, Esq. 

Mrs. Olive... 

Yen. Archdeacon Owen Evans 

Pepyat Evans, Esq. ... 

Misses Grifiath 

Mrs. Gwynne-Hughes... 

Miss Rickard 

Sir Lewis Morris 

R. M. Thomas, Esq. ... 

Mrs. W. J. Williams ... 

T. Morse Thomas, Esq. 

D. C. Evans, Esq. 

Colonel Gwynne-Hughes 

Rev. Sir George Cornwall, Bart. 

W. LI. Williams, Esq., M.P. 

J. Lewes Thomas, Esq. 

Miss Penman 

Miss S. A. Evans 

David Gethin, Esq. 

B. A. Lewis, Esq. 

Carried forward ... 

£ <. 


. 10 

3 8 

2 2 

2 2 

2 2 

. 2 2 

1 1 

. 1 1 

1 1 

1 1 

1 1 

. 1 1 

1 1 

1 1 

1 1 

1 1 

1 1 










. 6 

1 1 

. 2 2 

. 1 1 


. 10 



. 12 

. 7 


. 1 1 

. 10 



1 1 

. 7 


. 1 1 

1 1 

1 1 

. 6 

. 7 


1 1 




. 7 


. 7 









Digitized by VjOOQIC 



Brought forward 
Rev. T. Lewie 
Miss H«nry 

Ernest Collier, Esq. ... 
D. Hamer, Esq. 
John Lewis, Esq 
W. R. Evans, Esq. ... 
Mrs. Stacey Jonee 
Mrs. John Snow 
A. Llewelyn Davies, Esq. 


To subscriptions received as per list 


By Hire of Wagon 
„ Ditto Furniture ... 
„ Ditto Assembly Rooms 
„ Ditto Brake 
„ Ditto Rooms 
„ Gratuities to Assistants 
„ Secretary's Postages 
„ Ditto Sundry Disbursements 
„ Paid Reporter 
„ Excavating for Remain** 
„ Choque Book 
,, Balance in Hand ... 

Examined and found correct. 

A. Llewelyn Davibs, Auditor. 

Alan Stepney- Gulston, Chairman of Local Committee. 

A* $. 



1 1 



1 1 







1 1 



1 1 

MO 18 

£ 8. 


60 18 




6 4 

1 1 



1 7 

4 4 


1 5 


1 11 


1 10 



42 11 


JB60 18 

February 25th, 1907. 

Digitized by VjOOQ IC 


Digitized by VjOOQ IC 








- ■? 

•| • • 

g S « 9 
9 5^ 

-° ^ g 


1 1» 




Digitized by VjOOQ IC 


O ."t: .t; ^ 9 







Digitized by VjOOQ IC 


iaetotetos; anH Botitti of Soo&s;. 

Leland*s Itinerary in Wales. Arranged and Edited by Lucy 

ToDLMiN Smith. London : George Bell and Sons, 1906. 
It was a Lappj idea of Miss Tonlmin Smith to pnblisli the portion 
of John Leiand's famoos Itinerary that relates to Wales in a volume 
to itself; and as she has added some — though by no means all — of 
the notices of the Principality contained in the Collectanea, the book 
will prove almost indispensable t() the Welsh antiquary. The text 
has been collated with the original manuscript, which is now in the 
Bodleian, so that we probably have as perfect an edition as it is 
possible to produce. In the mere reproduction of an important 
volume like the Ithieraiy, this is much to be thankful for, but it is 
hardly sufficient for an exacting age. Leland*s bald topographical 
details are in many places no more than parts of a badly-articulated 
skeleton, the bones of which require to be decently stuffed and 
clothed by a painstaking and encyclopsBdic editor. Miss Toulmin 
Smith has not attempted to bring Leland up to date. She has 
preferred to leave him pretty severely alone, her notes being for the 
most part oonBned to trifling textual details. The really important 
work required in a modern edition of the Itinerary is that which 
should be given to a careful examination and identification of Leland*s 
topographical forms ; and for this Miss Toulmin Smith obtained the 
assistance of Dr. Gwenogfryn Evans. We much regret that we 
cannot speak with unqualified praise of the manner in which that 
gentleman has executed his task. There are plenty of perfectly 
obvious identifications which are hardly required for the enlighten- 
ment of the densest of Saxons. What is gained by a note identi- 
fying ' Place Newith ' with * Plas Newydd, ' unless it is that it 
affords an opportunity for Dr. Evans to introduce his topographical 
fad of a barred * d * for the regularly-used double * d * of ordinary 
Welsh orthography ? It is the same affectation of superior accuracy 
that doubtless leads Dr. Evans to identify ' Mouthey* with ^ Mow- 
ddwy' (with the usual barred 'd'), though the modern spelling is 
universally 'Mawddwy.* Leland's *Gurnay* is throughout given 
by Dr. Evans as * Gurvei,' whereas the ordinary style is * Gwirfai * (or 
* Gwyrfai *), which indeed is the spelling adopted by Leland himself 
on another occasion. The castle two miles from Usk, called by 
Leland ' Trergreg,* is identified by Dr. Evans as * Tre y grug,* but 
that form was never in use for the well-known manor of the lordship 
of Usk known as ' Trergrug.* * Gogarth * is not the Welsh name 
for the Great Ormc*s Head, but for a particular part of that pro- 
montory. It is, however, not so much the erroneous identifications 
that we regret — for these, after all, are not numerous— so much as 
the many really difficult place-names in the text that are left uniden- 
tified altogether. There is hardly a page that does not contain 
some word that calls for explanation, for which no explanation is 
even attempted. Thus, on p. 43, Leland says that a point marking 

Digitized by VjOOQ IC 


the ntmost limits of Wales in one direction was * Port Hojger by 
Holihed in Anglesey,' which enables ns to identify the name with 
that of the Forth Wygyr of the Triads, and to locate it somewhere 
on the northern coast of Anglesey. It should have been pointed 
out that Leland's ' Lng Harneis' was more frequently called ' Leigh 
Hames.' The river at Wrexham, now covered over in its coarse 
throngh the town, and so in danger of being forgotten save when 
it takes its revenge npon the olfactory nerves of the citizens dnring 
hot weather, is given by Leland as the *Wenbro,* which a note 
should have explained was intended to represent ' Gwenfro/ In 
the neighbourhood of Wrexham, too, a branch of the Pnlestous 
was seated at a residence called by Leland *' Marsche.' Surely Dr. 
Owenogfryn Evans should have known that these were the Pule- 
stons of Berse. 

Notwithstanding the blemishes which we have pointed out for 
correction by those who possess the book, and many others which 
our space will not allow us to indicate, we can honestly recommend 
our members to obtain it. Indeed, we heartily trust it will find its 
way into popular favour, so as to enable a fresh edition to be 
produced ; and we would then recommend the able editor to obtain 
the assistance of the one man in Wales (or elsewhere) capable of 
nnravelling Leland's conundrums in notes that would prove at once 
the admiration and the despair of every Welsh antiquary, Mr. 
Egerton Phillimore. 

Edward n im Glamorgan: Thb Stoet op the Dov^npall op the 
PiEST Prince op Walks, etc. By the Eev. John Griffith. 
London, 1904. Price 5s. 

This is a book of 257 pages (with 57 additional pages of Appen- 
dices), which would have been all the better for considerable com- 
pression. There is still some obscurity about the events of the last 
few months of the unfortunate Edward II's life, and, as most of 
that time was spent in Wales, it was a happy idea on Mr. Griffith's 
part to study the episode of the King's wanderings from the point 
of view of the Welsh historian. We cannot say that he has succeeded 
in advancing our knowledge of the deepening tragedy of the 
monarch's death ; but the story was well worthy of reconsideration, 
and, if possible, of reconstruction, in the light of Welsh history and 
tradition. Instead, however, of writing his book in a style that was 
appropriate to the dignity and pathos of the events which he records, 
Mr. Griffith has adopted a method which we cannot but regard as 
unworthy and inappropriate. He is a perferrid Welshman of the 
most " Nationalist" type, and, like many another, thinks the proper 
medium for the display of his patriotism is abuse of the other fellow. 
Such hysterical emotion as Mr. Griffith too frequently indulges in 
may be charitably regarded as an excusable incident of the National 
Eisteddfod, but is quite out of place in a serious work of history — 
and we trust that Mr. Griffith is desirous of having his little book 
regarded as such. The chief merit of our author is that he recognises 

Digitized by VjOOQ IC 


the sovereign importance, in an inquiry snch as be is instituting, of 
reliance upon original authorities for his facts. He has worked through 
the printed volumes of the " Patent and Close Rolls" with diligence 
and good results, and this must be counted unto him for righteous- 
ness ill a sphere where the imagination still exercises too potent 
a sway. He is exceedingly fond of quotation; but, amid many 
" authorities" who are frequently not authoritative upon the points 
upon which they are quoted, Mr. Griflfith rightly places in his first 
rank the late Bishop of Chester and Professor Tout ; and if he is 
not sufficiently careful in his use of so charming and picturesque a * 
volume as Mr. 0. M. Edwards's Wales, it must \^ admitted that the 
passages therefrom add to the eminent readableness of his book. Why 
Mr. Griffith has encumbered his volume with chapters upon **The 
Ancient Gods of Glamorgan," " The Picts and P and Q Celts," and 
others that have no possible connection with his subject, it would be 
difficult to conjecture ; they may be skipped by the reader with no 
loss of interest in Eang Edward's fate, and a considerable saving of 
his time and patience. 

The author, notwithstanding his diligence, has not been able to 
throw any fresh light upon the events that led up to the unfortunate 
Eang's capture. English authorities are inclined upon good grounds 
to regard Llantrissant as the place where he fell into the hands of 
his enemies. Mr. Griffith, relying upon a chronicle which he thinks 
was written by a Welshman, argues with much plausibility in favour 
of Penrhys, in the Rhondda Valley; though he prints a note 
producing a hitherto unnoticed authority whicb specifies Neath 
as the scene of the King's surrender. Nor has Mr. Griffith been 
more fortunate in penetrating the darkness that broods over 
the shocking murder of the King. He contents himself with 
a long extract from Bishop Stubbs' preface to his edition of the 
Chronicle which that great authority attributed to Thomas de la 
Moor, but which is now recognised as the work of Geoffrey le Baker, 
and rather tamely continues : " Now that the archives of the Vatican 
and of the Continent generally are rummaged and calendared by 
English experts, under the supervision of the Master of the Rolls, 
we may hear of other documents bearing on the historic doubt " — 
the mystery of the King's death. We may inform Mr. Griffith that 
there lurks at the Public Record Office the record of some judicial 
proceedings which arose incidentally out of the crime, and in which 
Edward's strong partisan, Rhys ap Gruffudd, plays an interesting 
part We are in hopes that this valuable contribution towards the 
elucidation of one of the minor points in our history may be given 
to scholars through the medium of this Journal. And we also trust 
that with enthusiasm unabated, but with style somewhat more 
chastened, Mr. Griffith may give us further evidence of his un- 
doubted capacity for the popularisation of history. Wo should like 
to have made some remarks upon the Appendices, which are the 
most valuable part of his book, but our space is exhausted, and we 
can do no more than recommend them in general terms to our 

Digitized by VjOOQ IC 

^n:h»^0l00t» (I{»mIrr^nHtH. 


JULY, 1907. 


By G. G. T. TREHERNE, Esq. 
{Bead duriff^g L(mgha/me Excursion, Augutt 16th, 1906.) 

Eglwys Cymmyn. 

The ancient little mountain church, Eglwys Cymmyn, 
is of singular interest in its situation, history, construc- 
tions, and dedication ; and, indeed, in spite of its rude 
and unpretentious appearance, presents in itself an 
epitome of Welsh ecclesiastical history. It stands 
in a commanding position in a circular " rath," or 
encampment (of about 250 yards in diameter, faced 
with stone and surrounded by an outer ditch and 
rampart), which forms the centre of — and is probably 
connected historically with — a group of earthwork forts 
contained within the territory formerly known as 
Swydd, or Cwmwt, Talacharn, and now represented by 
the comparatively modern Lord-Marchership of Laug- 
harne. Space forbids further reference to this group of 
forts, of which I have had careful surveys made, and 
from which I hope, with the aid of pick and shovel, to 
extract much information ; but now I can call atten- 
tion only to the important promontory fort of Pencoed, 
in this parish. Lewis and Carlisle both refer to the 
parish as the scene of a great battle and subsecjuent 
treaty, as alluded to by Sir John Price in his History 

6th skr., vol. VII. 17 

Digitized by VjOOQ IC 


of the Welsh Wars. I cannot find any work by Sir 
John Price with this title, neither can I find any refer- 
ence to Pencoed in any of his works. Cath Pencoed 
is one of the three decisive battles fought by Rhodri 
Melynog in the eighth century, and in the farm called 
Pencoed, in this parish, are two fields, called respec- 
tively Pare y Castell Vawr and Vach : the former 
containing an important promontory fort, with a broad 
and level field adjoining on the north side, called Pare 
yr Hedd — ** The Field of Peace" — which has been tmns- 
mogrified by English scribes into ** Pease Field." This 
Battle of Pencoed, or Cath Pencoed, opens out a wide 
field of inquiry, upon which I must not enter to-day, 
except to say that I am glad to observe a note on p. 206 
of Dr. Henry Owen's new volume of Pembrokeshire on 
Cath Pencoed, which, although it does not directly 
mention our Pencoed, indirectly tends to favour the sug- 
gestion that this was the scene of the battle in question. 
This Territory or Lordship of Laugharne, bounded on 
the south by the sea, on the east and north by the River 
Taf, and on the west by the Pembrokeshire frontier, was 
until the reign of Henry VIII included in the County 
of Pembroke. The parish of Eglwys Cymmyn is co- 
terminous with the manor of the same name, held of the 
Superior Lordship of Laugharne, and the church and its 
surroundings probably occupy the site of the head- 
quarters of the chief of the territory or tribal district 
which was subsequently converted into a Norman manor 
and an ecclesiastical parish. The name Eglwys Cymmyn 
in itself is remarkable, as offering a key to the history 
of the church. Taking first the generic " Eglwys,** and 
bearing in mind that in the Clergy List of to-day, 
" Llan ' appears as the proenomen of more than four 
hundred Welsh churches, and '* Eglwys" of only some 
half dozen cn^cient churches, it is curious that no serious 
explanation of this remarkable fact has, so far as I am 
aware, been attempted. 

As a result of much inquiry, and personal visits to 
most (if not all), of the ancient Eglwys churches, I am 

Digitized by VjOOQ IC 


inclined to offer the following explanation, which may 
at all events call attention to the paradox, and possibly 
result in further and better suggestions. The early 
missionaries, who presumably spoke Greek as the 
ecclesiastical language, would naturally on landing in 
this country ask the chief man of the* district for a 
piece of land on which to build their little church, and 
this they would naturally call their ** Eccleeia," of which 
" Eglwys" would be the Welsh expression. The mis- 
sionary having built his church would gather round 
him his followers, who in turn would form themselves 
into a monastic settlement, or religious tribe, and would 
occupy an enclosed territory, or *' Llan," the prototype 
of the modern parish. Thus we get the full title of the 
church of the religious tribe as Eglwys Llan Teilo, (or 
whoever the saint might be) Sant, ** Tne Church of the 
Religious Tribe or Community of St. Teilo." By a 
natural process of abscission the head and tail of the 
lengthy sentence would perish, leaving the kernel sur- 
viving as " Llan Teilo." If, however, this explanation 
is accepted, the question remains, why do any ancient 
churches retain the name of ** Eglwys" and omit that of 
" Llan." I suggest that " Eglwys," as applied to an 
ancient church, denotes the chapel-royal of the head 
man of the religious (or, indeed, secular) tribe ; and my 
inquiries into the incidents of the few ancient " Eglwys" 
churches remaining tend to confirm this suggestion. 
Confining our attention more particularly to the 
church now under consideration, we find that, apart 
from the circular fort in which the church is situated, 
and which probably formed the headquarters of the 
chieftain, we have adjacent to the north-east rampart 
enclosing the churchyard the remains of the old buildings 
of " Manor Court," a name which is still retained by the 
farm, although new buildings were erected further from 
the church late in the eighteenth century. Next we 
haye a custom, surviving to within the last few years, 
for the principal parishioners to maintain and repair 
each an allotted portion of the churchyard wall or 

Digitized by VjOOQ IC 


rampart ; and you will find still remaining on the face 
of the wall the initials of each farmer incised and 
marking the limit of the particular portion allotted to 
his care : a custom which may well be a survival of the 
liability attaching to the tenants of an ancient Welsh 
manor, to repair the walls of their lord's fortress 
(Seebohm's Tribal System in Wales, p. 12). 

This leads me to the specific name ** Cymmyn." As 
you are aware, the church contains a remarkable bi- 
lingual Ogam stone, notes on and illustrations of which 
will be found in the Arch. Camh., 5th Ser., voL vi 
(1889), p. 224. It is also described by Mr. Romilly 
Allen in his Monumental History of the nritish Church, 
S.P.C.K., 1889, pp. 76, etc. We have the good fortune 
to-day to have Professor Rhys with us; and in the 
hope that he may explain to us the speci^i details 
of value of this famous — and in some ways unique — 
monument, I will only say a few words by way of 
introduction, and confine myself to the part taken by 
this stone in the history of the church. When I first 
found this stone, in or about the year 1880, it was one 
of two steps on the right of the path leading from the 
entrance-gate to the south porch of the church (the 
other still remains in situ), which gave access from the 
pathway to the higher level of the churchyard to the 
east of the pathway. It was brought into the church, 
but subsequently removed and lost sight of, till dis- 
covered a second time in the rectory garden. On the 
occasion of repairing the nave in 1901, under the care 
of the Society for the Protection of Ancient Buildings, 
much and anxious consideration was given to the best 
mode of putting an efiectual check on the stone's 
erratic propensities ; and it was finally decided to build 
the strong oaken chest for it in which it now rests 
under the west window, securely safeguarded by lid, 
bar, and padlock, while giving the reverent inquirer 
easy access and fairly easy view. This stone com* 
memorates Avitoria, the daughter of Cynin ; and 
Professor Rhys some time since formed an opinion, 

Digitized by VjOOQ IC 


which he tells me he has no reason to alter, that in the 
name ** Cymmin*' we have a mis-spelling of ** Cynin ; " 
and although we cannot find any documentary evidence 
in support of this theory, it must be rememoered that 
we have no document earlier than 1248, the date of 
a conveyance of the Manor of Eglwys Cymmyn by 
the then Bishop of St. David's, Thomas Wallensis, 
as a marriage portion for his niece : so that if we take 
the date of the stone to be not later than the fifth 
century, we have a period of at least 700 years in 
which the transposition of " n" to " m" may have taken 

In a MS. in the British Museum of Edward IIFs 
time, the church is described as ^' Ecclesia de Santo 
Cumano." On the other hand, in s^n Inq, post-mortem 
of 1 Edward II, which I have very carefully examined 
at the Record OflBce, the parchment is so rubbed that 
the third letter of the word *' Cymin" may quite well be 
an " n" and not an " m.*' The transmutation of *' n" 
and ** m" is, of course, not uncommon, e.g., to quote 
iDcal instances, *' Penfro" and " Pembroke," " Lampeter" 
and "Llanpedr," but it is objected that there is no 
instance of such a mutation between two flanking 
vowels. The name of a church in Radnorshire, Llan 
Anno, which is sometimes found as ** Amo," is the only 
instance of this which I have come across, and it is not 
very convincing; but we have no time to-day for 
etymological discussions. At any rate, it is beyond 
question that Cynin, whether or no he gave his name 
to the church, was a very considerable person in the 
district ; and it is probable that the church which we 
have come to see stands on the site of his chapel-royal. 
We have in the parish a farm called ** Pare Cymmin," 
which, if Professor Rhys's theory is correct, should be 
" Cynin." (As an instance of English free translation, 
the farm immediately adjoining the churchyard on the 
west is called Common Church, a translation of the same 
character as that which converted the neighbouring 
hamlet of " Rhos Goch" into '' Red Roses.") At Tavern 

Digitized by VjOOQ IC 


Spite, three miles or so to the westward, we have Castell 
Gynin, a few miles to the north-east of us we have a 
church and parish called Llanginning, and in that 
parish a farm known as " Llangarth Gynin ; *' then a 
little further east the River Ginning, which we crossed 
to-day in coming from St. Clears, and at Trawsmawr, 
still further to the north-east, we have a stone which 
we are to see on Friday, which is inscribed ** Cunegni," 
and which probably records Cynin's burial. In the 
parish in which Trawsmawr is situated there are three 
farms bearing his name ; and far away in Cardigan- 
shire, in the parish of Llanbadarn Fawr, he has given 
his name to a district, ** Brogynin," in or over which 
he presumably had interest or influence. Now, who 
was this " Cynin V Mr. Fisher has very kindly given 
me his full Notes and References, but I have not 
time to quote them now ; suffice it to say that Cynin 
appears to have been a distinguished member of a 
distinguished family, the saintly family of Brychan, a 
fact to which I shall have to araw particular atten- 
tion when we are visiting the Pare y Ceryg Sanc- 
taidd ; and that he is said by Rees (in his Essay) 
to have been a Chorepiscopus (whatever that may 
precisely mean) of the fifth century. At all events, 
he appears to have been a leading member of the 
Church Militant in his day ; and the fact that Llan- 
ginning is described in the Myvyrian ArchcBology as the 
Church of Cynin, ** a' i Weision neu a' i Feibion" (his 
servants and his sons) suggests that it was his monastic 
foundation, as distinct from his headquarters and chapel- 
royal at Eglwys Cymmyn. The author of an Ode to 
King Henry VII, given in the lolo MSS. 314, suppli- 
cates Cynin, amongst other saints, to grant the King a 
long life, and Lewys Glyn Cothy (fifteenth century) in 
his poems frequently invokes this saint. The Irish form 
of the name, " Coinin," appears as the name of a bishop 
in the Martyrology of Donegal. It is interesting to 
remember that three miles or so to the north of 
Eglwys Cymmyn we have '*Tygwyn ar Daf" (now 

Digitized by VjOOQ IC 


Whitland), the site of Paul Hdn's famous monastery ; 
that David and Teilo were amongst the students in the 
monastery ; that Eglwys Cymrayn is one of the several 
parishes mentioned as the birthplace of Teilo ; that the 
church of the adjoining parish of Kyffijf appears as a 
Teilo Church in the Book of Llan Dav ; and it does not 
require much eflfort of the imagination to suggest that 
Eglwys Cymmyn has been honoured with the presence 
of these leaders of the early Church ; and that the place- 
name Brogynin may suggest (dates permitting) that 
Cynin accompanied Saints David and Teilo to the Synod 
of Llanddewi Brefi. I cannot find a trace of any local 
saint of the name of ** Cymin." 

Another suggestion for the origin of the name Eglwys 
Cymmyn is the ** Church of the Communion," and in 
support of this there is a tradition that on Communion 
Sundays, " in olden times," a flag was hoisted on the 
church so that the people might flock from far and 
near : a tradition consonant with the pre-eminence which 
seems to have distinguished the church throughout the 
ages, and which may have arisen from the fact that the 
church was served by a bishop, who in the early days 
of the Church would alone have had authority to con- 
secrate and administer the sacred elements. Another 
suggestion, ** The Church of— or on — the Common," is 
scarcely worth notice ; and still less a suggestion by 
Carlisle that the incription on an Elizabethan chalice 
(1574), '* Poculum ecclesia de Eglos Skymine," gives the 
correct name of the church, *' Sky mine" meaning 
** bleak," the church standing on high ground, bare of 
trees. This chalice, I regret to say, disappeared thirty 
years ago, and all efibrts to recover it have failed. A 
similar chalice, and of even date (1574), still exists in 
Cynin 's other church, Llanginning. In later times, the 
church was attached to the Benedictine cell of Monckton, 
by Pembroke, founded by Arnulph de Montgomery, who 
presumably created the Norman Manor of Eglwys Cym- 
myn. Monckton (being held of the alien Priory of Seez 
founded by Arnulph s fether, Robert), and its dependent 

Digitized by VjOOQ IC 


churches, on the occasion of the frequent wars with 
France fell into the hands of the Crown ; consequently, 
Eglwys Cymmyn was presented by Henry VI to his 
uncle, " the good" Duke Humphrey, who in turn gave it 
to St. Alban's Abbey. It is now in the gift of the Lord 
Chancellor. On a panel at the west end of the church 
you will find the names of all the rectors of the church 
whom we have been able to ascertain, from 1329 to the 
present day. Among them is Philip Marios (1389), of 
whom I shall have something to say when we visit 
Llandawke on leaving Eglwys Cymmyn. He subse- 
quently became Vicar of Ciistle Martin, co. Pembroke, 
which was also attached to Monckton Priory ; Michael 
Owen (1677), of whom you will find mention on a tablet 
in Laugharne Church; John Evans (1730), the no- 
torious author of a scurrilous pamphlet defaming 
Griffith Jones, of Llandowror ; and who turned adriu 
his curate, Peter Williams, the editor of the first 
Welsh Annotated Bible, who was born at Laug- 
harne and buried at Llandefeilog, which we visit to- 

With regard to the construction of the church, I 
have here a plan prepared by Mr. Weir, who was 
appointed by the Society for the Protection of Ancient 
Buildings to superintend and carry out the recent 
repaii-s, and which shows very clearly the probable 
dates and order of building. The oldest detail in the 
church is the small square-headed window in the north 
wall, to the west of the north door, which does not 
seem to have been glazed, and the original use of which 
is doubtful, but it may possibly have been the window 
of an anchorage. The wall between this and the interior 
of the church was apparently thickened at the time of 
the vaulting of the church, so as to make the interior 
surface level for carrying the vaulting, which seems to 
have been added late in the fourteenth century, when 
the window in the north wall against the pulpit was 
inserted. In the south wall and to the west of the 
porch is a low archway, now walled up, which it is 

Digitized by VjOOQ IC 


difficult to account for, unless it were a barrow-hole, 
through which the earth on which the vaulting was 
built was removed. This is Mr. Micklethwaite's sug- 
gestion. To the east of the south porch is an ancient 
doorway, probably the priest's door to the older church. 
There are traces of an older west window, the present 
window having been substituted at the time of the 
thickening of the north wall and the vaulting of the 

The present chancel is modern, having been built in 
1877-78. The chancel-arch is of curiously rude con- 
struction (the details are shown in Mr. Weir's plan), 
and is very similar to that at Llandawke Church, 
which we shall see this afternoon. Both churches have 
a smaller arch in the east wall of the nave, and to the 
north of the chancel-arch, giving access to the rood- 
loft. Here this arch was walled up at the time of the 
rebuilding of the chancel. At Llandawke the archway 
and a portion of the stairs still remain. 

It is noteworthy that under the entrance slabs of the 
north doorway, which appears to be older than the 
southern porch, we found five water-worn *' nine-pin'' 
stones, or ** muUers/' similar to but smaller than that 
forming the Ogam stone. Similar stones were found 
under the chancel-arch, and a large stone, of very much 
the same dimensions and quality as the Ogam stone, 
was found built into the east wall of the nave, to the 
south of the chancel arch, and there it remains. These 
stones are all shown in Mr. Weir's plan, which is safely 
deposited in the church chest. As will be noticed, on 
looking at the gable of the west wall from the outside, 
the roof covering the vaulting was at one time of a 
more acute pitch, and very likely covered with thatch. 
The present bell-cote was probably added when the 

E resent out;er roof was built. The font is old, and may 
e an adaptation of an original Norman font, cut down 
to its present dimensions. The base is new. Against 
the east wall of the nave, and over the chancel arch, 
is a tablet in memory of Sir John Perrott, at one time 

Digitized by VjOOQ IC 


lord of the Lordship of Laughame and of the Manor of 
Eglwys Cymmyn. It is worth notice, for the quaint- 
ness of the composition and the neatness of the letter- 
ing. The large slab in the chancel, to the north of the 
chancel-arch, commemorates the Shewen family of 
Rhosgoch, in this parish. The name appears several 
times in the registers of Llanelly parish church. In 
the church chest there are two coins found in the 
churchyard — a silver halfpenny of Edward I reign, and 
a copper Bristol token. 

On the north wall of the nave are the remains of 
four successive mural paintinffs. The oldest shows 
traces of polychromatic figured design ; over that red 
Tudor lettering and scroll border, and over that again 
two paintings in black lettering— one in English, the 
other in Welsh — of the Ten Commandments. The new 
memorial lectern is the work of Mr. Jack, and will 
repay inspection. 

In the tabernacle, on the south wall of the chancel, 
will be found a small cruet, of English glaze ware, 
3f ins. in height, which was found embedded in the 
south wall of the old chancel (near where the taber- 
nacle is hung), on the occasion of its rebuilding in 
1877-8. Its use is unknown, and the authorities of the 
British Museum know of only one other in England, of 
which a photograph and description will be found 
hanging on the south wall of the chancel, close to the 
cruet. On the north wall of the chancel will be found 
a facsimile representation in colour of a mural painting 
of Queen Margaret of Scotland, which still exists on 
the north wall of Binstead Church, by Arundel, co. 
Sussex, and the story of the painting is told in the 
printed description hanging by the side. This painting, 
and the Margaret Memorial Window lately erected in 
the east wall, are connected with the singular dedica- 
tion of this and the two neighbouring churches of 
Llandawke and Pendine, in honour of St. Margaret 
Marios : a unique dedication, said to have been given to 

Digitized by VjOOQ IC 


them by Sir Guy de Bryan, a great warrior, statesman, 
and church builder of the fourteenth century. He was 
Lord Marcher of Laugharne, fifty -seventh Knight of 
the Garter, and one of the chief benefactors of Tewkes- 
bury Abbey, where he lies, in a fine canopied tomb in 
a chapel built by him and dedicated to St. Margaret of 
Scotland, who was ancestress and patron saint of his 
family. Margaret Marios was the daughter of his 
sister Margaret, who married Sir Robert Marios, a 
resident landowner in this parish. The Scottish royal 
saint was grand-niece of the Confessor, who — and whose 
family — held St. Margaret of Antioch in special re- 
verence, and in whose honour the Confessor dedicated 
his first church at Westminster. The east window 
commemorates this remarkable Communion of Saints^ 
and portrays the three Margarets of Antioch, Scotland, 
and Marios. The time allotted for this Paper does not 
allow further reference to this remarkable and, indeed, 
unique dedication, but further particulars will be found 
in a pamphlet which I wrote some few years ago, and 
which can be obtained from the Rector at a cost of 
sixpence, which goes to the Margaret Memorial Fund. 
The window is the work of Mr. F. C. Eden, and has 
been erected chiefly at the expense of those bearing the 
name of Margaret throughout the Empire. 

A facsimile of Queen Margaret of Scotland's famous 
Gospel- book is preserved in the church chest. 

Parc-y-Ceryg Sanctaidd. 

The field in which we are standing is one of two 
bearing the name of Parc-y-Ceryg Sanctaidd, or *' The 
Field of the Holy Stones." The other and smaller field 
of the same name, with the addition of " Bach" (little), 
lies to the west of the larger field, and is separated 
from it by two hedges and a trackway. It, too, con- 
tains a longitudinal mound which, so far as I know, 
has never been examined. This field is partly in the 

Digitized by VjOOQ IC 


parish of Llansadurnen and partly in that of Llan- 
dawke, the parish boundary running parallel to the 
road and cutting the field in half lengthways ; the 
stones which we are looking at stand immediately on 
the boundary line. 

The road from the church and village of Llansadurnen 
on the south-east, which now is brought into the 
present high road by a sharp turning to the north- 
west, and forms the eastern boundary of this field, 
used formerly to cross the field on a line with the parish 
boundary. Immediately to the north of the smaller 
field, and separated from it by the road, is a ruined 
cottage, with the strange name ** Tavern diflas," and 
the field immediately in the north of the ruins is called 
" Pare difleis." This field contains a tumulus, on the 
hithermost side of which, and close to it, a stone axe of 
dolerite was, two years ago, found in ploughing, and 
is now in the possession of the British Museum. 

If, as I venture to suggest, "Tafarn diflas" is the Welsh 
form of the English " Cold Harbour" (Tafarn, Latin 
Tahema ; Diflas, '* insipid, worthless"), and " Tafarn" 
as an ancient place-name, indicates the line of a Roman ^ 
road, it is interesting to note that an ancient trackway ' 
leads from Tafarn diflas in the direction of Cwmbrwyn 
and its Roman remains which we visited this morning, 
and to this day affords the shortest route between the 
two places. 

I read a Paper on these stones on 22nd August. 
1903, which was fully reported in the Welshman of the 
27th August of that year, and led to some corre- 
spondence. To-day, time permits of only a condensed 
resumS of what I then said. 

I must premise by saying that the wall which you 
see built round the remains is of recent date, and was 
built to protect our treasures from cattle and other 
obtrusive creatures. 

When my attention was first called to these remains, 
some years since, all that I could see was a low mound a 

Digitized by VjOOQ IC 

J^AKCf Albt), ANt) LLANbAWKBl. 269 

few inches only in height, of darker green colour than the 
surrounding humus, and marked by the four amorph- 
ous stones standing in the position in which you see 
them to-day. The circular stone, with the cup or 
hollow in the centre, then lay at the north-west comer 
of and beyond the mound, and at its north-east corner 
lay the panelled stone with the incised circle and cross, 
which you see now placed on end slantwise, and in 
front of the larger upright stone which is placed erect 
behind it. Nothing else was to be seen until, in the 
summer of 1890, Mr. Edward Laws and I set to work 
to investigate the site. We commenced operations by 
digging carefully round the edge of the mound marked 
by the four stones, and found that they marked the 
corners of a fragmentary building of roughly-dressed 
stones, put together without mortar. We next drove a 
sectional trench, starting four feet or so from the south 
side of the enclosure, and digging down into the bed- 
rock some 3 ft. deep. We carried our trench across 
the enclosure from south-west to north-east, without 
finding the slightest trace of any burial. We found 
that the enclosed space consisted chiefly of loose stones 
thrown or fallen together, and mixed with these we 
found a few smooth water-wora pebbles, in size and 
shape like potatoes, small and big. These were all we 
found, with the exception of a small piece of white 
quartz or crystal, of the kind usually known as St. 
David's diamonds, and of about the size of a walnut ; 
also a small fragment of burnt red clay, of the size of 
a marble. Under the circular stone and covered by 
from 8 ins. to 12 ins. of soil, we found what turned out 
to be the lower portion of the cross-marked stone 
already referred to. This larger stone was lying recum- 
bent on its back, and looked at first sight very much 
like the lower half of the stone already discovered. 
This surmise has proved to be correct, Not only do 
the details of the fracture along the upper edge of the 
larger stone correspond (except where by weather or 

Digitized by VjOOQ IC 


other agent fragments have been removed) with those 
of the fracture along the lower edge of the smaller 
stone, but on the upper part of the larger stone traces 
of a cross and circle similar to that in the smaller stone 
can be detected (more easily by the eye of a camera than 
by the human eye), a portion of the upper segment of 
which is wanting, and is found on the upper stone. 
The bigger, you will observe, is also panelled, 
but, owing to its being very much more weathered and 
damaged, the details are difficult to decipher. It was 
thought better not to attempt to re-join the stones, but 
to place them in their present position, so as to enable 
the spectator to form a fairly accurate notion of their 
original appearance when forming one stone. The four 
amorphous blocks were left in their original position. 
The loose stones found in the mound were, as you see^ 
placed around the mound, making its boundaries, 
and in the limestone slab on which the upper part 
of the incised stone rests a hollow was made, in 
which the small stones or pebbles above referred to 
were placed. 

Now, what are we to say as to the origin of these 
stones ? Local tradition tells us that in old days coffins 
on their way to burial were wont to be rested on the 
cross-inscribed stone, and sprinkled with holy water 
from the cup in the round stone ; this is referred 
to in Miss Curtis's book, Antiquities of Laugharne, 
Pendine, and their Neighbourhood ; and the custom of 
resting coffins on convenient stones, or by roadside 
crosses, was not infrequent or unreasonable, especially 
in days when the journey was long and the road 

Probably, however, these stones belong to an age 
long prior to the use of coffins, although they may well 
have been subsequently used for such a purpose. 

The result of my investigation tends to confirm a 
suggestion made to me by Mr. Romilly Allen, that we 
may possibly have here the remains of a building 

Digitized by VjOOQ IC 

SANCTAIDD, and LLANDAWfefi. 271 

similar to the so-called altar stations (** Altoir") which 
stand to this day in the Cashel on the Island of Innish- 
murray, off the coast of Sligo. (In illustration of this 
possible connection, I may mention that Zimmer, in his 
Celtic Church in England, p. 35, gives '*Altoir" as one 
of the Irish words derived from the Latin through a 
British medium, and so indicating that Christianity 
came to Ireland from Britain.) 

The Innishmurray remains are fully described and 
illustrated in Mr. W. F. Wakeman's Book of Survey 
of the Antiquarian Remains in the Island of Innish- 
murray. I vol. London, 1893. 

These Altoirs consist of quadrangular heaps of stones, 
5 ft. to 7 ft. long by about 5 ft. wide and 3 ft. in height, 
and built without mortar, with a large stone, or 
** Dalian" (in some cases there are two such stones), 
standing upright in the middle or at one end of the 
altar, the Dalian being generally inscribed with a circle 
and cross. On the top of these structures are found 
smooth water- worn stones or pebbles, of various sizes 
and descriptions, in some cases incised, and which are 
locally known as *^ cursing-stones." It is, or was in 
quite recent years, the practice of the inhabitants, when 
in a prayerful mood, to go round these altars from left 
to right with the sun, thus forming the Irish '*desiul," 
or holy round ; or, if revenge was in their mind, they 
would reverse the operation by going against the sun 
" widdershins," turning the stones as they went, and 
uttering a curse against the particular object of their 
hatred. On one of these altars called '* Altoir beg " 
(the little altar), illustrated at p. 71 of Mr. Wakemans 
book, are a considerable number of small water- worn 
pebbles, apparently taken from the sea-shore. On others 
the stones are larger, and in some cases incised. 

I have here a curious stone which I found two or 
three years ago on the farm of Beefs Park, in the 
adjacent parisn of Marros, and which is seemingly 
of the same character as some of the larger cursing- 

Digitized by VjOOQ IC 


stones found at Tnnishmurray. If this stone is of that 
class, it implies the existence in former days of an altar 
station in that neighbourhood ; but possibly (although 
this is less likely) it may have been removed from here 
thither.^ I had every intention of visiting Innish- 
murray this summer, as in such cases seeing is believing, 
and it is diflScult to compare two objects without seeing 
both. Unfortunately, I have had no time to make this 
rather difficult journey, but I have been fortunate in 
making the acquaintance of Mr. Cuthbert Harrison, of 
Sligo, who has visited Innishmurray and phographed 
(he is a professional photographer) the remains ; and he 
has not only sent me some specimens of his work, but 
has kindly undertaken to make another visit to the 
island, in order to take photographs of further details ; 
and hopes to be able some time next month to come 
over and visit Pare y Ceryg Sanctaidd, so that we may 
compare notes. There do not appear to be at Innish- 
murray any stones resembling our circular cup-stone (if 
** cup-stone" is the correct term for it). Cup-stones, as 
we know, are exceedingly rare in Wales, although 
we have on Pendine Head, two miles to the south- 
west, a flat recumbent stone with two cups. Pendine 
Head, let me say in passing, well deserves a visit, 
which I regret extremely that time now forbids. 
We must hope that members will be so satisfied 
with what they have seen in Swydd Talacharn that 
they will individually, if not collectively, pay another 
visit. I can promise that their curiosity will be re- 

The cup, or hollow, in this circular stone measures 
about 8 ins. in diameter and 6^ ins. in depth. Possibly 
the stone may be wholly unconnected with the other 
relics, and have formed the base of a roadside cross 
standing on this ** Holy Ground." 

The four amorphous blocks resemble similar stones 

^ This stone is now in the Welsh Masoum at Cardiff. 

Digitized by VjOOQ IC 


placed at one end of one of the Innishmurray Altars, 
and on which devout worshippers were wont to kneel. 

If these remains prove to be akin (analagous, if not 
actually homologous) to those at Innishmurray, it fol- 
lows that the builders of both must have been con- 
nected in circumstances or idea, or have had knowledge 
of each others' method of work ; and here we may 
possibly have a clue through the medium of the multi- 
tudinous Brychan family. At Eglwys Cymmyn, as we 
have seen, we were brought vis-a-vis a distinguished 
member of this saintly family; St. Cynin. In the next 
parish, between us and the sea, we have a farm called 
Pare Cynog, containing Ffynon Cynog, orCynog's Well, 
and as a place-name Merthyr Cynog, a name which 
appears in Glamorganshire, and indicates presumably 
the possession of the relics of the Saint. The name 
" Cynog " appears amongst the numerous progeny of 
Brychan. It has been suggested that ** Toch,'' in 
" Castle Toch," the name of an adjoining farm which 
we passed on our way hither from Eglwys Cymmyn, is 
a corruption of Doch or Doc, a shortened form of Cadoc, 
another eminent Brychanite. The same origin has 
been suggested for **dawke" in Llandawke, the name 
of the church which we shall next visit ; and we have 
other Brychanite names in the neighbourhood, such as 
Brynach, Elidyr, Clydwyn, and Pab. Several field- and 
place-names in this locality also denote an Irish (to use 
a popular phrase) origin, and amongst the many nebu- 
lous circumstances which surround the Brychan story, 
his connection with Ireland is sufficiently clear. Next, 
the connection of St. Columba with St. Molaise of 
Innishmurray is, I believe, generally accepted ; and we 
find on Innishmurray traces of Columba and his so- 
called twelve apostles, such, for instance, as Reilig (the 
"resting-place" of) Columcil, Reilig Odrain, etc. Among 
the twelve apostles whom Columba took with him to 
Scotland were Rhun, whom Skene in his Four Ancient 
Books of Wales, vol. i, p. 52, identifies as one of the 

0TH 8ER., VOL, Vn. 18 

Digitized by VjOOQ IC 


sons of Brychan ; and Aedan, whom Professor Rhys, in 
his Celtic Bintain, refers to as the son of a daughter of 
Brychan ; and we know that Rhun, or a man of that 
name) is said to have been buried at Llandevaelog, 
which may possibly be (there are two parishes of this 
name) the Llandefeilog in this county, which we hope 
to visit to-morrow. 

Now, this suggested origin of these remains seems to 
be worthy of careful considemtion by those of our 
members whom I have the pleasure of addressing to- 
day. This theory, if found to be justified by the facts, 
would seem to account for the origin of some, at all 
events, of the many solitary pillar stones, or Menhir, 
found throughout the Principality. The loose walling 
forming the altar would naturally disappear in the 
course of ages, leaving the monolith standing alone to 
tell its tale. The inaccessibility of Innishmurray has, 
fortunately, preserved its precious relics for us ; and if 
these relics now before us are in the result found to be 
akin to those on Innishmurray, may we not confidently 
look forward to other traces of similar remains, espe- 
cially in those districts of Wales where the Brychanite 
influence prevailed, as is the case in that portion of 
Carmarthenshire in which we now find ourselves. 

Since this Paper was written, I have received the 
following letter from Mr. John Ward, the Curator of 
the Welsh Museum at Cardiff : — 

** 1. Parc-y-Ceryg Sanctaidd. — The hypothesis you refer to was 
suggested by Miss Curtis's view of the Grist Cross at Laug- 
harne. She shows the lower fragment of a raediteval shaft 
inserted into a circular base-stone, much like your " cup-stone*' at 
the above. Circular bases are unusual, but I feel sure I have 
seen several examples. 

" By the half-buried road on Merthyr Mawr Warren are the 
remains of a wayside cross. If my memory serves me aright, 
the base-stone is circular, with a circular socket, only, unlike the 
Grist base, it is decorated with a thirteenth-century moulding. 
The usual form is square or octagonal. 

"It cannot be doubted that the parish boundary through 

Digitized by VjOOQ IC 


Parc-y-Ceryg Sanctaidd perpetuates a former road : in fact, we 
found some indistinct traces of such a road. A cross by its side 
would almost certainly rest upon a mount of steps, and the square 
structure with the four great corner-stones suggests the plat- 
form, or lower step-course. To put it another way — it equally 
tits the theory of a mediaeval wayside cross with the theory of 
an ' altoir/ 

" We know that in pre-Reformation times funerals halted at 
wayside crosses for a paternoster, etc ; and we can well imagine 
that after the cross was thrown down, or fell from neglect, and 
even after the road ceased to be used for ordinary traffic, forcef 
of long habit would keep alive the custom of funerals passing by 
the site, and making the halt for the Lord's Prayer. The cross 
faded from memory, and the stones, from the circumstances just 
given, came to be known as ' resting-stones.* 

"The relation of this supposed mediaeval cross to the pre- 
Norman crossed slab on the site is a little difficult. So far as I 
am aware, the pre-Norman crosses were sepulchral ; but they 
were by no means always associated with churches ; yet there is 
no evidence (that I am aware of) that any of them were origin- 
ally regarded as simply wayside crosses, although it is con- 
ceivable that they were often raised near lines of traffic. We 
may imagine that our stone commemorated some early ' saint,* 
and that in later times it was replaced by another of a form 
which would then more strongly appeal to the passer-by as a 
cross. Or, possibly, the old stone, from some cause or other, 
was broken, as we now see it, and this led to the erection of the 
second structure. 

" This second structure would probably take the form of a 
crucifix, or it may have had a coped head, containing a crucifix 
on one side, and, say, St. Mary on the other. 

" Of course, this is only hypothesis, but the more hypotheses 
the more likely is one to prove correct.'* 

Furthermore, since this Paper vras written, Mr. 
Cuthbert Harrison has visited the Parc-y-Ceryg Sanc- 
taidd with me ; and so far as he could judge from the 
little that remains to be seen, he was emphatic in his 
opinion that, had he come upon these remains in his own 
(adopted) country, he would have considered them to 
be of the same class as the '^ altoirs" on Innishmurray. 
The central large stone, or Dalian, however, he states, 

18 --^ 

Digitized by VjOOQ IC 


is very much larger than any at Innishmurray, and 
there is nothing there in the least resembling the 
circular stone with central cup which we have here. 
He kindly gave me photographs taken by him at 
Innishmurray, and hopes to take and send me more 
this next summer : which with his permission I will 
send to the Journal for publication, together with 
photographs of our relics here, for the purpose of easy 
reference and comparison. 


A visit to this secluded and picturesque little church 
follows in appropriate sequence our visit to the far older 
and historically more interesting church, Eglwys Cym- 
myn. We saw in the Margaret Memorial Window 
there a representation of Sir Guy de Bryan, holding a 
model of Llandawke Church in his hands in the act of 
dedication. Among the list of rectors there we noticed 
the name of " Philip Marios" as Rector in 1389 ; and 
here we have the reputed eflSgy of Margaret Marios ; 
and there is good reason for believing Philip and 
Margaret Marios to have been brother and sister, 
children of Sir Robert Marios and his wife Margery 
(or Margaret), the sister of Sir Guy de Bryan. Each of 
these two churches is said to have been dedicated by 
Sir Guy in honour of " St. Margaret Marios ; " and the 
tradition that this little church was built in connection 
with a small religious establishment founded by Sir 
Guy, and presided over by his niece, seems to be justi- 
fied by appearance and circumstance. There is no trace 
of any village near the church, or in any part of the 
parish. The Rectory House, which is modern, stands, 
as you see, close to the churchyard, and may mark 
the site of an ancient conventual building ; while the 
fine timber surrounding the churchyard, the pond, 
the extensive range of buildings now belonging to the 

Digitized by VjOOQ IC 


Erincipal farmhouse or mansion of Llandawke, the farm 
eing co-terminous with the parish, all tend to suggest 
a settlement, or Home of Ancient Peace. 

The living is in the gift of the Nanteos family, as 
also is the neighbouring church of Pendine, which is 
held with Llandawke. 

Miss Curtis, whom I have already quoted in my 
account of Parc-y-Ceryg Sanctaidd, writes (in 1880) 
that a manor house once stood close to Llandawke 
church, and that two members of the de Bryan family 
were buried in the church. She refers to the finding 
of a gold urn of early date, and to various local legends 
incident to the church and its surroundings. She also 
refers to the eflSgy as representing the foundress of the 
church, and relates a tradition that, one day, when she 
(the foundress) was returning from her house at Broad- 
way (a mile or more to the south of Llandawke) from 
arranging afiairs regarding the church, she was attacked 
by robbers, who cut her into three pieces ; and that to 
commemorate her martyrdom her eflSgy was divided 
into three separate pieces. This story was doubtless 
invented to account for the fact that the eflSgy is 
divided into three parts by two sharp divisions, which 
it is diflficult to account for. They are too sharp and 
regular to have been caused by a blow, and there are 
no traces of saw-work. When I first saw the eflSgy 
(which I have the authority of Abbot Gasquet for 
saying, is undoubtedly that of a religious lady of rank 
of the late fourteenth or early fifteenth century), many 
years ago, it was lying covered with moss and tangle in 
the south-west corner of the churchyard, and in 1903 
was placed in its present position at the expense of the 
late Mr. Thomas Harries, who owned the whole parish 
(consisting of one farm and under thirty inhabitants), 
and who also, with characteristic generosity, paid the 
expenses of the work done for the preservation of the 
relics in Parc-y-Ceryg Sanctaidd. 

Of the construction and architecture of the little 

Digitized by VjOOQ IC 


church, the nave of which measures 83 ft. by 17 ft., 
there is not much to be noted. The simple chancel- 
arch and the opening in the east wall of the nave, 
originally giving access to the rood-loft, are very 
similar (either copied the one from the other or the 
work of the same builder) to what we saw at Eglwys- 
cymmyn. Generally speaking, the building may well 
be of late fourteenth-century (Sir Guy de Bryan died 
in 1391) work, and notwithstanding restoration, pre- 
serves much of its simple charm and seemliness. The 
windows, particularly that in the south wall of the 
chancel, are worthy of notice. The font seems to me 
to be of older date than the church, but it is difficult, 
if not impossible in Wales, to compute the dates of 
works of Art by English standards. A notable in- 
stance of this is furnished by the effigy where we find 
details of the lady's attire (for instance, the tight 
sleeves with buttons along the outside seam), identical 
with those in the dress worn by her namesake Margaret 
de Camoys, as shown by her brass in Trotton Church, 
CO. Sussex, and who died in 1310, or nearly one hundred 
years before our foundressi A beautiful drawing of the 
effigy has been most kindly made for me by Miss 
Edwards : who, I should like to be allowed to hope, 
may be induced some day to include Carmarthenshire 
in the good work she is doing in Pembrokeshire, in 
making drawings of the ancient monuments throughout 
the county. And now, before coming to (in the eyes 
of archseologists) the most valuable possession of the 
church, I should like to invite suggestions as to the origin 
of the name Jj\a,ndaivke. It is variously spelt " Dawg" 
or " Dawke." In does not appear in the Taxatio. 
In the Valor it is Llandawke ; in the Liber, Llandawk 
(without the final e). Speed's Map of Carmarthenshire, 
1662, gives it as LlanaacA. Leland gives Llanrfou^ 
and lAsindotighe as a place-name in Glamorganshire, 
possibly identical with one of the two churches of that 
name (also called Llandoch, or Llandocha), both dedi- 
cated to St. Dochdwy, who is said by Rees {Essay^ 

Digitized by VjOOQ IC 


p. 219) to haye been a Bishop of the sixth century, to 
have accompanied Cadvan to Bardsey, and to have 
had the care of the diocese of Llandaff during Teilo s 

If " Dawke or " Dawg" can be made out of *' Doch," 
we may have here another instance of the connection of 
Teilo with this district. The Glamorganshire Llan- 
dough appears variously as Llanc?ocA, docha, doghe 
(1314), aochar, doche, dochey. Other suggestions are, 
that in Dawk we have a shortened form of Oudoceus, 
who succeeded (Rees, p. 253) his maternal uncle Teilo 
(again) as Bishop of Llandaff; or of Cadog, or Cadoc, 
whose name not infrequently appears in the shortened 
form Doc (another Brychanite, by the way), who founded 
Llangadog, in this county. The Toch, in Castell Toch^ 
— the name of a neighbouring farm to which reference 
has already been made — suggests a similar origin. 
Another suggestion is that '* Dawke" is a corrupt ren- 
dering of "Dog" (in Dogmael), "Dog" standing as 
the name of the saint, with " mael " as an added adjec- 
tive. We have "Dog" also as the first syllable of 
"Dogfan," appearing in the Cognacio as a son of Bry- 
chan, said to have been slain at Merthyr Dogfan, in 
Dyfed, or Pembrokeshire, and that a church (of which 
no trace is left, even the site being unknown) was 
consecrated in his memory. And now we come to the 
Ogam stone, any description of which I am glad to be 
able to leave to Professor Rhys, who has already written 
more than once about it, and from whom we may hope 
to-day to hear his matured opinion. I will merely say, 
that when I first saw the stone, many years ago, it 
formed a stepping-stone into the church, whence it was 
removed to the vestry, under the tower, and in 1903 
to its present and, I hope, final resting-place. A mirror 
has been placed against the wall opposite and along 
the back of the stone, so that both sides may be seen 
without removing the stone. 

^ " T" and " D" would, in certain circumstances, be interchange- 
ablo in Welsh. 

Digitized by VjOOQ IC 


The stone is noticed in — 

1. Ogam^ Inscribed Monuments of the GaedhilL R B. Brash, 

London, 1879, p. 347, with illustrations. 

2. Lapidarium Wallics, J. O. Westwood, Oxford, 1876-9, p. 92 

with illustrations. 

3. Arch, Camb,, 1867, p. 343 ; 1874, p. 19; 1875, p. 413. This 

last was the date of the last visit of the Society. 

4. Y Cymmrodor, vol. xviii, 1904, p. 21. Professor Rhys on the 

Welsh Englyn, where the Latin inscription is given as an 
instance of the uge of " Hexameters" in the Welsh Englyn, 
as also is that on the Eglwys Cymmyn stone. 

5. The Gentlemaris Magazine for January 7th, 1838, vol. ix, 

p. 44. . 

6. Lectures on Welsh Philology. John Rhys. London^ 1877, 

p. 298. 

7. Ogham Inscriptions in Ireland, Wales, and Scotland, Sir 

Samuel Ferguson. Edinburgh, 1887, p. 118. 

Digitized by VjOOQ IC 




By Professor J. E. LLOYD. 

The story of Carmarthen is, in certain of its aspects, a 
familiar one, which has often been told and which 
scarcely needs re-telling on the occasion of this — the 
second — visit of the Association to the historic centre 
of the Dimetian country. Its legendary connection 
with the wizard Merlin, its importance as a military 
station in Roman times, and the part it played as a 
royal borough, protected by a strong castle, during the 
ages of conflict between Welsh and English, have often 
been discussed. There is, however, one period of its 
history, and that not the least interesting, to which 
little attention has been given ; and it is in the hope 
that I may enlist your interest in this neglected period 
that I venture to offer the following observations. 

I refer to the interval between the Roman occupation 
and the emergence of the town (in 1137) as an important 
stronghold of Norman power, carried by storm in that 
year by Owain and Cadwaladr, the sons of Owain 
Gwynedd. There is, of coui'se, a great lack of historical 
material for those years ; but the judicious use of what 
we have will enable us, I think, to reach some conclu- 
sions which will fairly fill the gap between the Mari- 
dunum of the Romans and the Carmarthen of the reign 
of Stephen. The first point is the identification of 
Carmarthen with the Llandeulyddog of the well-known 
list of the Seven Bishop-houses of Dyfed. In that list, 
which is known to us from the Dimetian Code of the 
Laws of Hywel Dda,^ Llandeulyddog stands sixth ; it 
was supposed by Rees, the author of the Welsh Saints, 

^ Ed. Aneurin Owen, vol. i, p. 668 (Llann Denljdawc) ; vol. ii, 
pp, 790, 869. 

Digitized by VjOOQ IC 


to be " in the southern part of Pembrokeshire' V and 
Aneurin Owen, in his edition of the Laws, could only 
suggest it was Llandudoch or St. Dogmaels. But in 
the Liber Landavensis there is a reference to " lann 
toulidauc ig cair mirdin,"^ and the allusion, slight as it 
was, did not escape the keen eye of Dr. Gwenogvryn 
Evans, who, in his Index,* locates this church definitely 
at Carmarthen. It thus becomes plain that Carmarthen, 
when the walls of its Roman fort had been dismantled 
and its first military period came to an end, began life 
again as an important ecclesiastical centre. For the 
bishop-houses of this list were not ordinary churches or 
ordinary episcopal manors ; what is recorded of them 
shows that they were churches of special distinction and 
ample resources, served by groups of clergy who in- 
herited monastic traditions. Whether the name "es- 
gopty" may be taken as proof that each had originally 
its bishop, is perhaps open to doubt ; but it is certain 
that, when the list was drawn up in its present form, 
Llandeulyddog had an abbot, a man of wealth and high 
social standing. Now an abbot implies a body of 
dependent ecclesiastics, and thus we may proceed to 
include the place under the general denomination of 
** Clas"; it belonged to that type, of which there are so 
many examples in the early Middle Ages, both in North 
and South Wales, viz., the primitive monastery still 
retaining some monastic features, such as the title of 
abbot, but converted in practice into a group of secular 

Llandeulyddog had an endowment of lands, and this 
partially accounts for its appearance in the Liber Landa- 
vensis. According to the legend of St. Teilo, Teulyddog 
was onQ of his disciples,* and the church of LlandafiP, 

1 Welsh Saints, p. 253, 

^ Ed. Gwenogvrjn Evans, p. 62. On p. 124, " rairdin" baa 
dropped out. 3 p, 409. 

-* Ibid., p. 115 (Toulidaac). The "dd" is not only to be naturally 
inferred from the Old Welsh form, bat is actually found in the 
poems of Lewis Glyn Cothi. — See the Oxford (1837) edition, p. 49 : 
** Mac Teilaw iddaw ; mae Telyddog." 

Digitized by VjOOQ IC 


under the energetic leadership of Bishop Urban, was 
for claiming as the inalienable property of that see, not 
only all the Teilo churches in South Wales, but also 
those which bore the names of his disciples. The fight, 
in this case against St. David's, was waged for many 
years, but without success ; with the death of Urban 
in 1133 (or 1134), all the spirit died out of it, and St. 
David's was suflFered to enjoy without question its 
authority over the Teilo churches west of the Tawe, 
and, among them, over Llandeulyddog. 

Until the death of Rhys ap Tewdwr in 1093, this 
district was not aff*ected by the Norman Conquest. 
But, immediately after that ciniel blow to the hopes of 
the men of South Wales, Norman invaders poured into 
Ceredigion, Dyfed, and Ystrad Tywi. While the west 
of Dyfed was given by Rufus to Arnulf Montgomery, 
the first builder of Pembroke Castle, its eastern half 
was bestowed upon William fitz-Baldwin, who was 
a cousin of Gilbert fitz-Richard (the first Norman Lord 
of Ceredigion), and succeeded his father, Baldwin de 
Meules, as Sheriff of the County of Devon.^ William, 
it may be remarked, was not the only Devonshire man 
who joined in the attack upon South Wales during this 
reign ; Richard of Grain ville, the conqueror of the 
Neath Valley, was of Bideford ;* the fitz-Martins, who 
settled in Cemais, came thither from the neighbourhood 
of llfracombe,* and the founder of Whitland Abbey 
was a John of Torrington.* The natural issue of the 
grant to William was the building of a castle to secure 
the new lordship, and this was placed, not at Car- 
marthen, where the "clas" were left for the time in 

1 Brut y Tyurysogion, od. Ab Ithel, 8. a. 1094 (i-eally 1096) ; 
Roand, Feudal England, p. 330. 

^ He gave a reut of twenty shillings ia Littlehara, hard by Bide- 
ford, to Neath Abbey when he founded this house in 1130. — Afonast, 
Angl, V, 259. 

3 Owen's Pembroleshire, ©d. Dr. Henry Owen, pp. 430*2. 

* Mona^t. Angl. (new edit), v. 591 : Royal Charters ofCarmnrHien, 
ed. Alcwyn C. Evans (1878), p. 73. 

Digitized by VjOOQ IC 


undisturbed occupation of the old ** caer" or fort, but a 
mile lower down the river, at Rhvd y Gors. For a few 
years, therefore, nothing is heard of Carmarthen, and 
the place it occupied later as the chief stronghold of the 
district is temporarily filled by Rhyd y Gors. The 
history of this castle was so short that, so far as I 
know, no remains of it have survived ; not even its site 
is certainly known. In the general upheaval of 1094, 
when the Welsh of Deheubarth rose in revolt againpt 
their new masters, it was one of two castles (Pembroke 
being the other) which survived in Dyfed and Ceredi- 

fion. In 1096, William died while the revolt was at its 
eight; this so discouraged the garrison that they 
abandoned Rhyd y Gors, probably retiring by sea to 
their Devonshire homes. The district then for a few 
years relapsed into Welsh hands. Henry I, indeed, 
recognised the claim of Richard fitz- Baldwin to succeed 
to what his brother had held in Wales, but Richard 
made no attempt to enter into possession until 1105, 
when he gave orders for the rebuilding of the castle.^ 
Hostilities now followed with the Welsh lord of Kid- 
welly, a commote which marches with Dyfed from the 
mouth of the Towy to Abergwili. Hywel ap Gronw 
attempted to destroy the rising fortress which so 
seriously menaced the peace of his borders, but in 
vain ; in 1106 he was himself slain by the treachery of 
one of his own men, acting in concert with the garrison 
of Rhyd y Gors.* In this way Norman supremacy in 
the district was assured. 

This is the last mention of Rhyd y Gors Castle, which 
at this point drops silently out of history, together 
with the claims of Richard fitz-Baldwin. Though 
Richard lived until 1136, and continued to be one of 
the great men of his county, he played no further part 
in the affairs of South Wales, and never again put 
forward any claim — so far as is known — to exercise 
authority in the valley of the Towy. When light is 

^ Bj-ut y Tywt/sogion^ s. a. 1102. 
' Ibid,, 8. a. 1103. 

Digitized by VjOOQ IC 


next thrown on the affairs of the district, in 1109, 
Walter of Gloucester, sheriff of that county, and one of 
Henry's active officials, is found at Carmarthen,^ and it 
is clear that the first steps are being taken for the 
building on the spot of a royal stronghold, to supersede 
Rhyd y Gors, and to keep Welsh prince and Norman 
baron alike in check in the interests of the Crown. We 
next come to the year of the rising of Giniffydd ap 
Rhys, who, in 1116, rallied around him the youth of 
South Wales in an effort to regain the lost crown of 
Deheubarth. Enthusiasm was on the side of Gruffydd, 
but the movement did not commend itself to the more 
experienced Welsh leadera, who knew the strength of 
Henry's position, and many of them took the king's 
side. Among these was Owain ap Caradog, a chieftain 
of Cantref Mawr, who, in return for his loyalty, was 
entrusted with the defence, during a particular fortnight, 
of Carmarthen Castle. It chanced, unhappily for him, 
that the attack of Gruffydd was delivered during 
Owain's term of office as guardian of the Castle. 
Advancing incautiously to meet the foe, he found 
himself without support, and was overwhelmed and 
slain.* The enemy forthwith destroyed the " rhag- 
gastell," or outworks, but failed to capture the **twr," 
or keep : a distinction which shows that the first Castle 
of Carmarthen was of the type commonly found in this 
age. It had a mote, or mound, crowned with a tower 
or donjoUj and surrounded by a ditch ; while an outer 
court, or bailey, adjoined the ditch, having its own 

Later in the same year,^ when Owain ap Cadwgan, 
of Powys, had taken the field against the insurgents, a 
party of them is found fleeing for protection to Car- 
marthen, where they were treated with some indul- 
gence. Owain himself showed them no mercy ; but as 
he was carrying them off as prisoners, he was overtaken 

^ Bnit y Tywyiogion, b. a. 1106, p. 88 (Gwallter ncheluaer Kaer 
2 Ibid., pp. 124, 120. 3 Pp. 136, 138. 

Digitized by VjOOQ IC 


by Gerald of Windsor — with whom, since the affair of 
Cenarth Bychan, he had been in bitter feud — and a 
force of Flemings from Rhos, and, being thus taken by 
surprise, was forthwith slain. The chronicle explains 
how Gerald and his Flemings happened to be at 
Carmarthen. They were there to meet the King's son 
— the young William who was heir to the Crown, and 
who, being now thirteen years of age, had on March 19th 
of this year received at Salisbury the homage of the 
great men of the land.^ The incident is one more 
iUustration of the fact that Carmarthen had now 
become a centre of royal administration for South- 
West Wales. Homage was no doubt done to William 
here by the Norman and the Welsh magnates of the 
district, whose jealous watchfulness of each other's 
doings had not allowed them to travel to Salisbury. 

The new castle of Carmarthen had not been placed 
within the limits of the Roman fortress, but on a height 
a few yards to the west, on the very brink of the river 
Towy. The little settlement of foreign tradei-s and 
artisans which, under the name of a **burgus/' or 
borough, was generally planted at the foot of an 
important castle, also lay outside the old fort to the 
north of the castle. Thus arose a distinction which 
persisted until 1764 between Old and New Carmarthen, 
the Vetus and the Nova Villa, or Civitas.* New 
Carmarthen was a royal borough, of which the first 
known charter dates from the reign of John,' but which 
had no doubt from the beginning such special privileges 
as would attract to the spot the settlers so necessary 
for the comfort and ease of the garrison. Old Car- 
marthen remained an ecclesiastical preserve, still under 
the protection of the venerable and mysterious Teuly- 
ddog.* In the corner of the old fort nearest to the 

^ Florence of Worcester. 

2 Spnrreirs Carmarthen, 1879, p. 24. 

5 Royal Charters, p. 1. 

^ See, for instance, No. 135 in the Cartulary of Carmarthen 
Priory, in which ** Thomas, prior monasterii Saucti Johannis Evan- 
geliste de KermerdTn, dominns Veteris Yille de Kermerdyn," 
claims a number of franchises. 

Digitized by VjOOQ IC 


town and Castle a new church was built in honour of 
St. Peter, and this was the ordinary resort, no doubt, 
of the burgesses and the garrison, though Llandeulyddog 
continued to be the mother-church, with the right to 
receive the tithes of its extensive parish, which included 
Llan Gain, Llanllwch, and Eglwys Newydd. About 
1120, however, an inoportant change took place. It 
was the policy of the Norman conquerors to establish 
in their lordships, in close proximity to the principal 
castle, a monastic house, generally a cell of some 
English or foreign abbey, which would secure for theui 
on the spot the advantages of spiritual support in their 
long and weary struggle with tne Welsh. It was also 
their policy to break up where they could the organisa- 
tion of the " clas," an institution which made no appeal 
to them, since its monastic features had become attenu- 
ated almost beyond recognition, and it corresponded 
to no ecclesiastical type of high repute with which they 
were familiar. Thus Henry I was but following in the 
footsteps of his marcher vassals, when he gave Llan- 
deulyddog to the great abbey of Battle, in Sussex, 
which his father had founded in memory of the victory 
of Hastings. The chronicle of the abbey says that, in 
the time of Abbot Ralph (110M124) the King, of his 
great love for the Abbey, bestowed upon it " a certain 
church in Wales, founded in honour of St. Peter the 
Apostle, and situated in the city called Carmarthen 
(Chsermerdi), with all its appendant rights, to be freely 
and quietly possessed for ever. He also gave another 
church, founded there in remote ages in honour of St. 
Theodore the Martyr (a bold shot at the unknown 
Teulyddog !), and land therewith not far distant which 
is called Pentewi ; because he thought this would be 
advantageous, it being very fruitful in corn."^ 

The next figure who appears in the history of 
Carmarthen is Bishop Bernard, of St. David's. Bernard 
was the first Norman bishop of that See, appointed in 

1 Chronic(yn Monasterii de Bdlo^ LondoD, 1846, pp. 55, 56, 

Digitized by VjOOQ IC 


1115 in direct furtherance of Henry's policy of subju- 
gating the Welsh church to the royal power, so that its 
influence might no longer be used on behalf of the 
Welsh in the struggle between the two races. The 
Bishop had a manor at Abergwili, and no doubt often 
visited Carmarthen. He was anxious to strengthen 
the new colony established there, and thought that, 
instead of the little cell of Battle,* a more imposing 
foundation might be placed there, and one, too, more 
dependent upon himself He was in favour at Court, 
having been one of Queen Matilda's chaplains, and he 
used his position to press this matter persistently upon 
the king, to the annoyance of the monks of Battle. 
At last, on the occasion of the election of a new abbot, 
in 1125, he carried his point. The King transferred to 
Bernard the Carmarthen possessions of Battle, and 
gave that Abbey compensation in Hampshire.* Bernard 
now instituted at Llandeulyddog a house of Regular 
Canons of the Augustinian pattern, under the govern- 
ment of a prior. There was a re-dedication of the church 
to St. John the Evangelist, and the official style of the 
place now becomes '* the church of St. John the 
Evangelist and St. Theuloc of Carmarthen." Teu- 
lyddog's name is in this form so abbreviated that some 
have supposed that the real patron was St. Teilo, but 
there can be no doubt in the mind of anyone who has 
carefully followed the sequence of events. 

The Cartulary of the Black Canons of Carmarthen 
has, fortunately, been preserved in a seventeenth- 
century transcript, now in the Peniarth collection. It 
was printed from this MS. (Hengim^t MS. 440^) by 

^ That a cell was actually established is shown by the langnage 
of the ChroDicIe : '* abi jam fratres ad Deo serTiendam adunati 
fuerant"(p. 61). 

2 Ibid.j pp. 61, 62. See also J. H. Round's volame of ^iicten^ 
Charters, edited for the Pipe Roll Society (vol. x, 1888), pp. 27, 28. 

3 Described in Arch, Camb,, 4th Ser., vol. ii, p. 105. Not being a 
MS. in the Welsh language, it is not calendared in Dr. G-wenogvryn 
Evans's Report, 

Digitized by VjOOQ IC 


Sir Thomas Phillips, of Middle Hill, in 1865 ; but the 
edition was a very limited one/ and copies are now not 
easily obtained. This is, perhaps, the reason why so 
little use has been made of the material here available 
for the history of the priory, borough, and district. 
The documents belong in the main to the fourteenth 
century, but there are some of earlier date, and a few 
which go back to the first years of the priory's exist- 
ence. Henry Us charter of 1176-1184, confirming to 
the priory the gifts of earlier benefactors, is known 
from the collection of Royal Charters relating to the 
town and county, published by Mr. Alcwyn Evans in 
1878.^ But in the Cartulary several of the original 
grants are preserved, and with their aid it is possible 
to tell the story of the first endowment of the house. 
The nucleus was furnished by the first Henry's grant 
of the **Vetus Civitas" of Carmarthen; with the 
churches of St. Peter and of St. Teulyddog, the Castle 
chapel, and all other chapels attached to these two ; 
and in addition a carucate of land at Pentewi, i.e., 
Pentjrwyn, in the parish of Llanstephan, near the 
outlet of the Taf. In the Cariulam^ is a letter addressed 
\rj Bishop Bernard to Maurice fitz-Gerald, who was 
Lord of Llanstephan,^ warning him that '* Pentewi " 
has been given to the canons, and that he must on no 
account interfere with it. The bishop was himself a 
donor to the priory. He gave, no doubt, out of the 
lands of the See, two carucates in Cymau,*^ a couple of 
miles to the west of the town, and thus was established 
the connection commemorated to this day in the name 
Maes y Prior. One of the knights of the Carmarthen 
district, named Alfred Drue, whose lands lay between 

^ The list of names of subscribers (iDolnding Jesas College, 
Oxford, and the Royal Institution. Swansea) suggests that only 
twenty-three copies appeared. The printer was John Lowe, of 

* It is contained in an " inspeximus'* of 31 Henry III (pp. 4-6). 
8 No. 36. 

* Qir. Camb., Dt Behus a Se Gestis, lib. i, cap. 9 (Works, i, 59). 
« No. 26. 

6th 8KB., VOL. VII. 19 

Digitized by VjOOQ IC 


the Towy and the Cy wyn, gave the church of Llangain, 
with one carucate of land. This became the property 
known as Maenor Grain.^ Lastly, a person whose 
identity is somewhat disguised in the charters, but who 
may safely be identified with Bledri ap Cydifor, ances- 
tor of the Lords of Cilsant, gives, between 1129 and 
1134, four carucates in Eglwys Newydd, or Newchurch.^ 
Bledri appears in the charters of the Priory as 
''Bledericus Latimerus," i.e., interpreter, the Welsh 
*' lladmerydd." It is thus suggested that he acted as 
an intermediary between the authorities at Carmarthen 
and his fellow countrymen, and the idea derives con- 
firmation from other sources. Bledri was one of the 
Welshmen who adhered to the King in the commotion 
of 1116 ;^ he was entrusted with the defence of a castle 
belonging to one " Robert Lawgam,"* or " Courtemain," 
as he is called by the compiler of Brut y Saeson, who 
may be the same as the '* Robertus cum tortis manibus," 
mentioned in the Liber Landavensis,^ but is otherwise 
unknown. Th6 site of the castle is also not easy to 
identify, but it may have stood at the mouth of the 
Cywyn.® Bledri further appears as a partially Norman- 
ised Welshman in another historical source for this 
period, to which I wish to draw special attention, as it 
is very rarely used for Welsh history — I mean the Pipe 

1 No. 34. 

^ See the oonfirmatioQ by Henry I (No. 33). The grant was 
renewed by BleHri's Ron Gruffydd iu the time of Bishop Dayid 
fitz-Gemld (No. 32). i.e., between 1148 and 1176. 

^ Brut y Tywy$ogion, ed. Ab Ithel, p. 126. 

^ The 'Maw^n*' of the Red Hook of Hergeat (ed. lilies and Evans, 
vol. ii, p. 297; is a mistake for " lawgam " See the text of the 
(older) Mostyn MS, 116, as given by Dr. Gwenogvryn E/ans, in his 
Report (vol. i, p. 59). 

^ Ed. Evans, p. 93. 

• Abercywyn appears in Lib, Land,, p. 124, as *'Aper couin." 
If the second part of the name were written **oouaT," it might easily 
yield the ** cofwy*' of Mostyn MS. 116 and the Red Booh Brut y 
Saeson has "comnyn" (Myvyrtan Arduiioloyy, second ed., p. 673, 
col. 2). 

Digitized by VjOOQIC 


Roll of the thirty-first year of Henry I.^ In this record, 
which shows what payments were due from and made 
by sheriffs and similar officers to the Royal Exchequer 
in Michaelmas, 1130, a glimpse — ^alas, too brief! — is 
afforded to us of the state of affairs in South- West 
Wales towards the close of Henry's reign. I will only 
cite a few entries of special interest from the Car- 
marthen section of the Roll : " Bledri the Welshman 
owes twenty shillings in atonement for the murder of a 
Fleming by his men."* A little lower down, he appears 
among a number of knights who owe various sums for 
the last aid due to the King.^ Alfred Drue, the bene- 
factor of the Priory, has his place in the list, though it 
is said the debt was incurred in the time of his father, 
Anschetil. He has only just succeeded to his father's 
fief, and still owes sixty shillings in respect thereof.* 
" Bleddyn of Mabudrud (the region around Pencader) 
and his brothers owe seven marks of silver for carrying 
off the daughter of BUdri by force."* And lastly, 
"the men of Cantref Mawr owe 40 shillings for the 
slaughter of a man of the Bishop of Salisbury."® The 
powerful Roger of Salisbury is shown by a charter of 
Kidwelly Priory to have held the commote of that 

' Edited for the Reoord Commissioa bj Josepb Hanfcer in 1833. 

^ ** Blehericas Waleasis debet xx solidos proconoordia Flandreasis 
qaem homines sui interfeoerant" (p. 89). 

^ " Bleliericus Walensis debet i roarcam argenti de eodem aaxilio" 

^ '* Alnredas filios Anschetil drine i maroam argenti de eodem 
auxih'o de tempore patris sai" (ibid.) : " Alaredas filias Ansdietil 
Drine debet Ix solidos pro terra patris sui" (p. 90). 

* ** Blehien de Mabaderi et fratres sui debent vii marcas argenti 
pro filia Bleheri quam vi rapuerunt" (p. 90). For the situation of 
Mabndrnd, which was one of the seven commotes of Cantref Mawr, 
see the volume of Appendices to the Report of the Welsh Land Com- 
mission (fiondon, 1896), p. 442, and cf. Lewis Dwnn, Heraldic 
VisitatioTiSj i, 148 (Glan Blodenn), where the text has '' Mab 

• ** Homines de Oatmanr [probably for Ca? maur] debent xl 
solidos pro homine episcopi Saresburiensis quern occiderunt" 
(p. 90). 


Digitized by VjOOQ IC 


name in the middle of the reign of Henry ;^ and as Kid- 
welly and Cantref Mawr were contiguous between 
Abergwili and Merlin s Hill, it is not surprising that 
conflicts should have arisen such as gave occasion for 
the imposition of this fine. 

Thus, when the " Lion of Justice," who had kept 
South Wales so well under his control, died on Decem- 
ber 1st, 1135, Carmarthen had attained the position it 
was to hold for so many centuries as the chief military 
and administrative centre of the Crown in these regions. 
Its castle was built, its borough settled, its priory 
endowed, and its oflScers installed for the transaction of 
the King's business 

1 Monast. Anglicy vol iv, pp. 64-5. The grant here recorded was 
made before the death of Bishop Wilfrid, of St. David's, in 1115, 
and the promotion of Prior Turstin to the dignity of Abbot in 1122. 

Digitized by VjOOQ IC 



By Professor JOHN RHYS» M.A., D.Litt. 

Lewis, in his Topographical Dictionary (vol. ii, dated 
1844), mentions a chapel of ease dedicated to St. Mary, 
and thence called Capel Mair, in the Carmarthenshire 
parish of Llangeler ; but he says that even then the 
chapel had ** been entirely demolished." He adds the 
words : — ** A monumental stone, bearing an inscription 
in rude characters, and said to be in the Welsh lan- 
guage, is still remaining. *' From this it does not 
appear that Lewis had seen the stone. 

In the year 1855 Westwood {LapidaHum Wallice, 
p. 93) was informed by Longueville Jones that the 
stone " was broken to pieces by the farmer who occupied 
the land some years previously, because people tres- 
passed on his land to see it." 

The Rev. E. L. Barnwell wrote in the Journal of the 
Cambrian Archaeological Association for 1872, p. 67, 
that he was informed by Mr. R. Randall Roberts, that 
the latter " was unable to find any trace of letters or 
oghams on the stone, which is near Capel Mair." He 
adds that ** some of the residents say that it had some 
characters on it, and that a wax (?) impression was sent 
to a gentleman in London, whose name could not be 
ascertained. It is," he goes on to say, ** in a farmyard 
near Capel Mair, where it was originally found. The 
stone is about b^ ft. long, and 2 ft. broad." He then 
mentions a ** copy kindly sent" to him by Mr. Spurrell of 
Carmarthen ; and, in passing, he states that the stone 
was originally found in 1828. 

In September, 1875,1 wrote to the Journal of the 
Cambrian Archaeological Association, p. 371, that I called 
on Mr. George Spurrell at Carmarthen, and that he gave 
me the reading of the Latin version as deca bar- 

Digitized by VjOOQ IC 


BALOM I FiLivs BROCAQN-, while the Ogam was Deccai- 
hanvdihdis. I recognise the latter as my own translitera- 
tion of the scoring which he showed me. I began my 
note by stating that Mr. Spurrell informed me, " that 
some time a,go he handed to one of our leading archaeo- 
logists a detailed account of the inscribed stone at 
Capel Mair.*' This was probably the copy which Mr. 
Barnwell mentioned in 1872, as received by him from 
Mr. Spurrell, a reference which I had forgotten, if I had 
ever noticed it. 

In June, 1876, an important note was sent to the 
Journal^ and pubh'shed in the July number, p. 236, by 
the Rev. Aaron Roberts, then Vicar of Newchurch, to 
the following effect: — '* About the year 1828 there was 
an inscribed stone near St Mary s Chapel, Llan Geler. 
The inscription was obliterated some years ago by 
a meddlesome bucolic. Fortunately, however, the 
Rev. David Morgan, Knightsford, Newchurch, at that 
time Viciir of Llan Geler, took a sketch of the stone 
and inscription thereon. One, in Roman capitals, was 
DECA BARBALOM FiLius BROCAGN. On the ridge above, 
or rather sideways, was an inscription in Ogham. As 
this latter appears in the copy I have I cannot make 
anything out of it. The sketch by Mr. Morgan was 
found among the papers of the late Captain David 
Davies, Trawsmawr, by his executor, Mr. George 
Spurrell, to whom I am indebted for my ability to 
place it on record.'' Captain Davies was probably in- 
terested in sculptured and inscribed stones : at any 
rate several were brought together in his grounds — 
see Westwood's Lap. WallicSy pp. 88, 89. 

It was, I think, after the publication of this letter of 
Mr. A. Roberts, that I made his acquaintance at Aber- 
gwili under circumstances which I have forgotten : I 
believe it was also from him that I got another 
reading, beginning with DECAPARBEILOM. About that 
time I visited Capel Mair, and failed to find anybody 
who could tell me anything about the inscription. It 
is true that a stone was shown me which was alleged 

Digitized by VjOOQ IC 


to have on^been inscribed and then to have had the 
writing on^t effaced, but I could discover nothing 
calculated to corroborate that story. 

From these statements it seems that the inscribed 
stone was known at Capel Mair about the year 1828, 
whether that was the date of its discovery or not. 
It was broken to pieces some years before 1855, and 
the reason for the tenant's action is said to have been 
the fact, that people who came to see it trespassed on 
his land. It cannot have been the stone in the farm- 
yard shown to me and others. 

There is evidence which will be mentioned presently 
that the obliteration story does not apply to the stone 
with the inscription deca, etc. Of this stone a copy 
came into the possession of Mr. George Spurrell, who 
gave it, or a copy of it — probably the former — to Mr. 
Barnwell, who passed it on to Mr. West wood for his 
Lapida/iium WallicB, where it has been figured in 
Plate 47. But Mr. Spurrell kept the original copy, or a 
copy of it — probably the latter — as he was able to give 
Mr. A. Roberts a copy — a bad one, as the latter gentle- 
man suggests. I suspect Roberts's reading of it was 
partly to blame. 

At all events, the copy which Spurrell allowed me to 
tnmscribe cannot have been a bad one, though I forget 
what it looked like. Mr. Spurrell may have given 
away more copies, but they must have been copies 
direct or indirect of the one made by Morgan, the 
Vicar of Llangeler. In fact, there seems to be no trace 
of the existence of any copy independent of the one by 
Morgan. For the version decaparbeilom is easily ac- 
counted for as a misreading of one of Spurrells copies, 
by neglecting the lower portion of the bipartite B 
so as to bungle it into P ; and similarly El was guessed 
out of a carelessly formed A with its top possibly 
square. So we are confined to the one copy, namely, 
that made by Morgan, or at any rate handed down 
by him. 
Now, Morgan 8 copy having, as already suggested. 

Digitized by VjOOQIC 

296 THE OA^^lL MAlB StON£. 

passed through the hands of Spurrell, Barnwell, and 
Westwood, has been deposited among the Carmarthen- 
shire ** Rubbings for West wood's LapidariumWaMicB'' in 
the Bodleian Library. The shelf-mark of the volume is 
" MS. Top. Caermarthenshire, a, l,"and our document 
is on an open sheet of notepaper, there paged (in pencil) 
258. Below the sketch of the stone come jottings in 
pencil by Westwood, one of which mentions the fact 
that Lewis alludes to the stone, and how it was broken 
to pieces ; also stating that this copy reached West- 
wood from Barnwell in July, 1871. In the right-hand 
bottom corner, in red ink, one reads as follows : ** From 
George Spurrell, Carmarthen, May 4th, 1871." This 
was probably written by Spurrell when sending the 
paper away to Barnwell. In the left-hand bottom 
corner one reads the following, Fig. 2 : 

" Found about 

1828, when 

Rev. David Morgan 

(Of Kuightsford) was Vicar." 

This is in black ink, but apparently in the same hand 
as the entry in red. I take it that it was made by 
Spurrell when the copy came into his possession. We 
have an older hand in the description of the copy itself: 
" Representation of a stone found near Saint Mary's 
Chapel in the Parish of Llangeler, Carmarthenshire." 
This may be in Morgan's own hand ; at any rate, it 
probably dates before 1855 and the final smashing of 
the stone. To render all this clear to the reader, I 
have had the sheet of notepaper photographed, in- 
cluding West wood's attempt (in pencil) to read the 
Ogam scores in the wrong direction. The sheet has 
been cut in two for reproduction, as Figs. 1 and 2. 

The capitals have the appearance of forming a very 
s^couvixte facsimile of the original. Fig. 1. The extremely 
bipartite look of the B must have been suggested by the 
original, and the form of the R is also well known. 
The straggling shape of the M is familiar elsewhere, and 
so is the sloping top of the F ; but the bottom portion 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 



H ^ 

S "s 

n u 

s s 

GO >> 

^ II 



B |.§ 

02 la 

•H O 



Digitized by VjOOQIC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

triB OAtBL MAttt STONE. 29^ 

of the s has incorporated with it some accidental scratch 
which was no portion of the letter as cut on the stone. 
Lastly, a horizontal i at the end of a line was to be 
expected. I treat the reading i\a making the following 
epitaph : piLh^^ROCAGm ^^^'^ Decabar Balom, name 
and epithet, in the nominative case, without the usual 
ending us added on to make them into Latin as 
Decabarus Balomus, 

It will be observed that when the copy was made, 
the stone appears to have been broken at its lower end. 
This did not touch the Roman lettering, but it did the 
Ogam, near its commencement. For, as usual, the Ogam 
read in the direction contrary to the Latin, and a very 
jagged breakage is suggested as occurring across the 
face of the stone, in such a manner as to sever the first 
vowel from the consonant following it. As it stands, 
Morgan gives the following scores : — 

II Mil \ III MM . I MM . . . I II MM, 

^TT ' '"" I ' Mill III ' II I ""• III! 

Deht ca i ban Vatobd i s 

Now, the third symbol is not such : it has been copied 
as if sloping backwards to meet the breakage, and 
form a sort of delta with it. Had it been a letter 
it could only be h, but it should really be counted 
with the 1_LL following, and with it form MM, that is 
another c. We should thus have JJ_LJ_L!_U., cc. It is 
possible that the inclination of the two consonants was 
intended to be diflferent, vn\_J_LM , in order to make 
the reading easier. In any case we should thus have 
decc corresponding to the dec of the Latin, in fact, 
Deccaiban corresponding to decabar. Here one cannot 
hesitate long between n and r. In Ogam they are 
respectively ppTTf and ////^ , so one can hardly doubt 
that the r has it, or avoid concluding that the con- 
tinuations of the r scores on the left of the edge had 
been worn away or so damaged that they escaped the 
eye of the copyist. This suggests the meting-out of 

Digitized by VjOOQ IC 

298 THfe CAPEL kAltl STONlS. 

similar treatment to the Ogam j, 6, and the regarding 
it as originally r m ; but one has no excuse here for 
doing so, as the spelling with b would have to be 
treated as the more correct, as will be pointed out 
later. For the present, suflSce it to say that both b 
and m would here have their mutational value of v. 
Then, as regards the initials of balom and of the Ogmic 
valoby one need only mention the fact that in Late 
Latin 6 had the values cf b and v from the fourth 
century down. It follows here that whether you wrote 
BALOM in Roman letters or valob in Ogam, the pronun- 
ciation was approximately valov, possibly for an earlier 
walob; but this means allowing the vowels a and o 

Erovisionally to stand. The latter is suggested by the 
atin spelUng balom, and the other is adopted from 
the copy of the Ogam version, though it fails conspi- 
cuously to fill the space between the Ogam scores on 
both sides of it. The guesses made thus far may be 
represented as follows : — 

I I ■ I I ■ I I I I I I I I I I ///// ■ ' , ■ I I ■ I I M 

.1.1 ...... I . ///// I I I < I I M I "'1.1111 

De c ca i bar Vatobd i s 

One or two points may be mentioned in relation to 
the making of the copy of the Ogam legend; for 
instance, there are traces of its having l^en taken 
down in pencil before it was inked, and there is evi- 
dence of at least one correction : the last score ot 

III I II — r is preceded by a score made in pencil ; the 

pencilling appears to have been rubbed out, but the 
groove made by the pencil remains, and can be detected 
in the original photograph. Before proceeding any 
further, I have the pleasure of introducing a fact or two 
of another order. One day, in the month of March, 1901, 
a Mr. Jones, a Welsh undergraduate of this University, 
and a native of the neighbourhood of Llangeler, called 
and left me a copy of two pieces of the lost stone, with 
lettering as follows : — 

Digitized by VjOOQ IC 




Fig. 3. — Sketch of the two Existing Fragments of the Capel Mair Inscribed 
Stone made by Mr. Jones, of Llangeler. 

It did not appear to me then how the Ogam on Jones's 
larger piece would fit the Morgan copy ; so it was put 
by till the other day, when I received from the Editor 
of this Journal a cutting from the Western Mail for 
January 23rd, 1900: it consisted of a letter, signed 
'* Thomas Williams, Oakland, Drefach, Llandyssil." 
Mr. Williams provided his letter with a sketch of the 
bigger fragment as below : — 

Fig. 4. — Sketch of the bigger of the same two Fragments given by 
Mr. Thomas Williams in the Wettem Mail for January 23rd, 1900. 

Digitized by VjOOQ IC 

306 THb OAPEL MAlii, STOKB. 

The diflFerence between the two sketches suggested 
to me that possibly neither was quite accurate, and I 
guessed that the original scoring intended was — 

1 1 

1 1 

-^ = o6c/, or III , I' =u6c?, 

the d being the beginning of Morgan's syllable dis. 
That is how I would have fitted the fragment into the 
Morgan copy of the stone ; but how mistaken my guess 
has proved will be seen immediately. 

After speculating thus far, I wrote to the Rev. W. 
James, of Llandyssul, and to Mr. Thomas Williams, 
making various enquiries about the fragments. The 
result was that Mr. James went to Capel Mair with 
James Jones, the mason, who found the fragments 
in 1900, when he was engaged as head-mason on 
the outbuildings of Tan y Capel, a farmhouse within 
earshot of the supposed site of old Capel Mair. His 
men were pulling down the old cowhouse, built 
about 1828, in order to lay the foundations of an- 
other. One of them called his attention to a piece of 
stone with letters. He (James Jones) told the man to 
throw it aside, as well as any other piece that might 
turn up. He had heard of the Capel Mair stone, and 
thought that the whole stone might be discovered, but 
only the two pieces came to light. This is a portion of 
the mason s account, as reported to me last February 
by Mr. James. The discovered pieces were, for a time, 
kept at the farmhouse, but latterly they appear to have 
been deposited in the grounds of the New Church or 
new Capel Mair (opened in September, 1899), namely, 
** just near the foundation stone." Mr. James and the 
mason discovered that both pieces were there till quite 
lately ; but they failed entirely to trace the smaller bit. 
The stones belong to Colonel Lewes, of Llysnewydd, 
the owner of Tan y Capel, who has been so extremely 
obliging as to send me the bigger fragment for me to 
study it at leisure. 

My friend Professor Sol las describes the fragment 

Digitized by VjOOQ IC 



as grey fine-grained siliceous grit, and it measures 
1 ft. 4^ ins. long by about 11 ins. wide, and 2 ins. 
or 2^ ins. thick. The surface is rough and very un- 

It will have been noticed that the top of the stone, as 
represented in the Morgan copy, is impossible as a piece 
of sketching : some of the lines appear to me to be drawn 
to supersede the others, that is, the copyist tried to cor- 
rect himself. The top of the stone is more as represented 
in Mr. Jones's sketch of the bigger fragment, except 
that he makes the stone rather too much broken to- 
wards the right-hand corner, and that the left-hand 
X5orner should appear more rounded, which, however, is 
of no consequence here, as it had no writing. All these 
points will be better understood by glancing at the 
following sketch made from my rubbing with the aid 

Pig. 5. — Extant Fragment of Capel Mair Stone : View of Front. 
Scale, i linear. 

Digitized by VjOOQ IC 


of photographs. The greatest surprise to rae, how- 
tver, was to find Morgan's ' ' { { ] j | ... all on the top 

edge of the stone ; so that his . ■ ■ 1 i ■ i should end 

at the corner. Thus Morgan turns out to have put all 
the writing on a single straight line, which may have 
been done from motives of convenience; but when one 

moves his ' ' [ j 1 1 1 1 1 1 1 to the top edge, one perceives 

how greatly his copy of the Ogam on the side edge is 
out of scale. Where, then, comes the bigger fragment 

with the Ogams represented by Mr. Jones, as » ' i ' ' ? 

The vowel, rightly copied by him, may have been misread 
by Morgan as yj-j- = v, whicn was very easy to do, because 
the edge there slopes away very gently : it requires 
careful observation to see that the imaginary line of 
the edge is crossed by the scores. Thus it. is with 
Morgan's j-pp that the fragment begins : Morgan has 
marked one notch in the interval which follows between 
his j-j-| and his Tr = ^; but, as already indicated, the 
single notch by no means fills the gap, and there the 
fragment has what Jones has copied as a jm sloping 
parallel to the vowel notches (Fig. 3) ; but those two 
scores slope much more, and cannot be read, in my 
opinion, as anything other than the left half of a // = 5^. 
From the imperfect j-f on to the corner, the- edge is all 
gone, having evidently been hammered oflp; and it leaves 
a crack which defines a flake that has not come off. So 
we have to supply the two consonants jT ^^^ T ^'^^^ 
Morgan 8 copy, and an intervening o from his balom : 
thus we get — 

If, as I hope to show presently, we have here the 
Goidelic etymological equivalent of what is given in 
the Latin version as BALOM, we have to complete the 

Digitized by VjOOQ IC 


legend as hugloh, with the initial 6 of balom ; or else 
with another spelling , . . [ } [. V . . | ] . , vuglob, with the 

jj7j which the lost bit copied by Mr. Jones serves to 
supply : at any rate, there seems to be no use for that 
bit elsewhere. This means that Morgan in his copy has 
accidentally left out the Ogam for v, namely, yj-i > unless 
one should rather regard the omission as a part of the 
error which has yielded us his yjj for [ j ] • That is to 
say, for 1 1 1 1 1 1 he has given us only the first of the two 
sets of three digits. The insertion of it goes now some 
way to lengthen his line of scores on the side edge of 
the stone, and to bring his copy so much the nearer 
to scale. 

The vocable in question, whether vuglob or some 
similar spelling, I regard as having occupied the edge up 
to the right-hand corner of the stone ; but the original 
corner and the adjacent part of the edge of the top of 
the stone have been hammered off. One seems, how- 
ever, to detect on the top near the present corner of the 
stone just the left ends of an Ogam J-L= d; the broken 
edge towards the left would have supplied room for 
' '^^ = i, after which we come to traces of 


I I I I -7 ^* 
ese consist of the upper ends of the scores reaching 

towards the edge, for lower down the back of the stone 
the hollows representing them are very faint and ill- 
defined, because the stone shows signs of having scaled 
there : in fact, there are still bits there which are not far 
from getting loose. This is not all, for the s scores are 
followed by a final vowel j {-n-lr*, the notches of which 
are not hard to trace. It is strange that Morgan did 
not copy them as part of the writing. The piece of the 
edge with these vowel-notches thins out somewhat in a 
weage-like fashion, so the notches are to be seen from 
the front as gaps in the rim, but the bottom of each 
hollow has been smoothed and rounded. The com- 
parative thinness of the edge made them look un- 
like the other vowels on the stone ; certainly unlike 

Digitized by VjOOQ IC 


the jjj which Morgan copied as j-pp, where there was 
no proper edge at all. This is probably the explana- 
tion why Morgan did not regard what followed his 


1 1 1 1 1 

1,1, ■ . ■ . as forming part of the lettering. Lastly, 

it is of some importance for the reading to mention 
that these vowel-notches showing from the front of the 
stone diflPer in that respect decidedly from the upper 
ends of the scores for 5, for these latter are situated 
distinctly more on the back. The whole of the Ogam 
on the top of the stone may be approximately repre- 
sented thus : -^ V:V:^ , m !f n = ^^^^• 

Helped by the conjectural emendations which I have 
indicated, and the corrections warranted by the frag- 
ments, the complete version in Ogam may be represen- 
ted as having originally read as follows . — 

2 €^ CL c A / /^''r 

Fig. 6. — Ogam Inscription on the Capel Mair Stone restored. 

Now that I have done all I can to establish the 
Ogam text of the inscription, it may be pointed out 
that the fragments found in the cowhouse walls in 
1900 establish the substantial accuracy of Morgan's 
copy as to the scores which he jotted down. He failed 
mostly in omitting scores which a more experienced 
reader could have readily deciphered; but in some 
instances portions of the lettering seem to have been 
merely overlooked by accident or carelessness, such for 
example as the initial letter of Vugloh. The two frag- 

Digitized by VjOOQ IC 


ments, however, besides enabling us to correct Morgan's 
copy, prove that the story as to a stone which had the 
writing on it erased, did not apply to this monument. 
What happened to the latter was, that, some time 
after it was discovered and copied, it was broken up to 
be used in the building of the walls of the Tan y Capel 
cowhouse ; and in spite of the mason's vigilance in 
1900, the walls of the reconstructed outhouse probably 
contain the whole of the stone except the two frag- 
ments here in question. 

Settinor out from the Latin version of the inscription 
DECABAR BALOM FiLivs BROCAGN-, one may mention first 
the^t Brocagn-i h the genitive of the name which meets 
us as Broccdn in the hagiology of Ireland, and Brychan 
in that of Wales : neither of those names requires any 
further notice at present. We then come to balom, 
which, as already suggested, was pronounced Valov, 
This we cannot be wrong in identifying without hesi- 
tation with the modern Irish falamh or folamh (with 
mh sounded t;), meaning ** empty, void, vacant, poor, 
without means." The Scotch Gaelic is also falamh (pro- 
nounced faP-uv) and means likewise " empty, void, in 
want, unoccupied." In Medieval Irish, the tovmfalumh^ 
occurs, meaning ** empty"; see Stokes's "Book of Lismore 
Saints" (in the Anecdota Oxoniensia for 1890), 4707. 
This use of a word meaning poor and devoid of woidly 
goods was meant to be complimentary, and reminds one 
of Caiman Bochty that is, *' Colman the Poor," on a tomb- 
stone at Clonmacnoise ; see Miss Stokes's edition of 
Petrie's Christian Inscriptions in the Irish Language, 
Vol. i, p. 16, Plate ii. Fig. 4. Thus the Latin epitaph 
may be rendered " Decabar the Poor, son of Broccd,n." 

The Ogmic version has been approximately estab- 
lished as Deccaibar Vugloh disi, where the first diffi- 

1 The Manx form is follym, *' empty," with an unmatated m .: 
some other instances ot the kind occur in that langoage. Tlie 
most probable explanation is ih&tjollym owes if s tn to the inflaence 
of the adjective cDrresponding to Irnsh fo-lomm, ** bare." This 
would not be surprising, considering that some scholars seem to 
regard falumh as a form of folomm. 

Digitized by VjOOQ IC 


culty is as to the etymological equivalence of balom 
and Viiglob. Assuming that equivalence, we have to 
suppose the former to have passed through a precedinfjf 
stage, Baglom, with a soft spirant g liable to be eli- 
minated, which was here done in the Latin spelling, 
just as sinum occurs for signum in the Whithorn 
inscription ; see The Academy fi>r 1891, September 5th, 
p. 201. 

In this word vagloh we seem to have u form in- 
volving the Celtic prefix vo, which makes in later 
Goidelicyb,/u,ya, and in Welsh ivo, gwo, Modern go or 
gwa : the rest of the word seems referable to the same 
root as the Greek y\d<l><o or yXvffxD^ " I hollow out." 
With the Irish word has been identified in The Englyn, 
p. 73» the Nennian word guoloppum, "an empty space," 
and ccctguoloph,'' a space empty of war, that is, an in- 
terval of peace." The digraph pp could stand probably 
for either j'f or v, just as tt did for either th or d, and 
ph bad also either the sound of jf' or oft; in Medieval 
Welsh. So here, doubtless, the pronunciation intended 
was guoloV'Um" and cad-ghuolov. Moreover, the passage 
suggests a neuter substantive rather than an adjective ; 
but an adjective could readily be formed by means of the 
termination j^o-s, |^a, io-n, namely, uolob-j^o-, uolob-j^a; 
Needles to say this would be in an early stage of the 
language, for later it would yield the form gueilyv (to be 
written gweilyf), according to the analogy ofheinif, heini, 
"active, agile," from ho-gnim, of the same formation 
as Irish so-gnim, and gweini, *' the act of serving," from 
uO'ffni : compare Irish fogniu, ** I serve." The form 
gweilyf is not attested, but we have the variant 
gweilyiy with d forf{ = v), a substitution not unknown 
in other Welsh words. Dr. Davies, citing gweilydd 
from the Welsh Laws, explains it as " Vacuus, inanis, 
voluntarius, avrofiaro^.'' In Aneurin Owen's edition it 

^ We seem to hayo this in anolof, anolo, *' iDeffectire, yoid, nse- 
less;'' made up of golof, golo^ with an intensive prefix an. See Silvan 
Evaos's Geiriadur, also Dn Dayies's Dictionary : both cite instance« 
from the Welsh bekws, 

Digitized by VjOOQ IC 


occurs, for instance, in Volume IT, Bk. xiv, Chap, xxiii, 
6-8 (pp. 664, 665), where the editor has translated it 
"absolver." Moreover, words ending in yd frequently 
have an optional ending i, as in the case of Dewyi, now 
Detail *^ St. David," and trefyA or trefi, ** towns." So here, 
Dr. Davies, under vacuus, gives gweili, gweilydd, anolo^ 
and other adjectives. The first is, in fact, a living word 
in parts of North Wales, especially Lleyn, where one says 
trol Weill, " an empty or unladen cart," ceffyl gweili^ " a 
spare horse," and the blank pages at the end of a book 
are sometimes called dalennau gweili. For these details 
I am indebted to the unpublished dialect studies of Mr. 
Glyn Davies, of the University Library, Aberystwyth. 

The name decabar maybe compared in part with 
Calabar or Catahor — it is hard to say whether the 
ending has a or o — in an Ogam inscription in Co. Water- 
ford : Brash has it at p. 266. It is at the first glance 
tempting to identify this with the name written Cath- 
barr in the Book of Leinstev (fo. 324^ 338**), meaning 
*• battle-head," and as a common noun, ** a helmet;" the 
second element being 6aiT, '' head or top," as in Barri- 
vend'i on the Llandawke stone (Journal of the Camb. 
Arch. Assoc, 1907, p. 77) ; but the second r stands in 
the way, and we seem to have here the same element ns 
in -FaZfeAar, mentioned in O'Curry's Manners and Cus- 
toms of the Ancient Irish, iii, 158. In my Paper (read to 
the British Academy) entitled '* Studies in Early Irish 
History," p. 29, I suggested that we have an early 
form of the plural of this name in that of the tribe 
called in Ptolemy's Geography OveXKdfiopot, whom he 
places in the south-west of Ireland : see Muller's 
edition, pp. 76-8. He cites the pseudo-Ethicus and 
Orosius HS calling them Velahri, which is probably to 
be emended into Velahori. We have a later form of 
the singular — feminine, however, not masculine — in the 
Llandyssul inscription VELVOR filia BROhO. Here Vel- 
vor would, in a normalised orthography, be Velbor, 
derived from a far earlier Velabora. But the combina- 
tion Broh' for what would in GoidtJic have been 


Digitized by VjOOQ IC 


BrocC'y shows that the inscription was in Latin of the 
Brjthonic rather than of the Goidelic description. 
That is, in Brythonic mouths the tendency of later 
Goidelic to substitute a for o had been avoided, so that 
we have the latter vowel here, as in the early Goidelic 
Vellahor-iy while the a is found established in Falbhar 
just as in our DECABAR and Deccaibar. The meaning 
of the element 6or-, bar- is doubtful, but with regard 
to the other, decay one may perhaps venture to suggest 
that this may be of the same origin as the Latin dectiSy 
** ornament, grace, honour, glory," and of the tribe- 
name of the Decantae of the extreme North ; also of the 
" Decantorum arx," the Deganvjy whose ruins stand 
near Llandudno and the river Conwy. 

The names here in question raise a number of difficult 
questions which I cannot discuss at present; but I 
may call attention to one or two more points connected 
with the spelling. The Latin version nas a single c in 
Brocagniy where the pronunciation was that of a hard ife, 
and also a single c in Decahar, where the c was mutated 
into the Goidelic guttural spirant ch, which in some 
Ogam inscriptions is represented by the digraph cc : 
that is how we have the c doubled in the Ogmic spel- 
ling Deccaibar. The ai of this last is of more doubtful 
standing. It would be rather a violent emendation to 
treat the notch for the vowel a as an error in Morgan's 
copy, though it would simplify the form into Decxibar. 
But on the whole one has, I think, to accept the ai, and 
I should be inclined to treat it as a digraph for the 
vowel g, which the author of the epitaph thought was 
the vowel sound which he detected in the syllable fol- 
lowing the stress. Ai^ cb, and e have, roughly speakings 
one and the same value in Old Irish glosses; and what 
may perhaps be still more to the point is that we have 
ai in other Ogam inscriptions, to wit, in such names as 
Bivaidonas^ Dovaidonay and others mentioned in my 
paper on the Kilmannin Inscription in the Journal of 
the Irish Antiqiiaines for 1907, pp. 65-7. 

That paper gives another inatapce of the use of th^ 

Digitized by VjOOQ IC 


syllable ' HI 1 1 1 1 1 1 =■ dis, which has puzzled me so 

long in Morgan's copy of the Capel Main Ogam. Fo^ 
the Kilmannin epitaph begins with ddisi, which I 
provisionally analysed into ddis-i, and took to mean 
below, or here below, with ddis derived from Is, " lower." 
For I ventured to equate ddis- with the modern Irish 
thioSt " below, beneath," which Old Irish scribes wrote 
<&, while, as I thought, giving the t one of the sounds 
of th. But since then Dr. Stokes^ has written to roe 
that if anything is certain in Irish grammar it is that 
the aspiration in this and similar forms is merely "Middle 
Irish,* not older, say, than the eleventh century. So 
for the present I give up the attempt to explain the 
etymology of ddisi or disi. But I am inclined to think 
that the word meant, if not ** here below,'' at any rate 
" here." The Capel Mair Ogam might accordingly be 
taken to have conveyed some such meaning as *' Dec- 
caibar the Poor (lies) here." Perhaps, however, some- 
body will find in disi a verb of rest and repose. 

Lastly, the Ogam version belongs to the same class 
as the Kilmannin one, which I have ventured to regard 
as dating from the seventh century. It is only a guess, 
and both inscriptions may prove to be somewhat later. 



1. My attention has been kindly called by Professor Ed. 
Lloyd to my rendering lletfer by " weak-kneed " in my fonner 
paper, p. 91. It should have been half-vnld or aevii'Savcige, as 
proved by a passage in " Buchedd Gruffudd ab Cynan "; see the 
Myvyrian Archaiology of Wales, II, p. 586, where we have 
Llymminawc lledfei- given in Latin as Saltus ferinus, with the 
fidjective ferinus meaning "of or belonging to wild beasts," and 

1 In proposing an etymology for the Welsh word blew '* hair," I 
forgot that it had been dealt with in Stokes's Urkdtischer Sprach^ 
ichatZf p, 187, where it is referred by Bezzenbergci to the same 
origin as the Greek (f>koi6t, the " rind of trees, peel, bark, bass." 
For Sanskrit grtva read griva, " neck," and for RusFian yrtva read 
gritfa, " mane." 

Digitized by VjOOQ IC 


deriving from the simpler adjective ferus/' wild^' lintamed." 
This word ferns was borrowed into Welsh, as proved by the 
instances given by Dr. Davies and Dr. Pughe under ffh'. It is 
needless to say that there is another Welsh word ffir, namely, 
that which is used in North Wales for " ankle ; " but Pughe's 
fferu, meaning " to congeal, to liecome rigid with cold," is pro- 
nounced and properly written fferriL His spelling of it enables 
him to refer it to one and the same origin with the two 
words ^^, for he would seem to have made the meanings con- 
verge on that of *• dense** or " solid/* while Dr. Davies regarded 
the adjective as more or less synonymous with cadam. Exami- 
nation of the uses made of the word would probably result in 
proving it to have retained more or less closely the sense of the 
Latin ferus. Meanwhile, I have chanced on the compounds 
kadfer and llawffer (Skene, ii, 56, 143). 

2. Apropos of Cian of Nanhyfer, I am I'eminded of Nennius's 
" Cian qui vocatur Guenith Guaut" who was one of those who 
were distinguished "in poemate Britannico/' On the wliole, I 
am disposed to think that that the latter was an earlier man 
than our Neveru Cian. I may mention that there was also 
a Cian after whom Llangian in Lleyn is called. Rees. ia 
his Welsh Saints, p. 302, associates him with Peris of Llan- 
beris, and mentions his day as December llth. All this only 
makes it rather more diflScrdt to say who Cian of Nanhyfer was, 
or to settle the question whether he was a Goidel or a Brython, 
I may here mention that the other name Cu-Duilich, in Welsh 
Gynddylig or Gynddilig, weis borne by a saint of whom Rees. 
p. 281, writes as follows : — " Cynddilig, a son of Cennydd ab 
Gildaa His memory has been celebrated in the parish of Llan- 
rhystud, Cardiganshire, on the 1st of November/* 

3. The difficulty as to the Scots of Nanhyfer coming dros nor, 
" over sea," depends a good deal on the place of the battle ; for 
without coming from Ireland they might be voyaging from 
Neyem or Newport to.some place on the coast of North Wales.' 
The weight, however, of historical opinion inclines to South 
Wales, and the correct date, I am told, is 1081. 

4,. In connection with Pont y Gini, Professor Lloyd also 
states that there is a large farm called Gim in the parish of 
Llanengan in Lleyn. I should like to be assured whether it is 
Gim or T Gim , for the presence or the absence of the definite 
;irticle may prove to be a difference of some importance. 

Digitized by VjOOQ IC 





{Continued from pagt 34.) 

CHAPTER II (oew^tViuoi).— SECTION II. 

The survey of Bromfield and Yale, known from the 
name of the surveyor by whom it was executed as 
Tidderley's Survey, and now at the Public Record 
OflSce, has next to be dealt with. It is undated, but 
the internal evidence points to its beings taken at the 
end of the reiga of Henry VIII. First comes a quaint 
description of " the towne of the Lyons, oderwise 
named le Holte." *' The said towne standythe yn Wales 
witheyn the lordshipp of Bromefeld and Yale one di' 
[le., half] mile ffrome the ent'yng towards yt over a 
stouyn [stone] brydge whiche partithe England and 
Wales. The same beyng an auntient Borough towne, 
but slenderly builte with tymber worke and the 
buyldings stondyng yn distance^ the one flFrome the 
oder on the este side of whiche towne ther stondithe 
the castell equaly withe the said towne. And yn in 
the same towne ther are ffaire Hawles' whereas the 
kyng*8 tenn*" of Wales witheyn the said lordshipp 
haven justice ministred to theyme. And in one of 
theyme the Meire and Burgises of the said town do use 
to sytte and kipe theyr courtes whiche Meire and Bur- 
geses do clayme to have dyv's p'vileges and lib'tyes by 
the grannie of a charter made unto theyme by [blank]. 

y We should, perhaps, suppose some such word as '* eveu," or 
'' equal,'' to have been intended here. - 

^ The first of these fair halls was the Welsh ooart*hoase in the 
castle precinct, and the other the Holt Town Hall. 

Digitized by VjOOQ IC 


" And yn the s tid towne are yerely kept two feires, 
one at the feast of sent Barnabe [llth June] And the 
oder yn the feast of St. Luke [I8th October] yerely 
and eu'y Ffrydey a comen markett yn the said towne. 
The Toll therof beyng set for II5. l)y the yereto the 
Bailyesof the said towne. And yn the said towne ys a 
maire, two bailies, two leve-lookers and coronor."^ 

Next follows an equally picturesque and valuable 
account of '* the castell of the Holte : " " The said 
cabtell stondithe yn the este side of the said towne and 
northe from the utter warde of the same. A gate 
howse beyng builte withe tymber worke abowte a 
Ix paces from the castell, wherunto is joyned upon the 
este side stabulls belowe ccc fote in lenght and ou' 
[over] the same stables faire loftes for haye. And west 
from the same gate a tfaire barne conteynyng in lenght 
XXXV paces and in widenes a xiij paces westward 
adioyning to the same a faire courte howse* of tymber 
for the kyng 8 justices and officers do sytte yn at tymes 
of sessions and courts to be kepte ther for all the 
kyng's tenn^ witheyn the lordshipp of Bromeflfeld and 
Yale beyng witheoute the ffraunchise of the towne of 
the Holte. And at the weste ende of the same courte 
howse a pale and a quyke sett hedge stondyng towards 
a parke adioynyng to the said castell named the Litell 
Parke witheyn which bwilding hedge and pale there ys 
a courte betwene that and the mote of the said castell. 
At the whiche zouth ende of whiche courte there ys a 
garden encloseid w** a pale and hedge and at the est 
ende of the same courte a doffe [dove] house of tymber. 
And adioynyng to the mote upon the lefte hande 
ent^yng the castell a som' howse of quyke sett trees 
like an arbor made withe a fframe of tymber. And as 

I <« Leve-lookers." It is almost certain that these were the two 
oflBcers yearly elected, who were afterwards called "appraisers.** 
There were " leave-lookers*' in the borongh of Denbigh also, who in 
1827 were orderf»d to inspect the slanghter-honses* See Williams's 
Records nf DefnJbigh and its Lordship^ p. 161. . 

. 2 ,Xhe Welsh court-house, or lordship court. See note 2, page 311, 
aad elsewhere. • . ; • 

Digitized by VjOOQ IC 

m coir!rrY Denbigh. 313 

ye enter ynto the said oastell the waye lyythe zouth 
esteward. The said castell stondeth equally withe the 
towne very strongly bwilte upon a rocke. The waye 
ffrom the said gate to the enteryng ynto the first warde 
conteynythe ffiftye paces, the first entye beyng ou' 
[over] a bridge of tymber and under the same a drye 
mote. The said warde beyng a square towre^ strongly 
bwilte with two strong gates yn eyther side one beyng 
XXX** fote betweene the said gates, and at eu'y gate a 
purtecules of tymber, and ffrome the same warde ent^yng 
the castell a bridge of tymber xx*^ flPote of lenght and 
ten in bre**the. The mote beyng ffiftye ffete depe 
underneth the same bridge. The saide towre^ stonding 
yn the midds of the mote betwene the saide two 
bridges. And then ent'yng the said castell two oder 
stronge gates, beyng fortye flfote betwene the same 
gates withe a portcules of iron for the inner gate, a 
courte paved of fy ve square beyng c . . . [a blank after 
the first c] ffote over witheyn tne inner gate. The 
said mote compassyng ffbure squares of the same castell, 
and the ryver of Dye [ = Dee] runyng by the este side 
of the said castell upon the wall side on the fyveth 
[fifth] square therof, tne mote beyng one hundred flfote 
yn breathe and metyng withe tne saide river on 
eyther side. And witheyn the saide castell ther are 
all bowses of office mete for a prynce to kepe his house 
yn. And on the lefte side of the inner gate ther ys a 
flfaire steres [stairs] of stone work vii fote wide goyng 
up to the haule, a streight steiie, and on the right 
hande of the steire heade the hawle, and at the lower 
ende therof a buttrey, a pantrey, and a flTaire kychen 
withe a drawght well yn the same. And a large 
chymney in the lefte side of the saide hawle, And at 
the ou' [over = upper] ende of the haule on the right 
side of the same ent'yng the greate chamber, and so 
directly two oder chambers rownde abowte bwilte 

- ^ This refers to the square Exchequer tower, separated from th^ 
castle proper bj a moat, whioh also snrroanded it on every side. 
The boilding first described was the outer [or ^* utter'^ gate*boase. 

Digitized by VjOOQ IC 


beyng the highest storye castelljke, with chymneys yn 
eu'y [every] of theyme. And underneth the said 
hawle and chambers thre stories for lodgeyngs and 
howses of office, amongst which ther ys a horse myll a 
stable for xx horses and ou' [over] the leadds ther is a 
faire waike and a goodly p'especte [prospect]. The 
castell beyng bwilte fy ve square, and att eu'y square a 
rownde towre of flfy ve stories highe, and on eu'y story 
a chymney and owte of eu'y the same towers a steire 
up to the leads, and from the leadds two seu'all 
wyndyng steires downe to the tower p'te [part] of the 
courte. And aswell the castell and the fyve towres as 
the utter warde beyng builte with ffrestone playne 
aishelar and embateled. And a secrete narrow wey 
goyng owte of the same courte downe a steres of stone 
and vawted [vaulted] with stone ynto the saide ryver 
of Dye whereto the warde and dore ys of iron. Two 
squares of the leadds of the said castell nedith to be 
emendyd, the reste are well repayred. The parpwynte 
[parapet] of the utter wall nyne ffote thycke, the 
inn[er] wall [blank] ffote. The said castell beyng more 
strongly bwilded with stone and tymber then [than] 
stately lodgeing or conveyant [convenient], 

" And adioyninge the mote on the weste side of the 
said castell an arbor and adioynyng the same on the 
zoUthe side over a bride [? bridge] a litel p'ke [park] 
paled abowte lyyng upon the said river of Dye, beyng 
replenyshed with xvi dere of auntler and rascalP yn the 
kypyng of S' Barye Acoton [Acton] knyght, and 
[blank] Pylleston, the arbage [herbage] whereof ys 
worthe yerely xxxs. 

" The saide castell and towne lyythe %ve miles 
ffrome the Cytye of West Chester,* and yn no clene 
ay re, but yn a sou re countrey. The ryver also beyng 

^ Rasoal, *' a term of the ohase. Certain animals not worth 
liantiiig were so called. The hart, till he was six years old, was 
accoanted r(wcay/tf.**— Skeat" " A dear lean and out of season." — 

* Actually; Holt is aboat 8^ miles from Chester 

Digitized by VjOOQ IC 


a darke muddy e water the lere [i.e., look] of the zoyle 
beyng a redde erthe." 

On folio 3 is the following description of Mersley 
Park : " In the ffranchese of the holte and witheyn one 
mile of the said castell ther ys a faire p'ke beyng thre 
miles abowte the same being paled Rownde w"* M pale 
w^ pke is more yn lawnes and playnes then [than] 
cou^'te. The midd® of the said p 'ke beyng cou'te with 
okes and smale Tymber witheout any oder Cou'te. 
And in the zouthe ende of the same cou'te a pretve 
lodge for the kyper well bwillte [so in MS. for ** built ']. 
All the growl ids of the said p'ke beyng level! And very 
goMe and depe pasture ground Replenyshed with Ixx 
dere of Auntler eight score dere of Rascall and foui- 
score fiiiwnes. Th arbage [herbage] therof worthe to 
be sett beside flpyndyng of the game yerely [blank]. 
The fejrping wherof is graunted by the kyng* maiestye 
his graces lett s patent to George cotton knyght for 
t'me of his liffe."^ 

The followingr is also worthy of note : "The bayly- 
wickes of hewlingtoii, hem man'm [that is, Hem manor] 
and Rydley are p'cell of the demayns of the said 
Castell which iij p'cells are of the yerely value of [blank] 
herafler p'ticularly sett forthe yn this boke of Survey 
the same lyyng by the said Castle being a veiy good 
fertile grownde both for medowe pasture and Errable 
londe which are letten to dyu's p'sons for t'me [term] 
of yeres by the kynge's letters patent as yn the same 
Baylywykes Aperithe. ... 

" Ther be thre very ffayre powles* [pools] lyyng 

^Mersley Park was the great park in Allingtou (long ago dis- 
parked) attached to Holt Castle, of which an aooonnt is given in 
the " History of the Townships of the old Parish of Ghresford," 
pp. 145-7, Areh. Camh., 1906, pp. 195-7. It was coterminous with 
the OoDlmon Wood, Holt, bat paled off from it. 

' These were the fishpools in Frog Lane for the sopply of Holt 
Castle. They occupied the site of the meadow nearly opposite 
Esphill, which nieadow has still a very irregular and uneven surface. 
The field, called '* Fishpoolfield," in which many of the burgesses of 
Holt had distinct ** acres, ** probably adjoined the 6shpools. 

Digitized by VjOOQ IC 

316 TttE TOWN OP HOLt, 

witheya one quarter of a myle to the Castell which 
haven bene heretofore replenyshed with ffreshe water 
ffishe the same being lyke to be distroyed for lack of 
scowryng and seeing vnto wherby the kyngs highness 
taketh no yerely p'fBt. 

**Thre miles frome the Castell of hclte ther ys a 
grownde which ys a myle and di* [that is, a haU] 
Abowte callyd I'he Warren^ Wheryn ther hath bene 
game of cunneys [conies] and the Custodye therof 
corny ttyd to one Edward Breerton [of Burras Hall] 
withe the yerely ffee of 1x5. xd. tharbage [the herbage] 
wherof of late . . . ys graunted to the said Edward for 
t'me of . . . for the yerely rent of [blank] so that the 
said Warren ys conu'tyd to A yerely iFerme and no 
game of cunneys theryn kepte. The same grownde 
beyng one myle and di' Abowte all couerte and over 
growen with brakes and Thornes except Thre akers 
therof thorowgh which grownde there are llire highe 
comen wayes." . . . 

*• Witheyn one myle of the castell and townt of the 
holte ther ys a faire comon* beyng thre quartrs of one 
myle Abowte of good pasture grownde whiche the bur- 
gesses of the holte clayme to have to theyme and to 
theyr heirs yn flfee by a charter made to theyme by 
[blank], sometime lord of Bromfelde and Yale, the 
same comon beyng adjoynt to the p'ke of marslie." 

Also, ** ther ys a ffre Chapel witheyn the Castell of 
the holte, of the kyngs Majestyes gyfte, of the yerely 
vaylue of Ten poundes, beyng yeven to one S' Thomas 
Birde, clerk, wherunto ther doth belong the tythe of 
s'ten [certain] land, lying witheyn the Baylywyke 
of Burton called Pastelande, and Tire Boroughe, the 
same beyng worthe by the year iii/i."* 

^ This warren was Pare CwiiiDg, in the townghip of Bieston, 
perhaps nsed for the snpplj of rabhits (conies) to Holt Castle, ns 
well as for the diversion of the lord or his officer (see my History of 
the Country Towruhips of the old Parish of Wrexham^ pp. 155 and 
159. 2 The Common Wood of Holt 

• This iigure is somewhat vaguely written. In Norden's Survey 
of the manor of Borton the lands belonging to the casUe chapel ar^ 
described as worth £10 in the time of Henry VIII. 

Digitized by VjOOQ IC 


Whatever is not quite plain in Tidderley's description 
of the castle will become clear when, in a future chapter, 
plans and views will be given, and other particulars 
furnished relating to the same. 

Meanwhile, it may be well so far to anticipate the 
further and more minute description of the castle and 
its precincts, so promised, by saying that an inquiry^ 
was made at Holt on the 30th January, 158f, by 
Roger Puleston and George Olive, Esquires, by virtue 
of a writ to them, and to Sir Hugh Cholmondeley, 
Knight, directed. Eight witnesses were examined, but 
the only points that need be noticed here are these : 
John Ledsum, of the town of Lyons, aged 72, said that 
" the first chamb' going up the store [stair] on the right 
hand of the gate coming into the inner Court the Con- 
stable did use to leade his prisoners to a tower adioyn- 
ing to the same, w'^^ tower also the Constable used to 
keepe his prysoners in, and so ascending to the said 
stores the Chamb' over the Gate was comonly cauled 
the Constables Chamber, w*^** his deputy lodged in, and 
thother chamber on the left hand of the said lodging 
the Constable used to laie his wood and coale in, and 
wherein also bedds were set"; and so had been used for 
sixty years. Richard Roydon, of the town of Lyons, 
Gent., aged 68,^ testified that " the tower, w*"^ hath a 
chymney in yt, and adioyneth is the chamber w*'** the 
Constable had his wood and coale in was also app^'tain- 
ing to the Constables oflBce, w*^** he hath knowen for 
these xlvi yeares or therabouts, and further saith that 
when Mr. Edward Aimer, being deputie steward, did 
lye in the said Castle Thomas Powell, who was then 
Cunstable, did locke and keepe all the foresaid roomes 
from the use of the said M' Aimer." William Kitchen, 

^ I owe my knowledge of this, as well as of the other inqnirj, 
presently to be named, to my friend, Mr. Edward Owen. 

^ See page 319, where, on 80th January, 158^, Richard Roydon's 
age is given as 60. Both documents are correctly copied^ bat there 
ia an evident error in one of them, in the respect noted, 68 bein^ 
written for 60, or contrariwise. 

Digitized by VjOOQ IC 


of the town of Lyons, aged 56 years, said that the 
"chamber over the gate coming into the inner court 
hath bene alwaies reputed and cauled the Ounstables 
Chamb', and the next chamber, w** was called the 
second chamber, and the tower having a chymney in 
yt, one Bernard Bewley, who served as deputie cun- 
stable, did use and occupie." . Robert Powell, alias 
Smyth, of the town of Lyons, smith, aged 42, testified 
to the taking downe of two portculleses of yron, w***in 
the said Castle by the com'ndem*^ of Edward Hughes, 
esqui*"/' Launcelot Bates, of the Holt, aged 60, de- 
posed that ** he hath hard [heard] one iron doore being 
belowe in the house towards the Riu' of Dee (in the 
tyme of M' Hughes) ys taken awaie." Being examined 
as to certain outhouses, he said that xxx"® yeres agoe 
[they] were ruynows, but since M' Hughes his tyme 
they are well repaired, and the same nowe holdeii by 
M' Hughes and his assignes. And that he knoweth 
about 4 yeres agoe ther was a slaughter house went to 
decaie." John Bewley, of Allington, aged 56, formerly 
a horse-smith within the Castle precincts, spoke of 
certain outhouses therein, which he remembered, since 
'taken awaie, but by whom he knoweth not"; but said 
also that "to his nowledge the said castle ys in better 
rep^'aco'n then [than] yt was xlviii yeres agoe, when he 
first did knowe the same"; and Richard Symkins, of the 
Holt, aged 74, testified that *' the said Castle ys in 
better rep^'acon [reparation = repair] then yt was when 
M' Hughes came to yt." 

The second inquiry before alluded to was taken in 
Holt parish church, 9th January, 30 Eliz., 158^, before 
Owen Brereton, Thomas Powell, Ralph Ellis, and John 
Salusbury, the plaintiflF being Launcelot Bostock, touch- 
ing the right of Thomas Lother to the utter [Outer] 
Gatehouse. Some of the depositions are interesting 
from the light they throw on the condition of the castle 
and on other matters. For example : *' John Dauyes of 
Earles" [Erlas Hall], aged 42, knew Bernard Bewley, 
servant to Richard Eaton, Constable of th^ Castle, 

Digitized by VjOOQ IC 


dwelling in the utter gjitehouse ; he knew also Thomas 
Lother dwelling there, as servant to Richard Eaton, 
Constable ; he remembered a letter coming to his 
roaster from Launcelot Bostock, begging that Lother 
tnight dwell in and have the use of tlie utter gatehouse 
in variance ; and Richard Roydon, of Holt, aged 60 
years (see note 2, page 317), remembered John Picker- 
ing, the porter, with rooms in the inner gate on the 
right hand going into the said castle. Pickering never 
dwelt in the utter gatehouse, which the Constable 
used for prisonere for debt and misdemeanours, and 
used the gaol within the body of the castle for felons 
and murderers. The castle ditch belonged to the Con- 
stable, etc. 

It may seem to the reader that the foregoing parti- 
culars of the inquiries at Holt on 30th January, 158^, 
and 9th January, 158^, have been dealt with some- 
what out of place. But many of the witnesses at the 
inquiries were old men, and were examined specially 
as to what they remembered of the castle : four of them 
speaking concerning its condition thirty, forty-six, forty- 
eight, and even sixty years before. Mention of these 
testimonies at this stage is, therefore, on reflection, not 
so much out of order or by any way of anticipation as 
it would at first seem. 

At the time of Tidderley s survey, the afternamed 
streets and lanes in Holt were already well established : 
Frog lane. Castle street, High street, " Brigestrete," 
Wrexham lane, mylne lane [Mill lane], GaTlow tree 
lane (the Holt end of Francis lane, properly Franchise 
lane), Chester lane and " Werrock lane, ' which appears 
to be another name for Chester lane, and should be 
spelled " Weirhook lane." So also are named " the 
devyn," or " devon" [brook] ** the hogmore " [now 
" Hugmore''] area, and *' the underwood called * Cor- 
nysh.' " There were four shops under the Town Hall, 
and there was a horse-mill, more particularly described 
hereafter. The "espyes" were fields somewhere near 
where Esphill now is. Already, besides the free bur- 

Digitized by VjOOQ IC 


gesses and those holding demesne land by twenty-one 
years' leases, was a class of forty years' leaseholders^ 
perhaps men to whom were granted burgages and lands 
which had escheated to the lord. And there was 
nothing in the way of a free burgess holding leasehold 
land also. Many of the burgages had two or more 
burgages ; and some, it would seem, had a single large 
house occupying the site of several adjoining burgages. 

In Appendix I will be found a list of the tenants ot 
Holt at the time of Tidderley's survey. 

I have not yet been able to find a convenient place 
to speak of the known constables, chaplains, and other 
officials of Holt Castle, and so supply a list of them in 
Appendix II to this chapter. 

And in what follows this paragraph of tlie same 
chapter, it will be fitting to say what remains to be 
said touching the history of the lordship generally from 
the time of Henry VII onwards. 

In 1534, Bromfield, Yale, and Chirkland were 
granted by Henry VIII to his illegitimate son, Henry 
Fitzroy, Duke of Richmond, then only fifteen yeai-s oi 
age ; but the lad died about two years afterwards, and 
the lordship came back to the Crown once more. 

Permant says that in the reign of Eidwurd VI 
Bromfield and Yale were **in possession of Thomas 
Seymour, Lord-Admiral, and turbulent brother to the 
Protector Somerset. He made the fortress of Holt 
subservient to his ambitious designs, and formed there 
a great magazine of warlike stores. His deserved but 
illegal execution again flung Bromfield into the pos- 
session of the Crown." The statement, so put, implies 
that there had been a grant to Admiral Seymour of the 
lordship as such : a most improbable circumstance, con- 
sidering all that is otherwise known. But the Patent 
Roll of 28 Henry VIII (1536) proves that on the 2nd 
October of that year there was conceded to Geoi^Q 
Cotton and Thomas Seymour, a gentleman of the Privy 
Council, the survivorship of the office of.**ma3t^ 
steward" of the Castle of Lyons alias Holte, and of the 

Digitized by VjOOQ IC 


manors or lordships of Bromefelde and Yale," void by, 
the attainder of ^William Brereton (Chamberlain of 
Chester, executed 17th May, 1536), with a fee of £20 
yearly, which was always the fee of the seneschal, or 
chief steward, as it still is. Nq doubt, Seymour as 
seneschal could procure the appointment of the Con- 
stable of Holt Castle, and would thus have the control 
of the Castle itself, making it, as Dugdale says, *'a 
magazine of warlike provision"; but the offices of Steward 
of Bromfield atld Yale and Constable of the Castle were 
distipct,;.and a grant of either or both of them did not 
involve a [grant <if the lordship. There is also a later 
grailt (dated 1st Edward VI) to Seymour of various 
manors and l^nds ip the lordship ; and this grant, no 
doubt, led Pennant (or Dugdale, on whom he probably 
relied) to the mis<(^ken assumption, which his words 
imply, that Thomas^ Admiral Lord Seymour of Sudeley, 
was Lord of Bromfield and Yale. 

The next event in the history of Holt, so far as the 
subject of this -chapter is concerned, was the accession 
of Queen Elizabeth; in the first year of whose reign ten 
of the burgesses of Holt were bound over, each in the 
sum of £100, to stand to such order as should be made 
in the Court of Exchequer at Westminster touching 
the revival of **Decaies" of rent which had grown 
ivithin the town and liberties. Then, in the fourth 
year of thet Queens reign^ by virtue of a commission, 
directed on 1st August, -1561, to. William, Marquis of 
Winchester (Lord Treasurer), Sir iiRichard Sackville, 
Knight (Under Treasurer), and Sir Walter Mildmay, 
Knight (Cheincellor of the,, Exchequer), these, three 
appdintea as .sub-oommissioners (to make a surVey oi 
the lordship,, revive decayed rents, and compound with 
the tenanjis), Robei^t l^oulton (then Auditor of Wales), 
John Puleaftoii, John Gwynne, John Trevor (of Tref- 
^.lyn), and Robert Tarbridge, Esquires, who thereupon 
proceeded to. make stich a survey and agreement as 
was from ttiem^ requi'red. . So far as the town and 
franchise of Holt were concerned, the burgesses were 

6th SkB., VOL. vit. 21 

Digitized by VjOOQ IC 


called upon, according to a decree made in Hilary term 
in the fifth year of the Queen's reign, to pay their 
ancient rents, which were more by £12 8^. lOa. than 
their existing rents, and to perform other things which 
need not be here particularised. This device seems to 
have been confirmed in the Court of Exchequer on the 
6th June, fifth year of James I (1607). The town was 
also charged with the collection of the borough rents. 
But some simple folk — free burgesses ignorant of their 
rights — took forty years' leases of their lands from the 
Queen; and in 1620 the burgesses asked that their 
descendants might have the benefit of the charter 

rnted to them by Thomas, Earl of Arundel, of which 
the fifth year of Elizabeth's reign the burgesses 
obtained a confirmation. 

In the survey of 4 Elizabeth (1562), the aftemamed 
streets and lanes in Holt are named : Chester lane, 
'* Warrhooke lane,*' **ffrog lane," Castle street, Wrex- 
ham lane, Cornish lane, gallow tree lane, " hogmore 
lane," Pepper street, Bennets lane, *' the pavement 
leading to Comon Wood," "mooregate," and **over- 
whart street." I have never elsewhere found any 
mention of the street last indicated in Holt : it was 
probably a cross street, " overwhart" having the mean- 
ing of over against^ or crossing. " Hogmore" is always 
so spelled. There appears as yet to have been no 
house at Cornish, or " Cornis," as it is once called. 
''Hodhill" was near the burgage of Thomas Crue, 
which burgage is now represented by Holt Hill. The 
high cross, "high greene," and '* litle green," by fish- 
poolfield, are also mentioned. ^* Knight s wood next 
Wrexham Lane" is referred to, also " the pool of Dee" 
[stagnum Dee], The " devon" brook is often named. 
The basement of the Town Hall is still occupied by 
four shops, and the site of the horse-mill described as 
containing 60 yards, and having the Dee on the east 
side of it, the land of John Pickering on the west, 
Saunders liey on the north, and the land of Edward 
Aimer, Esq., on the south. We may compare these 

Digitized by VjOOQ IC 


names with those given in Tidderley's survey (on 
page 319). 

There will be found in Appendix III to this chapter 
a list of the tenants of Holt in 1562. 

Queen Elizabeth seems to have neglected to appoint 
most of the ancient officers of the lordship, so as to sav« 
expense, and obtained money by granting leases of 
many of the demesne lands, or by selling them. In the 
second year of her reign she sold outright not merely 
the chantry lands belonging to Holt Church, but also 
the tithes of those other lands which had been appro- 
priated to the use of the chaplain in the chapel of Holt 
Castle. As to the first of these, mention will be made 
when, in a later chapter, the church of Holt has to be 
described, but this is the place to speak of the last- 
named. Those lands have already been discussed in 
my Country Townships of the Old Parish of Gresford^ 
p. 134, and a reference to them and to the free chapel 
in the castle is contained in Tidderley's Survey (see 
before, p. 316). But it is necessary now to enter into 
further particulars. The tithes of the chapel within 
{infra) the Castle are mentioned in April, 1451, and 
the chapel itself is indicated in the earlier plan of the 
same, hereafter to be considered. Also, in the will of 
John Roden, Serjeant-at-Arms, dated 6th March, 15 If, 
the " fre chapell of the castell of the Holt" is named, but 
only in such a way as to imply that the testator was 
farmer of the lands attached to it. On the 14th June, 
1548, Edward VI leased [the tithes of] certain lands 
and tenements called bourd [board] lands and passe 
[" passe" for *' past," that is '* pastus"] lands in the 
townships of Burton, AUington, Merford and Hoseley, 
" lately parcel of the possessions of the chantry or free 
chapel within the Castle of Lyons," to Thomas Barrett, 
John Wrighte, and John Coldewell. On the 20th 
January, 15f^, the Queen sold these tithes to John 
Norden and Clement Roberts and their heirs, who no 
doubt speedily conveyed them to other persons ; and 

21 « 

Digitized by VjOOQ IC 


the said tithes came ultimately into the possession of 
the Trevors.^ 

, . The chapel of the Castle was " free/' because the lord 
of Bromfield and Yale not merely provided it, kept it 
in repair, maintained the chaplain by the allotment of 
lands, the tithes whereof went to his support, and 
appointed that official, but held him and the chapel 
free from the jurisdiction and control of the Arch- 
deacon, as well as from the control of the incumbent of 
the parish in which the castle stood. 

It seems improbable that the chapel in the Castle 
was used as such after the tithes appropriated to it 
had been leased, and certain that it was not so used 
after they had been alienated. 

The Castle chaplains known to me are so few that 
they can soon be named. 

On the 24th January, 151^, Sir Anthony Byrne was 
granted the chaplaincy of the Castle, in the same way 
that William Alom formerly held it. To Anthony 
Byrne, son of Ralph Byrne, Sir Williain Roden, Rector 
of Gresford, bequeathed (24th June, 1526) a breviary, 
a samite hood, and a surplice. And, according to the 
late Mr. Ellison Powell, Henry VIII, in the 29th year 
of his reign, granted to Thomas Byrde, clerk, the free 
chantry or chapel of Holt Castle ; and this chaplain was 
still in possession at the time of Tjdderley^s Survey 
(see p. 316). 

On the 4th June, 1610, in the 8th year of James I, 
the King granted the title of Prince of Wales and Earl 
of Chester to Henry, his elder son, and then, or shortly 

^ The lands, still called " boardlands," ont of which the tithes 
were due, or some of thera, belonp^d to John Trevor in 1523, as 
appeal's by a canons entry on folio 39 of the Survey of 4 Qneen Eliz. ; 
from which we learn that at the time of the survey John Trevor had 
a capital messuage and thirteen acres of pasture pertaining thereto 
in IVIerford, Hoseley and TrovalliD, premises anciently belonging *' to 
the free chapel of the castle of the lordship of holt," and obtained 
by exchange, formerly the lands of Jenkin ap David Griffith and 
Jbhn'ap John ap Robert, as shown under the seal of GHstle Lyons, 
5th July, 15 Henry VIII. 

Digitized by VjOOQ IC 


afterwards, gave him the lordship oi Bronifield and 
Yale, the tenants, free and leasehold, whereof yielded 
a mize of 600 marks (£400) to the said Prince ; having 
already, on the King's accession to the throne and to 
the lordship, given a mize of like amount to James. 
After Prince Henry's death, the Principality of Wales 
and the lordship of Bromfield and Yale were granted 
(3rd November, 1616) to Prince Charles, afterwards 
Charles I, whereupon another mize of 600 marks was 

On the 3rd of March, 16^, a commission was made 
out to John Norden the elder, and John Norden the 
younger, the Prince's own surveyor, Sir Richard Smith 
being in personal attendance upon Charles, to make a 
new survey of Bromtield and Yale, and other the Prince's 
possessions in Wales. The survey of Holt was begun 
on the 11th April, 1620, and is very valuable. It will 
be dealt with in another chapter. 

Meanwhile, I may say (as explained in my History 
of the Country Townships of the OldPaiish of Wi^exham, 
p. 40) that James I, as a device for raising money, on 
the 27th January, in the 22nd year of his reign, 162|, 
leased Bromfield and Yale, for ninety-nine years, to 
commissioners, who were empowered to sell escheat, 
leasehold, and demesne lands ; and to make, in con- 
sideration of the payment of a sum of money (amount- 
ing in some cases which have come under my cognisance, 
to twenty-five years' purchase, calculated on tne exist- 
ingrents) freehold or fee-farm estates. 

Under the powers of this patent, or commission, 
many of the manors, demesnes, leasehold lands, and 
rents were sold. Thus, the Earl of Bridge water acquired 
the manor of Ridley, in Isycoed, at a reserved rent of 
£11 Is. 4fd. Also, on 3rd July, 1628, the commis- 
sioners, or patentees, of James I, as they may be called, 
conveyed Mersley Park, together with the Broadland 
and ^Dushy land, for £2,000, at a reserved yearly rent 
of £20, to the same John, Earl of Bridgewater, the 
two last-named pieces being apparently taken out of 

Digitized by VjOOQ IC 


common wood. We see, in short, one process by which 
landlordism, on the large scale, was being built up. 
But many leaseholders also purchased, under this 
patent, the lands which they held by forty years' leases, 
and became small proprietors. 

When Charles I was beheaded, Bromfield and Yale 
were treated, for ten years, as part of the public estates 
of the Commonwealth ; and a survey was then made, to 
which there may be occasion to refer hereafter. But, 
save for these ten years, the lordship has remained in 
the hands of the Crown since Charles I became King, 
and is still vested therein, being administered by the 
Commissioners of Woods, Forests, and Land Revenues. 
A steward is appointed for the same by the King, on 
the recommendation of the Prime Minister ; which 
steward receives the ancient salary of £20 yearly, and 
vacates his seat, if a Member of Parliament, on accept- 
ing the oflBce. 

William III intended to have granted the lordrfxip 
to William Bentinck, Elarl of Portland, and a mighty 
pother was made : Sir William Williams, Sir Roger 
Pules ton of Emral, Mr. Robert Price of Gilar, and 
Sir Robert Cotton of Combermere, with others, ap- 
pearing before the Lords of the Treasury to oppose the 
suggested grant. The concession proposed was there- 
upon withdrawn, and the reserved rents, when they can 
he identified, are still paid. 

So ends this account of the lords of Holt. 


(See p. 320.) 

List of the Tenants of Holt at the time of 
Tiddbrley's Sukvey. 

1. — Free Tenants. 

John Rodon Lancelot Prestlond 

John Knyght John hychyn [Hutchen] 

John Alford William ap John 

Richard Baker William Pate 

Digitized by VjOOQ IC 



Robert Abathowe [Ab Atba] 

James Gbytam [Cheetham] 

David Wyld 

Thomas Grue 

WiUiam Woddall 

John Williams 

Ralph Pnlforth 

John Adjo 

Robert hychin [Hntohin] 

Joan Balfer, late wife of John 

Owen Breerton, gent. 
Thomas ap hoell 
John ap Riohard 
John Griffith 
William ap William 
Roger Wylkynson 
John Aimer, gent 
John Tayler 
John Chethame 
John Hanson 

David ap O . . • 
William Owdall [Udnl] 
John maddoc ap Jollvn and 
Thomas maddoc ap Jolljn 
Ralph Bnlkelej 
hngh hanky, sen 
hngh hanky, jnn' 
Thomas Belott 
John Clnbbe 
Riohard ap Jenkyn 
Robert Aldford 
Thomas Barbor 
Richard Hanson 
fflorenc lother 
William Smythe 
John Roydon 
Thomas Edgworth 
Thomas Arodon [Rodon] 
John Erthley 
Edward Aimer, esq. 
Ralph Rawlins, chaplain 

2. — Tenants at Terms of Yeabs at samb Time. 

John Alford 
Richard Baker 
John Pykering 
Anthony Crewe 
Ralph Pnlforth 
William Crewe 
William Woddall 

Thomas ap hoell 
fflorenc lother 
John ap Won [? Gwion] 
Robert Aleford [Aldford] 
Richard Hanson 
Thomas Arodon 
Robert hyohen 

3. — Tenants at Will at same Time. 

William Pate 
Robert hychin 
William beggewvke 
William ap William Segge- 

John Pnlforth, lancelot pnlforth 
and William pnlforth 

John Pnlforth and William Pnl- 

Lancelot Polford 

Brian Bayte 


(See p. 320.) 

Resident Cokbtables or Holt Castlk 

The first Constable of whom I have any note was David 
Byton ap Llewelyn, of Upper Eyton, in the parish of Bangor 
is y Coed His grandfather, Ednyfed a^ Gruffith ap lorwerth, 
of Eyton, was a famous bard, who was living in the twelfth year 

Digitized by VjOOQ IC 

32 8 THE TOWN OF HdLt, 

of Edward III, when he appeared in the Court of the rhag- 
lotry. And this David Eyton ap Llewelyn was the *' Dauid de 
Eyton, Constable of the* Castle of Lyons," who occupied that 
post in 1391. He was an early example of the policy,* often 
pursued, of appointing the head, or at least a member, of a great 
Welsh family to an important Welsh office. 

In the tenth year of Henry V, Dr. A. E. Lewis infonns me, 
Egbert Corbet was Constable of Holt. 

I possessed the names of three other Constables — Lancelot 
Lothar, Lancelot Bostocke, and Thomw Powell — but Mr. Edward 
Owen supplied me with the names or two more, obtained from 
an "Exchequer Deposition" of 1591, which referred to Richard 
Eyton as being succeeded by I^uncelot Bostocke in the con- 
stableship of Holt Castle. It mentioned, also, Lancelot Lothar, 
Thomas Powell, David Price of Yale, and Sergeant Eyton as 
having been previous Constables. This would seem to imply 
the following order: Lancelot Lothar, Thomas Powell, David 
Price, Serjeant Richard Eyton [for Serjeant Eyton and Richard 
Eyton were presumably one], and Lancelot Bostocke. And this 
order corresponds with what is otherwise known. Some com- 
ment on these five names may not be unacceptable. 

Lancelot Lothar was Constable of Holt in the twenty-first 
year of Henry VII, and in the tenth and eleventh years of 
Henry VIII. He was appointed Constable, it may well be, 
after the execution of Sir William Stanley, and in that com 
nmst have received Henry VII when that monarch visited Holt 
on the 17th July, 1495, on his way to visit his mother, Margaret 
[Beaufort], she having married Thomas Stanley, Earl of Derby, 
brother of the above-named Sir William Stanley, of Holt 
Mr. Hughes, of KJhmel, believes Lancelot Lothar, or Lowther, 
to have been a son of Sir Hugh Lowther, by his wife Anne. 
daughter of Sir Lancelot Threlkeld, and there is much to .be 
urged in favour of this suggestion. Constable Lothar (for so he 
seems to have spelled his name) was Deputy-Receiver of Brom- 
field and Yale in 1519. There is at St. Asaph the office-copy 
of the will of a Lancelot Lothar, of the parish of Gres- 
ford, dated 19th April, 1578, proved 19th June in the same 
year. He desired to be buried at Gresford, spoke of "Elyn 
nowe my wife," of his son, Thomas Lothar,* and of his cousin, 

^ A sagaciops policy; not so much followed,. however^ between the 
reien of Henry IV and that of Henry VIZ. 

*^ There was a Thomas Lothar to whom Constable Richard Ejtou 
granted lodjfing in " the uttergate heuse*' of Holt Castle (see before, 
T>. 318). He was afterwards employed in some capacity by Mr. 
I'albot, of Grafton. This information I owe to Mr. Edward Owen. 

Digitized by VjOOQ IC 


Roger Wynn, who cannot be identified with certainty. How- 
ever, this testator appears to be another and later Lancelot 
Lothar. Lancelot Lothar, the Constable, had undoubtedly two 
daughters, one of whom, Catherine,* married the first Thomas 
Powell, of Horsley, his successor in the constableship, and the 
other, Elizabeth, became the wife of John Heynes, Keceiver of 
North Wales, whose daughter, Anne, married the second Robert 
Davies, of Gwysanney, in 1620, one of the burgesses or free- 
holders of Holt. Mr. Hughes tells me that there was a double con- 
nection between the Lowl^ers and Davieses, which will be shown 
sufficiently by a note at tlie foot of this page.^ Mr. Hughes also 
tells me that Ltmcelot Lothar, the Constable, married Elizabeth, 
daughter and heir of Raudle Minshall. This particular Con- 
stable was very popular, so that about and soon after his time 
we meet with such names as Lancelot Powell, Lancelot Lloyd, 
Lancelot Roydon, and Lancelot Pickering, denominating connec- 
tions of his in the second generation ; also with the names 
Lancelot Maddocks, of Marchwiel ; Lancelot Calcott, of Wrex- 
ham ; Lancelot Lewys, of Gwersylit, Lancelot Phillips, Lancelot 
Aldford, of Holt; Lancelot Yardley, of Holt ; Lancelot Hanson, 
of Holt ; Lancelot Broughton, of Eyton ; Lancelot ap Ellis, of 
The Court, Wrexham, and many others. These Christian names 
gave rise, in many cases, to corresponding surnames, so that we 
get presently John Lancelot, of Wrexham ; John and Edward 
Lancelot, of Caeca Dutton; William Lancelot, of Pickhill, to 
mention no more ; and " Lancelot " has been ever since a not 
wholly uncommon surname in this district. The fee of Lancelot 

There was anotlier Thomas Lothar, son of George Lothar, deceased i 
who held land in Holt in the year 1562, the widow, Alice, of the 
said George Lothar being then married to John Salasbary, gent. 
William Woodall, gent., had also at the same tiaie, by right of the 
said Alice, widow of George Lothar, a lease of some land in Holt. 
And in the same year a *' fflorenc lother^' was a burgess of the town 
(see Appendix I, p.j3271, . 

^ One of her sons was Lancelot Powell. 

Catherine, dau.=j= Robert =(2) Elizabeth, dau. of George Lowther, and relict 
of George Davies. of John Heynee, Receiver of North Walea, 
Ravenscroft. who died 27th May, 1591. Elizabeth was 
I buried at Mold, 11th March, 1636. 

Robert Davies j^Anue, dau. of John Heynes. She was buried at Mold, 
buried 27th SOth Aug., 1636. 
Jan., 1688. 

Robert Davies ; died Sept, 1667. 

Digitized by VjOOQ IC 


Lothar, as Constable, was £10, to which must be added his fee as 
Deputy-Receiver. And there were free lodging and various 

For Thomas Powell, of Horsley, the next Constable, son-in- 
law to his predecessor, Lancelot Lothar, the reader may be 
referred to my History of the TovmBhips of the Old Parish of 
Oresford, Powell pedigree, opposite p. 118. While he lodged in 
the Castle, Mr. Edward Aimer, the Deputy- Stevirard, also lodged 
there (see before, p. 317). 

David Price, of Yale, followed Thomas Powell, apparently, 
but I can attach no date to him or identify him. 

Richard Etton, Serjeant-at-Arms, the next Constable named, 
appears to have been Richard, third son of John Eyton, Esq., of 
Watstay, but this identification is not without doubt To him 
was leased, "about 7 Eliz.," by the Crown a meadow called 
" Constable's Meadow," still so called, near Coed Evan, in the 
manor of Oobham Isycoed (township of Dutton DifiFaeth). 
There was also a suit concerning this meadow in the thirty- 
fourth year of Queen Elizabeth. William, son of Serjeant 
Richard Eyton, was baptised at Ruabon in December, 1577. 

Lancelot Bostocke, High Sheriff of Flmtshire in 1574, the 
last Constable, appointed as such in November, 1585, of whom 
we have any trace, was son of Robert Bostocke, formerly of 
Churton, by his wife Jane, daughter of Richard Roydon, of 

pedigree of bostocke of holt. 

Robert Bostocke (son of Robt. ^Dorothy, dau. of Sir Geoi^ge Cftlveley, of Lea, 
Bostocke, of Churton. Baokford parish, Cheshire. 

Lancelot Bostocke, Constable of=Fjane, dau. of Richard Roydon, of Holt, by Anne 
Holt Castle, * 'the pensioner. ' his wife, one of the daughters of the first 
I Thomas Powell, of Horsley. 

Qeorge Bostocke of Holt ; will dated=T=Dorothy, dau. of Hugh Calveley, of Lea, 
17th Sept., 1627, proved 1628. Cheshire. 

1. I 2. 

Jane, dau. and heir of = George Bostocke ofsKatherine, dau. of Hugh Jones, 
David ap Edward, of Holt ; will dated of Wrexham, widow of Ed- 

Dinbren, o. «. p, 3rd Aug., proved ward Jones, of Wrexham. 

(HaUUm MS.) 8th Oct., 1668. (HdUton MS,) 

Dorothy. "Thomas Yale, son of Mary.=j=Thomas Williams of PlAs Jenkin 

in Dutton, parish of Holt, and 
of Abenbury, son and heir of 
Roger Williams, of Oswestry. 

the first Thomas 
Yale, of Plas yn 

Lancelot Williams, alias Bostocke, 2nd son, devisee of his unde Geoiige 
Bostocke ; buried at Holt, Ist Jan., 166i. 

George Bostocke, of Holt, was pardoned, 5th Oct., 1694, for killing John Roydon. 

Digitized by VjOOQ IC 


Holt, gent.^ His only son was George Bostocke, of whom 
more in the next chapter. Mr. Hughes, of Kinmel, suggests that 
Lancelot Bostocke, the Constable and " psnsiouer," belonged to 
the '' Band of Gentlemen Pensioners," gentlemen of blood and 
coat armour, instituted by Henry VIII, and now known as 
" His Majesty's Honourable Corps of Gentlemen-at-Arms." 

Sir Richard Lloyd, of Esclus, who defended Holt Castle for 
Charles I during the Civil War was, of course. Governor of the 
same, but he was so much else that his name is not put in this 
list of Constables as not standing in the direct line of succession, 
which had been brought to an end some time before, and I 
propose to deal with him in a future chapter. 

Nor does it seem fit to give here any list of the seneschals or 
receivers, whose place would rather be in a history of the Lord- 
ship generally ; but I should like to say a few words concerning 
one Receiver in particular, Edward Hughes,* who actually lived 
at Holt Castle. He was High Sheriff of Denbighshire in 1582, 
and died on the 23rd October, 1592. His daughter, and ulti- 
mate heiress, Mary, became the fourth wife of John Massie, of 
Coddington^ son of Roger Massie of the same. It has not been 
found possible to discover with anything like certainty the 
paternity of this Edward Hughes, but Mr. Edward Massie, of 
Coddington, and Mr. Hughes, of Kinmel, state that Robert Cooke, 
Clarencieux (1567-1592), granted him the aftemamed coat-of- 
arms : Gnles, a fret argent on a canton o?' a pheou of the first, 
and he is then described as " of Denbighshire, servant to Mr. 
Dudley." He also became Receiver for the counties of Chester 
and Flint. In his will, proved at Chester in 1592 (a summary 
whereof Mr. Hughes, of Kinmel, has given me), the testator 
leaves everything to his wife, Ann Hughes, trusting that she 
will behave as a good mother to his children, and begging his 
loving and worthy friend, " Mr. Roger Puleston, of Emmerald 

1 For the above-named Richard Bojdeo, see note on p. 31? ; and 
the Bostockes of Holt obtained their lands in the franchise, or part 
thereof, by descent from him. In 1627, a Richard Roydon was 
living in Castle Street, Holt. Mr. E. B. Royden tells me that 
Richard Roydon, father-in-law of Lancelot Bostocke, Constable, had 
besides Jane, five other daughters — Dorothy, Maud, Mary, Anne, 
and Alice, and that he was the son of Thomas Roydon, son of 
Richard Roydon, son of William Roydon, English bailiff of Wrex- 
ham in 1467. 

^ Mr. Edward Massie informs me, on the authority of Sir Henry 
Maxwell Lyte, that Edward Hnghes was appointed Receiver for the 
Grown in September, 1568, and that in March, 1569, he received 
authority to inhabit Holt Castle, and to have twenty cartloads of 
wood out of Mersley Park. 

Digitized by VjOOQIC 


[Eiural], Esq.," and his cousin, Mr. JoUu Tuieston, of Llwyn y 
Knottie, to aid and assist her, the will being witnessed by John 
Roydon, Thomas Crue, Richard Case, and John Leeche. Mr. 
E. B. Royden has also supplied me with the summary of a 
case in the Court of Exchequer, Trinity Term, 1598, in which 
Christopher Hughes, son and heir of the late Edward Hughes, 
Esq., late Receiver-General of the revenues of North Wales and 
of the county and city of Chester, sets forth for himself, as well 
as on behalf of Anne and Margaret Hughes, daughters of the 
said Edward Hughes, that his father, at the time of his death, 
possessed various free lands in the parish of Holt, worth about 
£20 yearly, certain copyhold lauds there worth £7 yearly, and 
personal estate worth about £2700; and by his last will appointed 
Anne« his wife, aged about 66, his sole executrix, and directed 
her therewith, together with £800 then in his house, to satisfy 
the money due from him to the Crown, or sell his lands for that 
purpose, and for the benefit of his unprovided children. When 
the said Edward Hughes died, 25th October, 1592, he was in- 
debted to the Crown in £1,822 odd, whereof the said Anne paid 
within a year aBout £66, leaving £1,162^ unpaid. Afterwards, 
about Bartholomew-tide, four years past, the said Anne Hughes 
married^ John Roydon, gentleman [of Isycoed], who entered upon 
the lands of the late Edward Hughes, got hold of his personal 
estate, paid into the Exchequer £670 only, leaving £800 unpaid, 
and, affirming that the said personal estate would be insufficient, 
sought to sell the lands of the deceased. About four years 
past, Roger Roydon and John Taylor were authorised to receive 
csrtain arrears of revenue owing to Edward Hughes, and they 
collected about £650, which they had accounted for to the 
Exchequer, and the complainant begged for a subpa^na against 
John Roydon, Roger Roydon, and John Taylor. In reply to 
these ex parte statements, John Roydon declared in Michaelmas 
term, 43 Queen Elizabeth, that Thomas Crew, of Holt, gentle- 
man (who was son-in-law of the said Anne Hughes), and John 
IjCach, since deceased, were, by commission, authorised to receive 
the Crown revenues of North Wales, Cheshire, and lordship of 
Denbigh, up to Michaelmas then last past, and rendered up 
their account to the Exchequer on behalf of Anne Hughes, but 
detained thereof the respective sums of £691, £99, and £30 
odd, for which John Roydon, as husband of the said Anne, was 
answerable, who prayed for a subpcena against Thomas Crew, 
and Thomasine, widow of John Leech,* both **very rich and 

^ There is something wrong in the arithmetic here. 
' Anne, widow of Edward HugheR, was John Roydon's 2nd wife. 
3 John L«eob, of Holt, gont, was living on the 27th Jane, 1598, 
when he was about iO years old. Lands and goods in the possession 

Digitized by VjOOQ IC 


wealthy persons/' while he (John Koyden) was in hia eetatd 
decayed by reason of the charge he had been at in finding out 
the falsehood of the account aforesaid. Thomas Crew and 
Thomasine Leech gave in their answers on 13th April, 1601. 
One would like to know what was the name of the daughter 
of Anne Hughes who married Thomas Crue. Strange also it is 
that no mention is made of Edward Hughes' daughter, Mary, 
who married John Massie, of Coddington. The Massies still 
quarter her father's arms, and regard her as his heiress. Perhaps 
the other children named in the pleas died without issue, and 
Mrs. Massey would not associate herself in "the complaint" 
with her brother, Christopher Hughes, with her sisters, Anne 
and Margaret Hughes, or with Mr. John Royden, and so is not 
mentioned in the bilL 

It is. perhaps, worth while to refer to a bit of Welsh verse, 
formerly in the Shirburn Collection (Report on Welsh AtSS,, 
voL ii, part ii, p. 648, by Dr. Gwenogvryn Evans), addressed to 
John Salusbury, heir of Lleweni, when he overcame in the field 
Captain O. Salusbury, of Holt, in 1593. It does not now seem 
possible to identify either of the persons just named. "Mr. 
Salsburie his Cham be''" in Holt Castle, is mentioned on 27 
June, 1598; and, according to Mrs. Slopes, a Captain Owen 
Salusbury was slain in Essex Gallery, London, 10 February, 
160®, at the rising of the Earl of Essex, and was buried at St. 
Clement Danes, Strand. 

(See p. 323.) 

Tenants of Holt in the Fourth Year (1562) of 
Queen Elizabeth. 

(1) Free Tenants. 

Lancelot Yardley William Kethyn, for life of Robt. 

Edward Davje Kethyn, his father 

Thomas Crewe, senior Randle Pulforde 

Brian Bate John hjlchyu [Hutcheon] 

John Wilkinson Launcelot Baker 

John Grifif[ith] Thomas Edgworth 

John Nuttall Thomas Crewe, gent. 

of Mr. John Roy don, formerly those of Mr. Edward Hnghes, were 
seized on the 80th March, 1598, to satisfy the claims of the Crown, 
by Sir Richard Treror, Roger Puleston of Eroral, Esq., and Mor^n 
Broaghton, Esq., the estimated valne of the whole being £od I 13s. 4d.; 
a total which included an ifcem of J&300 for plate, goods, household 
sin if, and jewels. 

Digitized by VjOOQ IC 



William Woodall, sen' 

William Woodall, gent 

Thomas Caloott 

Richard Rodon, son of Thomas 

John Crewe 

Thomas Yaixleley 

Balfe Bainvile and Edward Tay- 

Edward Allmer, esq. 

Susanna, lately wife of . . . Han- 

William Bird and Richard Bird 

John Greene 

Thomas Billot 

John Knight 

William Pate 

Ridiard Aldforde 

William ap Batha and the wife of 

Ralph ap Atha 
William Smyth 
Ralph Bulkeley 
John Stokeley 
Owen Brereton, esq. ["William 

Briereton" crossed out] 
Launcelot Bamston 
John Clubbe 
Peter Rodon 
Katherene, lately wife of John 

Wife of Geoffry Smyth 
Thomas Powell 
Edward Puleston 
William Woodall 

(2) Tenants for Term op Years and at Will. 

John Pickering 

Heirs of Thomas lowther [John 

Salesbury's name crossed out] 
Launcelott Prestland 
Launcelott Hanson 
Edward Davies 
Edward Crew 
.... Pova 
Handle David 
Thomas Pulforde 
John ap Griff[ith] 
John Ledsam 

William Kethyn [Gethyn] 
John Goz ap Richard 
Randle Pulford 
John Hugh Grififfith] 
William Godson [now dead, John 

Godson his son] 
Launcelott Baker 
William Woodall, gent, [by right 

of Alice his wife, lately wife 

of George Lowther] 
Richard Rodon, sun of Thomas 

Thomas Edgworth 

David Wilde and Edward WUde 

Edward AUmer 

Joan Pulford, widow, late wife of 
William Pulford and Thomas 

Launcelott Philipps 

David Gyttyn 

John Princeston .[crossed out and 
John heynys substituted] 

John Alforde 

John Rodon 

John ap JeuNi ap dd 

John Stokeley [crossed out and a 
name, illegible, substituted] 

Thomas Maddock 

Peter Rodon 

Thomas Powell, gent. 

Edward Jones, gent. 

1 William Woodall [crossed out 
and John Heynys substitu- 

Edward Puleston 

Digitized by VjOOQ IC 




In writing an article on this ancient parish church — a 
church wnich to-day stands as a noble living link 
between the Welsh Church of a thousand years ago 
and that of the present time — I propose to deal chiefly 
with matters of historical and monumental interest 
rather than with the architecture of the building, 
which latter I could only but very imperfectly describe. 

The present church, replacing no doubt an earlier one, 
is generally attributed to the thirteenth century, and to 
which period belong the lower portion of the nave walls 
— with the recesses for altar-tombs — and the greater part 
of the tower; the chancel and south aisle being decidedly 
later. The church consists of a nave, chancel, soutn 
aisle, western tower, and north transept. The north 
transept is probably built on an older foundation, and it 
is difficult to say to what period it belongs. The eastern 
portion of the south aisle is used as the Consistory Court 
of the Diocese, and up to fifty years ago was separated by 
a screen from the rest of the aisle. In old documents 
this part is also often described as the " town chancel." 
There are two vestries — one being a recently- built choir 
vestry. The nave is divided from the south aisle by five 
lofty arches, resting on massive buttresses of a severely 
plain character. 

The length of the church from entrance door to the 
east window is 170 ft, while the width of the fabric 
is 50 ft. 

The singular site of St. Peter's, outside the walls of 
" Kaermerdin '' (to use the spelling of the early 
Charter-rolls), and between these and the old City 
of Carmarthen — ^now forming the eastern portion of 
the town, is a matter of much interest. To quote from 
Archdeacon Be van's lecture on St. Peter's in 1884 : 

Digitized by VjOOQ IC 

336 ST. petbr's oauRou, oakmabthbk. 

" It would be interesting to know the relations that 
may have existed between St. Peter's and the older 
church of St. Teilo, previously to the appropriation of 
these churches to the Priory of St. John, Carmarthen, 
in Henry the First's reign (1100-1 135), and ascertain how 
it was that St. Peter's became the parish church rather 
than the other ; to define the relations,, ecclesiastically 
speaking, between the old City of Carmarthen, to 
which St. Peter's was more particularly attached, and 

Fig. 1. — Interior of St. Peter's Church, Carmarthen : View looking Eafit. 
{Excelsior Co., Carmarthen Photo.) 

the new town, outside whose walls the church stood ; 
and to discuss the question whether St. Peter's was in 
any way responsible for the spiritual cure of the in^ 
habitants within the walls." Unfortunately, it seems 
very difficult still to clear up this interesting matter. 

The visitor on entering St. Peter's for the first tim^ 
cannot but be struck with a certain noble impressiveness, 
in spite of the plainness of the interior, and many 
architectural defects which alterations made in different 
periods have brought about. 

Digitized by VjOOQ IC 

ST. PBTBR's church, CARMARTHEN. 337 

Considering the antiquity and importance of this 
church, its central position in the diocese, as well as 
its contiguity to Norman castles and influences, one is 
disappointed not to find any traces of medisBval or 
later embellishments. The quaintly-carved gargoyles 
of the tower alone remain of this class of work. 
Probably this defect — one which applies to so many 
of the churches of the district — is due to the absence 
of good local stone suitable for the purpose. The earlier 
church on this site may possibly have been burnt and 
destroyed in the fierce Welsh conflicts of the thirteenth 
and fourteenth centuries ; but had it been much more 
ornate than the present one, we might have expected 
some traces of it to have been handed down to us in 
portions of the stone work. 

The first mention of this church occurs in the 
Annals of Battle Abbey, to which it was given by 
Henry I (1100-1135) in the earlier years of his reign. 
The gift included, as well, one other church of an earlier 
origin, named the Church of St. John the Evangelist 
ana St. Teilo, and the old City of Carmarthen. As 
St. Peters appears to have had chapelries attached 
to it at this time (Newchurch and Llangain), it favours 
the belief of many authorities that even at this time 
the church wiis an old one. Bernard, the first Norman 
bishop of St. David's, was appointed in 1115. He 
seems early in his episcopate to have taken steps to 
get the king — with whom he had considerable influence 
— to exchange St. Peter's for some other possessions in 
Hampshire, with the view, no doubt, of appropriating 
the living in favour of the newly-founded Priory of St. 
John, Carmarthen. The bishop seems to have been 
much interested in this Priory, and gave it an endow- 
ment. The king's consent was obtained about 1125, 
though it was not until after Bernard's death that the 
transfer was confirmed by a grant from Henry II in 
1180. One of the witnesses to this deed was Peter 
de Leia, Bishop of St. David, and formerly Prior of 
the Cluniac Abbey of Wenlock, Salop. In connection 

6th 8ER., VOL, VII. 22 

Digitized by VjOOQIC 


.9T. Peter's ohu«ch, Carmarthen. 

with this arrangement jthere is extant a very curiously 
;i;irorded deed^ showin;^ that, as the results of the com-: 
plaint of one Richard ap John, Vicar of St. Peter's in 
1278, to the Bishop of St. David's (Richard de Carew) 
against, the Prior of St. John, for making too scanty 
an allowance to him the Vicar, the Prior agreed to 

Fig. 2. — Coat-of-Arms from St. John's Priory, now in South- 

East Wall of St. Peter's Carmarthen. 

{Exceltior Co.t Carmarthen Photo.) 

pay the Vicar in future ten marks a-year, the payment 
to be made quarterly. The deed is dated at Lamphey, 
April 4, 1278, and in it occurs the first reference to 
the vicarage. This arrangement with the Prior, by 
which the Vicar of St. Peter's got paid a small pension 
of £6 1 3s. 4d. out of the tithes, instead of a certain 
portion of the latter being assigned to him, bore very 

Digitized by VjOOQ IC 

ST. pxtbr's chuuch, oaiImakthun. 339 

UQsaitisfactory results when the Priory was dissolved 
by Henry VIII, The latter directed that the lessee of 
the tithes 3hould continue to do what the Prior had 
done in the past— pay a slightly increased stipend of 
Sn to the Vicar ; and, unfortunately, this arrangement 
had continued to modern times, in spite of the greatly 
increased value of the tithes, the latter being now 
worth nearly £1,000 a year. 

Through the kindness of Mr. T. W. Barker, the 
Diocesan Registrar, I am enabled to give the following 
extracts from the oldest diocesan registers on some 
early appointmenta to 8t, Peter's by the Priory authori- 
ties. Under date December 20, 1403, David Robyn 
was made Vicar on the presentation of the Prior and 
Convent of St. John's, Carmarthen; July 12, 1408, 
William Sty ward, Chaplain, was presented to St. 
Peter's by the same authorities ; February 22, 1486, 
John David, M. A^, was collated to this church ; July 3, 
1600, John Harry to Vicarage vacant by death of John 
ap David on presentation by Prior, etc., of Carmarthen ; 
January 1, 1501, Admission of Sir David Webbe to 
Vicarage vacant by death of John Harry. 

In 1394, we find by a Charter-Roll of Richard II, 
one Thomas Rede, of Carmarthen, received permission 
to assign certain lands for endowing a chaplain *' to 
daily celebrate Divine service to the honour of the 
Blessed Marv, in a certain chauntry, anciently founded 
within the ohurch of the blessed Peter, of Kermerdyn," 
etc. It is difficult now to identify the position of this 
I^ady Chapel; but on the supposition that the earlier 
church was in a cruciform shape, this chauntry may 
have been on the south side, and was possibly absorbed 
when the south aisle was added. On the south wall 
may be noticed a niche for a holy-water stoup. Speed's 
Map of 1610 pictures a church almost identical with 
that of to-day ; but the fact that the centre of the roof 
of the nave is not in line with that of the chancel 
favours the idea that either the chancel arch was en- 
larged, or a south aisle was added in the fourteenth or 

Digitized by VjOOQ IC 

340 BT. pbtbr's church, carmarthsn. 

fifteenth century. The steps leading to the rood-loft 
were noticed in the wall behind the pulpit, when 
alterations were being carried out some nfty to sixty 
years ago. The remains of the ridge of a higher roof 
may be noticed on the north side of the tower, but 
whether this lowering was done by Nash when re-roofing 
and re-ceiling the church in 1790 is a matter of con- 
jecture only. Before referring to just a few of the 
numerous monuments which are in the church, it may 
not be amiss to touch on some historical and social 

Just as the parishioners to-day have a warm affection 
for this ancient House of God — oonsecrated as it is to 
them by the worship, the hopes and fears, the joys and 
sorrows of many generations of their forefathers — so 
in days gone by the same reverent care was displayed. 
In 1557 we find (thanks to the Rev. G. Eyre Evans' 
careful researches into the Old Minutes of the Cor- 
poration) by an entry " that in consequence of the 
decay into which the parish church had gone for want 
of care to provide material," an annual rate of j£20 
a year (a large sum for those days) was to be levied 
on the parishioners, and the churchwardens were to 
account to the Mayor and the Council for the same. 
This attachment further evidences itself in many wills, 
both where bequests are left to the church or vicar, or 
where directions are given for burial there. An ex- 
tract from one or two wills will illustrate what I 
allude to. 

Amongst the leading citizens in Queen Elizabeth's 
reign was Alderman Edward Myddleton, whose beautiful 
autograph signature might have been noticed in the 
Old Corporation Minute Book, kindly lent for the 
exhibition at the Assembly Rooms, Carmarthen, in 
August, 1 906. Besides being Mayor (1583), he seems to 
have been a merchant- trader, ship-owner, patriot and 
educationist, and preceded Robert Toye as Mayor, the 
latter being one of those who had petitioned Queen 
Elizabeth for a grammar school for Cfi^rmarthep, From 

Digitized by VjOOQ IC 

ST. pbteb's church, carmakthen, 341 

Mjddleton's will, dated October 6th, 1537, we give the 
foUowing extract : — 

" Edward Myddleton, Alderman of the To.wne and 
Cpuntie of Caermarthen, being sicke in body— T <lesire 
to be buried in the Church of Carmarthen named St. 
Peter. To the repair of the said church I give 208. ; 
to the repairs of the Bridge, lOs. ; to the poore, lOs, ;. 
to the Free Schools of Carmarthen, 10s. yearly for ever, 
out of the rente of my two Houses in Water Street, 
within the Towne of Carmarthen ; to my servants, 
Thomas Cook and Griffith Adkins, the one haufe of my 
Bark named * The Margett,' which is betwixt me and 
Griffith Howell, etc.'' 

Or, again, observe the curiously-expressed desire ot 
a certain Dame Margaret Lloyd to be buried at St. 
Peter's without the expensive ostentation so usual at 
that time. The tablet to this worthy lady may be 
noticed in the north transept. Her will, dated December 
27, 1755, contains the following : — " If I should depart 
this life in the County, of Ca^*aigan, my Executor shall 
provide a carriage to carry me to the Parish Church of 
St. Peter's, Carmarthen ; and I beg that my corpse may 
be there laid in the same grave with, or as near as 
may be to my late dear husband, and my late dear 
daughter, Elizabeth Evans ; and that a funeral sermon 
shall be preached, and the text taken from the 88th 
Psalm, and the latter part of the 13th verse ; and I 
hereby desire that my coffin may be made of good 
oak, without any ornament or covering ; and that 
instead of scarfs or hatbands at my Funeral, that twelve 
poor old women be clothed with black bays gowns, a 
yard of flannel on their heads, and each a pair of 
gloves, and that they walk before my corpse to my 

From the middle of the sixteenth century this church 
becomeis closely identified with many of those whose: 
names and lives live in history 

Had you en1;ered the church in February, 1555, you 
might have witnessed the strange, sad spectacle of the 

Digitized by VjOOQ IC 

342 ST. Peter's chdkoh, carmarthkn. 

Bishop of St. David's, Robert Ferrar, being handed over 
by the Sheriffs of the County as a prisoner to the 
custody of his successor in the Diocese, one Henry 
Morgan ; and on six subsequent occasions Ferrar was 
brought up here for examination. Here he received 
his final condemnation shortly before his martyrdom at 
the market cross, on March 30th, 1555. 

In the answers which Ferrar gives to some of the 
charges brought against him by his enemies, there are 
frequent references to this church. For instance, he 
states that (1) he had preached right often at Car- 
marthen, the latter being described as " an English 
Towne and the chiefe of his Diocis ; (II) that ** while 
sitting in the Church in Carmarthen with the Chan- 
cellor to hear causes, and seeing the Vicar with other 
priestes, with song and lights bringing a corpse up 
to the Church, he called forthwith the Vicar and 
priestes and rebuked them in open courte as cormorants 
and Ravens flying about the dead carcase for lucr€ 
sake;'' and (HI) *'that George Constantine having 
pulled downe without any authority the Communion 
Altar in Carmarthen Church, appointing the use thereof 
in another place of the Church, not without grudge of 
the people,' he, the Bishop, fearing tumult, "commanded 
the Vicar to set up the Communion Table (for the time) 
neare to the place where it was before." The Bishop 
evidently shows by this last answer that he wished 
to make changes cautiously, when the congregation 
were so conservative in their ideas. In November, 
1576, the body of Walter Devereux, K.G., Earl of 
Essex, a native of Carmarthen, and father of the great 
Earl, was brought here for burial, the Earl having 
died in Ireland. The funeral sermon was preached 
by Richard Davies, Bishop of St. David s, the coad- 
jutor of Salesbury in getting the New Testament 
translated into Welsh. The remains of the Earl are 
believed to be underneath the site of the organ, and 
Di)uovan, in hh Excursions Through Wodes^ in 1804, 
mentions that the Vicar, the Rev. W. Higgs Barker, 

Digitized by VjOOQ IC 

ST. Peter's church, Carmarthen. 343 

gave him the following description of the coflfin, as 
observed by the sexton when preparing a fresh grave 
near the chancel : "The outer coffin was of oak, but of 
singular construction. It was somewhat cymbiform, 

Eointed at both extremities, and strongly bbund with 
oops of iron. The inner shell was lead, in which the 
body lay, embalmed in a pecular sort of spirituous 
liquor, that had retained its purity in an astonishing 
manner, and was scarcely diminished in quantity since 
the time the body was enclosed in the coffin, being 
nearly full when first opened." Evidently the Irish 
concoction was very good for the purpose, or the remains 
could not have been so well preserved at the end of two 

Some half century after the burial of Essex, viz., on 
Sunday, September 11th, 1625, we find the stern 
Bishop Laud preaching the Assize sermon before the 
judges ; and once again, on October 9th of the same 
year, we find Laud occupying the pulpit of St. Peter's. 
On Sunday, August 10th, 1684, there was a State 
service in honour of the visit of his Grace, Henry, Duke 
of Beaufort, Lord-President of the Council in Wales. 
Whether we consider its dignity or its pageantry, it 
was probably unrivalled in the history of the church. 
One Dineley, who acted as his Grace's Secretary, gives 
many details of this visit. His Grace was accompanied, 
not only by his own retinue of noblemen, including his 
son, the Earl of Worcester, Sir John Talbot, Sir William 
Rice, and many othei-s, but by the Carmarthenshire 
militia, a great number of the gentry, as well as by the 
mayor and aldermen in their formalities. The pro- 
cession of the Corporation itself was in those days much 
more imposing than it is to-day, for it included — as a 
modern writer tells us — not only the mayor, recorder, 
aldermen, and common councillors, but bailiffs, cham- 
berlains, serjeants-at-arms, serjeants-at-mace, sword- 
bearers, beadles and constables, all in quaint costumes. 
The Bishop of St David's (Lawrence Womack) preached, 
we are told, a learned sermon, '' after which his Grace 

Digitized by VjOOQ IC 

344 ST. pbtkr'm chukgh, oarmartbek. 

and whole company were nobly entertained in town by 
the deputy-lieutenant and gentlemen of the county at 
the loagings prepared for him. After which, and even- 
ing prayer, his Grace and company took a view of the 
** key and towne, and were nobly coUationed." 

Dineley gives many details of the monuments that 
are still at St. Peter's, as well as an interesting and 
unique view of the church from the south side. In this 
view a ** bone-house/' situated to the left of the 
Spilman Street porch, is shown, as mentioned by Arch- 
deacon Thomas at the last meeting. 

From the Register we learn that Sir Richard Steele, 
the essayist, was buried in this church on Septem- 
ber 4th, 1729, his remains being placed in the vault of 
liis wife's family, the Scurlocks. This is situated in 
the Consistory Court ; and here, in July, 1876, the 
vault was accidentally laid open, and the cofl&n of Steele 
was noticed in a very decayed state. The writer 
noticed that the skull was very well preserved, and 
bore a periwig, with a bow of black ribbon tied at the 
end. It may be added that the churchwardens had a 
small leaden coffin made for the skull, and after in- 
scribing the name outside, this was placed back in the 
vault. The latter is now covered over by the tiling of 
the Consistory Court, the spot being about 8 to 10 fl. 
from the entrance porch. Of events in the eighteenth 
century there is little to record, though it is interesting 
to recall that John Wesley was present on July 13th, 
1777, and again on August 22nd, 1784, at the Sunday 
morning services, and commended the sermons. In 1797, 
many of the French prisoners en route from Fishguard 
were detained for a snort time in the church, owing to 
the want of room in the prison and town hall. In the 
last century Bishop Thirlwall nearly always occupied 
the pulpit on Christmas morning, as well as on many 
other occasions ; whilst amongst others who preached 
there were Bishop Tait (afterwards Archbishop), when 
Bishop of London, and Archbishop Benson. 

The Registers date from 1671 ; the bells from 1722, 

Digitized by VjOOQ IC 


though nearly all the latter have been reciust since, and a 
full peal provided. The communion plate is compara- 
tively modern, the exception being the cover of a chalice, 
inscribed : " Poculum Ecclesie Santi Petri Carmarthen, 
1 577." The old colours of the 23rd Royal Welsh Fusiliers 
were impressively handed over (which now adorn the 
chancel in this church) to the custody of the Mayor, for 
preservation, in the year 1846. Many Peninsular 
veterans were present on the occasion. 

Of the numerous monumental remains, a few only 
can be noticed. The oldest is undoubtedly the stone 
cofl&n-lid discovered some fifty years ago, in one of the 
recesses in the north wall of the nave. For a long 
period the latter had been concealed by the tiers of 
pews, which ran back against the wall, until the church 
was re-pewed at the period referred to. 

The inscription on the bevelled edges at the head 
and on the right side is now much worn and difficult to 
decipher. Fortunately, in Spurrells History ofCanfnar- 
then, we have the letters as noted some years ago, viz. : 


The late Mr. Spurrell attributed it to the eleventh or 
twelfth century. On the surface there is a head, in 
relief, with a floriated cross below, as shown in Miss 
Edwards' admirable sketch. Mr. Edward Lawes care- 
fully examined the lid at the last meeting of the 
Cambrian Archaeological Association ; and judging from 
the mode in which the hair is dressed inclined to the 
opinion that the figure represented a civilian of the 
latter part of the tnirteenth century, or early in the 
fourteenth century. Possibly the lid belongs to the 
coffin of some important man originally buried at the 
Priory of St. John or at the Grey Friars' Monastery, 
and removed here at the Dissolution ; while on the 
other hand the original resting-place of the coffin may 
have been the recess where it was found. 

In an adjoining recess there is a half-length stone 
effigy of a man with one arm folded across his chest. 

Digitized by VjOOQ IC 



Fig. 3.— Sepulchral Slab of Richard in St. Peter's Church, 


{Drawn by Mist Emily JET. Edward^,) 

Digitized by VjOOQ IC 



but of the history of this little is known. In the 
chancel wall, near the vestry door, there is a portion of 
an early but much-worn effigy, plastered into the wall. 
Originally, it seems to have covered a vault under or 
near the chancel arch, where the organ now is. 
The most interesting monument in St. Peter's 


Fig. 4.— Tomb of Sir Rhys ap Thomas in St. Peter's Church, 

{Excelsior Co,^ Carmarthen Photo.) 

undoubtedly the stately tomb of Sir Rhys ap Thomas, 
K.G., in the south aisle, with the recumbent effigies of 
Sir Rhys in chain armour, and his second wife, Janet, 
surmounting it. Sir Rhys was the third son of Thomas 
ap Gruffydd) of the illustrious House of Dynevor, and 
famous for his successful efforts in supporting the claim 

Digitized by VjOOQ IC 

348 ST. Peter's church, Carmarthen. 

of Henry Tudor, Earl ot fiicbmond, afterwards Henry 
VII, to the throne of England. 

Sir Rhys died in 1527, aged 76, and was buried at 
the Grey Friars' monastery of Carmarthen. On* the 
dissolution of the monastery in 1537, the remains of 
Sir Rhys and his wife were removed to St. Peter's, and 
were placed within the communion railings on the 
north side. Here they remained from 1539 to 1865. 
Unfortunately, Sir Rhys' banner, armour and helmet, 
as well as the iron railings, disappeared at the time, or 
after the removal to the church. The indenture in the 
Record Office as to the dissolution of the Grey Friars' 
monastery at Carmarthen, dated August 30, 1539, 
after stating that one " John Trahern " was warden of 
the Convent," " Makethe mencyon of all the stufis of 
the Grey Frieres of Karmardein receyved by the Lorde 
Visitor, under the Lorde Prevey Scale for the King's 
Grace, and delyvered to mv Lorde William, Bishop of 
Seinthe David and Thomas rrichar. Vicar of Carmarden, 
to se and order to the King's use with the House and 
all the appurtenances tille the King^s pleasure be further 
knoweing, and * Mr. Meyer ' to have ye oversithe of the 
same." Among the " stuflfe " referred to are ; 

Item. '* A Paule of clothe of tussey for the Erie of 
Richemunde tumbe." (This fine tomb was removed to 
St. David's Cathedral from the monastery.) 

Item. " A goodlye tumbe for Sir Rhys ap Thomas 
with a grate of yron abouthe him." 

Item. '* A stremer banner of his armys with his cote 
armer and helmit." 

In the autumn of 1865 it was deemed desirable to 
have the tomb removed to a more convenient site 
under the arch between the chancel and Consistory 
Court ; and the expense of removing and restoring it 
was borne by the fourth Baron Dynevor, a descenoapt^ 
of Sir Rhys. 

It was with much interest that one watched the 
opening of the tomb on September 11, 1SQ5, and the 
discovery of the remains of the old warrior knight. 

Digitized by VjOOQ IC 

ST. pbt£r's chuech, cabmarthen. 349 

The skull and most of the bones were found under a 
slab level with the floor of the chancel. The remains 
were carefully collected, and later on plax^ed in a stone 
sarcophagus within the tomb. The interior of the body 
of the tomb was filled up with pieces of stone, some of 
them coloured, earth, mortar, and portions of old 
tesselated pavement. 

The tomb may be said to follow after the design, on 
a much simpler and more modest scale, of that of 
Henry the Seventh's tomb at Westminster Abbey : the 
monarch whom Sir Rhys served so faithfully in his 

The inscription around the tomb was added at the 
restoration, and commences — " Here rest the remains 
of Sir Rhys ap Thomas, K.G., who fought at Bosworth 
Field, and of Dame Eva, his wife," etc. (It should be 
Dame Jknet.) A few years ago the tomb was once 
more removed to its present position, so as to provide a 
more convenient site for the organ. 

In the south wall, opposite the back of the organ 
and adjoining the monuments to Bishop Ferrar and 
Greneral Sir William Nott, as well as Sir Richard Steele's 
tablet, may be noticed a stone slab inserted in the wall, 
and bearing a carved armorial shield (see illustration, 
p. 338). Tnis was found in 1878, when repairing some 
cottages adjoining the site of St. John's Priory, the stone 
being used as a hearthstone in a cottage, but fortu- 
nately with the carved face underneath. From Spur- 
relFs History of Carmarthen^ we find that it carries "the 
armes first borne by Henry V, when Prince of Wales, 
and by other Princes of Wales, up to Edward VI . The 
shape of the shield indicates the middle of the fifteenth 
century-T-^emp. Henry VI — and the arms are those of 
his only son, Edward. The latter was born in 1453, 
seventeen years after the destruction of the Priory by 
fire. Possibly his arms were placed on a part of the 
building not restored until after his birth." 

In a niche in the chancel is a beautiful recessed 
rnonum^nt — being the eflSgy of a kneeling lady. It is 

Digitized by VjOOQ IC 


the monument of Anue, the liady Vaughan. who is 
portrayed kneeling, owing to her having been found 
dead in the attitude of prayer at her beapide. There 
is the following quaint epitaph beneath to this chari- 
table lady : 

'* Kinde Reader nnderneath thii Tomb doth Ije 
Choice Blizar of mortalitie 
By carefall proTidence Create wealth did store 
For her relations and the Poore. 
In Essex borne bat Mpent her gainfnl dajes 
In Terra Coed to her eternall prajse 
Where bj her loanea in spit of adrerse fates 
She did preserve men*8 persons and estates. 
A Create Exemplar to onr nation 
Her to imitate in Life and action 
Would joa then know who was this good woman, 
*Twas rirtnoDS Anne, the Lady Vanghan." 

** She died May 15, 1672. Being aged 84 years.'' 

It may be mentioned here that below the chancel 
and Consistory Court the church is so honeycombed 
with vaults containing the remains of old citizens and 
representatives of old Court families, that St. Peters 
may well be called the " Abbey " church of the district. 
Space will only allow of my calling attention to a few 
other monuments of interest Some of the most 
beautiful and delicate sculpture work in the church 
may be noticed on the mural tablet affixed to the wall 
behind the pulpit in memory of George Lewis, Armiger, 
who died December 21, 1715. The cherubs' heads in 
white marble are worthy of careful inspection. The 
handsome memorial pulpit of carved wood and stone was 
the gift of the family oi the late Rev. Latimer M. Jones, 
B.D., who was for fourteen years the devoted Vicar of 
the parish. The Lych-gate at the entrance is the 

Earishioners' memorial to the same Vicar ; while the fine 
rass lectern is the gift of the relatives of the late 
Mr. Valentine Davis, Registrar of the Diocese. The 
latter gentleman was a munificent contributor to the 
repair and beautifying of St. Peter's. 

The living of St. Peter's is in the gift of the Bishop, 

Digitized by VjOOQ IC 

ST. peteh's ohxjbch, cakmarthen. 351 

and is worth £282 with a vicarage. This is not a large 
income for the vicar of so large and important a churcn ; 
while it was very much less than this until the middle 
of the last century. This poverty of the vicarage may 
be directly traced to the very much one-sided settle- 
ment referred to in this article, when, for the sake of 
peace, in 1278, Richard ap-John, the then Vicar, agreed 
to take ten marks yearly from the Prior of St. John 
for his share of the tithe. 

Much might be said about the many thousands of 
parishioners who sleep — ^high and low, rich and poor — 
in the broad Grod's acre surrounding the church, but 
space will only permit me to conclude with a few lines 
by Henry Kingsley on another parish church, but which 
may well apply to St. Peter's. 

" Eight hundred years of memory are crowded into 
this dark old church, and the flood of change beats 
round the walls and shakes the doors in vain, but never 
enters. The dead stand thick together, as if to make 
a brave resistance to the moving world outside, which 
jars upon their slumber." 

Digitized by VjOOQ IC 



By thb Rkv. E. A. FISHBOURNE, M.A. 

The church of Gresford is fortunate in many ways, 
but most fortunate in that so much of its ancient glass 
remains to show us something of its former glory. The 
east window and one in the north aisle chapel are 
complete, while others have remnants of great interest. 
The great size and lofty position of this east window, 
filled with beautiful glass, seen beyond and above the 
dark oak screen, render it the most striking object in a 
church of no mean beauty. 

As it was restored by Messrs. Clayton and Bell in 
1867, the window having become quite unintelligible, 
filled with confused remnants of glass, and much of it 
also having been destroyed, it appears to me very 
desirable to put on record what was done at the time, 
hence the following notes. 

The church was practically rebuilt in the closing years 
of the fifteenth century ; the window is therefore 
in the latest style of Perpendicular, with flattened 
arch. It is 21 ft. in height by 14 ft. in breadth, and 
is of seven lights. The glass was given by Thomas 
Stanley, Earl of Derby, in the year 1500, when the art 
of glass- painting was at its best. The following is a 
description of the window as it is to-day. In the 
tracery above there is a Tree of Jesse, culminating in 
the Virgin and Child. The tone of tJiis part of the 
window is rather darker than the rest. Along the 
upper half of the lights there is a row of six single 
figures, the seventh light containing two, enclosed in 
vesicas of broad yellow rays. Above them are small 
attendant angels. In the centre light stands God the 
Father. He is triple-crowned, and holds the orb and 
sceptre. Towards the left is the sitting figure of God 
the Son. He also is triple-crowned, the lowest being 

Digitized by VjOOQ IC 


the crown of thorns. His hands are extended, and 
show the wound prints. His right foot is placed on a 
globe. Between His knees the Virgin Mother is seated 
on a lower throne, her hands upheld in front, and her 
right foot is also placed upon the globe. Further to the 
left is St. John the Evangelist, and next to him the 
Blessed Virgin, carrying on one arm the lily, and on the 
other either the gillyflower, or the palm of light which the 
angel brought to her from Paradise. On the other side 
of the central figure is God the Holy Ghost, seated 
and wearing a triple crown. Further to the right are 
the angel Gabriel and the Blessed Virgin, designating 
the Annunciation. 

Beneath these figures there are five rows of three- 
quarter figures, one hundred and eight in number — 
apostles, martyrs, virgins, angels, and seraphim. These 
rows are marked off* by the corresponding words of ** Te 
Deum laudamus," making it what is called a Te Deum 
window. Positive colour is very sparingly introduced 
into this lower division, and consequently the upper 
figures stand out with all the greater brilliancy and 
splendour. The whole conception is very fine, and calls 
forth the admiration of all experts. 

As for its history since it was erected, we know 
but little. Two windows only, on the north side of 
the church, appear to have been destroyed in conse- 
quence of the orders of 1547, for there were seventeen 
still remaining in 1574. In 1634, Anthony Lewis, of 
Burton, bequeathed ** a somme of one hundred pounds 
to mende and make clean the fayre, costly and curious 
painted glasse windowes in Gresford Church that I 
sawe was falling in decay, to be mended neatly with 
couUered glasse, where a head, arme, body, legg or 
coate of tne Personages be broken or inscripcion gone, 
to mende them art like in shape and proper couller." 

It must be remembered that this represented a 
bequest of from five to six hundred pounds at the 
present value. 

After this repair, it would seem that all windows 
were allowed to perish from sheer neglect; and pro- 
6th sbb. yoL. vn. 23 

Digitized by VjOOQ IC 


bably it was by mere good fortune that the Madocks 
chapel window lived through this time. The east 
window must have been patched anyhow with the 
pieces which fell from it from time to time, and, 
perhaps, with fragments from other windows. It was 
described in 1845 as being " entirely filled with confused 
remnants of painted glass, from the Abbey of Basing- 
werk, in Flintshire. A figure of a pope, with triple 
crown, and one of the Virgin, are perfect, also heads of 

Those now who remember it before 1867 say that it 
was impossible to make out any design, and that a 
considerable portion of the lower part — about two- 
thirds — waa completely destroyed, and replaced by 
plain glass. 

When Archdeacon Wickham undertook the renova- 
tion of the church, this window, together with the east 
window in the Madocks chapel, was sent to London ; 
and its restoration by Messrs. Clayton and Bell was so 
excellently carried out that it is difficult even for an 
expert to distinguish in every case the new from the 
old. Now for the object of these notes. 

While reasoning lately on the d priori ground that 
it was improbable a Te Deum window would be erected 
to express the All Saints idea, I became convinced that 
the words from the Te Deum were inserted by Messrs. 
Clayton and Bell, and that, consequently, all the angels 
and cherubim were also new. This was but a slender 
thread to go by, but it led to definite results. To set 
the matter at rest, I wrote to the firm for information, 
on the chance of some record having been kept. They 
most kindly gave me all the information in their power ; 
and as they had but little to go upon before them, they 
sent Mr. Clement Bell to inspect the window, and 
point out to me what was new and what was old. 
They wrote : '* We think the words from the Te Deum 
were certainly new, as we have no knowledge of a 
window being made in this country in illustration of 
the Te Deum until very long after the date of the 
original east window glass; although attempts have 

Digitized by VjOOQ IC 

efltJKOH OF All SAlNtS, GRfeSFOBD. 365 

been made to prove the contrary on the evidence of 
saints, martyrs, etc., as in the case of the sculptured 
figures on the west front of Wells Cathedral." 

It is evident, therefore, that true principles of 
•* restoration" were not thoroughly understood forty 
years ago, and the window, which originally expressed 
only the thought of the adoration of all the saints (the 
church itself being dedicated under the name of All 
Saints) was altered to express a rather wider idea, not 
quite so appropriate as the first. 

Mr. Bell informed me that the Tree of Jesse in the 
tracery was also a new idea, though some five or six 
heads, and some of the drapery here and there are 
old, which were worked into the new design. Of the 
great figures, he said, that of the Virgin on the left is 
entirely old. The St. John is almost entirely new ; 
only the feet, and a portion of the green robe near the 
feet, are old. There is nothing to show that the original 
figure was St. John, but who else could be more 
suitably placed there ? In the next light the head of 
the Virgin is new, and to this I wish to direct par- 
ticular attention. In a diagram of the window which I 
have, Messrs. Clayton and Bell call this subject (see 
pp. 352-3 for description) The Assumption ; but as the 
Virgin is seated in front of the seated figure of our 
Lord, and each has the right foot on the same globe, it 
appears to me that this subject represents the divine 
Son displaying His Mother to the universe as Queen ot 
Heaven ; and that originally the figure was a crowned 
figure, as it is in other places in the church — above the 
porch door, upon the font, on one of the misiHres, and 
in the window of the Madocks chapel. 

To strengthen this surmise, I may state that in the 
north-east window of the Madocks chapel we have 
represented there — first, the funeral of the Blessed 
Virgin, with its episode of the wicked Jew ; then her 
burial in the Vale of Jehosophat, where, in the clouds 
above, the sacred Trinity are seen taking part: the 
divine Son with His hand raised in blessing, and the 
Holy Spirit sending down His divine influence. The 

Digitized by VjOOQ IC 

356 NOTftS ON tHE fcAST WtNDOW Of TH* 

figure of the Father is now altogether missing. Next 
to this is the Assumption, where, "clothed with the 
sun," and the girdle falling, the Blessed Virgin ascends 
to heaven, surrounded by angels. After this comes the 
Coronation in heaven, now almost entirely shattered 
and confused. 

The scene, then, in the great east window carries the 
subject on to the final stage of her glory : where, seated 
enthroned as Queen of heaven and of the universe, she, 
as " the Mediatrix of Intercession," is " placed between 
Christ and the Church." [See Encyclical Letter of 
Pius IX, 1849.] 

All this fits in with the supposition I have put 
forward elsewhere : that the great object of veneration 
in this church was an image (probably a wonder-working 
image) of the Virgin, which stood in the now empty 
niche of what was then the Lady Chapel, an object of 
devotion to pilgrims, from whose " oflFrryngs . . . the 
churche of the sayd parysche was strongely and 
beautyfully made erecte and buylded." 

From this digression I return to the description of 
the present condition of the window. 

Of the central figure — God the Father — the face is 
new ; the remaining two have a good deal of new work, 
including the faces ; and about two-thirds of the lower 
figures are new, which include all the angels and 
seraphim, and all the words from the Te Deum. 

An interesting suggestion was made by Mr. Bell : 
among the angels there is one figure without a halo, 
wearing a cap and an ermine tippet. He is of opinion 
that this represents the donor of the window — Thomas, 
Earl of Derby. 

Archaeologists, I am sure, will agree with me that it 
is well thus to put on record what little is known of 
the history of this beautiful window, possibly now 
unique, in order to prevent mistakes in the future. 1 
append a diagram of the lowest division, giving the 
names of the saints ; it will show in what way it has 
been altered. 

Digitized by VjOOQ IC 



JU . . . . 

^00 QQ OQ GO 

ja 3 S a 



Digitized by VjOOQIC 


^Ircbaeologtral ^otta ann (Qiimta. 

Remains of A.ncient Bkidgb, Carnarvon. — On Angast 5th last, 
Mr. J. Issard Davies wrote informing me that the demolition of 
Messrs. Pierce and Williams' drapery stores, for the purpose of 
erecting a new Lloyd's hank, had disclosed one or two perfect 
arches not previonsly known. 

I took an early opportunity of going over to Camanron to examine 
the discovery. 

The site of the new hank is on the north side of East Gate Street.f 
between the railway catting and the town wall, and is abont 120 ft. 
distant from the East Gate. 

The arches are situated under the pavement and roadway, imme- 
diately in front of the new pt*emi8e8. The face of the ancient 
walling was exposed when the ground was being excavated for the 
purpose of obtaining foundations and building walls below the level 
of the adjoining streets. 

The accompanying illustration shows the ancient work in plan, 
and elevation, together with sections of the two arches. 

The entire face of the eastern arch, A, was visible, bat only about 
half of the western arch, B, as the remaining portion was nitnated 
in front of the adjoining premises. It will be noticed that the 
crown of the latter arch is more depressed than that of the former. 
Although probably there was an original difference between the 
heights of the two arches, I think the difference has been increased 
by subsequent compression of the western arch. Each archway 
consists 01 an inner order of massive ribs, supporting an outer order. 
When I visited the plaee three ribs only were visible in connection 
with each arch. The foreman employed on the building, however, 
told me that there was a repetition of similar ribs extending nnder 
the roadway. The sections of the two arches differ slightly. In 
both oases the outer order is chamfered on the face. All the ribs of 
arch A are of a sqnare section, while the outer rib of arch b is 
double-chamfered, and its ribs are further apart than those of 
arch A. The chamfers of the outer order of arch A were stopped 
above the springing-line. The foreman told me that the arches, or 
responds, started several feet below the level at which I saw them. 
He further added that he had to go down to a depth of about 18 ft. 
below the level shown on the drawing, to obtain a good foundation. 
He considered, for this depth, that the ground was made, though it 
is quite possible it was, in reality, a river deposit. 

The foundations of a return wall, at right angles to the bridge, 
could be traced for a distance of 15 ft. 

When Camarron was first constructed as a walled town there 

Digitized by VjOOQ IC 



were two main entrances : the one tbe East Gate — Forth Mawr, or 
Land Gate ; the other seaward, known as the West Gate, or Forth 
jr Anr. The High Street, or Stryd Fawr, connected these two 
gateways. The East Gate Street is the continuation of the High 



ANCiEivrr wSll. 



-r REAxXiNS or Xncieint bridce 


Street without the walls, and terminates in an open space, where 
several roads meet, which has been known at varioas periods as 
the Oatmeal Market, Fentice Grounds, and Turf Square. In 
Leland's Map of 1610 the town is shown, excepting for one small 
neck of land, surrounded by water. 

The River Cadnant, on the east side, passes below the road 
between the Square referred to above and the East Gate, and flows 

Digitized by VjOOQ IC 


into tbe Menai Straits. Leiand shows a bridge — possiblj intended 
for a drawbridge — connecting tha East Gate with some outworks. 
With reference to these oater defences, W. H. Jones, in Old 
Kamarvon (an undated book), tells us, on p. 103, that there were 
two bastion towers, and while rebuilding the cabinet- maker's shop 
in East Ghite Street, some years ago, the workmen came across the 
foundation of one of these towers, and the prodigious thickness of 
the walling and solidity of the work necessitated the use of gun- 
powder to remove it. 

The other tower must have been opposite this one, and un- 
doubtedly the foundations will be discovered when the houses here 
are rebuilt. The archjes lately discovered are those of a bridge on 
the land side of the outer gateway. The bed of the Cadnant, I am 
informed, is now coiiBned, underground, to a position immediately 
to the west of the site of the new bank. In Old Kamarvon, p. 86, 
we are told that the course of the Cadnant is difficult to trace, as 
it has been arched over. 

I think there can be little doubt that formerly the bed of the 
river was spread over a much larger area, and that the arches lately 
discovered carried the road over soma of its branches or swampy 
margins. In Old Karwirvoriy p. 85, we are informed that many 
years ago the river was diverted at some distance above the town, 
and that this so greatly reduced the flow that it was fouud practi- 
cable to fill in the bed of the river immediately above an old bridge 
which connected Tnrf and Castle Squares. 

29th August, 1906. Harold Hughbs. 

Annual Mbbtino op the Association. — The Annual Meeting of 
the Association for this year will take place at Llangefni, Anglesey, 
on Monday, Augnst 26th, and foar following days, under the 
presidency of Sir Richard H. Williams-Bulkeley, Bart 

Digitized by VjOOQ IC 


OCTOBER 1907. 


By Propessob ANWYL. 

The present Paper, like those of the writer which have 
preceded it, aims at giving a succinct account, on the 
basis of the scattered information hitherto recorded, 
of the condition of man in Carmarthenshire in pre- 
historic times, so that future investigators may have, at 
the commencement of their task, a bird's-eye view of 
the material already obtained. The counties of Wales 
with which the writer has already dealt in this way are 
Breconshire, Carnarvonshire, and Cardiganshire, A 
comparative study of the prehistoric antiquities of the 
Welsh counties from this point of view has the advan- 
tage of bringing into relief the salient characteristics 
of the diflFerent great epochs of early civilization, and of 
showing the effects upon that civilization of similar 
geographical and climatic conditions.' For this purpose, 
the consideration of the antiquities of the present day 
county areas is but a conventional one, and the modem 
county areas are only adopted as geographical units for 
the sake of convenience, and in order to' prevent over- 
lapping in the arrangement of the facts. In the present 
instance the consideration of prehistoric man in Car- 
marthenshire is but a small part of the larger problem 
of the life of early man along the north coast of what 
is now the Bristol Channel, and indeed of that of the 

6th skb., vol. vil 2i 

Digitized by VjOOQ IC 

Ji62 tflE EARLY SfiTtLEltS OF CAllMAlltHftt^. 

life of man in the Bristol Channel district and in South- 
West Britain generally. We might even go further, 
and say that we have here before us the wide and 
difficult problem of man in Western Europe generally, 
at a time when Britain was joined to the Continent, 
and when Europe itself was linked, by means of land- 
bridges, to the north coast of Africa. It may be said at 
the outset that in this remote period the area now 
covered by the Bristol Channel was a fertile plain, 
watered by a river which ultimately flowed into the 
sea near Cape Clear. The conditions under which man 
lived in Carmarthenshire and elsewhere at this remote 
period will be shown later. 

Before we proceed, however, to deal with the life of 
prehistoric man in Carmarthenshire, perhaps it might 
be well to explain what modern Carmarthenshire 
is. In his well-known work on Pembrokeshire, Mr. 
George Owen expresses a complaint that Carmarthen- 
shire had in his time encroached on Pembrokeshire. 
His words are : " but in all this tracte betweene the 
both shires, Carmarthenshere hath encroached upon 
Pern brokesh ere ; makeinge itselfe lardger and demin- 
nisheinge Pembrokeshere." In his Taylor^s Cushion he 
attributes the encroachment to Sir Thomas Jones, 
Knight of the Parliament for Pembrokeshire. In this 
connection, as bearing on the topography of Carmar- 
thenshire, it may noted that nearly the whole of 
Cantref Gwarthaf — one of the "seven hundreds of 
Dyfed " — and half of Emlyn, another hundred, are now 
parts of Carmarthenshire. The portion of Cantref 
Gwarthaf not included in Carmarthenshire is Efelfre, 
or Velfrey. The district of Elfed, as Mr. Egerton 
Phillimore points out, was in Cantref Gwarthaf, not in 
Emlyn. It was in this cantref, too, that the district 
of Pelunyawc (Peuliniog) — called after a Peulin (of 
Cape] Peulin), or Paulinus — had its situation. This 
district of Peulinyawc, in the Red Book of Hergest, in 
the story of Kulhwch and Olwen, is wrongly called 
Pelumyawc. It should be observed that in later times 

Digitized by VjOOQ IC 


the "cymmwd," or "commote," of the old Welsh came 
to be the lordship of the manor, while the cantref in 
turn became the mediaeval ** hundred." There is one 
district of Carmaithenahire — Derllys — which is thought 
by Principal Rhys to have a Goidelic name, the equi- 
valent of the Irish Durlas. At the present day it forms 
a modern hundred of Carmarthen. As an indication of 
the artificiality of the modern boundary on the west 
side of Carmarthen, it may be noted that the Church 
of Castell Dwyran is in Carmarthenshire ; while, on 
the other hand, the churches of Llanfallteg, Llandyssilio, 
and Llangan are in Pembrokeshire. Castell Dwyran 
and Egremont churches, again, are situated in a re- 
markable kind of peninsula of Carmarthenshire, which 
projects into Pembrokeshire. It is three miles in length, 
but has a neck whose breadth is only a quarter of a 

Our leading authority on Welsh topography, Mr. 
Egerton Phillimore (in a note in Owen's Pembrokeshire, 
p. 199), points out that, in the beginning of the twelfth 
century, the eastern portion of Dyfed was (roughly 
speaking) bounded by a line drawn from the Towy at 
Carmarthen to the Teifi at Llangeler, and including 
within Dyfed both these parishes ; and similarly those 
of Penboyr, Trellech, and Abernant. Previous to about 
750 A.D., Dyfed included Cantref Mawr ; that is to say, 
the portion east of the boundary line of modern Car- 
marthenshire north of the Towy. There is no evidence, 
Mr. Phillimore says, that in post-Roman times Dyfed 
included any part of Ceredigion (now Cardiganshire). 
We are further told that the district of Cedweli 
(Kidwelly) was a commot of the third hundred of Ystrad 
Tywi (said to have been called Cantref Eginog), and 
that it obtained its name as a tribal derivative of 
Cadwal,just as the name Arwystli comes from Arwystl. 
The name Cadwal is the Welsh equivalent of the Irish 
Cathal, from an original Catu-uallos. This district of 
Cedweli was bounded on the east by Carnwyllon and 
Iscennen ( the latter a *' cwmmwd " of Cantref Bychan). 


Digitized by VjOOQ IC 


CarnwjUon lay between Cedweli and Gwyr, or Gower, 

the latter being bounded on the west by the Llychwr. 

Llanelly, it may be noted, was included in the ancient 

Carnwyllon, the name of which is still preserved in 

those of two farms called " CarnawUon," and ** Cam- 

awllon Fach," near Pontyberem, on the Gwendraeth 

Fawr. Gwyr, or Gower, which lay between the rivers 

Llwchwr and Tawe, Brycheiniog and the sea, was in 

the old district of Ystrad Tywi, and consisted of the 

modern hundreds of Swansea and Gower (English 

Gower), and a part at least of that of Llangyfelach (or 

Welsh Gower). In this name (/yfelach, again, Professor 

Rhys detects a Goidelic survival of a name equivalent 

to the Welsh Cyfeiliog, as he also does in Tachlowmon 

for the older Telich Clowmon, near Llandeilo. The three 

commotes of Cedweli, Carnwyllon, and Gower made up 

the third "cantref" of *' Ystrad Tywi," the other can- 

trefydd being " Cantref Mawr" and •* Cantref Bychan." 

As the result of some later division, however, the lowest 

of the three commots of Cantref Bychan, that ot 

Iscennen, came to be substituted for Gower in the 

grouping. On the other hand, it may be noted that 

the Deanery of Kidwelly consisted of Cedweli and 

Carnwyllon, but did not include Iscennen. It included 

the parish of Llangyndeyrn (a daughter-church of Llan- 

dyfaelog), but not tnat of Llanddarog, which was in 

the Deanery of Ystrad Tywi (later on known as that of 

Llandeilo and Llangadock), and therefore in Cantref 

Bychan (Owen's Pemhrokeshirey p. 206). Another name, 

Talacharn, now the Welsh designation of Laughame, 

was originally, as Mr. Phillimore points out, in all 

|)robability that of a district, the old name of the site 

of the castle or town of Laugharne being Aber Coran. 

This old commot of Talacharn doubtless lay between 

the estuary of the Taf and the eastern boundaiy of the 

hundred of Pen fro, which ran from EglwysFair, on the 

Taf, to the coast at Amroth Castle. Its northern 

boundary was probably the Taf, so that it would thus 

include Llanddowror, but not St. Clears. The name 


Digitized by VjOOQ IC 


Talacharn is probably made up of " tal" — a forehead or 
end — and "acham" — an intensive of "earn," a cairn, 
just as " achas" is the intensive of " cas." 

Another point of topography that should be noted is 
that the lordship of Llanstephan is approximately iden- 
tical with the old Welsh commot of Penrhyn, bracketed 
with that of Derllys, which bounded it on the north. 
Both of these districts formed part of Cantref Gwarthaf. 
It may be further mentioned, too, that Llanddowror had 
at one time a double name, Llandeilo Llanddyfrwyr, or, 
as it is spelt in the Book of Llan Dav^ Lanndubrguir. 
From these considerations, it will be seen that modern 
Carmarthenshire, which is treated as a unit for the pur- 
pose of the present Paper, consists substantially of the 
ancient Welsh division of Ystrad Tywi, but with two 
exceptions that are of importance : 1st, that it does not 
include the commot of Gower ; and 2ndly, that it 
includes, with the exception of the little district of 
Velfrey, and possibly the district of Peuliniog (not yet 
identified), the whole of Cantref Gwarthaf, the largest 
of the seven hundreds of Dyfed. The whole of the 
two chief cantrefs of Ystrad Tywi, viz., Cantref Ma wr, 
north of the Tywi, and Cantref Bychan (south of it), 
are in Carmarthenshire. 

With regard to the references already made to pos- 
sible traces of a Goidelic population, in addition to the 
Ogham inscriptions, Professor Rhys mentions not only 
Derllys (Durlas), but also Llethrach (from Leitir), 
identified by him with the Irish Leitrioch or Leatracha 
Odhrdin (of St. Oran), now Latteragh, near Nenagh, 
county Tipperary. Of other possible Goidelic traces, 
whether of survivals from the Bronze Age, or of later 
settlers in Roman or post-Roman times, it may be noted 
that there is a Llwyn Gwyddel in Lampeter Velfrey, 
and a Pant y Gwyddel in Llanfyrnach. Cerfciinly, 
from the evidence of the Ogham inscriptions and of 
Nennius, Goidelic settlements, whatever may have been 
their origin, existed in some districts in post-Roman 
times. Nennius (Hist Brit, Section 14) says : " Filii 

Digitized by VjOOQ IC 


autem Liethan obtinuerunt in regione Demetonim et 
in aliis regionibus, id est, Guir et Cetgueli, donee 
expulsi sunt a Cuneda et a filiis ejus ab omnibus Brit- 
tanicis regionibus." The situation of these districts 
seem to suggest, as in the case of the D^si, a settlement 
from the sea. 

After this preliminary topographical statement, we 
come now to the subject of the present Paper, namely, 
the life of man in Carmarthenshire in prehistoric times. 
It is a commonplace of anthropology that the forms 
which man's development took were largely determined 
by geological, geographical, climatic, and economic con- 
siderations. In some parts of the area, such as at 
Coygan, near Laugharne, we have most valuable re- 
mains of animal life, which go back to as remote a time 
as any similar remains in Europe. In other portions of 
the area, the earlier vestiges have disappeared, and the 
presence of early man is a matter of indirect inference. 
As a rule, it is only when his most convenient materials 
were of stone that traces, of him are still distinctly 
visible. When his shelter, where he had it, consisted 
of the trees of the forest, nothing now remains to 
reveal his former presence with any degree of cer- 
tainty. Of late, however, special attention has been 
called to the survival of prehistoric hearths ; and the 
search for these, to which reference will be made later, 
has opened up a new and fruitful field for inves- 

The distribution of early man, like that of man in all 
ages, was conditioned by economic considerations, and the 
governing considerations were the accessibility of food 
and water. The geologist, the zoologist, and the botanist 
could, from a joint survey of a given district, give a very 
shrewd guess as to the places where early man would 
be likely to cluster his communities together, and those 
for whose possession the stress of competition would be 
greatest. In the remotest times mining formed no factor 
in the distribution of the population, but the " Gogofau" 
of Dolau Cothi show that there came a time (when, is 

Digitized by VjOOQ IC 


very uncertain) when mining (not improbably for gold) 
had its share in the economical development of the 
county, even in the dim period of antiquity. 

Though some of the later problems of the population 
of Carmarthenshire in early times have, for certain 
reasons, already been anticipated, yet it will be con- 
venient in the remainder of the Paper to trace in order 
the great stages of prehistoric civilization, and to see 
what relics they have here left behind. In the case of 
the Coygan cave we have traces of the conditions of 
life under which lived the cave man of Palaeolithic 
times, much as we find them in the caves of the Vale 
of Clwyd and of other districts where they afforded 
shelter to man and beast ; while in the " kitchen- 
middens" near Pendine we have the remains of early 
man's diet of shell- fish. The caves of the Bristol 
Channel area continued in use during Neolithic times ; 
and consideration will be given to these and to the 
conditions which they reveal in connection with that 
period. In the case of stone implements it is not 
always easy to be sure of the period to which they 
belonff, because the introduction of bronze, and even of 
iron, aid not mean that the use of stone was abandoned 
for implement-making, especially in the districts which 
were least accessible to commerce, or were least econo- 
mically flourishing. This caution should, consequently, 
be borne in mind in considering the following records 
of the finds of stone implements and other relics ap- 
parently of the Stone Age. In Arch. Camb. for 1851 
(p. 334), there is an account of a stone celt found on 
the Henllan demesne, and exhibited at the Tenby 
Meeting of the Association by J. Lewis, Esq., of Henllan. 
A stone hatchet was also found in a rab-quarry, em- 
bedded in the rab at Llan, in the parish of Llanfallteg. 
In Arch. Camb. for 1853 (p. 262) there is an account 
of " Y Gam Goch '' (a fortress certainly much later than 
the Stone Age), by Mr. John Williams, of 127, King's 
Road, Brighton, wherein mention is made of certain 
triliths on a small scale, said to be still visible there. 

Digitized by VjOOQ IC 


It is not impossible from their cromlech-like character 
that they may prove to be of the Stone Age, and older 
than the fortification itself. In Arch. Camb. for 1856, 
p. 103, there is an account of a cromlech known as 
GwS,l y Filiast, Carmarthenshire; and, on p. 107 of 
the same volume there is mentioned a circle of 8tx>ues 
called Buarth Arthur, as well as another called Meini 
Gwyr. It is of interest to note that the writer who 
makes reference to them is the late Mr. T. Stephens, of 
Merthyr, the distinguished author of The Literature of 
the Kyminf. This cromlech, like others, probably has 
come down from the Stone Age, but stone circles aro 
usually the relics, not of the Stone but of the 
Bronze Age, and generally surround the tumuli of that 
period. Again, in Arch. Camb. for 1858, p. 371, there 
is a Paper by the late Mr. T. O. Morgan, of Aber- 
ystwyth, on a series of cairns on Craig Cwm Twrch, 
which are designated on the Ordnance Survey as Cam 
Carnau, Carn, Cam Fawr, and Cam Fach respectively. 
On the line of these cairns, Mr. Morgan says that there 
is an immense stone called Maen Prenvol, or Penfoel, 
near Lluest y Bwlch and Esgair Ddu on Waun Cellan, 
which appears to have been the capstone of a cromlech, 
but to have fallen from its original position. It is 16 ft. 
in length, and 24 ft. in circumference, and lies upon a 
moated tumulus of earth. About two yards from it 
was a walled erection and some scattered stones. Mr. 
Morgan thought that the whole once formed a crom- 
lech. The existence, however, of the moated tumulus 
of earth suggests that further investigation is needed 
before this view is adopted. In Arch. Camb. for the 
same year, p. 371, mention is made of a monolith called 
Hirfaen Gwyddog, which stands 16 ft. above ground ; 
but we are under no necessity of assigning this to the 
Age of Stone. In Arch. Camb. for 1864, in the account 
of the temporary Museum at Haverfordwest, reference 
is made to a stone celt from Llethr, in Brawdy parish ; 
but as there is another from a tumulus near Llanrhian 
(which tumulus is most probably from the Bronze Age), 

Digitized by VjOOQ IC 


it IB not impossible that both really belong to that 
period. In Arch. Camb. for 1875, p. 415, we have an 
account of a stone celt found on Caerau Gaer, in the 
parish of Llanddewi Velfrey ; and again in Arch. Camb. 
for 1876, p. 236, there is an account of a very curious 
cromlech at Ffynnon Newydd, in the parish of Llan- 
gunnog. The three supporting stones are said to form 
a parallelogram-shaped chamber, open on the west, while 
the capstone leans on the northern support, with one 
end resting on the ground. This is called **Twlc y 
Viliast." A few yards to the east, there is, we are 
told, a semicircular rock known as " Bord Arthur." In 
Arch. Camb. for 1877, p. 81, the late Rev. E. L. Barn- 
well has a Paper on " Early Remains in Carmarthen- 
shire," dealing mainly with " Y Clawdd Mawr" in 
Cynwyl Elfed. Of this the writer says : " The object 
was evidently that of defence from attack from the 
opposite heights, or to command the road in the valley 
below ; and neither of these motives could have acted 
on a leader whose great end was to get over the ground 
as soon as he could." Mr. Barnwell thought that the work 
" was probably connected with the adjoining Megalithic 
remains, formerly of a much more extensive and impor- 
tant character than they are at present." Here, again, 
it would be extremely rash to assign these remains to 
Neolithic times. In view of their elaborate character, 
it is impossible not to suspect that they are of a much 
later period than the Stone Age ; and it would not be 
surprising if they proved to be Late Celtic, like Y Garn 
Goch and Tre'r Ceiri ; or they may be even later, as 
Mr. Phillimore suggests, in Owen's Pembrokeshire. 
Owing to this uncertainty about the period to which 
they belong, judgment should be suspended until a 
fiirther investigation of them is made. Mr. Barnwell 
thought that a certain group of stones formed a crom- 
lech, but he remarks : ** It is very rare to find the 
actual supporters of a capstone more than four." Mr. 
Barnwell held that the chambers of Clawdd Mawr were 
once covered up. 

Digitized by VjOOQ IC 


In the article in question we are further told that, 
on the way to Ystrad, on the left-hand of the road 
leading to Carmarthen, are four stones, one of which is 
smaller than the others. The stone to the right is of 
coarse grit ; the small one and the stone next to it are 
of quartz conglomerate, the largest one being of old red 
sandstone. The three largest ones formed the walls of 
a chamber, and may have aided in supporting the cap- 
stone. Their denudation is complete ; nor is there the 
slightest vestige of the former mound. 

These are Mr. Barnwell's words: — "Within the 
grounds of Ystrad are one or two pillar-stones, one of 
which was said to have been Roman, but is an ordinary 
menhir. They are not remarkable as regards dimen- 
sions. No other remains exist near them. They may 
perhaps have been ancient boundary stones, but are 
more likely to be ordinary meini hirion." On the left- 
hand of the road from Llanboidy Church to Dolwilym 
is a more important group (Fig. 5), concealed by a high 
and thick hedge from the road. The stones lie in a 
field called "Parcy Bigwrn," a portion of Pensarn Farm. 
The original chamber is easily made out, though only 
two of its stones remain erect. The fallen ones, with 
the exception of one, have not been removed, so that 
their original position, when upright, is easily ascer- 
tained. The stones average about 7 ft. high above the 
ground, with an average thickness of 3 ft. ; the longest, 
that lying apart under the hedge, measuring more than 
8 ft. The chamber was nearly perfect within human 
memory, and seems to have been broken up about sixty 
years ago. It had no doubt been deprived of its cover- 
ing of earth or stone ages before, as our informant never 
saw any indications of such a mound, although the 
cromlecn or chamber was perfect in his early days. 
This man, John Jones, of 80 years of age, a man of good 
character, had lived close to the spot all his days. His 
memory was remarkably clear, and his veracity never 
suspected. He does not remember the covering stone 
in its original horizontal position, for at the time he 

Digitized by VjOOQ IC 


speaks of, it had been tipped over and shifted from its 
western bearer, one end resting on the ground. He 
had, however, often been told by his seniors that it was 
once horizontal, and known as " The Table" — a term 
that proved its former position. Six horses and ten 
men were required to draw the stone. From all indi- 
cations the chamber in question was a cromlech, and as 
such may well have belonged to the Stone Age. The 
cromlech may be regarded as a kind of artificial cave in 
which burial generally of a number of bodies took place, 
as it did in the natural caves. Even natural caves in 
France have been found to contain burials of thePolished 
Stone or Neolithic period, and the traces of funeral rites 
found in them were identical with those found in arti- 
ficial chambers. In some instances the chambers are 
only partly natural. Sometimes, in France, they have 
been simply excavated out of the rocky ground to a 
certain depth, and covered up with a large stone slab. 

Mr. Barnwell quotes some observations of M. Bert- 
rand on the cave-burial of Belport, in France, discovered 
in 1876, when some quarry men laid bare a cavern ; and 
also of M. Duport on the famous cave called " Le Trou 
de Frontal,*' found at Furfooz, in Belgium, and described 
in that writer's work called VHomme Pendant les Ages 
de la Pierre (p. 195, Second Edition). These were 
burials in a place of shelter — or recess — rather than in 
a cave. With the type of chambered cromlech in 
question, Mr. Barnwell compares the Henblas cromlech 
of Anglesey. As for the distribution of cromlechs in 
Europe, the best statement is that of Sir John Evans, 
at the Stockholm International Meeting, when he said 
that " their distribution depended on the distribution 
of their materials." To the foregoing may be added 
the Llwyndu cromlech, that has lost its capstone, 
which is near the road from Carmarthen to Llan- 

In the Arch. Camh. for 1878, p. 321, there is a 
reference to the Lampeter Meeting, at which Miss 
Johnes, of Dolau Cothi, exhibited some objects found 

Digitized by VjOOQ IC 


in her neighbourhood, such as a stone celt and some 
spindle- whorls, which may possibly have belonged to 
Neolithic times. In the Journal of the Society for 
1879 (p. 55), there is an interesting article on "Pre- 
historic and Other Remains in Cynwil Gaio," by the 
Ven. Archdeacon Thomas, M.A. Mr. Worthington 
Smith and Archdeacon Thomas had availed themselves 
of an invitation of the Rev. Charles Chidlow to go to 
spend a few days at Caio, for the purpose of explorin^f 
some curious remains on Craig Twrch, and some cists and 
barrows on the hills of Mallaen. None of these appear, 
however, to belong, with any certainty, to the Stone 
Age ; but some may belong to the Bronze Period. In 
the Arch. Camh. for 1884, the Rev. E. L. Barnwell has 
given an account among the cromlechs of South Wales 
of Longhouse (p. 141), Llanwnda(p. 137), and Dol wily m, 
near Whitland, and of the latter a picture is given. 

Valuable light is thrown on the conditions of life of 
prehistoric man in Carmarthenshire, notably on the sea- 
border, by the investigations of Mr. Edward Laws, 
Professor Boyd Dawkins, and others, into the caves of 
South Pembrokeshire, and the adjoining caves of Car- 
marthenshire. Mr. Laws has embodied the results of 
his researches mainly in his well-known volume on the 
History of Little England beyond Wales. In this he 
deals with the bone-caves of Hoyle's .Mouth, Caldy 
Island, Coygan, etc. ; and at this stage it would be well 
to pause for a moment to consider, in the light of his 
discoveries, what were the conditions of life on the north 
shore of the Bristol Channel in the remotest times. 
This account is well given in the words of Professor 
Boyd Dawkins, who says that the islands and cliflEs of 
South Wales were hills overlooking a vast fertile plain, 
occupying what is now the Bristol Channel, where 
ample sustenance would be found to feed the herds of 
elephants, horses, and reindeer. The Towy and similar 
valleys would form adjuncts of this ancient plain, and, 
so to speak, extensions of it. Mr. Laws gives it as his 
opinion that Hoyle s Mouth was inhabited by man in 

Digitized by VjOOQ IC 


Palaeolithic times. At the Coygan cave he found an 
awl and two flint flakes in tne undisturbed earth 
beneath the stalagmite, associated with the bones of 
the rhinoceros, ana therefoi'e of the Palaeolithic Age. 
Professor Boyd Dawkins also suggests that the same 
may have been the case with the Gower caves. The 
latter, in his work on Cave limiting, says : — ** We must 
therefore picture to ourselves a fertile plain occupying 
the whole of the Bristol Channel, and supporting herds 
of reindeer, horses, bisons, many elephants and rhino- 
ceroses, and now and then being traversed by a stray 
hippopotamus, which would afford prey to the lions, 
bears, and hyaenas inhabiting the accessible caves, as 
well as to their great enemy and destroyer man. It 
appears, too, that prehistoric remains are occasionally 
dredged up from Carmarthen Bay. A large river prob- 
ably flowed into the sea past Lands End.'' On 
p. 6 of his book, Mr. Laws says : — *' We never came 
across human bones or human handiwork in the Hoyle 
Cave that were attributable to Palaeolithic Man." In 
a letter to himself from Professor Boyd Dawkins, which 
Mr. Laws quotes, he says : — " I never dug out any flint 
or horn-stone implements with my own hands in asso- 
ciation with Pleistocene beasts in this cave. I believe, 
however, that Mr. Ayshford Sandford found them in 
association with bear, under the stalagmite and near 
the entrance, on the right-hand side, along with frag- 
ments of charcoal and splinters of bone : these I con- 
sider Pleistocene. Hoyle's Mouth seems to me to have 
been used by hyaenas during the old Stone Age. In 
Neolithic times it became both a dwelling and a ceme- 
tery for men." Similarly of the Little Hoyle : " In 
Pleistocene days this was a hyaena den." A rich 
Neolithic harvest was here found by Mr. Laws and his 

Of the caves of this series Mr. Laws says : — ** The 
most interesting ossiferous cave in West Wales is the 
Coygan, near Laugharne, in Carmarthenshire ; and, as 
this comes within the zone of the modern county, an 

Digitized by VjOOQ IC 


account of it, based on the investigations of Mr. Edward 
Laws, may be given here : * It is excavated from an 
outlying hill of mountain limestone, which stands about 
a mile from the sea, flat marsh land and sand burrows 
intervening. There can be little doubt that in com- 

Earatively recent days the sea washed the foot of this 
ill. The entrance to the Coygan is extremely low and 
narrow, but soon opens out into a lofty and extensive 
chamber.' " So far as Mr. Laws knew, there had been 
no discovery of Neolithic remains in this cave (but this 
is probably accidental). It was deemed by the late 
Professor Rolleston to be the most perfect instance of 
a hyaena den he had ever met with. Mr. Laws found 
hyaena bones in position. The other remains were 
similar to those found in Black Rock and Caldy, but 
were more plentiful, in good condition, and much scored 
by teeth-marks. Mr. Laws further says : — "In addition 
to these ordinary cave-bones, I had the good fortune 
to find under rhinoceros bones which were overlaid by 
stalagmite, a piece of bone, whittled and rounded into 
the shape of an awl, lying alongside of two flint flakes: 
one of which had indubitably been manipulated ; the 
other was a pebble, which had been broten, whether 
by natural or artificial means it is impossible to say. 
These are in the Tenby Museum, and constitute the 
sole proof of Pleistocene Man in West Wales discovered 
by me." The Pleistocene fauna appear to have been of 
three classes : Northern, Temperate and Southern ; but 
the curious state of things is, that as these remains are 
found in the closest association together in the caves, 
there can be no doubt that they ranged the land to- 
gether. As it is important for the purpose of picturing 
the life of man in the remotest times in Carmarthen- 
shire to know with what animals he lived, some of these 
may be enumerated. We have, first, the Northern fauna, 
the first of which is the mammoth or elephas primi- 
geniu^, which fed on the woody fibre of trees, for 
example the larch. Mr. Laws points out that the 
Pembrokeshire mammoths of the caves were mostly 

Digitized by VjOOQ IC 

tHE EAitL^ SEWLerS O^ CAtlMARTbEN. 3^5 

calves. Next we have the woolly rhinoceros {R. Ticho- 
rhinus), whose nostrils were divided by a long ridge. 
The British rhinoceros had a thick woolly coat composed 
of short hair of a cinereous grey colour, from 1 in. to 
3 ins. long, with here and there a black hair longer and 
stiffer. The rhinoceros fed, like the mammoth, on the 
twigs of the larch and other trees. It can scarcely be 
doubted that both these beasts roamed at will at one 
time, right into the Towy Valley. Other Northern 
beasts whose remains have been found in the West 
Wales caves are the reindeer and the elk ; but in 
English caves there have been also found remains of the 
musk-ox, the lemming, the tailless hare, the marmot, 
and the Arctic fox. Of the beasts of the Temperate 
group, the following have been found : the wolf, the 
fox, the cave and the brown bear, the horse, the ox, 
the bison, the Irish elk, and the red deer. All these, 
except the cave-bear, have survived from the prehistoric 
period. The cave-bear had some points of contact with 
the polar bear, though generally he is considered the 
prototype of the American grizzly. The animals of the 
Southern group, which roamed in the area of the Bristol 
Channel, were the lion, the hyaena, and the hippopota- 
mus. The cave-hyaena was of a heavier type than that 
of South Africa. At that time Britain was joined to 
the Continent of Europe, and migration was constant. 
The junction of Britain to Europe probably made a 
great difierence to the climate ; and further, some of 
the animals in question may have been able to adapt 
themselves, as man has done, to the zone in which 
they chanced to live. The coast of the Bristol Channel 
is surrounded by a belt of submerged land. It is not 
improbable that the first settlers of Carmarthenshire 
formed the northern fringe of the men of the Bristol 
Channel area, who penetrated into the adjoining val- 
leys. Though there are no remains from Carmarthen- 
shire itself to illustrate the life of these men, the re- 
mains of the adjoining Pembrokeshire caves afford 
abundant indications of its character. From the Little 

Digitized by VjOOQ IC 


Hoyle in Longbury Bank, Penally, the following re- 
mains were unearthed by Mr. Laws, Mr. Wilmot Power, 
the late Professor Rolleston, and the late General 
Pitt-Rivera, in 1876, 1877, and 1878 : (1) The remains 
of certainly nine, if not eleven, human beings ; (2) large 
quantities of the bones of domestic and wild animals ; 
(3) birds ; (4) shells ; (5) pottery ; (6) charcoal ; (7) 
stone and bone implements. These were mixed up with 
black earth and angular stones in a sort of hotch-potch. 
The precise explanation of this hotch-potch is doubtful ; 
but as for the crania themselves. Professor Rolleston 
said that they were dolichocephalic, with a remarkably 
low cephalic index of 69, and with a pear-shaped contour 
when viewed from above, due to a rapid tapering from 
the level of the parietal tubera forwards. Among the 
natives of Carmarthenshire measured by some of my 
colleagues at Aberystwyth and by myself, I remember 
none that was found with so remarkably low a cranial 
index as 69* 

The picture of the life of Neolithic Man in the 
northern side of the Bristol Channel is best completed 
by the following account given by Mr. Edward Laws, 
who has studied the conditions of his life with the 
closest attention. In the History of Little England 
Beyond Wales, Mr. Laws says, speaking of Eiirly Man's 
weapons : '* The projectile weapons were javelins, ar- 
rows tipped with flint or bone, and slings ; their side- 
arms, polished stone celts, some heavy and some light, 
set in wooden handles. Their clothing consisted, no 
doubt, partly of cloth, for a carding-comb found in 
Hoyle's Mouth and the stone spindle-whorl from Stack- 
pole proves that they were weavers. Still, the nu- 
merous flint scrapers show that the preparation of hides 
was a very important business ; while the bone needle 
found in the Little Hoyle is well adapted to sew skins 
together. Poundera and mullers of com for rubbing 
corn into meal are found." They probably had wheat, 
barley, oats, and rye. The cultivation of these was 
probably women's work. These men of the Neolithic 

Digitized by VjOOQIC 


period turned out strong, serviceable ware. In the 
Little Hoyle Mr. Laws found shards made of old red 
sandstone, ground fine and mixed with clay. This ware 
WAS not turned on a wheel, but fashioned by the hand. 
They were rich in oxen, sheep, goats and swine, all, 
however, of a small breed. The horses of these men 
were comparatively scarce, but they had fine dogs, and 
one from the Little Hoyle was as large as a mastiflf. 
These hounds hunted the brown bear, the red deer, the 
roebuck, hares, and foxes. The wild boar seems to have 
been scarce, and wolves and beavers are conspicuous by 
their absence. The woods were inhabited by black 
game, and the hill-sides with partridges. The same 
learned archaeologist remarks that oysters, cockles, 
periwinkles, whelks, pectens and the like were used in 
great numbers ; also an occasional crab, but he found 
no remains of lobster. The fish was probably collected 
by women and children. Some of the cave-men took 
fish, e,g.^ the conger-eel, ray, and angler fish. A dug- 
out canoe (either of the Neolithic or the Bronze Age) was 
also found close to the Hoyle's mouth. 

Coming now to further remains that may be Neo- 
lithic, we may note the following. In Arch. Camh. for 
1893 (p. 157) there is an account of the exhibition of 
a nether millstone found by Mr. Stepney-Gulston on 
Carreg Sawdde, near Llangadock, in 1871. Further, on 
p. 163, there is a Paper on the Craig Derwyddon Bone 
Caves (near Pant-y-llyn, Llandybie, Carmarthenshire), 
read upon the spot on August 11th, 1892, by Mr. Alan 
Stepney-Gulston, of Derwydd, to whom archaeology in 
Carmarthenshire owes a deep debt of gratitude. This 
Paper deserves attention. In 1878 Professor Rolleston 
visited these caves, and collected all the information 
then available. Mr. Stepney-Gulston quotes Professor 
Rolleston as saying : ** Many years ago — in fact, in the 
month of August, 1813 — ten or eleven skeletons were 
found in a cave near Llandybie. One skull from the 
find we have in the Oxford University Museum. It is 
filled with crystalline loaf-sugar-like stalagmite, which 

6th bkb,, vol, Til. i^ 

Digitized by VjOOQ IC 


has, of course, preserved it in its original outlines . . . 
This skull was carried oflF by the Lord Dynevor of the 
time being, and by him was transferred to the hands of 
Dean Buckland, and so into our Museum . . . the rest 
of the human bones, together with the bones of elk and 
wild boar, were re-interred in a pit dug for their re- 
ception close by. The site of this pit I hope to identify." 
Mr. Alan Stepney-Gulston had, however, by careful 
investigation, been able to explore and excavate the 
ancient sepulchre of the place. His words were as 
follows: **This piece of the 'living rock' which you 
see here still standing was, it seems, left as a mark of 
the whereabouts of the actual site. The vault itself, 
lying to the north side, was entered by a lateral 
opening, some 30 ft. in length, which had become so 
entirely blocked up (whether through the silting-up 
action of time, or perhaps through the direct action of 
those who chose this solemn retreat as a sepulchre, 
must remain a matter for speculation), that the work- 
men were not aware of even the existence of the cave 
until they broke into it from above in the ordinary 
course of their workings." The part of the cavern which 
was used for sepulture, and which was entered from 
the north side, seems to have measured from 16 ft. to 
17 ft. in length by 12 ft. to 13 ft. in width, and was of 
an ovate form, the irregular vaultings of the roof 
averaging about 4 ft. high in the centre. 

** There were twelve skeletons in all, the first seven 
lying with their feet towards the entrance, and their 
heads towards the west. In juxtaposition were three 
other skeletons, placed transversely, with their heads 
lying towards the south ; and lastly, at a point about 
10 yards further into that part of the cave that 
extended towards the south, were two other skeletons 
of great size, lying also with their heads towards the 
south. It is remarkable that they all lay with their 
faces turned upward, and with their heads brought 
slightly forward on to their breasts, the skulls in every 
case resting on a solid ledge of rock, some 6 ins. higher 

Digitized by VjOOQ IC 


than the level plane upon which the rest of the skeleton 
lay, and the arms extending flat down each side of the 
body, which was laid straight out and face upwards, 
the whole of the floor being covered by what is described 
as fine sand, one of the skeletons only having been 
subjected to the incrustation of the stalagmite referred 
to by Professor Rolleston." Mr. Stepney-Gulston then 
proceeds to give an account of what had happened to 
these " precious prehistoric relics.'* To the skull de- 
posited in the Oxford Museum reference has already 
been made. Several others of the skulls were taken 
away by a gentleman of the name of Wrey, then living 
at a place called Thornhill, some four miles distant ; 
which place was sold in 1880 by a Miss Fosset, when all 
traces of the skulls were lost sight of, a huge stalag- 
tite only being still to be seen as an ornament upon the 
lawn there. 

3. A portion of the bones, together with the stalag- 
mite found there, were burned in an old lime-kiln, 
which pre-existed on the site of the present kiln, now 
marked with a stone, showing the date of 1823: namely, 
ten years later than the find. 

4. The whole of the remainder of the bones, Mr. 
Stepney-Gulston was told, both of the human remains 
and also of the elk-horns and teeth of the wild boar, 
which latter were only found among the dihris which 
stopped up the mouth of the cave, were thrown away, 
and were gradually covered up by the *' talus'* or " spoil" 
from the quarry. Mr. Stepney-Gulston also says : — 
"I have also been informed that certain 'copper' 
utensils were known to have been found together with 
the skeletons ; and I have great hopes, should this 
prove to be a fact, that one or more of them may yet 
be recoverable. If bronze implements were found, it is 
probable that we may have here some unburnt burials 
of the Bronze Age. All the skulls are described as 
being exceedingly large, and there exists a tradition 
that, at the time of their exhumation, the hat of the 
largest-beaded bystander proved, upon trial, to be too 

25 » 

Digitized by VjOOQ IC 


small for the smallest of the ancient skulls.'' In a 
memorandum of September 7th, 1892, Mr. Stepney- 
Gulston states that by inquiry from an old quarryman, 
who was present at the find, the spot where a number 
of the bones had been re-interred had been identified. 
It would not be at all surprising if this proved to be 
a Late-Bronze Age or Late-Celtic burial. Mr. Stepney- 
Gulston has rightly called attention also to the im- 
portance of a thorough exploration of the Carreg Cennen 

In the Arch. Camb. for 1894 (p. 78), there is an 
account of the identification by Horatio Thomas (twelve 
years of age), son of Mr. Cerridfyn Thomas, B.Sc, of a 
finely-shaped, large-sized, and well-preserved celt, so 
smooth that it might almost be called polished ; 10 ins. 
long, 8 ins. round in the thickest part and 3j ins. along 
the knife-edge end. It was made of grey granite, in 
which some specks of felspar and mica were visible. It 
was found by the boy's uncle, Mr. John Morris, 
of " Rwyddfa Gatw" farm, in the pariah of Llanegwad, 
Carmarthenshire, in the first week of October, 1893, 
while extending a pond into the peaty soil adjoining. 

The most important contribution recently published 
connected with prehistoric Carmarthenshire is that 
published in the Arch. Camb. for this year, entitled 
"Note on the Discovery of Prehistoric Hearths in 
South Wales," by Mr. T. C. Cantrill, B.Sc, and Mr. 
O. T. Jones, B.Sc, B.A. ; the latter is, I am glad to say, 
a distinguished old student of the University College of 
Wales, Aberystwyth, and of the University of Cam- 
bridge. To some researches by the former I had the 
pleasure of referring in my Paper on '* The Early Settlers 
of Brecon. *' These hearths appear to have been prehis- 
toric cooking-places. I have had the good fortune to 
identify a new one of the kind in Cardiganshire, and on 
inquiry to find that there are more, which I hope at some 
time to describe. The general description of them is 
as follows : — These hearths consist of small heaps of 
broken and burnt stones, generally near streams, espe^ 

Digitized by VjOOQ IC 


cially where these arise from a strong spring close by. 
Occjisionally, the stream is found to have cut into one of 
the banks in such a way as to expose a complete section 
of the heap, which is seen to consist of a muss of stones 
— generally pieces of sandstone or grit — broken to the 
size of road-stone and evidently burnt, inasmuch as 
they were friable and reddened by heat. The inter- 
stices between the stones were found to be iSlled with 
fine soil, in which charcoal-dust and fragments were 
abundant ; the heap, of course, being covered with grow- 
ing turf. The mounds in question were associated with 
supplies of good drinking-water, and especially with 
springs. They have been found in Carmarthenshire, in 
the following places. Several have been found in the 
parish of Gwynfe, and seem to indicate the presence in 
that district of a flourishing prehistoric community. 
The following is a brief summary of the spots in Car- 
marthenshire where these sites have been found. For 
a fuller account, see Mr. Cantrill and Mr. Jones's 

Prehistoric Hearths in Carmarthenshire. 

1. In Carmarthenshire, E. bank of lane, 100 yards S.E. of 
Ty-brych Farm, IJ mile S.W. of Llanddeusant, near Llan- 

2. 400 yards W. of Llan Farm, 2 miles E. of Gwynfe, uear 

3. K side of Nant-dwfn, at PwU-y-fuwch Farm, 1 mile S.S.W. 
of Capel Gwynfe, Llangadock. 

4. S.E. side of stream, 400 yards N.E. of Pare Owen farm, 
2^ miles N.W. of Capel Gwynfe, Llangadock. 

5. Bank of small stream at foot of Cylchau, and 550 yards 
K by S. of Llwyn-y-Wennol Farm, 2 miles E.S.E. of Capel 
Gwynfe, Llangadock. 

6. Side of small stream, 250 yards N. of Llygad Llwchwr, near 
Forge Llandyfan, 4 miles S.E. of Llandeilo. A small flint flake 
was found a yard or so away. 

7. Side of same stream as N"o. 9, and 70 yards farther up the 
stream, 300 yards E.N.E. of Llygad Llwchwr. 

8 and 9. 450 yards N.W. of Llygad Llwchwr. 
10. S.E. side of small pond, in middle of the upper camp on 
Gram-G6ch, Llangadock. 

Digitized by VjOOQ IC 


11. Side of stream, 400 yards S. by, W. of Cwm-fTrwd Farm, 
three-quarters of a mile N. of Glanamman Kailway Station, 
Amman Valley. 

12. Edge of pond, 50 yards W. of Gelli-fawnen Farm, 1 mile 
W. by N. of Glanamman Railway Station. 

13. Side of path, 200 yards N.W. of Hafod Farm. Lower 
Clydach Valley, 4 miles N.W. of Pontardawe. 

14 and 15. S.W. side of stream, 300 yards S.S.W. of Tresgyrch- 
fach Farm, 3| miles N.W. of Pontardawe. 

16. Side of stream, 450 yards S. by E. of Tregib House, 

17. W. slopes of Cennen Valley, S. of Meusydd Mill, 2 J miles 
S.W. of Llandeilo. 

18. E. side of stream, 350 yards W. of Penrhiw, J of a mile 
E. of Derwydd Koad Station, 3 miles S. of Llandeilo. 

19. A few yards below a spring (marked and named on the 
6-in. map) close to Nant Gwyddfau, J mile S.S.E. of Garn- 
bica Farm, 1 J mile E, of Llandybie. 

20 to 23. Side of small stream, 300 yards S.W. of Cilcoll 
Farm, 1^ mile E. of Llandybie. 

24. 350 yards S. of Castell-y-Graig Farm, 1 mile W. by N. 
of Llandybie. 

25. 400 yards E. of Gelli Siffor Farm, 1 mile N. by W. of 

26. Within the southern edge of a wood, 170 yards E. of Gelli 
SifiTor Farm. 

27. (?) Side of stream at N. end of a wood, 400 yards S.K 
of Gelli SifiFor Farm. 

28. 300 yards N.N.B. of Glyn-gl&s Farm, 1 mile S.W. of 

29. 300 yards N. of Pl^s-Mawr, 2^ miles S.W. of Ammanford. 

30. 10 yards W. of the well at Llwyn Ifan Parry Farm, Banc- 
y-Mansel, 8 miles E.S.E. of Carmarthen. 

31. 350 yards N.K of Garn Farm, If mile N. of Ponty herein. 

32. Side of stream, 150 yards N.N.W. of Tor-y-coed-isaf Farm, 
J mile E. of Llaugyndeyrn, 5 miles S.E. of Carmarthen. 

33. 250 yards S.S.E of Blaenau Farm, If mUe B.NJbL of 

34. In a dingle between Cwm-y-dwrandCil-carn-fach Farms, 
I mile W.N.W. of Pontyberem. 

35. About 300 yards S.W. of Llwyn-gwyn Farm, 1^ mile S.W. 
of Llangain, 3 miles S.W. of Carmarthen. 

36. Side of stream, 200 yards S.E. of Pengelli-isaf Farm, 
li mUe W.S.W. of Llangain. 

Digitized by VjOOQ IC 


37. Side of stream, 100 yards N.W. of Pen-picillioii Farm, 
1 J mile N.E. of Llanybre, 6 miles S.W. of Carmarthen. 

38. 250 yards S.E. of Maes-gwyn Farm. IJ mile E.N.E. of 

39. Side of stream, 200 yards S.W. of Maes-gwyn Farm. 

40. 400 yards S.W. of Maes-gwyn Farm. 

41. On E. side of stream, 120 yards S.W. of Cwmllyfrau Farm, 
1 mile N.N.E. of Llanybre. 

42 and 43. Side of stream, 60 yards below Ffynnon-olcwm, 
\ mile E. of Llanybre. 

44. In a hedge, 100 yards S.E. of Ffynnon-dagrau, near the 
Vicarage, Llangynog, 5 miles S.W. of Carmarthen. 

45. 100 yards W.S.W. of Gelli Farm, 1^ mile N.E. of Llan- 
dilo-Abercowin, near St. Clears. 

49. In a coppice, 500 yards E. by N. of Llandilo-Abercowin 

50. In a thicket, 450 yards E. by S. of Llandilo-Abercowin 

51. 300 yards E.N.E. of Ty'r Gate Farm, I mile E. of Lower 
St. Clears. 

52. Side of stream, 50 yards N.E. of Broadmoor Farm, 1 mile 
S. of Lower St. Clears. 

53. N. bank of stream in deep valley (transversely the Pem- 
broke road) 400 yards S.E. of Parcau Farm, 1 mile S.W. of 
Llanddowror, St. Clears. 

54. E. side of small pond, 300 yards W. of Blaeu-gors Farm, 
^ mile S.W. of Llangynin Church, St. Clears. 

55. In corner of field and by side of stream, 400 yards S.E. of 
Sabulon Farm, 2 miles W. of Blue Boar, St. Clears. 

56. S. side of small pond, 350 E.N.E. of Forest Farm, li mile 
W. of Whitland. 

57 and 58. 150 yards E.S.E. of Coleman Farm, IJ mile W. of 

59. 80 yards S.E. of first milestone from Dryslwyn Ford, on 
the Castell Rhingyll road, W. of Llandeilo. 

60. Side of stream 280 yards N.E. of Crug-y-felin or Crui^-y- 
fifetan-fawr, 1^ mile E. of Eed Roses, 3 miles S. of Whitland. 

61. Side of footpath, 200 yards S.W. of Cwmfawr Farm, J mile 
N.E. of Eed Eoses. 

62. Side of stream, 600 yards S.S.E. of Eed Eoses, at head of 
stream which flows southwards between Westpool and Sich 
Farms. A strong spring breaks out 100 yards N. of the hearth. 

63. K side of stream, 150 yards S.E. of Mountain Farm, 
Tavernspite, 3 miles S.S.W. of Whitland. 

Digitized by VjOOQ IC 

384 trie tARLlr SBTTLEftS 01* CAftMARTBtBl^. 

We now come to the antiquities of the Bronze Age 
period. In the Arch. Camb. for 1851 (p. 159) there is 
a reference to Allt Cynedda, near Kidwelly, of a very 
perfect but ancient encampment, with two barrows or 
tumuli to the eastward of it. The larger of these two 
barrows was raised about 300 yards from the camp, and 
measured 56 ft. in diameter, but wns only elevated 
about 5 ft. from the surrounding turf. Two feet below 
the original surface of the soil there was a large stone, 
cut into a hexagonal figure like an old shield. The 
stone measured 8 ft. 4 ins. in length and 7 ft across, 
and 12 ins. to 15 ins. in thickness. A cist was found, 
in which there lay the bones nearly entire of a very tall 
human skeleton. The skull was almost perfect, but 
was singularly flat and depressed in front, with a cir- 
cular opening upon the left hemisphere, as if beaten in 
by the blow of a slingstone or pointed mace ; another 
chin-bone was very projecting. The teeth were entire, 
but had fallen out of the mouldered jawbone. This 
tumulus, which is called Banc Benisel, has a circular 
depression at the apex about 5 ft. or 6 ft. in diameter. 
In the Arch. Camb. for 1855 (p. 297), there is an 
account of the discovery of urns on Y Garn Goch. We 
are told that several urns were laid bare, the greater 
part of which were ornamented with a more complex 
and decorative pattern than is usually found in sepul- 
chral urns of this character. All these urns were in- 
clined outwardly, all at the same angle ; but this was, 
as the writer remarks, probably due to the pressure of 
the superincumbent central mass of stones. There was 
also an inner circle, not concentric with the outer one. 
Smaller urns were discovered in juxtaposition, which 
were supposed to have contained food for the departed 
spirits, for their support during their transit to their 
new abodes (p. 298). Mr. Babington observed that the 
position of the urns was usually inclined, Smaller 
urns (the so-called incense cups) were often found in 
close proximity to larger ones. In the Arch. Camb. 
for 1856 (p. 107) there is a reference to a circle of 

Digitized by VjOOQ IC 


stones called "Buarth Arthur," and another called 
" Meini Gwyr." These were mentioned by the late 
Mr. T. Stephens, of Merthyr Tydfil, in November, 1855, 
and may have been stone circles surrounding Bronze 
Age burials. In the Arch. Camb. tor 1879 (p. 155) 
there is an account of prehistoric and other remains 
in Cynwil Gaio, by the Ven. Archdeacon D. R. Thomas. 
In this article Archdeacon Thomas points out that at 
the base of Cerrig Cestyll lie the scattered remains 
of a cairn. Cairns, he says, are very numerous 
on the hill ; the largest, he says (p. 58), is that 
of " Y Garn Fawr," a great stone mound raised on the 
highest point of Craig Twrch. The base of the cairn 
appears to have measured 30 ft. in diameter, or, 
including the enclosing dyke, a diameter of 52 ft. The 
upper portion has fallen away, and another part has 
been employed in the construction of an abutting 
sheepfold (p. 58). At the base of the slope, on the 
western side of Cerrig Cestyll, is a group of no less 
than five cairns, of which only the bases now remain. 
All of them have been disturbed, and some of them 
almost entirely removed. They have no surrounding 
ditch, and their average diameter is about 25 ft. In 
one only was a cist found, and in that a double grave 
with a bottom of prepared clay, but no sepulchral re- 
mains of any other kind (p. 59). The most curious 
feature was the portion of a series of rough slabs placed 
edgeways close together, and pointing towards what 
was probably the most important portion of the cairii. 
A somewhat similar arrangement of stones, laid to rest 
on each other in rows, and sloping towards the cist, had 
existed, we are told, in Carn Trawshant on the Mallaen 
range. They had, however, been removed some fifty 
years before, and the cist exposed ; and all that now 
remains of it were the containing slabs of the grave, 
2 ft. 9 ins. in length and 2 ft. in breadth. The bed of 
the grave appears to have been a yellowish clay, from 
which all stones had been carefully removed, and this 
formed a layer upon the natural soil. '* West of this, 

Digitized by VjOOQ IC 


at no great distance, is a circular mound of earth, 25 ft. 
in diameter, and to all appearance undisturbed, and 
so presenting a most favourable field for further ex- 
ploration. A third mound, somewhat smaller, measur- 
ing 18 ft. in diameter, lay to the south of this last ; but 
it has been almost entirely cleared off." There is also 
mentioned the cairn called "Y Garn Fawr," to the 
north-west of the farmhouse of Bryn Aran. This is a 
large stone platform about 50 ft. in diameter, with a 
raised cairn in the centre, in which it is probable, 
Archdeacon Thomas says, the cist may be found undis- 
turbed, although the surrounding portion has been 
carted away for walling and road-metal. A smaller 
one of 25 ft. diameter, a little to the south, has been 
almost entirely carried away ; and near it is an ellipti- 
cal arch about 45 ft. by 36 ft. at the greatest length, 
formed by a stone rampart 6 ft. in width. In the Arch. 
Comb, for 1886 (p. 348) there is an account of a sepul- 
chral urn of rude pottery, exhibited in the Swansea 
temporary museum by Sir J. T. Dillwyn Llewelyn, 
M.A., F.L.S. This was found in Y Gam Goch, and 
bore upon it the impression of twisted thongs or rushes. 
It has thus the usual characteristics of Bronze-Age 
sepulchral urns. 

In the Arch. Camh. for 1890 (p. 41), there is an 
account by Mr. G. G. T. Treherne of the opening of a 
tumulus at Castle Hill, Carmarthenshire. In this 
account we are told that the tumulus lay in the south- 
east corner of a field called " Pare y Twmp," on the 
southern slope of a farm called '* Castle Hill," in the 
parish of Kiffig, Carmarthenshire. It was circular in 
form, and measured roughly 70 paces in circumference. 
It is 25 paces in diameter, and its depth to the clay 
floor was 6 ft. 6 ins. in the centre. Mr. Treherne gives 
the account as follows : — " We drove an adit 4 ft. wide 
from the south side, and found no trace of the usual 
stone circle. There was a thin floor of clay level with 
the field surface, apparently much trampled, and covered 
with a thin covering of black ash. Rather to the south- 

Digitized by VjOOQ IC 

tflfi l&ARVi SteMLERS Oi* CARMARTHEN. 38? 

east of the true centre there seemed to be an artificial 
depression or hole in the floor, and this was filled with 
black ash, fragments of charcoal, slight traces of disin- 
tegrated bone, and patches of red earth ; these last 
possibly the remains of the original urn. There were 
no implements, shells, or anything beyond the burnt 
materials. This was of a distinctly greasy character." 
The mound, Mr. Treherne says, is formed of earth and 
rubble (mostly rubble In the centre), not homogeneous 
with the soil of the field surrounding it. We replaced 
the ashes in situ, and partly filled up the trench. In 
connection with the same account, Mr. Treherne says 
that at a distance of 50 paces to the south, in an 
old red sandstone quarry, the party found a flint flake, 
evidently artificially worked. It is difficult, however, 
to be certain to what period the flint flake belonged. 

In Arch. Camb. for 1893, p. 89, there is an account 
of an artificial mound between the two lakes at Talley, 
or Talyllychau, to which attention was called by tne 
Rev. Charles Chidlow. An account of it was sent to 
Dr. R. Munro, author of The Lake'Diuellings of Europe, 
but all that he said of it was : " We are here dealing 
with a lake-dwelling, or fort, of unique character, pre- 
senting special features I have not hitherto observed in 
any of our Scottish or Irish crannogs. This mound at 
Talley is said to be riddled through and through with 
rabbit-holes, but these have brought to light no trace 
of human occupation." 

That Carmarthenshire shared further in the Late- 
Celtic civilisation is made highly probable by the dis- 
covery of various Late-Celtic objects just outside its 
two extremities. For example, some enamelled horse- 
trappings, which are now in the Cardiff Museum, were 
found at Seven Sisters, near Neath, and a fine Late- 
Celtic collar, similar to the one at Wraxhall, was found 
at LlandyssuL As they certainly were worked in Roman 
times, it is not improbable, too, that the Dolau Cothi 
mines, which are thought to have contained gold, were 
worked in Late-Celtic times. Then, and doubtless 

Digitized by VjOOQ IC 


earlier, the district lay on one of the trade-routes to the 
South of Ireland, and thus participated in the com- 
mercial prosperity of the South Wales coast. The 
Late-Celtic objects from Kyngadle, near Laugharne, 
are also an indication of the same type ; and it may 
well be that a thorough investigation of Y Garn Goch 
would tend to place it in the same Late-Celtic period 
as TreV Ceiri, where a Late-Celtic bead and some traces 
of iron were found. The only scientific way to deter- 
mine the true age of ancient remains is by patient and 
judicious excavation. It is to be hoped that one result 
of the present meeting will be to lead to a more 
thorough exploration of the ancient remains M the 
county. It is possible that the folklore of the county, 
both mediaeval and modern, if we only had the key to 
it, would yield valuable ethnological results ; but the 
difficulty is to distinguish in these stories the kernel of 
fact from the added embellishments. The reader who 
is interested in the folklore of Carmarthenshire cannot 
do better than consult Rhys' Celtic Folklore, where the 
ethnological bearing of the mediaeval and modern folk- 
lore is discussed. It is not impossible that each stratum 
oF the early settlers left in local tradition some memory 
of itself. It is to be hoped that the excellent Anti- 
quarian Society of the county will keep a careful record 
of all material that will throw light on the ethnology 
of the district. 

Digitized by VjOOQ IC 



{Continued from page 334.) 


It is proposed here to deal with Norden's Survey of 
Kolt^Harleian MS., vol. 3696). I have myself made 
many notes and extracts from this Survey at the 
British Museum, and for the rest am dependent upon 
a part copy transcribed for the late Chevalier Lloyd, 
collated, wherever required, by Mr. Edward Owen. 
The men sworn as jurors to assist John Norden were: 
George Bostock, Esq, ; David Speed, gent. ; Thomas 
Calcott, gent.; Thomas Pate, gent.; William Wyld, 
gent. ; Francis Pickering, gent. ; John Yardley, gent. ; 
John Wilkinson, Randolph Hutchins, Thomas Wilkin- 
son, Thomas Pulford, Roger Edgworth, John Wright, 
William Batha, Geor^^e Wright, Richard Vernon, and 
Roger Greene — seventeen in all, concerning whom it is 
noticeable that one only — William Batha — bore a Welsh 
surname. The same seventeen were also sworn in as a 
ury of survey for the manors of Hewlington and Cob- 
lam Isycoed. Some portions of their presentment will 
^e quoted in full, preserving the spelling actually used, 
and other portions will be omitted, or only briefly 

The jurors say that ** there is in the Towne of Lyons 
als Holt in the countie of Denbigh one Castle builded 
with Stone consisting of five Squares and of five Towers, 
covered with lead, having foure Gates^ at y® entrie into 

^ These were the onter and inner gates of the main entrance, 
together with the two portcullises, or else those two gates and the 
inner and outer gates of the Exchequer Tower, 

Digitized by VjOOQ IC 


y same, w** Castle is nowe in great decay and some 
parte of the Roofe thereof fallen dowue and much of 
the timber rotted, and y* lead likewise decaying and 
worne thinne. Neverthelesse the Lead^ of the said Castle, 
and other materialls, if the same should be demolished, 
are worth to be solde about . . . poundes besides ye Stone. 
There is adioyning to y* said Castle at the first entrance 
into y® same one Tower or building of Stone commonlie 
called y* Excheq' where y® Recordes touching the Lord- 
shipps of Bromtield and Yale are and have been vsuallie 
kept w^^ is likewise covered with lead. There are di- 
verse Howses of Office belonging to y® said castle, all 
builded w** timber as namely, one Gatehouse called the 
Outward Gate, Garners for Corne^ Barnes, stables for 
Cattell, Killne, Brewhouse and one Pidgeon howse all 
decayed conteyning by estimac'on . . . Bayes. There 
are within the precinct of the saide Castle certaine 
parcells of Land as namely one plott or parcell wherein 
the said Castle standeth called the Castle ditch one 
parcell called y® outward gate or Court, one garden 
place, one parcell called y® greene Court, and one par- 

^ Elsewhere, on the back of the ^ound plan of the castle, Norden 
inserts his estimate of the valae of the same, thns : — 

" The lead ouer the 5 towers con tayne ... ... ... ... 4,650 

The lead ouer the mayn lodginges yet remayninge and falne downe 

with the timber cont* ... ... ... ... ... 14,250 

Ouer the gatehouse that stands to little yse, p' est. ... ... 320 

which Although it be the most part very mnch worn and very thjn, 
yet onn with another it may be valned at iiiid p' foote which will 
amount vnto occl** or thereaboats.'' He adds : " Mnch of the timber 
abont the Castle is yet very sounde, but decayes daylie theawgthe 
[thronji^h] the defect of the Leades. One mayn floore fell the very 
night I came to the holt, the timber and Leade doth lye now very 
confusedly ; mnch of the reste is so weake as it is dangerous to 
adnentnre vpon it. To re-edifie will cost much new timber and 
Leade, the Lead that now is being worne so thyn that being oast 
new will yelde mnch drosse, as it doth now much duste; yet fit 
eyther timelie to be repaired, or the materialls to be taken downe, 
kepte, or soulde." 

* Against these words underlined is written : " These were leased 
to Kdward Hughes at viij«. iiijo?. p. ann. 

Digitized by VjOOQ IC 


cell called y® Orchard conteyning in y® wholle by es- 
timac'on Three acres.* There is one Howse or buildinof 
upon y* westside of the Court called Green court 
commonlie called y® Welshe Courthowse with a loft 
over y® same conteyning . . . Bayes w°** is appurtenant 
to y® said Castle where y* twoe great Leete Courtes for 
the Lordshippes of Bromfield and Yale have heretofore 
been vsuallie holden and kept, w°** are discontinued.* 
There is one stable place and an vpper Roome at the 
Northend af the said Welsh Courthowse with a smithie 
or smithes shoppe therevnto adioyninge demised to 
Thomas Crew Gent, for Forty yeares dat : Primo die 
Junii Anno Rn e Eliz. 35 ^ wherevppon y® yearlie Rent 
of Two shillinges is reserved heretofore charged within 
y* BaylifFes charge of the Manno' of Hewlington. There 
is also at y® southend of y® said Welshe Courthowse one 
Large Bay now vsed for a Barne in the holding of one 
William Burgeny or his ass's [assigns] w*'** hath been 
charged heretofore in the Bayliffes Accouraptes for y* 
Manno' of Hewlington." Concerning the last two sen- 
tences, a correction is made afterwards in the following 
words : '* Memorand : there is one Chamber and a 
Lowe' Roome vnd' the same now vsed for a Stable and 
a Smithie or Smithes fforge adioyninge to the North 
end of the Welsh Court howse in y® Towne of Lyons 
als Hoult w**is graunted to Thomas Crue gent, for 40 
years by Lease dated primo die Junij 35 Eliz. Rn'e 
w^ we finde to bee within the Survey of Holt as an 
Appurtenaunt to the Castle & within y* precinct of the 

^ Probably customary acres. If so, eqaal to nearly 6J statute 

' Two discrepant statements are made in tlie Survey of 1620 
concerning the holding of these leet courts /or the whole lordship: one 
statement made doubtless by the jurors, and the other by the sur- 
veyor. According to one statement, the two great courts in the 
year formerly held were at the time of the Survey discontinued ; 
according to another, they were still kept. As a matter of fact, 
they were discontinued, but from another point of view the eonrt- 
honse, however decayed, was still there ready to accommodate the 
Brom6eld and Yale tenants who owed suit and service there, if the 
steward or bis deputy should duly summon them. 

Digitized by VjOOQ IC 


same as by the particulars belonging to the said Castle 
in the p'sentment for y® said Castle appeareth. There 
is also another part of y® said Welsh Court howse at y® 
south end of y® same now vsed for a barne in the 
houlding of humphrey hanraer, Gentleman, in right of 
his wief, y* late wief of Anthony Burgeney at Will w** 
we find likewise appurtenaunt to y® said Castle and no 
part of J® Bayliffes charge of this Manner [Hewlingtou] 
as wee supposed/' Nevertheless, it is probable that the 
first presentment is right, for in the Survey of 1562 
these buildings ai'e declared to be in the manor of 

Before proceeding further with our extracts from the 
Survey, it is necessary to discuss the sketches and plans 
of the Castle and its precincts, given by Norden and 
others. These illustrations have appeared before, either 
in Pennant, Powys Fadog,^ or elsewhere ; but it would 
be impossible to present a history of Holt without 
giving therewith the illustrations just named; which, 
moreover, it seems to me, have never hitherto been 
studied with sufficient care, although they raise as 
many problems as they solve. 

The Castle is, of course, of a far earlier period than 
the earliest pictorial description of it; but, unfortu- 
nately, it has never come under the critical eye of a 
master of military architecture, such as that of the late 
Mr. G. T. Clark. But as its main features probably 
continued unchanged down to the early part of the 
seventeenth century, we may confidently commence 
our description of it with the accounts of eye-witnesses 
who had seen the Castle before it became the common 
quarry for buildings in the neighbourhood. It should, 
however, be borne in mind that not a single one of the 
early illustrations referred to (including Buck's view of 
17 . .) are characterised by the strict accuracy of a 
modern surveyor's drawings, nor were they intended 

^ I desire to acknowledge my indebtedness to Mrs. Lloyd- Veruey, 
of Clochfaen, for permission to reproduce, fron» Pourys Fadog, one 
of the plans, 

Digitized by VjOOQ IC 


i^ cdi/j^Tr t)fiNBmH. 


by their authors to be so. They are none the less 
of great value, and enable us to reconstruct the build- 
ing of the ^^'arrennes more perfectly than would have 
been possible from an examination of the heaps of ruins 
which now alone remain. 

The elevation and ground plan, marked 3 and 4, 
were made evidently by Norden in 1620, and represent 
the form and condition of the Castle and its precincts at 
the time of his Survey.- But there are two other illus- 



! fl !<.>.! 

W .1 '» . v.- 

Fig. 1.— Early Sketch of Holt Castle. 

trations of the Castle, preserved in vol. 2073, Harleian 
MSS., flF. 112 and 113, which I believe to be earlier 
than 1620. The first of these (Fig. 1) is a rough sketch 
of the exterior. Daniel King, of Chester, used this 
sketch in 1656, passing it oflF as his own, and made an 
engraving of it (see Harleian MSS. 2073, 594 B, 
fo. 126), with the intention of illustrating Camden's 
Britannia} It may be urged, reasonably enough at 
first seeming, that if King used the sketch, this would 

^ It did not appear in Gbngh's edition of that work, 1695. 

6TU 8£R., VOL. VII. 26 

Digitized by VjOOQ IC 

394 TriE TOWN OF HOLt, 

show, at any rate, that the Castle was as represented 
by him, he having probably seen the building before 
the siege. But the truth seems to be that when King 
first saw the Castle it was recently dismantled ; and as 
he wished to give a drawing of it, he used this sketch 
(probably collected by one of the Randle Holmes), 
which was the only representation available* to him. If 
this opinion be accepted, Daniel Kings authority for 
the drawing disappears. But the drawing itself re- 
mains, and the ground plan corresponds with it. And 
the more the one and the other are considered, the 
more will it appear that both represent an earlier 
arrangement than that which Norden represented : 
perliaps the arrangement, which Sir William Stanley 
found when the grant was made to him. It does not 
seem possible to specify the date more exactly. 

Assuming, then, that figures 1 and 2 show a much 
earlier state of things, in respect of the Castle, than 
figures 3 and 4, we will now proceed to discuss the first 
two illustrations. 

These (Figs. 1 and 2) reveal an irregular pentagonal 
castle of small size, enclosing a court, also in form a 
pentagon. At each corner, on the outside, but con- 
nected internally with the main body, was a round 
tower, higher than the battlements of the pentagonal 
portion. To one of the five towers, the next south- 
eastwards to the tower east of the entrance, was 
attached an external rectangular addition, of equal 
height with the tower, containing in its lower portion 
the chapel. On each of the towers, except on the 
chapel tower, was also a small conning- or watch-tower, 
which apparently contained a chimney. The entrance 
was between the two towers on the north side, and 
there was a wooden bridge thrown over the inner 
ballium between this entrance and the *' Chequers" or 
Exchequer tower, which, according to the plan, was 
then a low building. Another wooden bridge spanned 
the foss on the other side of the Exchequer tower, and 
led to the outer gate. Over the main entrance of the 

Digitized by VjOOQ IC 



Castle was sculptured a Hon passant-guardant (see ch, i, 
vol. 1906, p. 221). 

Coming now to the ground plan of the castle of the 

Fig. 2.— Early Ground Plan of Holt Castle. 

earlier date, as shown in Fig. 2, it may be well foi* the 
benefit of those who find old writing diflBcult to read, 
to give the description of the building, court, and 

26 « 

Digitized by VjOOQ IC 

396 THE TOWN OP HdLt, 

lowermost chambers in modern print. Here it is : — 
" The castle fine [five] square eu'y [every] way with 
square in the inside 51 foot betweene wall and wall, 
being the breadth of all the romes [rooms], about 
22 foot the in'er wall foot, the outerd [out- 
ward] wall .... foot, the castle three storyes high : 
the tower[s] fowre beside en [every] watch towre w"^** 
mak' tiue [five] chappell 12 foote broade and 15 foote 
Longe / the ditch 20 yards (and in some places more) 

Between the entrance and the tower flanking it on 
the east was the well-house. The whole of the side 
between this and the chapel tower was occupied in the 
basement by a stable, as was also the side between the 
chapel tower and the tower next southward or south- 
eastward. On the next side, towards the west, were 
** oflSce romes [rooms] for cook and Butler/* and on the 
remaining side, between the tower last named and the 
tower on the west flanking the entrance, was the 

Across the ground-plan of the interior court are 
written these words : " fro the court to the battlem** of 
the castle but two storyes high all these romes being 
under ground." 

At a later date, perhaps in Sir William Stanleys 
time, considerable structural alterations were made in 
the Castle. The Exchequer was converted into a strong, 
square tower with an upper room. The square addition 
to the chapel tower was, if the plan is to be truMed^ 
removed, and the tower restored to its round form. On 
the other hand, the next or south-east tower — that one 
whose base abutted on the river — was made wholly 
square.* And a well was dug in the middle of the 

^ It may be well to observe that my friend Mr. Edward Owen, 
after a harried examination of the rnins, does not agree with the 
above. The internal towers have completely disappeared, so that it 
is impossible to decide the particular point in question. Bat his in- 
spection of the Castle, with the varions illastrations in band, revealed 
so many discrepancies in the latter, not only from each other bat in 

Digitized by VjOOQ IC 



court. But, perhaps, the best way will be to set out, in 
fair modern print, the several descriptions contained in 
Norden's ground plan of the Castle and precincts, as 


^f' : I- - 

\ Sir; -t. , 


-^Vy r-'V' *'-*•'' 

\ /A- ,-:-'-' '"•• 

ri.r,!..|».--.. ♦,.,^•^f• 
, * r,/, J.///,/|».,..^'/r,/. (. 

Fig. 3.— Elevation of Holt Caatle in 1620. 

all from th^ rniDS of the Castle itself, that he is disposed to Consider 
the absence of the square projection from the chapel tower in one 
plan, and the transformation of a round into a square tower in 
another, as no more than the errors of the artists, who probably 
completed their sketches far from the spot they are intended to 

Digitized by VjOOQ IC 



shown in Fig. 4. The elevation, Fig. 3, needs no 

The four round towers are indicated on the plan 
(Fig. 4) as supplied " with manie Lodginges and Chim- 
neyes couered as is all the Castle that yet standes with 
Leade." Then with respect to the square tower, this is 
what is said : '* There is a vaulte under this square 


/Mw2« 4^1.^/. •«*«-^ 

n* t^^P^^-^^ 

Fig. 4.— Ground Plan of Holt Castle and Precincts in 1620. 

towre secretly to come to the river at 23, were is an 
Iron gate as is sayde."^ 

We come now to speak of the interior of the several 
sides of the pentagon. The side between the tower 

* In Tidderley's description (see Ch. ii, p.. 814) this secret narrow 
passage, yaulted with stone, is mentioned as leading oat of the court 
by steep stairs to the river, " whereto the ward and dore ys of yron." 
In the inquiry made at Holt Castle on 30th January, 158^ (see 
Gh. ii, p. 318), this " iron doore being belowe in the house towards 
the Riv' of Dee," is spoken of as having been taken away during the 
time of Mr. Edward Hughes, then receiver, and resident within the 

Digitized by VjOOQ IC 


east of the entrance and the next tower south-eastward 
was occupied by the hall, reached, as we know from 
Tidderley's account, by a straight stone stair, 7 ft. 
wide, leading up from the inner gate. The side be- 
tween the hall and square tower contained '' Butteries, 
Pantries, kitchene, etc.," and the side westward be- 
tween the square and the next tower included the 
''Great Chamber and other chambers"; while, on the 
fourth side, to the tower west of the entrance there is 
noted : " At this place timber, Lead and all inwarde 
materiall fallen down." And over the entrance, as we 
learn from the inquiry of 30th January, 158f , the con- 
stable's chamber, also u chamber adjoining in which the 
constable kept his coal and wood. 

At A (Fig. 4), between the near entrance gateway 
and the Exchequer tower, '*the mote [was] 30 foote 
deepe vnder the bridge and [there] was a drawbridge." 

As to the Exchequer tower (c), it is recorded : " In a 
lofte over c lye all the records, and was auntiently the 
Chequers," all being surrounded by ** A mote or 
Trenche. A verie deepe ditche within which the 
Castle stands hewed out of the same rocke wheron the 
Castle is most strongly situate." 

The outer gate of the Castle yard is also shown, west 
of which were first " Barnes" and next " The Shire hall 
longe out of use."^ Again, east of the outer gate and 
abutting on it were '' olde stables [and] cowhouses" ; 
while at right angles to, but detached from, these were 
other " olde buildings." Also, in the yard east of the 
Castle was ** a decayd doue-house fine square." 

An account of the Castle yard given on the plan is 
as follows : — **The Castle yarde of noe benefite to his 
highnes, for that it is a comon passage, as it is per- 
mitted for all the townesmen to driue there Cattle to 
the water hauing manie other Wayes, begininge now to 
Challendge this by prescription the outer gate standing 
nighte and day open, by which all kind of cattle and 

^ The Shire Hall, or Welsh Court Hoase was, as we otherwise 
learn, on the west side of the Green Court. 

Digitized by VjOOQ IC 


swyne haue free entrance." This account refers to the 
east part of tJie yard, and as to the north-west part 
we have this further description : — " A little peece of 
the Castle groundes unclosed wher the people now 
challendge freedom to sport in and doe already pre- 
scribe to be proper vnto them for that vse, so the 
P[rince] shall have litle or no profite by the same if it 
be not reformed ; the whole Castle yarde with the mote 
contaynes about 2^ English or Statute acres." 

West of the yard, the east portion of the " litle park 
demaynes in the time of H. 8, and in his owne vse 
stored with deere, now in the disposinge of the E. 
of Bridgwater, claymed as is sayde, in righte of his 
oflSce, being stewarde of Bromfeilde and Yale, as he 
hath also certayne groundes called the Pooles, som- 
tiraes fishpondes, now freme Land all worth per ano. 
St. xviZi. The fee of the stewarde in H. 8 time, both 
for Holte and Chirklande, is but x\li. ; but there are 
since added I know not" 

And with regard to "The Ryuer of Dee" is this 
note : — '* The overflo winge of this riuer is a great annoy- 
ance of the Prince 8 and other mens Lands, confininge 
being barred back by a Causeway at Chester" ; and as 
to which it may be said that many inquiries were made 
and commissions held, but no practical remedy ever 
devised, and ultimately the causeway was ordered to 

Meanwhile, it is most necessary to dispel an opinion, 
firmly and generally held at Holt, and based upon 
misunderstood and imperfect data, that the outer gate 
of the Castle stood in Castle Street. This opinion is 
due wholly to the fact that Pennant, in his reproduc- 
tion of the plan, omitted descriptions of buildings and 
areas, and especially failed to record the points of the 
compasSy all of which are given in the original (see 
p'ig. 4). Now, if the outer gate stood in Castle Street, 
it would be due west of the Castle ; but, as a fact, as 
shown in the figure, it stood due north of the same, and 
was on a line with the Exchequer tower, the ruins of 

Digitized by VjOOQ IC 


which can still be identified. The outer gate stood 
at the bottom of the present lane to the Castle, past 
the schools. This lane one would expect to be broader 
than now, for it was the chief approach from the town 
to the outside of the outer gate. In any case, the lane 
is old, for it is mentioned in Tidderley's Survey (end of 
reign of Henry VIII) in 1562 and in 1620.^ It opened 
on a road proceeding directly from the outer gate of the 
Castle to the ford above the bridge, and to the bridge 
itself, passing the back of Church Street and the front 
of the church tower. Traces of this old road are yet 
to be seen, a bit of it being in the western end, or new 
portion, of the churchyard, and it is mentioned in 

It has been urged in support of the view that the 
outer gate of the Castle was in Castle Street, that in 
Pennant's plan a ^ve-sided figure shown thereon is 
evidently the Town cross, but the cross is eight-^xdiedi ; 
and on referring to the original plan this figure is actu- 
ally marked as *' a decayd aouehouse fiue-square," 
standing within the Castle yard and outside the gate of 
the Castle itself. 

It does not follow from this that there was no com- 
munication from Castle Street to the inside of the 
Castle yard itself. A road is said to have been laid 
bare some yeai^ ago by Mr. George Redrope, 3 ft. 
beneath the surface. It started from a point opposite 

* In 1620 Sir Richard Trevor is described as haying a curtilage 

*• noare the Castle gate extending towards the Riuer of Dee.** 

Also, in tlie same year, Lawreuce Welles held a piece of land " neare 
ynto the Castle gate in a Lane leading from the pavem^ towardes the 
Riyer called Mill Dee." See also next note. 

^ In 1562 Edward Aimer is described as haying a messuage with 
curtilage near Castle gate, in length from ** the royal way leading 
from the castle of the town of Lions to the church or chapel of the 
said town, and in breadth from a stable of the said castle towards 
the way leading from the said payement towards the horse mill 
there." The payement was apparently the paved way or footpath 
near the cross, and the horse mill was by the river ; and it is quite 
clear, therefore, that the main Castle gate was not in Castle Street, 
but in the lane leading towards the Dee, 

Digitized by VjOOQ IC 


Mrs. Baker's, and entered the Castle yard at a dip in 
the crag, which could easily be closed, and passed pre- 
sumably beneath the rock where the higher ground 
still is to the front of the Exchequer tower. 

Pennant's plan is imperfect, and must be used with 
discretion and considerable hesitation. 

We are now prepared to continue our extracts from, 
or summaries of, the text of Norden's Survey. These 
will illustrate further the statements made by the sur- 
veyor on Fig. 4, as to the Little Park, Pools, etc. 

Rent, 30«. 

There is one parcell of land adioyniiig to ye Castle, 
commonlie called the little Park,^ nowe in the 
tenure of John Earle of Bridgewater, or of his 
assignes, conteyniug by esfcimao'on, great 
measure ... ... ... ... 9 0* 

One close or parcell of land within y* said Towne, 
called The Pooles, adioyning to the Highway 
or pavements leading from the Holt' towardes 
the Common Woode, nowe in the tenure of the 
said John Earle of Bridgewater, or his assignee, 
con teyning by estimac'on, great measure ... 7 0^ 

One meadowe called Crackstringes, aU, Crack- 
stones meadowe, conteyning by estimac'on ... 7 0^ 

^ This '' Little Park" was that now in three parts, called re- 
spectively ••Top Park" and "Bottom Park," extending between the 
Castle on the north and the Qus Works on the south, and between 
the riyer on the east and Castle Street on the west. 

^ In statute measure about 19 acres. 

8 On the south side of Common Wood Lane, next the pound and 
nearly opposite Esphill, is a meadow called "The Pools." The 
unusual configuration of the surface of this meadow suggests at once 
that fishponds had formerly been kept here, as indeed the Survey of 
1620 elsewhere declares, doubtless for the supply of fish to the 
Castle. The lane itself is cobble-paved, and on one or the other 
sides of it are five or six fields or meadows called " Pavement Field" 
or " Pavement Meadow," illustrating the statement of the Survey 
concerning " Pavement Lane," between Frog Lane and Common 
Wood Lane, and confirming the identity of the meadow now known 
as " The Pools" or *' Pool Meadow," with that described in 1620 
under the same name. There was another Fishpool field, probably 
adjoining this one, in which various burgesses of Holt had acres. 

^ Over 14| statute acres. 

Digitized by VjOOQ IC 


Odo other paroell of lande commonlj oalled the 
Heye, als. Saunders Hey, in the tenure of the 
said John Flarle of Bridgewater, conteyning by 
estimac'on ... ... ... 6 0* 

The jurore then go on to refer to the Friday market 
and to the two fairs, one held on St. Barnabas' day in 
suramer, and the other on St. Luke's day in winter 
(see Ch. ii, p. 312), the toll of which market and fairs 
was let for thirty-one years to Sir Peter Warburton, 
one of the burgesses of Holt, at a yearly rent of 22s., 
payable to the Prince. 

A list of the lands, houses, and one burgage set 
aside for the sustentation of a lady- priest to celebrate 
within the church or chapel of Holt is next given. 
These had been sold, and will be referred to in a future 
chapter. On the north of the main road from Wrex- 
ham to Holt, by Deevon Bridge, is a field still called 
" Priest's field." And in 1620, " St. Mary's lands" are 
named, which may be the " Mary's loons" of the Tithe 
Assessment Map to the north of the Bible meadow. 
These fields are, it is possible, part of the lands so sold, 
formerly belonging to Holt Chureh. The separate 
items of the yearly rents of these lands in my copy add 
up to £6 10^., but the total given seems to be £5 78. 
These lands, or part of them, as will hereafter be shown, 
had been bequeathed in 1523 by Thomas ap David ap 
Deio, of Holt. 

The jurors of 1620, in their presentment, next recite 
the terms of the charter granted them by Thomas, Earl 
of Arundel and Surrey, in the thirteenth year of 
Henry IV, declare the decay of rent since that time, 
and the revival of it in the fifth year of Queen Eliza- 
beth. They deny that any fine is due from them on 
the marriage of their daughters, stand upon their 
charter, and refer to " an extent made in the fifteenth 
yeare of Richard the Second," which extent, unfor- 
tunately, does not now seem to be in existence. Next, 
they say that " there is within the Towne of Holt, one 
Howse builded of timber & covered with shingles com- 

^ Oyer 12^ statnte acres. 

Digitized by VjOOQ IC 


raonly called the Townehall, where the Courtes I name- 
ly the Borough Courts] are holden and kept by the 
Maio' everie three weekes, and two great Leet Courtes 
likewise holden and kept by the Steward aforesaid for 
matters happening within the said Towne and Liberties 
are vsuallie kept, w** is the Princes bowse, the repaire 
whereof from tyme to tyme as occasion shall require 
belongeth vnto the Prince his Highness." The lower 
story or ground floor of the Town Hall was then divided 
into four shops. 

Then comes a list of all the freeholders and burgesses 
of Holt in 1620, with an exact and full description of 
their burgages, houses, and lands. I made a copy of 
this list and a summary of the description. This sum- 
mary wiH be presented towards the end of the chapter, 
as it seems desirable to give first a copy of the rental 
[crown rents] of Holt in 1620, and then of the peram- 
bulation of the town and franchise in the same year. 

The totall sum of the rente of the Holte is IzxiVt. 

xim. iud. 
Whereof to be dedncted for Certaine Landes grann- 

ted to the Earl of Bridgwater in fee, viZt. xii<. 

So remajnes with the toll of the Market and 


The BoroDgh Rent as appeares bj the Rentall, 

Ixii/t. iiitf. ixe^. 
Besides the Castell houses, yixis. iiWd, 
Besids for Crackstones meadowe and Saunders Heath 

p. ann., xxx^.j 
The Toll of the bridge and Market xxii«. 

Ixv^t. xii<. 


iiii^. \d. 

The above totals do not agree exactly with the sums 
which compose them : a slight error in the copy is to 
be presumed. 

The bounds of the town and liberties of Holt are 
described [fo. 43] in these words : ** The Towne of Lyons 
als Holt with the Liberties and Franchises of the same 
is meered and bounded as followeth. First from the 
Bridge called Holt Bridge w^ is the passs^e of the 
River of Dee divideing the Countie of Denbigh and the 
Countie of Chester, at the Northside of the said Towne 

Digitized by VjOOQ IC 


it extendeth itself along to the said River vnto a place 
called Moore Dee w** joineth vnto a Meadowe called 
the Lords Meadowe beeing parcell of the Manno' of 
Hewlington vppon y® soutnside of the said Towne and 
so along the skirt of the said Meadowe and other 
landes therevnto adioyning called the five acres and 
Tier garreggs vnto the lane called Hewlington lane 
vppon the west. Then it extendeth a long that Lane 
towardes the North vnto the lane called the Gallow- 
tree field lane, and so vpwardes or alonge that Lane 
Westward vnto another lane leading towardes the 
Manor of Iscoeyd vnto certen landes there adioyning 
vpon the said Lane called the Gorstifield in the tenure 
01 David Sped parcell of the Manner of Hewlington 
vpon y® East, and soe along the said lane, including a 
parcell of Land called Kae Stockley and two Tenements 
in the holding of Peers Spencer and John Goze [i.e., 
John Goch, Red John~\ adioyning to the vpper end of 
the said lane vppon the East. From the head of that 
Lane it then extendeth itselfe towardes the West vnto 
the Manner of Ridley, and excludeth one tenement and 
lands in the tenure of Richard Prestland and one Cot- 
tage and certaine other parcells part of the said 
Manner of Ridley lying upon the Northside the Lane 
which leadeth towards Crossyockin Lane upon the 
West. From Crossyockin Lane w** is called Hugmore 
lane or Wrexham Lane,^ and from the head of the same 
I^ne called Hugmore, then it extendeth itselfe againe 
towardes the North after the Hedge lying upon the 
west w^ divideth the Towneshippes of Gourton and 
Boras at the west and of the landes of Owen Jones Gent., 
Owen Breerton, George Bostock Esquires and William 
Botha [Batha] as they are particularly mentioned in 
the verdit or presentment of the said jury for Holt, 
and from the lands of the said William Batha it exten- 

^ The meaning here is that from the head of Croes locjn Lane, 
otherwise called Hagmore Lane, the bonndarj then extends itself, 
etc. The bead of Hagmore Lane was in Wrexham Lane. We are 
not to understand that the two lanes last named were one. 

Digitized by VjOOQ IC 

406 THt TOWN OF HOLt, 

deth along the lane leading from Borras to the Common 
wood upon the East^ and so includeth parcell of y* 
great Parke called Mercley [Mersley] Parke adioin- 
ing to the Mannor of Burton vpou tlie North ; w** 
parcell or parcells are called the Bushell [Bushy] and 
Broade land and extendeth from the vpper Parte 
of the said broad Land w**** adioyneth vnto the land 
leading from the- Common wood upon the North, 
towardes AUington vnto the landes called Woerhookes 
lying within the Mano' of Hem vpon y* said North 
parte. And also along the said groundes called Wer- 
hookes unto a passage or bridge over the Brooke called 
Devon Commonly called Werhook bridge, neare vnto 
the River of Dee upon the North, and so is bounded 
with the said River of Dee towardes the Eaat vnto the 
Holt bridge where first it begun." 

The foregoing account of the boundaries of the liber- 
ties and franchise of Holt is for the most part clear 
enough, knowing as I do many of the lands designated 
by names now forgotten, or almost forgotten. How- 
ever, let us understand that the present Parliamentary 
borough or parish of Holt is made up practically of the 
old franchise and the manor of Hewlington. The 
bounds of the old franchise are then recognisable 
directly upon the east, west, and north, and only 
somewhat uncertain on the south. But when we come 
to deal with Hewlington, the perambulation of that 
manor by the jury of 1620 will be presented, and then 
some of the ambiguity relating to the limits of the 
franchise on this side will disappear. Suppose, however, 
an attempt be here made to describe, in modem terms, 
the bounds of the liberties of Holt as they were in 
1620. The northern boundary of Hewlington will then 
itself become more intelligible. 

The boundary of the old franchise of Holt starts 
southward from the borough bridge along the Dee, 
which nears the liberties on the eastern side, until it 
reaches the first bend on the river above the Little 
Park and the meadow called the Moore Dee, along the 

Digitized by VjOOQ IC 


south side of which it proceeds, leaving Lord's Meadow 
in Hewlington on the left hand, and so by the north 
side of the northernmost Tir y garreg [^Land of the 
stone, now called " Tithe Garret "] to Hewlington Lane, 
which leads from Sutton to Holt, then northwards to 
Gallowtree Lane, and westwards by Gallowtree Lane to 
Franchise Lane (now called "Francis Lane"), and by 
that lane westwards on the whole to Hugmore Lane, 
formerly sometimes known as '* Croes locyn Lane," 
along which it proceeds in a northerly direction, until 
it marches with the eastern boundaries of Gourton and 
Burras RiflFii to Common Wood Lane, a continuation 
westwards of Frog Lane. Eastward along that lane it 
runs until it reaches the roadway to Red Hall, opposite 
the end whereof it strikes in a north-easterly direction, 
touching the AUington boundary, having The Lodge 
just north and Plas Devon just south of it, and so 
reaches the Devon near Wearhookes Bridge ; and fol- 
lowing the Devon and AUington boundary comes to 
the Dee, and so southward to Holt Bridge again. 

Next follows a list of all the freeholders and bur- 
gesses of Holt in 1620, with a summary of the descrip- 
tion in the Survey of their burgages, houses, and lands. 
But the names of the freeholders are here arranged 
alphabetically y^ so that they can better be referred to, 
and there is given, under the names of the more import- 
ant men a short history of the families to which they 
belonged : it being thought that this is the most con- 
venient place to present what could hardly be intro- 
duced elsewhere. There were sixty-five freeholders and 
burgesses in all, tenants of the Prince, namely : — 

Owen Brereton, Esq., was of Burras Hall (see pedigree of 
the Breretons of Burras, opposite page 162, of my History of 
the Country Townahips of the Old Parish of Wrexham), He 
had two burgages in Frog Lane, of which one was *' neere vnto ye 
Crosse," and the other " neere the Pinfold/' and about 130 cus- 
tomary — or 275 statute — acres of land within the franchise. 

George Bostocke, Esq . had his capital messuage somewhere 
in the town of Holt, which, with the bams, outhouses, garden, 
^ The arrangement in Norden's Sw^ey is not alphabetical. 

Digitized by VjOOQ IC 

408 THE TOWN O* HOLt, 

and orchard appurtenant to the same, represented the site of six 
burgages. He had also nine other several burgages, in 1620, in 
Pepper Street, Castle Street, and elsewhere in the town, and 
much land in the franchise, the exact area of which is uncer- 
tain, though it was undoubtedly considerable. He belonged to a 
notable family — the Bostockes of Churton — of whom no account 
is given, under Churton, in Helsby's Onnerod'a Cheshire, later 
than the time of Henry VIII, although they were seated there 
long afterwards, for George Bostocke, of Churton, Esq., was 
buried at Famdon 4th March, 165|.i The above-named George 
Bostocke was Mayor of Holt in 1620, and buried there 24th 
December, 1627, being son of Lancelot Bostocke, by Jane his 
wife, daughter of Richard Roydon, of Holt, Mr. H. R. Hughes, 
of Kinmel, confirmed this account of the parentage of George 
Bostocke, and sent me a pedigree copied from one of the Halston 
MSS., which I have abbreviated, added to, aud already presented 
in Appendix II, Chapter II. His will is dated 17th September, 
1627, whereby he bequeathed to his " nobell and good frend. Sir 
Robert Chumley, baronett," and his brother-in-law, " Henrie lea 
esquire," all his burgages, messuages, lands, etc., " in the town 
and liberties of lions, alias Houlte,*' in the holding of various 
persons named ; also " one House in the Castell Streete wherein 
Richard Roydon dwelleth," half an acre " in the place called the 
sent marie loundes vsed to the said house," one croft called 
** the kichen aker," and ** a croft called the intake ajoining" ; so 
that by the sale of a part thereof they might pay what he owed, 
namely £365 10s. to the daughters of Sir George Calveley, knight, 
deceased, unless (leorge Bostocke, his son and heir, or any other 
that at his death might be his heir or heirs, should take upon 
him or them the payment of the said sum. And he gave to his 
wife Dorothy [who was perhaps the *' Mrs. Dorothy Bostocke" 
buried at Holt 3rd November, 1678] the rest of his goods and 
chattels. The Mr. George Bostocke whose will has just been 
summarised appeai^s to have been followed at Holt by his son 
George. In any case, we find, a few years afterwards, a George 
Bostocke, of Holt, Esq., a captain of the local levies raised for 
Charles P, concerning whom Philip Henry writes in his diary, 

^ It may be permitted to give here a few other extracts from 
Farndon Registers relating to the Bostockes : 

3 Nov., 1620. Mr. George Bostocke, baried [of Churton]. 

19 May, 1682. Katherine, wife of George Bostocke, Esq., buried. 

30 Aog., 1634. Ann, da' of George Bostocke, Esq., buried. 

16 Oct., 1658. Mrs. Elinor Bostocke, widow, buried. 

* Mr. W. M. Myddelton, of St Albans, tells me that Mr. Bos- 
tocke, on 23rd April, 1663, certified that Thomas Sowne, of Isooed, 

Digitized by VjOOQ IC 


under date 14th August, 1663, thus : " Mr. George Bostock dyed. 
His death occasioned by a surfet of drink which hee took at 
y« time of y® quarter sessions at Uanrust, whether hee had 
bound over certain of y« Inhabitants in and about Wrexham 
who were deprehended at y® meeting, to their no small trouble. 
And now just before the Assize y^ lord hath taken him away 
by a remarkable stroke," etc. Whether he died of a surfeit of 
drink or not, Mr. Bostocke's death was certainly sudden. In 
his will, dated 3rd August, 1663, in which he describes himself 
as of " Plas Bostock within the Libertyes of Lyons als. Holt," he 
desired to be buried " in my vsuall burying place in my Chan- 
cell within ye parrish Church of Lyons ak. Holt." He gave to 
his " cozen John Pulford's wife of Wrexham" his mare ; to his 
friend, Richard Alport, of Overton, Cheshire, Esq., his " gray 
nagge ; 40« to his friend, John Jeffreys, of Acton, etc., and all 
his lands and tenements to his well-beloved nephew, Lancelot 
Williams, second son of his brother-in-law, Thomas Williams, 
of Abenbury vawr, gent, [the Plas Jenkin estate extended into 
Abenbury Fychan and Dutton Dififaeth, and included Cae 
Mynach], provided that the said Lancelot assumed the surname 
and quartered the arms of Bostocke ; and in defect of heirs male, 
to the third, fourth, or fifth sons of the said Thomas Williams, and 
to their heirs male respectively." Lancelot Bostocke, alias 
Williams, the devisee, died apparently unmarried and without 
issue, and was buried at Holt, 1st January, 166f . It is not 
possible to speak with any certainty as to the subsequent 
history of Mr. George Bostocke's estate. The " Mr. Robert Bos- 
tocke, of Iscoyd,** who died 13th November, 1670, may have 
come into the property, and been another younger son of Mr. 
Williams. Also a second Lancelot Bostocke was baptised at 
Holt in April, 1665. A Thomas Williams, of Sutton, Gent., 
probably the father of Lancelot Williams, alias Bostocke, re- 
nounced his interest in the will of George Bostocke on 14th 
October, 1664. This was perhaps the Thomas Williams, of Plas 
Jenkin, who was buried at Holt, 12th January, 16|f . However 
this may be, I have seen the will of another Thomas Williams 
of Place Jenkin, Esq., dated 20th February, 170|^, proved 17th 
March, 170|-, who directed his body to be buried in his chancel 
within the parish church of Lyons, alias Holt, and left all his 
landed estate, subject to certain legacies, to his son and heir, 
Peter Williams, who succeeded him, and to his lawful heirs 
male, or in default to testator's nephews, Lancelot Bostocke, Esq,, 
and John Evans, gent, equally to be divided among them, or in 

had been a foot-soldier, and Thomas Holt a sergeant in bis company 
in Sir John Owen's regiment. 

6Ta s£B, VOL. vn. 27 

Digitized by VjOOQ IC 


(iefanlt to his own right heirs. To his said sou Peter all his 
personal estate whatever, and the debts due to him by virtue 
of the will of testator's brother, John Williams, Esq., deceased, 
son Peter to be executor, and to be assisted by Mr. Richard 
Jhomas, of Borras Rififri, and Mr. Andrew Floyd, of Sutton. 
The testator bequeathed small legacies to his sister, Mary Evans, 
his nephew, John Evans, and his niece, Elizabeth Evans., 
Provision was made, in 1717, for the tuition of Thomas Williams 
of Plas Jenkin. After this I cannot find Plas Jenkin so much 
as mentioned, except as the name of a field. In order to cast 
some light on the connections of the Churton Bostockes, and as 
illustrating their relation to William Burganey, it may be said 
that in The Cheshire Sheaf for 1891, page 57, a letter is printed 
from George Bostocke, of '* Ohorton^* [Churton], dated 18th 
March, 1642, in which he speaks of his cousin, William Bur- 
ganey, as having a son at Oxford. This son the late Mr. J. P. 
Earwaker identified as William Burganey, son of William 
Burganey, and grandson of Anthony Burganey, of Pulford [and 
Holt], who matriculated from Corpus Christi College, 7th July, 
1637. In the same letter Mr. George Bostocke mentions "a 
kynsman of myne," " Mr. Bostockes sonne of Acton, a minister." 
This was Nathaniel Bostock, who matriculated, Mr. Earwaker 
found, from Brasenose College, Oxford, 28th March, 1617, 
aged sixteen, son of the Rev. Thomas Bostock, of Acton, Nant- 
wich. Mr. George Bostock says further that " my Cosin Burgayny 
[that is, William, son of Anthony], his granmother was my 
Grandfather Bostockes sister." Here is a clue for whomsoever 
has the opportunity to follow it up. The wills at the Chester 
Probate Court and the entries in the Farndon registers should 
also be consulted. 

William Batha. — This William Batha held seven parcels of 
land " on the vpper end of Common Wood, lying together, to- 
gether with all buildings therevnto belonging, conteyning 
Eleaven Acres*' (about 23 statute acres), possibly where Bed 
Hall now is. 

George Buckley had a burgage and curtilage near the bridge, 
and a curtilage near " the Church Ashe.'* 

Thomas Bithell had a burgage and a-half in Frog Lane, late 
of John Yardley. 

Richard Bithell, alias Howell, had also a burgage and a-half, 
late land of John Yardley. 

William Burganey had two burgages near the pinfold and 
seven acres (or nearly 15 statute acres) called " The Bottoms" in 
the lane leading from Frog Lane towards Common Wood. He 

Digitized by VjOOQ IC 


was, probably, the son of the "Anthony Burganey of Holt/'^ an 
inventory of whose goods, taken in 1611, is in the Chester Pro- 
bate Office (see also above, under George Bostock), and is to be 
identified with William Burganey, whose grandson, also named 
William Burganey (son of the William who died in 1732), 
married Rachel, daughter of Handle Holme, of Chester. The 
grandson of this last-named William Burganey, namely, John 
•Burganey, of Pulford (son of John, son of William), married at 
Gresford " Miss Anne Pate," of Croes Howel, in Burton, county 
Denbigh ; from which marriage the Burganeys appear to have 
obtained Llyn Tro and other messuages and lands in Burton and 

Thomas Calcott, Gent., had two burgages in High Green, 
one burgage in Smithfield, and another in Castle Street, Holt, 
but does not appear to have lived in any one of^ them. He had 
also a piece of land in or near Wrexham Lane, called ** Annes 
hey goch" [that is, " Ynys hey goch," or perhaps Ynysau cochion] 
(see next page). He was son of Handle Calcott of Caldecote, 
often pronounced "Calcott," or even *' Cawkott," Cheshire, by 
his wife, Jane, daughter of Alban Butler, and was Mayor of 
Holt in 1631. He married Dorothy, daughter of John Dod, and 
had a sou of the same name, probably the " Thomas Caldecot of 
Caldecot, Gent," who was buried at Farndon, 1st October, 1672. 
But there were so many branches of this family, and so many 
Calcotts bearing the same Christian name, that it is difficult to 
speak with any confidence. However, it seems certain that Thomas 
Calcott, or Caldecot of Caldecot, had a son William, who died a 
few years after his father (in December, 1677), and that the 
"Mr. Caldecott of Caldecott" mentioned on I6th June, 1690, 
was Robert Caldecott. I give four extracts relating to the 
Caldecotts from the Holt registers ; and there are many more in 
these and in those of Farndon which it is not worth while to 
reproduce : — 

Thomas, son of WUliam Caulcot, of Holt, Qent., bapt. 5th June, 1675. 
William Caldecotte, of Caldecotte, Qent., died in Holt, and was buryed in 

Farndon, 30th December, 1677. 
Richard Craven, of Ridlej, G't., aad Mrs. Caldecote, of Isacoyd, married 2nd 

June, 1699. 
Thomas Caldecote, of Holt, buried 30th January, 170J. 

Edward Clough, of Common Wood, a small holder. 

Edward Crew, Gent., had in Holt one burgage *' in Midding 
streete, neare the Crosse, where his Mansion house standeth" ; 
three other burgages in the same street, " wherevpon a Barne 

* Anthony Burganejr'p widow married, before 1620, Humphrey 

Hanmer, Qont. 


Digitized by VjOOQ IC 


standeth/' one parcel of land near the Cross containing one bur- 
gage, and the House called "y® Comon Backhouse*' (Bake- 
house), and 21 customary or nearly 45 statute acres of free 
lands, and about 6 acres of leasehold land called " the Ladies 
Landes." At the same time, an Edward Crew, possibly the 
same person, held a dwelling-house in Receiver's Street (now 
Queen Street), Wrexham, and much land in the same town. 
Edward Crew was buried at Wrexham, 19th January, 163|* 
More will be said as to the Crews of Holt in the next entry. 

Thomas Crew, or Crue, Gent — This Thomas Crue lived in his 
capital messuage near Holt bridge, with another burgage, belong- 
ing to him, adjoining it ; five other free burgages and one leasehold 
burgage; and among his lands were the Dovehouse Croft, the Stony 
Croft, "The Wallock conteyning six Acres'* (equal to about 12 J 
statute acres) ; " hilton croft or Cases Croft, of 2 [customary] 
acres" ; the Bottoms adjoining Chester Lane ; several parcels 
called "Gillwall," an acre adjoining called ''Agnes hey Gough'*; 
a parcel of land in a close called " The Espes," etc. It is clear 
to me that his house was that now represented by Holt Hill. 
The Crews, or Cnies, formed an important family, originating at 
Crew-by-Farndon, and establishing themselves at Holt, Wrex- 
ham, and elsewhere. I have in my possession sheafs of notes 
concerning them, which do not seem capable of being woven into 
a consistent or satisfactory pedigree. A Thomas Crew, Gent, of 
Holt, was aged 51 in 1597.^ Another of the same name was 
buried at Holt in October, 1(513, being perhaps he who dis- 
covered and destroyed the " Roraane monument" described in 
the Appendix to Chapter I ; and to the memory of another 
"Thomas Crue," who died on the 12th August, 1666, age 27; 
his kinsman, Silvanus Crue, of Wrexham, engraved the remark- 
able brass aflSxcd to the east end of the north wall in Holt 
Church, of which brass a reproduction will be given hereafter. 
Yet another Thomas Crue, Gent, was buried at Holt, 28th 

* Mr. Edward Owen enables me to go still further back, by supply- 
ing me with a reference to a complaint of William Holstooke, Gent, 
against John Oruwe of the town of Lyons, and also the answer of 
Thomas Crue to the same, wherein the respondent refers to an 
indenture, dated 7th May, 13 Hen. VII, 1498, between John Crewe 
and Robert Troutbeck, in view of the marriage of Thomas, sou of 
John Crewe, and Elizabeth, daughter of Robert Troutbeck. Thomas 
Crewe also speaks, in his answer of his mother, Margaret Further, 
in the Survey of Holt, 23 Hen. VII, the following names appear : — 
Thomas Crewe, William Crewe, and the heir [or heirs] of John 
Crewe. A Thomas Crewe, Gent., was also one of the jurors in the 
jury of survey for Holt, 4th year of Queen Elizabeth, and had pretty 
nearly the same lands as had the Thomas Orue of 1620. 

Digitized by VjOOQ IC 

IN couKtt Denbigh. 413 

November, 1699 (see also under Thomas ffoster's name, p. 415). 
And, on 28tb July, 1749, Samuel Crew, of Common Wood, Gent., 
was party to a deed wherein he is described as eldest son and 
heir-at-law of Thomas Crew, of Holt, Gent., deceased (buried at 
Holt, 28th June, 1741), and as having two sisters, Eebecca (with 
whom £600 had been paid as marriage portion), married to 
Thomas Dod, of Edge, Esq., and Christian, who afterwards 
married Mr. John Jones, of Pentref, and was a widow on 6th 
June, 1752. It is possible that Holt Hill, one of the houses of 
the Holt Crews, came to the Joneses of Ynysfor, Penrhyn Deu- 
draeth, through the marriage last named. Samuel Crew, Gent., 
then tenant of Cornish, was buried at Holt, 18th January, 1770. 
However, it is almost as dangerous to speculate as to the Crews, 
who, according to the proverb, were as " common as crows," as 
it is dangerous to speculate as to the Joneses. 

George Cowes had a burgage and curtilage in Frog Lane, 
also a close of land "neere vnto Devon," in 1620. 

Egbert Davies, Esqr. was of Gwysaney, near Mold, son of 
the first Robert Davies of the same. He married Anne, only 
daughter of John Heynes, by Elizabeth his wife, which Eliza- 
beth was one of the daughters and co-heiresses of Lancelot 
Lothar, of Holt, Constable of the Castle.^ He held, in 1620, 
three of the four shops underneath the Town Hall. On 28th of 
October, 1664, Thomas Speed declared by deed that what title 
he might have to a certain pew in Holt Church was subject to 
the right of Robert Davies, Esq., of Gwysaney, in the same pew 
(see Arch. Camb., 1878, p. 147). 

WiLUAM Davies held a burgage, wherein he dwelt, in Midd- 
ing Street. 

William Davies, tanner, had also a burgage in Midding 
Street, " an acre" adjoining, and a parcel of land in *' ifishpoole** 

Egger Decka held a curtilage in Smith field Green. 

Roger Edgworth held a free burgage in Wrexham Lane, 
wherein he dwelt, and a parcel of land in fishpool field. He 
had also two parcels of leasehold land in Hewlington, namely. 
Cunning's land and "ynys croft dyon," formerly in the tenure 
of Thomas Edgworth, The Edgworths constituted a noted local 
family, which branched out afterwards to Wrexham, Hoseley, 
March wiel, and elsewhither; Thomas Edgworth, of Bryn y grog, 
Marchwiel, becoming the first Mayor of Wrexham in 1857. In 
1784, the messuage, bam, and other property in Holt then lately 

^ It was through this marriage, perhaps, that Mr. Davies became 
a freeman of Holt, and acquired lands there (see under Richard 
Hooker, below). 

Digitized by VjOOQ IC 

414 THE TOWN Of HOLf, 

belonging to Mr. Thomas Edgworth, deceased, were ofifered for 
sale at the " Red Lion" there, the family thus denoting their 
final severance from Holt. There is a fuller account of the 
Edgworths on pp. 194-197 of my History of, the Town of Wrex- 
ham, etc. The will of the Roger Edgworth named in the list of 
1620 was proved at Chester in 1629. He was followed by 
another Roger Edgworth, " gent., and public notary," who was 
an official for one of the Parliamentary Committees during the 
Protectorate, and died in May, 1668. 

Sir Richard Egerton, knight. — He had a house at Common 
Wood, in occupation of Roger Green, and a little over 14 cus- 
tomary acres appurtenant there ; a burgage iu Smithfield Green ; 
3 other free customary acres, and various leasehold lands in 
Hewlington, of which a map is given in Norden's Survey, 
namely, Maddock's Moor, Dolvawr or Dolwern, and the Elties 
or Elthie. He was son of Ralph Egerton, of Ridley, Cheshire 
(who died in 1619), and grandson of another Sir Richard Egerton, 
whose widow, Mary, Lady i^erton, speaks in her will, dated 
18th October, 1597, of her lands in " Holte, als, the Towne of 
Lyons, Alington, als, Trevalyn,'* eta The Sir Richard Egerton 
of 1620 died at Ridley, 24th February, 1627, and was buried at 
Bunbury (see Ormerod's Cheshire). Peter Egerton, half-brother 
of Sir Richard, son of Ralph Egerton, was, possibly, the Lieut.- 
Col. Peter Egerton who helped to capture Holt Bridge for the 
Parliamentary party in November, 1643. 

William Fisher had a burgage in Midding Street, and 
3 customary acres of land. 

John ffL etcher had three burgages, forming the site of a 
house in Wrexham Lane, 10^ customary — or about 22 statute — 
acres of land iu Croes locyn Lane, and certain lands " of ancient 
demesne" set by lease to him. 

Thomas ffosTER, Gent., had one burgage in Castle Street and 
no other holding in Holt. There were two Thomas ffosters, the 
elder and the youuger, of Parkside, Allington ; see my History 
of the Townships of the Old Parish of Gresford, pp. 147 and 
179, where I have, by mistake, made the elder ffoster's wife, 
Dorothy, to be a daughter of Richard Roydon, of Holt. She 
was, in fact, a daughter of John Roydon of Jsycoed by his wife, 
Anne, daughter of Richard Chambers, of Sussex, as shown in the 
College of Arms pedigree. The Thomas flfoster of 1620 was 
probably the elder, and he whose will was proved at Chester in 
1636. Thomas ffoster, the younger, afterwards lived at Holt, 
where he bought various houses and lands, and served the oflBce 
of Mayor in 1642 ; his will is dated 23rd December, 1675, and 
was proved 17th January, 167f . He desired to be buried in the 

Digitized by VjOOQ IC 


upper end of the south aisle of the parish church of Holt ; and 
left the messuage in which he lived and all his messuages and 
lands in the town and liberties of Holt, purchased of Soger 
Edgworth, Mr. Samuel Davies, Thomas Taylor, Peter Taylor, 
subject to legacies, fo his wife Jane, whom he appointed sole 
executrix. He bequeathed to his cousins, the brothei*s and 
sisters of his cousin Handle Crue, of Holt, £160 ; namely, to 
Samuel Crue, £20 ; to William Crue, £100 ; to Dorothy, wife of 
John Gough (see below), £20, and to Elizabeth, wife of John 
Powell (see p. 421), £20, etc. ; he bequeathed also to his niece, 
Elizabeth, relict of Christopher Dutton, £4 a year for life, to 
issue out of the further hall field ; to his cousin, Elizabeth, wife 
of John Jones, of Darland Green, 40«. ; and besides some minor 
bequests, 625. yearly to twelve of the more aged and indigent 
poor of the town and liberties of Holt for ever, to take effect 
immediately after his own decease, and to be distributed every 
sabbath day in bread, the said 52«. to be secured on a field 
called " The two acres," alias ** The Espes." Then, after the 
decease of testator's wife, all his estate, goods and chattels, were 
to go to his cousin, Handle Crue, of Holt [a son, apparently, of 
a Thomas Crue of the same] for life, and afterwards to Samuel, 
son and heir apparent of the said Handle and his lawful issue 
male, or in default to Thomas Crue, Handle's second son, and Ms 
lawful issue male ; or in default to William Crue, a younger 
brother of the said Handle ; or in default to Handle's right heirs for 
ever. Mr. Thomas ffoster was buried at Holt, 31st December, 
1675, and his goods were valued on 5th January, 167|-, by Joseph 
Powell and Thomas Edgworth, at £345 17^. 8d. Mrs. ffoster, 
the widow, was buried at Holt, 5th June, 1689. 

Jane Gerard was daughter of William Aimer, Esq., of Pant 
locyn, and widow of Gilbert Gerard, of the same, son of Sir 
William Gerard, knight. She had, in 1620, two burgages in 
Castle Street, one burgage in Frog Lane, and 16 customary — or 
nearly 34 statute — acres of land in Holt. 

John Godson had a burgage near the bridge, next that of 
George Buckley, on the south side of Church Street. 

John Gough had two burgages in Frog Lane, and another 
John Gough, or the same, had a few acres of free land. 

HoGER Greene, besides being Sir Hi chard Egerton's tenant 
at Common Wood, had 12 customary acres of leasehold land 
between Common Wood and Wrexham Lane. 

Edward Griffith had, in 1620, "one Burgage wherevpon 
the Mansion-house late of William GriflBth standeth," one other 
burgage and 18 customary — or 38 statute — acres of free land. 

Richard Hooker, clerk, "holdeth one Aere and a half of 

Digitized by VjOOQIC 


Land wherevpon a Howse now standeth with all buildings to 
the same belonging, sometimes the landes of George Lothar, 
late the Landes of John Henry and Robert Davies Esquire and 
Anne his wief, and late of Edward Warmingham f also a kiln 
near thereto. 

Randolph Hutchins. — His capital messuage stood on the 
site of three burgages. He had one other burgage, 26^ cus- 
tomary acres of land, " in Huejmore and Cornish," 4 customary 
acres "in ffishpoolefield," etc. He died 17th July, 1624, leaving, 
by his wife Margaret, a son, Thomas (see Owen's Catalogue of 
Welsh M8S., etc., Part II, p. 135). This son, Thomas Hutchins, 
gent., and Dorothy, his wife, were parties to a fine levied 17th 
October, 1660, on a farm at Hugmore, now known as *' Hayes* 
Farm/' at the corner of Hugmore Lane and Wrexham Road. In 
the 23rd Henry VII, ** John hychen" was a freeholder of Holt, 
and '* hychen" is merely the Welsh way of spelling " Hutchen." 

John Jenison lield 6 customary — or about 12^ statute— acres, 
part of the 30 acres late the lands of Thomas Pulford. 

Owen Jones, gent, of Glan y pwU (see my History of the 
Country Townships of the Old Parish of Wrexham, p. 158, £md 
elsewhere. He had 9 customary acres of free land in Hugmore, 
and 5 customary acres of leasehold land. 

Thomas John Lewis, of Burton, had three burgages in Frog 

Edward Maddock had a dwelling-house in Pepper Street, 
his curtilage adjoining the pinfold, late the land of Peter Roy- 
don ; a burgage adjoining the garden of Greorge Bostock, Esq.; 
a parcel of land called " The Espes adioyning to the pavement 
leading from fifrog lane towardes the Common Woode,'' and one 
of the four shops under the Town Hall. 

William Nicholl had three burgages and J of an acre be- 
tween Smithfield Green and Fishpool field, late the lands of 
Edward Puleston. 

Thomas Pate, gent. — He had, in 1620, three burgages repre- 
sented by his house, and 5 customary acres near Hall field ; one 
other burgage and 27 customary acres of land ; a parcel of 
land " in hie greene," with cottage built thereon, lying in breadth 
between Wrexham Lane and the said green, containing half a 
burgage and the twelfth part of a burgage. And in the manor 
of Hewlington he had four closes of leasehold land called " Tier 
garregge," that is, '* Tir y garreg,^' or Land of the stone, now 
known as the ** Tithe garrets.^' A certain Thomas Pate, of Holt, 
gent, deposed, in 1597,^ that he was then fifty years old. 

^ Jankyn Pate, senior, Alice his mother, and Richard Pate are 
also named among the tenants of Holt in the 28rd year of Hen. VII, 
and William Pate among those of 4 Queen Elizabeth. 

Digitized by VjOOQ IC 

IN CotTN'TY Denbigh. 417 

The will of the Thomas Pate of 1620, mayor of the town in 
1618, was proved at Chester in 1621. Also, another Thomas 
Pate, of Holt, married Martha, one of the daughters of John 
Powell of Holt, the fighting Puritan. The Pat^s were a family 
of considerable local influence, established not merely in Holt, 
but in Allington, Burton, Wrexham, Farndon, Shocklach, and 
elsewhere. In Holt itself were two branches of them. On the 
11th April, 1625, William Pate, of Holt, gent., entered into a 
prenuptial agreement with Thomas Edge, of Hope Owen, Flint- 
shire, in view of his marriage with Grace Edge, which marriage 
soon after took place; and on 18th October, 1622, William 
Pate' settled, in trust for his wife, upon Thomas Edge and John 
Meredith, of Allington, gent, among other lands, *'Gwern 
Saeson," in Cobham Isycoed, " the tieth garregs" in Hewlington, 
and various quillets of land and meadow in Caeca Button. 
Later on, we find Ferdinando Pate, of Holt, gent, (whose wife's 
name was Mary), in possession of the abovenamed lands, which 
afterwards were purchased for the poor of Wrexham. The son, 

1 The abovenamed Grace, wife of William Pate, was buried 2l8t 
March, 1664, and her hasband 29th March, 1675. Alno the Thomas 
Pate, senior, who was buried 3rd March, 166|^, is described dis- 
tinctly as brother to William Pate, gent. ; so that the Thomas Pate, 
of Holt, who was buried 8th February, 170f, was very likely a son 
of Thomas Pate, senior, and nephew of William Pate. [Since 
writing the foregoing, I have seen a copy of the will of Thomas 
Edge, of Hope Owen, father of Mrs. Grace Pate, dated 26th Decem- 
ber, 1634, proved Ist March, 164J, wlierein, after bequeathing varions 
sums of money to his nephews and nieces surnamed Edge, he 
devised to his grandchild, Thomas Pate, son of William Pate, of 
Holt, all that dwelling-house, with the buildings and lands thereto 
belonging, *' in the hoult afoi*esaid for the purchase whereof I have 
paied Nyne pounds in earnest vnto John Presland and Lancelott 
Presland of the holt aforesaid ;" or if the said bargain came to no 
effect, then he bequeathed to the said Thomas Pate £86. Whether 
this bargain of sale was ever realised does not appear. But the 
testator left his wife, Custance, all his leasehold lands and tene- 
ments, she bringing up, maintaining, taking order for her learning 
and preferring, his granddaughter, Elizabeth Pate, daughter of the 
said William Pate : and if the said Elizabeth Pate declined to live 
with his wife, Custance, by reason of marriage, or any other cause, 
then the testator's will was that all his lands — leasehold and pur- 
chased — should be divided into two equal parts, whereof the one 
part should be enjoyed by his wife for the term of her life, and the 
other by said Elizabeth Pate, who, on his wife's death, should have 
both the parts, to hold to her and the heirs of her body, or in default, 
to the said Thomas Pate, and the heird of his body ; or ia default, 
to William Pate, son of the said William Pate.] 

Digitized by VjOOQ IC 

418 TUB TOWN O^ ttOLT, 

Ferdinando, of this Ferdinando Pate, was baptised 26th Auj^ust, 
1723, and died at Poulton, 26th January, 1812, being ninety 
years old ; and one of his sons, John Pate, of Chester, upholsterer, 
married 24th January, 1814, Sarah Pate, thus uniting the two 
Holt branches of the Pate family. This Sarah Pate was a 
daughter of the Thomas Pate of Holt who died 22nd Decem- 
ber, 1816, aged seventy -nine, and a sister of the Thomas Pate 
of Holt, mayor of the borough in 1819, who died 6th November, 
1823, aged fifty-two. Mary, daughter of the last-named Thomas 
Pate, by Sarah, his wife, married Moses Steven, of Chester, and 
the present Mr. Thomas Pate Steven is their grandson, being son 
of John Pate Steven, who died in February, 1875. The house 
of the Thomas Pate branch still survives opposite Ainsdale, on 
the Wrexham road, and will be described hereafter. 

Margerie Phillips had three customary acres of free lands 
in Hugmore, sometime land of Lancelot Pulford, and then late 
of William Batha; 12 acres of leasehold land between Common 
Wood and Wrexham Lane ; and over 5 acres called " says hey 

Francis Pickering, gent, had, in 1620, three burgages repre- 
senting the site of his mansion-house near the churchyard : the 
little Wallock (see Chap. I, p. 11) ; the " little annes hey " [*' Yr 
Ynysau*' — the holmes] on north side of Wrexham Lane ; and 
a parcel of land appurtenant to his mansion house, whereon a 
barn was built, adjoining the castle ditch. Mr. Francis Pickering 
was mayor of the town in 1632, married Margaret, sister of 
Robert Worrall, and died 3rd September, 1635, leaving a sou, 
Francis, who, a mere youth, was one of the garrison in Holt 
Castle when it was held for Charles I, and, after its surrender, 
was fined £70 by the Sequestration Committee. This Francis 
Pickering the second, leased for eleven years, on 24th Decem- 
ber, 1640, a parcel of land called "Annesse gouch " (Ynys 
goch — the red holme) then or lately in the tenure of Koger 
Edg worth, of Holt, to Thomas Baker, of Wrexham. He was 
still living in J 656. Yr ynys Goch is on the north side of 
Wrexham Road (see the map prefixed to Chap. I). A John 
Pickering was doorkeeper of Holt Castle in the 21st year of 
Henry VII, and in the 10th and 11th years of Henry VIII ; 
and in the 14th year of the first-named king there was ap- 
pointed, as Receiver of Bromfield and Yale, Sir Edward Picker- 
ing, who, as Mr. Hughes of Kinmel thinks, must have been son 
of Sir Christopher Pickering, of EUerton, Yorkshire, by his 
second wife (Ellen, daughter of Sir Richard Haryngton, knight). 
Sir Christopher's first wife was Mary, daughter of Sir Robert 
Lowther. The surname " Pickering " still survives at Holt 

Digitized by VjOOQ IC 


John Platt, glover, had a burgage and a half near the Town 

John Presland had a burgage and some land in Holt. There 
was a Richard Presland who lived in the manor of Isycoed, next 
adjoining the franchise, at the same time. Doubtless, John Pres- 
land was related to him, and the Preslands of Ridley and the Pres- 
lands of Presland and Wardle. Cheshire, were of the same stock, the 
eldest son — or one son at least, in almost every generation — being 
called Richard. When the Earl of Bridgwater purchased the manor 
of Ridley from the Crown, he seems to have ignored entirely 
the composition made by the forty years* leaseholders with the 
Queen's officers, and the right of renewal by the tenants of their 
leases, treating them as tenants-at-will, or giving them leases 
for lives at arbitrary fines. In October, 1622, the Earl granted 
a lease to Richard Presland, the elder, of the house and lands 
he then held in Ridley, for ninety-nine years, if he, the said 
Richard, Robert Presland his son, and Katherine his daughter, 
should live so long. Katherine Presland just named became 
afterwards the wife of Captain Edward Taylor (second son of 
Thomas Taylor, of Dutton Diffaeth, yeoman), a famous Par- 
liamentary officer, who had, with his wife, the reversion of a 
lease of one of the farms called " Parkey," in Bedwall. These 
Preslands and Taylors were, at the time of the Civil War and 
afterwards, strong Presbyterians. Richard Presland, the elder, 
had, among other children, Richard Presland, the younger, 
Nathaniel Presland, and Mary Presland. His widow, Katherine, 
became the second wife of Edward Thomas, of Wrexham, one 
of the local officials of the Parliamentary Sequestrators, who 
had, by his first wife, two sons, namely, Jonathan Edwards, 
who was afterwards Vice-Chancellor of Oxford, and Samuel 
Edwards, often mentioned by Philip Henry in his diary. Nearly 
all these particulars have been gleaned from Presland deeds 
perused by me. A Thomas Presland was bailiff of Hewlington, 
by Holt, in the 10th year of Henry VIII : perhaps the same 
Thomas who was one of the witnesses to the will of John Roden, 
Rector of Gresford, made 24th June, 1506. There was also a 
certain Lancelot Presland, son of John and Alice Presland, who 
was living in the 44th year of Queen Elizabeth ; and a John 
Presland and Lancelot Presland sold before 26th December, 
1634, to Thomas Edge, of Hope Owen (see p. 417, note), a 
messuage and lands in Holt. 

Makgaret Pova had a burgage in Wrexham Lane, between 
the burgages of John Read and John ffletcher. 

John Powell had a burgage in Frog Lane, Holt, where he 
lived, formerly belonging to Edward Aimer, William Aimer, 

Digitized by VjOOQ IC 

















a- S 










© "*^ =3 00 *"' 

.a --^ ij « 
o • -^ * 


3 X 


©I'd S c «> 

^ O y . 

Digitized by VjOOQ IC 








a.S ^,| §.3 









>lt, G 
d 26 
J bur 
,15 J 






= Eh-7^1 
of Th 
of H( 









rtha, bur'd John Powell of 
t Holt, 30 Holt, the Non- 
ug. 1699. conformist, and 
benefactor to 
the poor of 
Holt; died 
29 Sept. 1706. 







1715/, Mayor 25 Sept. 1720 
buried there, 29 July, 1807, 

Had issue. 

ishe of Howlte."-A. N. P. 
of Sutton, in the Parish of the 


well of H 
hester. So 
t, 28 April, 
It, 1747 ; 1 
[y, 1763. 

in the Pari 
%p Howell, 
k."— A. N 

•e "o 7^ JT i: 


== 2 -g 

— ?."|i- 





eS O (N 

Z O 9 



Geo. Row- 

Farndon ; 

Feb. 1685. 

born 23 







himself in his will " John a 
himself in his will " Harry< 
nvericke" for " Cynwrig," o 
Hugmore Lane." — A. N. P. 


-Frances. = 1 
land of 
m'd 16 

Oct. 16 


-Samuel, bo 


o - 







J of 






V V 


P^ C* « f 


owell ; 
>f his 
r, 26 




John P 
bur'd i 
time ( 








Digitized by VjOOQ IC 


and Jane Gerrard, and no other lands there. Nor was there, in 
1620, any other freeholder of Holt bearing that name. He was 
afterwards a Parliamentary soldier, his will being dated 18th 
November, 1644, and not proved until 3rd December, 1664. 
There was, it would appear, another John Powell of the parish 
of Holt living at this time, but he called himself "John ap 
Howell," and was not a burgess. Both these were derived, 
according to the late Mr. Ellison Powell (see annexed pedigree), 
from an earlier John Powell, who is said to have been a 
younger brother of the first Thomas Powell, Esq., of Horsley ; 
and so late as 1904, when I sent to the press my pedigree of 
the Powells of Horsley (in my History of the Townships of the 
Old Parish of Gresford, opposite p. 118), I accepted this deriva- 
tion without reserve. But further investigation gave rise to 
grave doubts. Mr. Ellison Powell spent an immense amount of 
labour in tracing the origin of these Holt Powells; and in 
justice to him, as well as to show the point of the criticisms 
about to be made, I print the accompanying abbreviated 
pedigree, compiled almost entirely from Mr. Ellison Powell's 
book, in which pedigree the two possible John Powells of 1620 
are indicated by putting their names in italics. Indeed, the 
pedigree is correct beyond doubt, if we start in the one case 
from John Powell whose will was proved on 15th December, 
1638, and in the other from Harry Powell, who is said to have 
been his brother. But, first of all, I cannot find any evidence to 
show that these two were really brothers: there is certainly 
nothing in their wills pointing to any such relationship Next, 
in 1589, the aforesaid Harry does not call himself " Powell" at 
all, but simply " Harrye ap John ap Howell,'* naming his 
brother " William ap John ap Howell," and his own two sons 
*^ John ap Harrye and "Rauf ap Harrye," although John ap 
Harrye, after his father's death, seems to have adopted ** Powell " 
as a surname, and become John Powell, the Parliamentary 
soldier aforesaid. But the fact that he took this surname does 
not prove that he was of the Horsley stock, nor does his father's 
name — " Harrye ap John ap Howell " — establish a derivation 
of the sort indicated. Finally, there is no hint in any of the 
wills of the Powells of Horsley, known to me, pointing to any 
relationship with the Powells of Holt, of either stock. I feel 
bound to make these criticisms, although the possibility — ^the 
bare possibility — may be admitted of some earlier will coming 
to light which shall prove the connection for which Mr. Ellison 
Powell contended : and am glad that these " historic doubts" did 
not occur to me during that gentleman's lifetime,^ and so have 

1 The fact that Joseph Powell, of Cornhill, London (son of Caleb, 

Digitized by VjOOQ IC 


possibly interrupted a long, valued, and most fruitful corre- 
spondence. A few notes of the principal bequests contained in 
the will (made 27th June, 1706, proved 15th March, 1707) of 
John Powell, of Holt (son of Alexander Powell) may be in- 
teresting here. The testator gave to the poor of Holt for ever 
" all that parcell of land adjoining southward to the lande of 
Thomas Passnage the Elder, of Holt aforesaid, Ralph Churton 
of Aldford, in the County of Chester, and other land along the 
Ditch to the land of Caleb Powell afores**, westward to a lane 
called Chester Lane, and on the north and east parts thereof to 
other lands I lately purchased of Mr. Eddowes, Ironmonger, of 
Whitchurch, in the county of Salop, together with the s<* parcell 
of land, the rents, issues, and profits thereof." He bequeathed 
also two other parcels of land, purchased from the said Mr. 
Eddowes, to ** Master Long, now a Nonconformist Minister of 
the lately new-built Chappell in Wrexham" [Chester Street] for 
life, and to his successors, ministers of the said chapel, for ever. 
He left all his personal estate to his kinsman, Mr. Thomas Crue, 
and to Mr. Thomas Billington, both of Holt, and appointed them 
his executors. Thomas Billington, gent., was buried at Holt, 
3rd April, 1734. The house of the Powells still stands in Frog 
Lane, Holt, although divided into three or four tenements. In 
1843 it belonged to Mary Powell, who had in the borough 
nearly 29 acres of land. 

. Thomas Pulford, of Barton, and Thomas Pulford, of Holt. — 
Thomas Pulford, the elder, of Barton, was buried at Farndon in 
April, 1628. The Pulfords were a wide-branching family, 
originating, doubtless, at Pulford, but connected mainly with 
Holt, Farndon, and Wrexham. An account of them is given on 
pp. 20, 34-36, and 186, 187 of my History of the Tovm of 
Wrexham, etc. To this account I might append many additions, 
but will only make a few here. About the year 1546, John 
Pulford, Lancelot Pulford, and William Pulford held at the lord's 
will 30 acres of pasture in 5 closes in Hugmore, John and William 
Pulford 6 acres of pasture on the north side of "gallowtree 
lane," and Lancelot Pulford 10 acres of land and pasture in 
three closes next " Comen Wood," and 5 acres of land and 8 of 
meadow on the south side of Wrexham Lane. And in 1620 
Thomas Wilkinson, of Farndon, Richard Vernon, and John 
Jenison had each a part of 60 customary — or nearly 127 statute 
— acres, once the lands of Thomas Pulford and of Alice and 
Joan Pulford, one of whom was the wife of William Pulford. 

grandson of the first Caleb Powell), bore in 1766 the arms of the 
Powells of Horsley, need not regarded as Qonqlosive. 

Digitized by VjOOQ IC 


In 1620, Thomas Pulford [of Holt] held by lease the Clayfield 
in Hewlington, in succession to John Pulford, whose estate he 
had. There was also a Eichard Pulford, gent., who had lands 
called '' Cae hicke'* in Holt, and died 28th February, 1630, 
leaving a son, Ferdinand (see Owen's Catalogue of iJie MSS, 
relating to Wales in the British Museum, Part II, p. 185) whom 
I cannot trace. A Mr. Thomas Pulford, of Wrexham, was in 
Holt Castle at the time of its surrender to Colonel Mytton. In 
the will of this Thomas Pulford (made 1st December, 1657, 
proved 22nd December, 1660), the testator speaks of his 
daughter, Katherine Weld, and of his only son, John Pulford. 
The John Pulford just named married Ursula, daughter of 
Alexander Walthall, of Wistaston, Cheshire, and had several 
children, among whom was Alexipider Pulford, of Wrexham, 
gent., whose mother, Ursula, married secondly George Gold- 
smith, of Wrexham, gent. Two of Alexander Pulford's sons 
were John Pulford, the Prothonotary, and the Rev. Thomas 
Pulford. These latter remarks are made so as to clear up some 
uncertainty, since dissipated, which I felt when writing the 
History of the Town of Wrexham, 

John Bead had four burgages in Wrexham Lane, whereon 
his mansion-house stood, 2 acres of land in " ffishpoolefield," and 
a parcel of land called Knight's Wood, adjoining Wrexham 

Thomas Rogers, alias Cooke, had one burgage in Frog Lane. 

John Rogerson had two burgages in Wrexham Lane, whereon 
his dwelling-house stood, and three customary acres of " ancient 

William Rogerson had a burgage and three-quarters of a 
curtilage in High Green, also IJ acre (customary) of free land 
in " little Annes goz " [Ynys goch]. 

" Roger Roydon, Esq., holdeth Two Burgages whereon his 
Capitall Messuage standeth in Castle streete*' (Norden's Survey, 
A.D. 1620). ** The same holdeth seaven Burgages adioyninge to 
the said Messuage where his oi*chard place is." "The same 
holdeth one Burgage in the said Castlestreete where his stable 
standeth late the landes of Thomas Billot ;" also the " Moore 
hall field" and " lefft hall field '' (28 acres, customary, the rent 
of the two fields being reckoned at one shilling an acre) ; also 
Mill hey and a meadow adjoining (the rent of which last three 
were reckoned at 2s. an acre (that is the site of nine burgages and 
46J customary — or over 98 statute — acres of land. " The same 
holdeth one parcell of land called Ridley wood contayning 
23 acres [=45 J statute acres] late the landes of Launcelot 

Digitized by VjOOQ IC 


Aldford.** And it is added in the Survey that the lands last 
mentioned were ancient freehold lands, granted by the charter, 
but then held by lease. Mr. Roger Roydon had also until 
recently held various parcels of leasehold land in Hewlington, 
containing 48 customary — or 101 J statute — acres ; which, how- 
ever, had been assigned on the 22nd March, 161f to Mr. David 
Speed. Mr. Roger Roydon was the eldest son and heir of John 
Roydon, Esq.,^ of Isycoed, by his first wife Anne. Roger married 
firstly Jane, daughter of Thomas Powell, of Horsley, Esq., and 
among his sisters were Dorothy, wife of Thomas fifoster, and Joan, 
wife of Edward Crewe, both named above. It is intended to deal 
with the Roydons at greater length when treating of the 
chapelry of Isycoed. , 

John Sivedale had an acre and a-half of free land, part of 
little " Annes hey goz," near Knight's Wood. 

David Speed, gent, had two burgages, whereon his mansion- 
house stood^ one other burgage, and over 17 customary — or about 
36 statute — acres of free land, besides the leasehold land in 
Hewlington mentioned above ; also the Gallowtree field there 
(on lease), and an estate at the Rossett. The Speeds were a 
notable Holt stock, although which house was the head of their 
estate there it is not yet possible to point out. John Speed, the 
antiquary, is said to have been born at Farndon in 1552, and to 
have been a member of this family. " David Speed, of the 
Holt, gent," was buried 11th April, 1633, and his will proved at 
Chester in 1639. This man it was who appeared on Norden's 
jury of 1620 ; and it is probable that to his nimble brain and 
sound knowledge is due the exposition of the case of the forty 
years' leaseholders as it stands in the preamble to the present- 
ment of the jurors of Hewlington ; the jury for Holt and 
Hewlington being composed of the same persons, and he being 
named among the jurymen, next after the mayor of Holt He 
was followed by another David Speed, recorder, who was, most 
likely, the David Speed who married at Farndon, 25th June, 

^ In the Survey of 4 Elizabeth, 1562, John Roydon, the father of 
Roger, is described as having in the town of Holt one messuage, 
two burgages, the fourth part of one burgage, an orchard, in which 
formerly were seven burgages, and five parcels of land called 
"morehalfield, lesehallfield, milnehey, harbors hey," and a parcel of 
meadow lying next said milnehey, lately in tenure of John Roydon, 
his fieither. Bat " morehalfiold ** was assigned, in November, 37th 
year of Queen Elizabeth, by John Roydon and Roger his son to 
John Hare. 

6th sbb., vol. vn. 28 

Digitized by VjOOQIC 


1623, *' Sydney P . . . bill" This second David made, with three 
others, the Parliamentary Survey of 16^. I have seen his 
nuncupative will, made I3th July. 1660, the day before he 
died, in which he describes himself as of Hewlington, appoints 
Sidney, his wife, his executrix, and mentions his son, Thomas 
Speed. The inventory of his goods was made on the 18th of 
the same month by Mr. Thomas Humberston and John ap 
Edward, who designate the deceased as " David Speede gent, 
of the towne of Hoult." His son, Thomas Speed, was buried at 
Holt, 9th November, 1664; and there was a David Speed, son of 
John Speed, gent., who was also there buried 26th December, 

Thomas Spencer, of Farndon, had 6 customary acres of land 
adjoining Knight's Wood [in Wrexham Road], formerly the 
lands of John Aldford. 

EoQER SucKLEY had a burgage next the bridge, next that of 
John Godson. 

Sir John Trevor, knight, had a burgage in Castle Street, two 
burgages and one curtilage in Cross Green, both formerly in 
the holding of Edward Aimer and Jane Aimer, widow; a 
parcel of land, containing by estimation half a curtilage, whereon 
a house was built, and about 31 customary acres of free land. 
He was of Pl&s Teg, county Flint. 

Sir Richard Trevor, knight, had a curtilage near the Castle 
gate, whereon a house was built, extending towards the River 
Dee. He was of Trefalyn Hall, and the elder brother of the 
aforesaid Sir John Trevor (see the Trevor pedigree opposite 
page 100 of my History of the Townships of the Old Parish of 

Richard Vernon had 6 customary acres of land, whereon a 
house was built, part of the 30 acres formerly belonging to 
Thomas Pulford. Richard Vernon's will was proved at Chester 
in 1629. 

John Welles had a burgage in Midding Street, between 
the burgage of William Davies and that wherein William Cork 
dwelled (see also under William Wilde). 

Lawrence Welles had a burgage in Castle Street, m the 
holding of Jane Warburton ; six free customary acres " beyond 
devon ;" five leasehold parcels containing 1 1 customary acres in 
" Cross yockin [lane] ;" and a leasehold piece of land " in Holt 
neare vnto the Castle gate in a lane leading from the pavem^ 
towardes the River called Mill Dee, wherevpon a Cottage 
standeth conteyning the third part of a curtilage." Mr. Hughes, 

Digitized by VjOOQ IC 


of Kinmel, tells me that a Laurence Wells married Elizabeth, 
daughter and heir of Owen floydon, son of Peter Roydon, son of 
John Roydon,^ a drover (see also under the next entry). 

" William Wilde, ^ent, holdeth nyne Burgages, viz., three 
wherevpon his Mansion howse standeth with the orchard, 
garden, and backside therevnto belonging, one Croft over against 
the said howse conteyning foure burgages, and two burgages are 
adioyning to the orchard of Randolph Hutchiiis." " The same 
holdeth one parcell of land lying neare devon platt^ adioyning 
to the pavement w<* leades to the Coraon Wood conteyning 
three Acres." The same holdeth one parcell of land at Bspehill 
called The Bspes, conteyning two Acres. The same holdeth 
a Curtilage adioyning to the howse called the pavement howse 
in the hie Qreene ;" making iu all nine burgages, and over 
28 customary — or over 59 statute — acres of free land. Mr. 
William Wilde belonged to a very ancient family, members of 
which were among the earliest mediaeval English settlers in 
Bromfield. John le Wylde, clerk, was one of the witnesses to 
an Allington charter of 1391, as I learn from Mr. Edward Owen. 
Richard de Wylde, son of John de Wylde, of Holt, married, 
according to Poivys Fadog (vol. iii, p. 91), Margaret, daughter of 
John Lowther, of Holt ; and their great-great-grandson, Thomais 
de Wylde, purchased the house and lands of leuan and Howel, 
sons of David Llwyd, forfeited for their share in Owen Glyndwr's 
rebellion. This account of the date of the forfeiture is not 
quite in accord with chronological facts, but in the 23rd year of 
Henry VII {seq, 1508), Thomas the Wylde, and others were 
actually in possession of land in Hewlington, "formerly the 
land of Madoc ap leuan ap Madoc, Jankyn his brother [and] 
leuan and Howel, sons of David Lloyd, John Wele was seneschal 
or steward of Bromfield and Yale in 1411, and one of the wit- 

1 Mr. E. B. Roydon has sent me a copy of the will of John 
RoydoD, of Holt, dated 20th May, 1560, and proved on the 16th 
Jnne following, in which the testator names, among other children, 
his son Peter. Mr. Roydon thinks that the father of the Elisabeth 
Roydon who married liaarenoc Wells may have been Owen Roydon, 
son of the abore-named Peter Roydon, son of John Roydon. I find 
that Peter Roydon released, on 15th December, 25 Qneen Elizabeth, 
1582, to Owen Roydon, his son, his Holt lauds. 

2 This shows that the brook crossing Common Wood Lane was 
called, in 1620, " The Devon ;" as a passage snbseqnently to be 
quoted (under John Wilkinson) shows that the same brook was 
then' called Devon, and not ** Ugg," at the point it crossed Wrexham 


Digitized by VjOOQ IC 


nesses of the Holt charter.^ The name of this family came to 
be spelled in later times " Weld " and "Wells," or " Welles," as 
well as " Wylde," " WUde," and *' Wilds." One branch remained, 
or became Boman Catholic ; and a Richard Weld, of Holt, was 
" presented " as " a recusant," and buried by night at Tarporley, 
Cheshire, 20th Aus^ust, 1626. Samuel Wilds, of Wilds Green, wcw 
mayor of Holt in 1624. And a " William Wild, son to Edward 
Wild, gent/' was buried at Holt 22nd February, 166}. Other 
Welds, amon«^ whom may be named Mr. Peter Weld and Mr. 
Rolph Weld, both of High Street, Wrexham, were Presbyterians. 
Thomas Weld, citizen and grocer, of London, and of Richmond, 
Surrey, in his will of 1678, speaks of " nephew Peter Weld of 
Wrexham," This Peter was buried at Wrexham, 16th July, 
1688. Mr. Ralph Weld, before the Restoration, Lieutenant 
Ralph Weld, buried at Wrexham, 28th August, 1681, was 
a fast friend to Philip Henry, leaving him £5 by his will ; 
which bequest was delivered by his nephew, also named 
Ralph Weld, probably the Rev. Ralph Weld, rector of Great 
Saxham, Suffolk, who died 21st September, 1721, leaving £100 
to Wrexham Grammar School. Tj^ Mr. Peter Weld, of Wrex- 
ham, who died in 1 688, is called tn the Parish Registers, at 
different times, " Weild," " Welds," and " Wells," but he described 
himself consistently as " Peter Weld.'* The representatives of 
the family who spelled their surname " Wells" and " Welles" 
are represented in the 1620 list of Holt freeholders by John and 
Laurence Wells. 

John Wilkinson had three burgages, where his dwelling- 
house stood, with barns, etc. ; a burgage in Frog Lane ; a bur- 
gage and a-half near the Cross ; a burgage near Cross Green ; 
and another burgage ; also " one parcell neare Devon bridge in 
Wrexham Lane;" a close called "The Espes," another near 
Devon, about 23 customary — or 48 J statute — acres of free land, 
an acre of leasehold land called " Y Pase," and 6 customary — or 
12 J statute — acres of leasehold land, including the little (Jallow- 
tree field, in Hewlington. 

Thomas Wilkinson, of Farndon, had a house and 18 cus- 
tomary acres, part of the 30 customary acres formerly of Thomas 
Pulford, etc., and 3 other like acres of leasehold land. 

John Wright had "one curtilage neere vnto the Crosse 
adioyning to y« pavement leadinge from the said Crosse towardes 

1 To this may be added that Jankyn Wylde and William to 
Wylde were tenants of Holt in the 23rd year of Henry Ylly and in 
1564 David and Edw^ard Wilde had nine bargages, eta, at Holt. 

Digitized by VjOOQ IC 

tN coxmrr t)BNBiGtt. 429 

frog lane where his dwelling howse standeth with all buildings 
there vnto belonging late the landes of Edward Puleston/' 
12 customary aeres of free land and more of leasehold land in 
the franchise of Holt, and Thomas Lothar*s estate in Hewlington, 
being a leasehold customary acre demised to Thomas Xiothar, 
deceased, on 10th December, 1580 (see also under next entry). 

Georgb Wright had 13 customary — or 25J statute — free 
acres of land whereon his dwelling-house stood, formerly the 
land of Thomas ap Madoc ap lolyn, late of Sutton, and 14 like 
acres of leasehold land called " Kay Robbin/' The will of George 
Wright, of Holt, was proved at Chester in 1623. Thomas 
Wright, of Holt, gent., was buried there in November, 1679, 
and Katherine, daughter of George Wright, was married about 
1693 or 1694 to Koger Roydon, of Calcott (Caldecote), Cheshire. 
On 16th November, 1641, Thomas Niccoe,of Holt, yeoman, and 
Jane his wife, sold, in consideration of £37 10s,, to Thomas 
Wright, of Holt, yeoman, a close of 3 acres adjoining the land 
of George Bostocke, Esq., on the east ; the laud of Arthur Wright, 
brother of the said Thomas Wright, on the west ; the land of 
Sir John Trevor, knight, on the north, and the King's highway 
on the south. In the will tf Joseph Wright, of Tarvin (dated 
26th February, 1767, proved 26th March, 1774), the testator 
speaks of his two messuages, etc., in Holt ; of his nephew, John 
Speed, of the same ; of his niece, Elizabeth Powell, widow [of 
John Powell and daughter of John Speed; see the Powell 
pedigree, before] ; of his niece, Mary Speed, etc. 

Thomas Williams had 7 customary acres of land, with house 
and other buildings, " in Crosse Yockin Lane." 

John Yakdley, gent., had, in 1620, three burgages, whereon his 
dwelling-house stood, one other burgage, and 17 cui?tomary acres 
of free land. The Yardleys were well known in Holt, Farndon, 
and elsewhere in the neighbourhood, about this time. In 1562, 
Lancelot Yardley was a free tenant of Holt. In 1597, Thomas 
Yardley, Mr. Edward Owen told me, had a dispute with John 
Roydon concerning the right and title to lands in Hewlington 
called " the Fourteen Acres," and land in Ridley Wood, late of 
John Yardley, plaintiffs father ; the point at issue being whether 
these lands were left in trust to Alice, John Yardley's wife. Sir 
George Calveley, knight, Thomas Calcott, and others, for the 
payment of his debts, and sold by them to the defendant, John 
Roydon, without condition of redemption. And I have since 
seen a series of depositions, furnished me in summary by Mr. 
E. B. Roydon, of Bromborough, relating to this dispute, which 
seems to have been a very complicated one. I need only say 

Digitized by VjOOQ IC 

iSO ttfE Hovmt Off rioLt, 

that Thomas Tardley, the complainant, of Orew and Famdon, 
was in 1597 about sixty years of age ; that when his father, John 
Tardley, died, he, Thomas, was about nine years old, and iu 
ward to John Roydon, his uncle. On the 29th day of the eleventh 
month, 1659 [January, 16fJ], John Yardley, of Holt,^ and 
Elizabeth Farmar, of Shrewsbury, were married at the place last 
named, after the fashion of Friends ; and in 1682 a piece of land 
in Cross Green, Holt — ^still called " Quakers' Yard " — formerly 
the properly of John Yardley, was vested in trustees as a 
Friends' burial ground. A " John Yardley, gent," was buried at 
Holt, 22nd January, 167^, and a " a Thomas Yarley,'' of Holt, 
on 14th March 14th, 167f • 

The Mayor and Citizens of Chester held 2| customary — 
or 5^ statute — acres of land, with a house thereon built, late 
belonging to Alderman Valentine Broughton, of Chester, de- 
ceased, and left by him to feoffees for charitable uses. 

The foregoing extracts, or summaries of extracts, give 
a most vivid and instructive picture of the town of 
Holty as it was in 1620. Many of the old burgages 
then remained, probably very much in the same con- 
dition as when first built and set out, each with its 
curtilage in front and with its croft behind, along Castle 
Street, Wrexham Lane (now Wrexham Road), near the 
bridge (now Church Street), Midding Street, Pepper 
Street, and Frog Lane (in which the pinfold was) ; also 
around Cross Green, Smithfield Green, and High Green. 
But in a great number of cases, two or more burga^es,^ 
had made way for larger houses, and for the gardens, 
stables, and other buildings appurtenant to them. It 
mattered not at all to the lord how far this process 

^ This was, perhaps, the John Yardley the younger, of Holt, son 
of John Yardley the elder, who on the 12th Janaary, 166f , left all hit 
estate^ezcept some trifling legacies — to his wife, Elizabeth ; which 
Elizabeth, his widow, on 18th December, 1663, renoanced all her 
interest in the executorship of her late husband's will to her trusty 
friend Charles Bradshaw, the yoanger. 

^ Mr. Wilde's house, orchard, stable, and crofb occupied the site of 
seven burgages ; and Mr. Boydon's house, orchard, and stable the 
site of nine burgages in Castle Street. There is no need to cite 
other examples. 

Digitized by VjOOQ IC 


was carried, so long as it was duly recorded that 
the holding of so-and-so represented, say, seven bur- 
gages, from which seven times the rents and ser- 
vices were to be rendered as were due from a single 
burgage. In 1620, 125 ancient burgages were remem- 
bered in Holt ; or rather the rents (" crown rents,*' as 
they are now called) were remembered. Of course, all 
these oldest burgages have long ago vanished, but I 
I myself recollect some queer - looking half-timbered 
thatched cottages in Church Street and elsewhere in 
the town, pemaps as old as the time of Norden's 
Survey, which gave a very fair idea of what an old 
burgage was like, but these have been much altered 
in recent years : windows enlarged, or the thatch 
covered with corrugated iron, so that all picturesque- 
ness is gone from them. The burgages were built 
across the width of long narrow strips, aoout two chains 
in length, and containing, so far as can be made out, 
a little over a rood of ground, the area varying slightly. 
A few Holt houses, although modern, represent, so far 
as their site and the size of their gardens are concerned, 
exactly the area of the old burgages, with their Cur- 
tilages and crofts. But we have to calculate with care, 
for in Norden's Survey we are told distinctly of ttuo 
adjoining houses which represented three burgages ; 
while in another case it appears as though three modem 
cottages stood on the site of tvx) burgages, the crofts at 
the back being divided into as many gardens as there 
are cottages ; and other disturbing factors have been 

Suppose we now try to picture to ourselves the open 
spaces, streets, and lanes in Holt at the date of Norden's 

And first let us take the open spaces. 

** Smithfield Green" denotes the same place that is 
still so called. 

Church Green is not named, but it is quite clear that 
what is now so called was in 1620, and later, known as 
Cross Green." It is easy to prove this. I have 

Digitized by VjOOQ IC 


432 THE TOWN Olf HOtf, 

seen the original deed recording the sale (28th Feb- 
ruary, 168^) to "the Friends" of the piece of ground 
at the comer of Bridge Street and Church Green, 
Holt, for the purpose of a Quakers' burial-ground. 
Now, this piece of land, the position of which is exactly 
known, is described as ** adjoyning thereto a greene 
called Crosse Greene." Unless there were once two 
crosses standing in the town of Holt, it seems very 
probable that the Cross Green was at an early date 
continuous with what is now known as " The Cross 
Bank ;" in other words, that it extended over the site 
of Holt Hall and gardens, and other intervening houses. 
However, there was only one cross in Holt in 1620, 
several houses being described, as though that descrip- 
tion were suflScient, as being '*near the cross." Never- 
theless, I do not doubt but that this cross, standing on 
Cross Bank, gave its name to Cross Green, now called 
Church Green, and that all houses between represent 
aucient encroachments, or enclosures, made before 

** High Green" was another open space in 1620. I 
feel certain, from the references to it in the Survey , 
that it was the enclosed triangular space between the 
point where the Wrexham Road and Gallowtree Lane 
(now Francis Lane) enter Holt, and it is still called, 
Mr. Edwin Bellis informs me, "The Intak.'* The 
Pavement House, and at least five burgages, stood 
around it, or **in" it, to use the exact preposition 
employed in the Survey. 

The identification of High Green, proposed in the 
last paragraph, with what is now known as " The 
Intak," may explain a fact which has always been a 
puzzle. Holt being a town laid out in accordance 
with a definite plan, and almost at one time, the 
streets in it are almost parallel to each other, or at 
least straight. 

The roads leading into it were also continuous with 
its streets, or opened on unenclosed spaces, with one 
conspicuous exception — the present Wrexham Road, 

Digitized by VjOOQ IC 



which enters Castle Street, Holt, at a sharp angle. 
But it is probable that in 1620 and before, Wrexham 
Road did not enter Castle Street as it does now, but 
debouched upon High Green, as other roads in Holt 
debouch on open areas. 

The reference to Wrexham Road just made leads to 
a consideration of the Holt roads, streets, and lanes 
mentioned in the Survey of 1620. 

" Wrexham Lane," as Wrexham Road was then 
called, contained three or four good houses, among 
them the Pate house and a few single burgages, one of 
which was the burgage of Roger Edg worth. The bridge 
over the Devon in Wrexham Lane was known as 
•* Devon Bridge." 

In Castle Street were many burgages, and at least 
one large house, that of Mr. Roger Roy don, with its 
extensive orchard (see before under Roger Roydon). 

Cross Bank is not mentioned under that name in 
the Survey^ but various burgages are described as 
*' near the Cross," amongst them that of John Wright 
(see before). 

Many burgages stood in Midding Street, amongst 
them those of EdwardCrue, gent., and William Davies, 

Also two burgages were in Pepper Street. 

The names "Pepper Street " and " Midding Street" 
have long ago been forgotten, "Green Street" having 
apparently absorbed them both. 

It looks as though Frog Lane contained more single 
burgages than any other street in Holt. Between Frog 
Lane and Common Wood Lane, by the Fishpools, the 
road was called " The Pavement." Three burgages in 
Frog Lane are described as " near the pinfold." The 
Powell house was here also. Further along the lane 
was " Devon platt " — the flat bridge over the Devon. 

Many burgages are described as " near the Bridge" — 
that is. Holt Bridge — and a large house, that of Francis 
Pickering, gent., was " near the churchyard," and a 
barn, appurtenant to his mansion-house adjoined the 

Digitized by VjOOQ IC 


Castle ditch. George Buckley had also a curtilage near 
" the Church Ashe. 

As to the lane leading towards the Castle, enough 
has been already said. 

" Hiefield Lane" cannot at present be identified. Sir 
Richard Egerton had, in 1620, 3 acres called ** Spencer's 
Acres" adjoining it. 

What is said by Norden concerning Hewlington will 
be reserved for another chapter. 

Digitized by VjOOQ IC 


laetotetDfit anH Jt^ottcest of Soofest. 

Thb History op the Dioobsb of St. Asaph : Gknbral, Cathedral, 
AND Parochial. By Ven. Arohdeacoa D. R. Thomas. New 
Edition. Part If 

Wb are glad to welcome Part I of a new edition of the History of the 
Diocese of St Asaph, by the learned and accomplished Chairman of 
Committee of the Cambrian Archaoological Association. The first 
edition of this valaable work was published in 1874. This new 
edition is not merely a reissae of the original work, bat contains a 
large addition of interesting material ; while the illastrations (if we 
may jndge from those gi^en in the Part before as) promise to be 
vastly superior in execution and archaoological interest to those 
included in the older volume. Amongst these may be mentioned 
an excellent reprodaction of the original Charter (preserved in the 
Shrewsbury Museum) granted by Bishop Reiner (1186-1225); two 
views of Guilsfield Church, nave and gallery, showing the ritual 
arrangements and pews of all shapes and sizes ; the nnique wooden 
font at Efenechtyd ; the dog-tongs preserved at Llanynys ; Capel 
Trillo in Rh6s, an unique illustration of the primaoval oratories 
(like that of Ghtllerus, at Kilmalkedar in Ireland) which formed the 
type of the earliest British Churches. 

As an instance of the care which has been taken by the author to 
bring the work up to date, we have, on page 21, an interesting 
reference to the English Church History Exhibition, held at St. 
Albans in 1905, when a copy of the Missale ad Usum EcdesioB Ban- 
goriensis, circa 1400, was shown, with the inscription : *' This Booke 
was geven to the hye alter of the Paryshe Churche of Oswestry by 
Sr Morys Griffith prist." 

Among the minor improvements is the relegation to the foot-notes 
of quotations from the original Latin and Welsh, the convenience of 
the reader being considered by an adequate and scholarly rendering 
in English in the text. 

Although the work purports to be a History of one of the Four 
Welsh Dioceses, there is much in the earlier part which is of distinct 
use and interest to all members of the Church of England at the 
present crisis. Chapter I dealing with the Origin of the See ; 
Chapter II describing the early foundation, constitution, and cas- 
,tom8 of the British Charch; Chapter III setting forth the land- 
marks in its early history, and the ecclesiastical policy of the Lords 
Marchers and others ; Chapters IV and V, giving the history of 
the annexation and subjection of the Welsh Churches to the 
Province of Canterbury, contain valuable statemente of historical 

Digitized by VjOOQ IC 

436 REVIEWS ANl) NOtlCfcS 0*" feOOfeS. 

fact, carefully supported by eTidenee, duly marsballed and set forth 
with sound judgment and in scholarly style. 

The author deals with the great and abiding evils of appropria- 
tion, commencing witb the Norman rulers, which has been the very 
bane of the Church from their time downwards : a sad story of the 
plundering of Church revenues, which was repeated generation 
after generation. He does not omit, however, while recording this 
grievous treatment of property bestowed for sacred purposes, to 
refer to the period after the Wars of the Roses as marked by " a 
spirit of church building and restoration that endeavoured vigor- 
ously to repair the damage of the past** ; and he mentions '* notably 
the Stanley series— as they are sometimes called — at Moid, Holywell 
(St. Winifred's Chapel), Holt, and Northop, to which may be added 
Gresford, Llangollen, and many others" (p. 70). 

It is worthy of note how differently ** gallant little Wales*' was 
regarded by those in power during the mediesval period, for the 
ordination of Welshmen to any but the lowest order was prohibited 
(p. 48) ; and, on the other hand, when the question of an Italian 
nominee of the Pope was raised, and '*a reservation" of the 
Bishopric of St. Asaph had been made for a foreigner — no doubt an 
Italian — this intrusion of Roman influence was opposed, and '* an 
Indult was issued to the Dean and Chapter, notwithstanding the 
reservation, to elect a bishop of their own, (u the people of Wale$ 
were too savage to be governed by a foreigner * (p. 6 J). 

This preliminary notice must close with the briefest allusion to 
the numerous side-lights thrown on the survival of Pagan customa 
(p. 22) ; mistaken renderings of Welsh words (p. 2) ; the value and 
importance of pilgrimages to St. Winifred's Well and elsewhere 
(p. 80). Page 30 contains a most instructive table of tithe appro- 
priations at successive intervals, covering 600 years, at the time of 
the Norwich Taxation, 1253; the Lincoln Taxation, 1291; the 
Dissolution of the Monasteries ; and the Commutation of Tithes, 

We have, on pp. 14, 62, and 153, instances of the free-will 
offerings to Parish Priest and Bishop, consisting of particular 
kinds of produce, varying in different localities. Such were ** blith 
y ddafad," or J^ctualia ; '* cnu'r person," the parson's fleece; 
'* blawd y gloch " and *' ysgub y gloch," the clerk's sheaf and 
flour, from each tenement in the parish, and ** offrwm rhaw," or 
spade money, made at the grave on the extended spade to the clerk 
for his services. 

We look forward to ihe issue of Part II, which is promised before 
the close of the year. 

Digitized by VjOOQ IC 


9[rc|)aeological JlBotesE anH (Bntvits. 

Pembrokeshire Association for the Preservation of Ancient 
Monuments : Annual Report, 1906. — A committee meetings of the 
Association for the Preservation of Ancient Monaments in the 
County of Pembroke was held on Taesday at the Temperance Ball, 
Haverfordwest. The Dowager Ladj Kensington presided; and 
amongst those present were Dr. Henry Owen, Mr. E. Laws, Mr. T. 
L. James, the Rev. J. Llewellii^ Mr. A. J. Wright, etc. 

The following report was read by the hon. secretary, Mr. J. W. 
Phillips :— 

" Llawhaden Castle. — Some farther repairs have been found neces- 
sary to the square tower, in consequence of a fresh crack above a 
window opening on the west side. The abutments, of which very 
little remained, and the arch above, have been rebuilt, and cement 
g^ont ran into all cracks. The ivy and growth on the walls require 
catting again, and your committee recommends that a sufficient sum 
be spent upon it next spring. The place still continues to be visited 
by a considerable number of people, and it would be of advantage if 
some safe means of ascending the octagonal tower could be devised, 
as the view it commands is well worth the climb. 

" Gilgerran Castle, — Your committee wishes again to call attention 
to this castle. Nothing has as yet been done, and its condition is a 
disgrace to the neighbourhood. 

** Castell Cock Castle. — Nothing has been done with this building. 
A small sum spent in removing the trees growing on the walls 
would preserve this interesting building from farther destruction. 

" Carew Castle, — Mr. and Mrs. Trollope, the owners of this his- 
toric castle, have most carefully and successfully preserved the 
eastern window in Sir John Perrot's banquettidg-hall from further 
decay. Railway iron was introduced at the top of the window, 
which relieved the pressure caused by the battlements, and prevented 
outwArd movement. The decayed places were built up with 
masonry, or strengthened with cement grouting. Substantially the 
suggestions made by Mr. Caroe were carried out. This window is 
safe, but many others require immediate attention if they are to be 
preserved. N.B. — The cause of this premature collapse was the 
jerry-builder who put deal lintels into the windows, while Sir John 
Perrot was confined in the Tower of London. 

" St. David's Cathedral : St, Nicholas' Ghapel.^ThiB chapel has 
been carefully restored and roofed over. The ceiling of carved oak 

Digitized by VjOOQ IC 

438 ARCH^OLoaiCAL notbs and queries. 

IB of good design, and the work has been skilfnllj execnted, but it 
does not harmonise with its sarronndings. 

^^ Herhranston Church. — This charch has been carefally restored. 
The early Norman font has been repaired and cleaned, a new bowl 
of Nolton stone, copied from the Rndbaxton font, having been 
added. This was necessary, as the original had been broken, and a 
wooden bowl, lined with lead, snbstitnted. 

'* St. Mary's, Haverfordwest, — The naye of this church has been 
reseated and reopened for divine service. The beautiful arcade on 
the north side, which had loug been disfigured by many coats of pahit, 
has been carefully and skilfully cleaned and repaired. The pillars 
were found to be loosely built, and new ashlar work had to be worked 
round the bases ; cement grout was run into each pillar, until it 
would hold no more, some of them taking as much as twenty-five 
pailf uls. The south wall of the tower had to be carefully shored- 
up, and the remains of the arch underneath taken down and 
rebuilt. Every stone of the old arch that could be used was in- 
corporated in the reconstructed arch. The walls of the tower on 
the south and west sides showed some very bad cracks. These 
have, where necessary^ been carefully cross-bonded and run full of 
grout. The chancel-arch has been cleaned, but has not yet been 
repaired. Much of the bases of the piers and some of tbe inner 
mouldings are of Roman cement, but funds do not yet admit 
of its restoration. The windows, except one, are all in a very bad 
state, but the restoration committee are quite unable to attempt 
anything more at present. The ancient carved bench-ends, with 
two of the original oak stalls, have been fixed in the chancel. The 
stone groining under the tower has been cleaned, but has not been 
restored. Some of the intermediate ribs are of wood, the floor of 
the clock-chamber above has been laid with wood blocks, and the 
walls and buttresses of the tower thoroughly repaired. 

" Ambleston Church, — This church is now under restoration, and it 
is to be hoped that its original features, especially the ancient font, 
will be carefully preserved. 

** Tregidreg Cross. — This cross has been removed from Tregidreg 
farm, and built into the wall of Mathry churchyard. 

'^ Mesur-y-Dorth Cross. — This cross, being small and dose to the 
roadside, is in some danger of being damaged by passing traffic, and 
should be protected. 

** Trekenny Maenhir, — This stone has been fixed upright again upon 
a strong concrete foundation in its original position. Some difficulty 
was experienced in getting the work done, as the stone weighed 
more than five tons. 

** Cilgerran Ogham Stone. — This stone will be protected where it 
is, as there is no room for it inside the church or porch. 

Digitized by VjOOQ IC 


**ir«wr» Ogham Stone. — This stone has been moyed nnder the 
aaspioes of your oommittee> and Professor Rhys has been able to 
read both the Ogham and the Latin inscriptions. It is to be desired 
that this stone, and the scalptared stone near by, should be moved so 
that they both can be seen. 

^^ Tenby Church. — The old Perpendicular font has been moved 
from the churchyard into St. Anne's Chapel by Mrs. Thomas Allen. 
In the course of the investigations by Mr. Edward Laws, another 
window in the church has been opened, making seven in all opened 

^^ Caldy Island. — This Ogham stone has been moved by the Rev. 
Done Bushell from its inconvenient position in the lower chapel to a 
much better site in the Priory Church, close to the place where it 
was first discovered. 

^^Pembrokeshire ArcluBological Survey. — This survey, which has 
occupied some members of the Association for many years, is now 
approachiug completion. As a record of the ancient monuments of 
the oounty, it will be of the greatest value. Pembrokeshire has in 
this instance again led the way. 

'* It is to be hoped that members will endeavour to find out what 
antiquities exist in their neighbourhood, and will inform the hon. 
secretary if any of thess require attention, or of any damage likely 
to be done to them.*' 

Banoor's ANTiQaiTiBS : Special Committee's Report. — Colonel 
Piatt, C.B. (the Mayor), presided at a recent meeting of the Bangor 
City Council, and proposed a vote of sympathy with the Dowager 
Lady Penrhyn and the Penrhyn family on the death of the late Lord 
Penrhyn, by whose death the Council, as representatives of the city, 
and Wales generally, had sustained a great loss. 

The motion, seconded by Alderman Thomas Lewis, was adopted. 

Alderman Mathews read the report of the special committee as to 
the sale of antique furniture recently reported. 

Tour committee have the honour to report to the Council that 
they think it extremely regrettable — 

1. That the sale should have been carried out at all, without 
sufficient inquiry as to the proper value of the articles sold. 

2. That the negotiations should have been carried through with- 
out the knowledge or sanction of the Council ; and 

8. That the clerk of the Museum Committee should have omitted 
to submit the important resolution of the committee for confirma- 
tion by the Council, and should not have reported to the City 

Digitized by VjOOQ IC 


Treasarer the manner in whiob the Town Olerk adrised that the 
parchase-monej shonld be devoted : though this latter fact maj be 
accounted for by the fact that the Town Clerk's advice was g^ven 
twelve months previously. 

Tour committee, therefore, recommend : — 

1. That no sale of any further effects belong^g to the Museum 
shall take place, except after the fullest inquiry and investigation as 
to the true valae of any articles proposed to be sold, and without 
the fall knowledge and sanction of the Council. 

2. That a proper schedule of all the effects of the Coancil should 
be made without delay, and that the University authorities should 
be asked to render such assistance as they can in the classiScation 
of such effects ; and 

3. That a letter be written to Mr. Duveen, in the name of the 
Council, thanking him for his offer to re-sell the furniture to the 
Council, but at the same time intimating that they are unable to 
accept such offer. 

Mr. Mathews moved the adoption, and Mr. Vincent seconded. 
The report was adopted without dissent 

Digitized by VjOOQ IC 




John Romillt Allen was the eldest son of Mr. George Bangh Allen, 
Barrister-at-Law of the Inner Temple, of Cilrhiw, Narberth, Justice 
of the Peace and Depnty-Lientenant of the Coantj of Pembroke. 
Born in London in June, 1847, he was educated first at King's 
College School, then at Bngbj, and at King's College, London. 
Having a strong mechanical bent, he was articled to Mr. G. F. 
Lester, Engineer-in- Chief to the Mersey Dock Board; and later on 
he was engaged in engineering work in Persia, and he has embodied 
some of his ideas on that science in his *' Design and Constmction 
of Dock Walls." Bat he had a still stronger inclination to 
archaeology, to which he devoted the energies of his after life. It 
was at the Carmarthen Meeting in 1875 that Mr. Romilly Allen 
joined onr Cambrian Association, and became a contributor to our 
Journal ; and in 1887 he took the place of Archdeacon Thomas 
as joint editor with Canon Trevor Owen, F.S.A., and finally 
became sole editor in 1892. He wrote much and ably on many 
subjects, as will be seen by the appended list of his articles, but his 
favourite subject was that of Celtic Art and Ornamentation, which 
he illustrated with his facile pencil. He was a keen and scientific 
archfldologist, and in 1889, on his appointment as Rhind Lecturer 
in ArchflBology in Edinburgh University, he took for his subject 
"The Early Elhnology of the British Isles, and more especially 
Scotland, treated from the point of view of Languages," which he 
published under the title of " Early Christian Symbolism in Great 
Britain and Ireland." Another work which appeared the same 
year (1889) was " The Monumental History of the Early British 
Church," published by the S.P.C.K. He was also the author of 
**The Early Christian Monuments of Scotland," and he edited The 
Reliquary till his death. In 1898 he was appointed Yates Lecturer 
in Archaeology at University College, London. 

Indifferent health tended to develop in him a tone of moroseness 
and an irritable temper, and a roughness of manner caused no little 
friction at times between him and the officers of the Association and 
the contributors to the Journal ; but all recognised his ability and his 
devotion to his subject. Of late years he had become much 
more mellowed, and his attendance at our Annual Meetings gave 
pleasure and instruction. His ready and lucid addresses on his 
tavonrite subjects on the excursions always commanded attention 
and respect. As Editor he maintained the high standard of the 
ArchcBologia Cambreiuis^ and was always jealous for the reputation 
6th sbb. vol. vn. 29 

Digitized by VjOOQ IC 


of the AssociaiioD. At the last AnnDal Meeting at Llangefiii a 
generons acknowledgment of his seryices was made, and a vote of 
Hjmpathy with his brother and sisters was passed nnanimoiisly. 

A List of Papers contributed to this Journal by 
Mr, RomUly Allen. 

1876 Notes on Forth Kerry Church, Qlamorganahire, with Special Refen;noe to 
the Churchyard Cross. 

1876 On an Inscribed Ogham Stone at Little Trefgame. 

1877 Camrose Church. 

1878 Pembrokeshire Churches, Johnston. 

1883 Crouses at St. Edren's Churcli, Pembrokeshire. . 

1884 The Past, Present, and Future of Archeology. 

1888 Notes on a Roman Steelyard and other Objects found at Strettbn Qrandi- 

*on, Herefordshire. 

1889 The Inscribed and Sculptured Stones at Llantwit Major, Olamorganshire. 

1889 Recent Discoveries of Inscribed Stones in Carmarthenshire and Pembroke- 


1890 On the Organisation of Archieological Research. 

1891 Notice of a Mediaeval Thurible found at Penmaen in Qower. 
1893 Celtic Art in Wales and Ireland Compared. 

1893 The Cross of Guidon, Gk>lden Qrove, Carmarthenshire. 

1893 lolo Morgan wg's Readings of the Inscriptions on the Crosses at Llantwit 

1895 Catalogue of the Elarly Christian Inscribed Monuments in ComwalL 

1895 Note on the Carew Inscription. 

1896 The Trawsfynydd Tankard. 

1 896 Catalogue of the Early Christian Monuments in Pembrokeshire. 

1896 Notes on Late-Celtic Art. 

1899 Early Christian Art in Wales. 

1900 Some Dolmens and their Contents. 

1901 Two Kelto-Roman Finds in Wales. 

1901 Some Carved Wooden Spoons made in Wales. 

1902 Old Farm Houses with Round Chimneys near St DavidV 

1902 The Chevron and its Derivatives. 

1903 Pre-Norman Cross-Bade at Llangyfelach, Glamorganshire. 

1903 Note on a Perforated Stone Axe- Hammer found in Pembrokeshire. 

1904 The .Cross of Irbic at Llandough, Glamorganshire. 

^905 The Discovery of an Early Christian Inscribed Stone at Treflys, Carnarvon- 

1905 Find of Late-Celtic Bronze Objects at Seven Sisters, near Neath, Glamor- 


Digitized by VjOOQ IC 




Aber Goran ^Laugharne), 364 
**Advowry, Persons in = W., Ar- 

ddelwyr ; L., Advocarii, 21 
Aldeford, John de, 24 ; Aiken 

(Hawlkyng) Church, 64 
Allen, J. RomiUy (Obituary), 441 
Aimer, Edward, 317 ; John, 23,26 ; 

WUliam, 26 
Alport, Richard, 401, 409 
Allt Cunedda, Camp with Barrows 

and Cist, 384 
Altoir (Innishmurray), Similar 

Stones at Eglwys Cymmyn, 

Amobr in Bromfield, 6 
Ancient Bridge, Carnarvon, Re- 
mains, 368, 369 
'^Annesse Gk)uch," Corruption of 

Annual Meeting of Association, 

Notice, 360 
Anwyl, Professor— The Early Set- 
tlers of Carmarthen, 361-388 
Archaeological Notes and Queries, 

Arundel, Earl of, Richard Fitzalan, 

11 ; Thomas, 13 
Beatrix of Portugal, Widow of 

Thomas, 16 

Baker, Jeffirey, 24 

Banc Bemsel (Allt Cunedda), 384 

Banc y Belli, 224-6 

Band of Gentlemen Pensioners, 331 

Bangor's Antiquities, 439 

Barre, Joan de, 10 

Barrett, Thomas, 323 

Basingwerk Abbey, Old Glass, 364 

Bastion, Traces at Carmarthen, 

Unique Example, 238 
Bates, Launoelot, 318 
Batha, William, 389 

Battle Abbey, Llandeulyddog given 
to, 287 

Beaufoit, Henry, Duke of, Visite 
Carmarthen, 343 

Bentinck, William, Earl of Port- 
land, his Grant Opposed, 326 

Bernard, Bishop of St. David's, 
Endows Priory of St. John, 
Carmarthen, 337 

Bevan's Chair, Madam (Llanddow- 
ror), 226 

Bewley, Bernard and John, 318 

Bird, Sir Thomas, Clerk, 316 

BitheU, Richard, 410 ; Thomas, 410 

Bledri ap Cydifor (Bledericus La- 
timerus), 290 

Bostocke of Holt, Pedigree, 330; 
George, 389, 407-9 ; Launce- 
lot, 318, 330, 409 

Boundary of Carmi^henshire, 363 ; 
Of Dyfed, 363 

Brandon, Sir Charles, Duke of 
Suffolk, 24 

Brereton, Owen, 318, 407 ; William, 
Chamberlain of Chester, 321 

Bridgewater, John, Earl of. Ac- 
quires Manor of Ridley, 325, 

Brigstocke, T. E., St. Peter's 
Church, Carmarthen, 336- 

Bristol Channel once a Vast Fertile 
Plain, 372, 373 

Brogynin, 261 

Bromfeud, 9 

Bromfield in Two Commotes, 2 ; 
Taken by Hugh Cy veilioc, 2 ; 
A Part of Powys Fadog, 3 ; 
Commote of Yale added to, 
3 ; "Amobr" in, 6 ; Welsh 
Names of Freemen, 6 ; Con- 
nected with House of Lan- 
caster, 10 

Digitized by VjOOQ IC 



Bromfield and Yale Grant Aid to 
£arl Warrenne, 4 ; Granted 
to John de Warrenne, 8 ; 
Sub-granted to William de 
Warrenne, 9; Vested in 
Crown, 17 ; Granted to Sir 
William Stanley, 17 ; to 
Henry Pitzroy, Duke of 
Richmond, 320 ; to Prince 
Henry, 325 ; Prince Charles, 
325 ; Part of Commonwealth 
EsUtes, 326 ; Grant to Wm. 
Bentinck, Earl of Portland, 
Opposed, 326 ; Thomas, Duke 
of Norfolk, Justice of, 17 ; 
Foresters of, 12 ; Tidderley's 
Survey, 311 ; New Survey, 
325 ; Sub-Commissioners for 
Survey, 321 

Bronze Age in Carmarthenshire, 
384-387 ; Stone Circles, Re- 
lics of, 368 

Bronze Celts (Coelbren), 137 ; Pa- 
tella and Strainer (Eyngadle), 
227-230; Torque (gold-pla- 
ted). 40, 52 ; Pin, 47 

Bryan, Sir Guy de, 267 

Buckley, George, 410 

Burgeny(Burganey), Anthony, 392; 
WUliam, 391, 410, 411 

Carmarthen in Early Norman Times 
—Professor J. E.Lloyd, 281- 

Carmarthen and Llandeulyddog 
Identified, 281, 282 : Old 
and New, 286 ; Castle or 
Caer at Rhyd y Gors, 284 ; 
Castle Defended by Owain 
ap Caradog, 285 ; Secures 
Norman Supremacy in Car- 
marthenshire, 284 ; First 
Charter, 286 ; Homage to 
Prince William, 286 ; Gerald 
of Windsor at, 286 ; Chapel 
of St. John the Evangeust 
and St. Theuloc, 288 ; Alfred 
Drue gives Llangain Church 
to, 289 ; Maes y Prior, 289 ; 
** An English Towne," 342 ; 
Castle, 236, 237; Bastion 
Traces, Unique, 238 ; Mayors, 

Carmarthen, St. Peter's Church — 
T. £. Brigstocke, 336-351 ; 
First Mention in Annals 
of Battle Abbey, 287, 337 ; 
Chapelries, Newchurch and 
Llangain, 337 ; Vicars of 
(Richard ap John, 1278 ; 
David Robyn, 1403 ; William 
Sty ward, 1408 ; John David, 
1486 ; John Harry, 1500 ; 
David Webbe, 1501 ; Thomas 
Prichatxi, 1539), 338, 339; 
Body of Walter Devereux, 
Eirl of Essex, Brought for 
Burial, 342 ; Richard Davies, 
Bishop of St. David^s, Prea 
ches, 342 ; Bishop Laud, 343 ; 
Bishop Lawrence Womack, 
343 ; John Wesley, 344 ; 
French Prisoners Detained 
in the Church, 344 ; Sir 
Richard Steele Buried, 344 ; 
Chalice Cover, 1577, 345; 
Stone Coffin Lid, 345 ; Effigy, 
347, 349 ; Monuments— Bis- 
hop Ferrar, General Sir W. 
Nott, Sir Richard Steele, 
349 ; Sir Rhys ap Thomas's 
Tomb, 347; Earl of Rich- 
mond's Tomb Removed to 
St. David's Cathedral, 348 

Carmarthen Meeting, 1906, Report, 
103-128; Excursions, 2ia- 
248 ; Mayor's Address of 
Welcome, 105, 106; Preei- 
sident's Address (Llanstep- 
han Castle), 108-118 ; Sub- 
scriptions to Local Fund, 
249, 250 ; Accounts, 253 

Cadwal = Irish Cathal, 363 

Caer at Rhyd y Gors, 284 

Caerlleon, Welsh Name for Holt 
Castle, 10 

Caerwent Roman Gateway, 184 

Calcott, Thomas, 389, 411 

Cdintref Bychan, 364 ; Gwarthaf, 
362; Mawr, 364; Ystrad 
Tywi, 364 

Cantrill, T. C— Geological Notes 
on Roman Remains, Cwm- 
brwyn, 176 

Capel Mair Stone — Professor J. 
Rhys, 293-310 

Cardiff Castle, North Gateway, 

Digitized by VjOOQIC 



Carmarthenshire, Early Settlers in 
—Professor Anwyl, 361388 ; 
Bronze Age, 384-387 ; Traces 
of Groidelic Population, 365 ; 
Prehistoric Hearths, 380-383; 
Modem Boundary Artificial, 

Carnarvon Ancient Bridge, 358, 
359 ; Oatmeal Market, Pen- 
tice Qrounds, Turf Square, 

Cam Comal, 132 

Camwyllon, 364 

Castell y Gaer, 248 

Castrum Leonum, Holt Castle, 9 

Cath Pencred, 258 

Cedweli (Kidwelly) Derivative of 
Cadwal, 363 

Chalicd, Elizabethan, 218, 246, 263 ; 
Cover, 1577, 345 

Charles of Bala, Thomas. 218 

Chilston, Sir John K., 24. 25 

Cholmondeley, John, 12 ; Sir Hugh, 

Ciau of Nanhyfer, 310 ; Ciml) = 
Ransom Money, 70 

Clawdd Mawr, 247 ; Late Celtic ? 

Clawdd y Fanweu, 132, 153 

Clear's (St.) Church, Norman Arch, 

Clive, George, 317 

Coelbren, Report on Excavations 
at — Col. W. U. Morgan, 
R.E., 129-174 ; The Gaer, 
136 ; Ramparts, 138 ; Layer 
of Logs, 1:38-143 ; Their Pur- 
pose, 149 ; Ditches, 149-153 ; 
Entrances, 153- 15H ; The In- 
terior, 156-159; Finds (chief- 
ly Roman), 159-167 ; Con- 
clusions— not a Walled Town 
— houses, if any, not stone ; 
traces of roads — remains 
typical of second or third cen- 
tury ; many memorial stones 
on line of road, 172 

Coffin Lid Stone, 345 

Coiniu, 262 

Coldewell, John, 323 

** Communion Sunday Flag," 263 

Corbet, Robert, 328 

Corpse Litter (Elor feirch), 219, 220 

Cotton, George, 320 ; Sir Robert, 
of Combermere, 326 

Coygan Cave, 373, 374 

Craig Derwyddon Bone-caves — 
Skeletons found. Mode of 
Burial, Large Skulls, 377- 

Crew (Cme;, Thomas, 391, 412; 
Edward, 411 

Cromlech, Ffynnon Newydd, 369 ; 
Pare y Bigwrn, 370, 371 

Curious Cmet English Glaze, 266 

Cursing-stones, 271 

Cwmbrwyn Roman Remains — John 
Ward, 175-212, 227-230; Ex- 
ploration of Interior, 186- 
200 ; The Finds : Quern, Pot- 
tery, Mortaria, Amphorse, 
200 ; Tank, 194 ; Roofing- 
slates, 188 ; Roman Glass, 
Coins, Bronze Fragment, 
204, 205 ; Flue-tiles, 189, 193 ; 
a MansiOf or Outpost of a 
Roman Fort, 208 

Cyfekch = Cyfeiiiog, 364 

Cynin, 261, 262 

Cynog Ffynnon, 273 

Cynwrig ap Codflawd, 4 

Cyttiau Gwyddelod, Tre'r Ceiri, 35 ; 
Compared with Primitive 
Houses in Ireland, 36 

Cyveilioc, Hugh, takes Bromfield, 2 

Dalian, 271, 275 

David, John, Vicar of St. Peter's, 

Carmarthen, 1486, 338 
Davies, Richard, Bishop of St. 

David's, 342 
Davies, Robert (of Gwysaney), 413 
Davies, Rev. W.. Paper on **Llan- 

fihangel Abercowin," 216, 

**Dawke" (Llandawke), 273 
Dawkins, Professor Boyd, Note on 

Tre'r Ceiri, 35-37, 124, 125 ; 

on Gower Caves, 373 
Decca, John, Forester of Bromfield, 

12 ; Roger, 413 
Dee River and Holt Castle, 400 
Derllys = Irish Durlas, 363 
Devereux, Walter, E. of Essex, 

brought to Carmarthen for 

Burial, 342 
Devon Bridge (Holt), 433 
Dinas Bran Castle granted to John 

de Warrenne, 8 
Dochdwy, St., 279 

Digitized by VjOOQ IC 



Dolichocephalic Skulls, 376 

Dove-house (Holt), 399 

Drue, Alfred, gives Llan gain Church 

to Carmarthen, 289 
Dyfed, Persistence of Irish Lan- 

gu^e in, 91 ; Boundary, 363 
Dynulle (Cinhinlle), 4 

Early Man, Distribution of, how 
Conditioned, 366 

Early Settlers in Carmarthenshire 
—Professor Anwyl, 361-388 

Eaton, Richatxi, 318 

Edgworth, Roger, 389, 413 

Edward II in Glamorgan— Rev. J. 
Griffith— Reviewed, 265, 266 

Effigy (Anne, Lady Vaughan) in St. 
Peter's, Carmarthen, 347, 
Llandawke, 277 

Eglwys Cymmyn, Pare y Ceryg 
Sanctaidd, and Llandawke, 
Notes on— G. G. Trehem»*, 
Eglwys and Llan distinguished, 
268, 259 ; Ogam, 260 ; Fon% 
266 ; Mural Paintings, 266 ; 
Stones similar to Altoir in 
Innishmurray, 271, 276 

Ellis, Ralph, 318 

Epigraphic Notes — Professor John 
Rhys - Llansaint. Llan- 
dawke, Nevein, Treflys, Lly- 
st>n Gwyn, 66-76 

Evans, Rev. Geo. Eyre, Llansaint, 

Exchequer Tower, Holt, 390, 394 

Eyton, David, 327 ; Richard, 330 

Fagan, Robert, Built St. Asaph 
Cathedral, 12 

Ferrar, Robert, Bishop of St. 
David's, Imprisonment and 
Martyrdom, 342 ; Monu- 
ment, 349 

Fishboume, Rev. E. A. — Notes on 
East Window, Gresford 
Church, 362-357 

Fitz-Baldwin, Richard, 284 

Fitz-Gerald, Maurice, Lord of Llan- 
stephan, 289 

Fitzroy, Henry, Duke of Richmond, 

Font, Eglwys Cymmyn, 266 ; Llan- 
dawke, 278 ; Llanfihangel 
Abercowin, 216, 217 
Base (Ystrad House), 213 

French Prisoners Detained in St. 
Peter's, Carmarthen, 344 

Funerals Halt for Lord's Prayer, 

Gaer at Coelbren, 136 

Gam Fawr (Craig Twrch), 386, 386 

Gam Goch, Urns, 384 

Gateways of Roman Cities and 
Forts, 183-186 

Gavelkind, Welsh, 6 

Gavells (Gafaelion), Holdings, 6 

Gelligaer (Walled Town), Aban- 
doned, A.i). 90, 172; Gate- 
way, 184 

Gerard, Sir William, 416 ; Gilbert, 

Goidelic Population, Carmarthen, 
Traces, 366 

Greene, Roger, 389 

Gresford Church, Notes on East 
Window, 362-357 ; Glass 
given by Thomas Stanley, 
Earl of Derby, 352, 365 ; 
Money Bequeathed for Win- 
dow, 353 ; Old Glass from 
Basingwerk Abbey, 364 ; 
Madocks Chapel, 354 ; " Te 
Deum" Window, 366. 
Sir Wm. Roden, Rector of, 

Gruffith, Thomas ap David ap, 24 

Gwelys (Gwelyau) = Holding of a 
Kindred, 6 

Gwestfa, 4 

Gwyddel, Llwyn, 366 ; Pant y, 
365 ; Traces of Goidelic 
Population in Carmarthen- 
shire, 366 

Gwynne, John, Sub-Commissioner 
to Survey Bromfield Lord- 
ship, 321 

Hanmer, Humphrey, 392 

Harry, John, Vicar of St. Peter's, 

Carmarthen, 338 
Hawton, 239 

Hearths, Prehistoric, 380-383 
Henry VII at Holt Castle, 23 
Hewlyngton, 7 
Hirfaen Gwyddog, 368 

Digitized by VjOOQ IC 



History of the Diocese of St, Asaph 
— Ven. Archdeacon D. R. 
Thomas (Reviewed), 4x$5 
Hodeslea (Hoseley), in Land of Earl 

of Chester, 2 
Hogmore, 322 

Holt— Town, Castle, Church, Fran- 
chise, and Demesne — A. N. 
Palmer, 1-34, 311-334, 389- 

Commote Welsh in Language, 
Customs, Tenure, adopts En- 
glish Customs, 1, 2, 5 

Norden's Survey, 389-404 
Holt Castle =Chastellion, Castrum 
Leonum, 9 

Welsh Name, Caerlleon, 10 

Date of Building, 10 

Annexed to Earldom of Ches- 
ter, 13 

Defended by Sir Richard Lloyd 
of Esclus, 331 

Richard II at, 13 ; Henry VII, 

Officials (21 Hen. VII), 23; 
Resident Constables, 327 ; 
Dove-house, 399 

Chaplains, 324 ; Free Chapel, 
324 ; Lady Priest, 403 
Holt, Charter granted, 13; Missing, 
13 ; Importance of, 14 

Charter, Thomas, Earl of Arun- 
del's, 26-31 

Bounds of Town and Liberties, 

Courts Leet, 20, 391 ; Welsh 
Customs, 4, 5 ; English v. 
Welsh Burgesses, 15 

Anglicising adjoining Parts of 
Wales, 24 ; Cymricised by 
Inflow of Welsh, 24 ; English 
Procedure ousts Welsh Me- 
thods, 21 

Illicit Opinions, 26, 26 ; Privi- 
leges of Burgesses, 15 ; Bai- 
liffs, 24 

Levelookers, 312 ; Names of 
Tenants, 31-34 ; Tidderley's 
Survey, 326, 327 ; Tenant's, 

"The Common Wood," 15; 
*»The Chekers" (Exchequer 
Tower), 16, 390, 394 

** Le Quarrer," 17 ; Werrock 
Lane( Weirhook, Warrhooke), 

Holt, Devon Brook, 322, 433 ; " Over- 
whart" Street, 322; Hog- 
more, 322 
" Parse" Lands, 323 ; Priest's 
Field, 403 ; Llewelyn and 
Grufl^dd Drowned under 
Bridge, 8 
Thomas Cholmondeley Custo- 
dian of Bridge and Passage, 

Howel, David ap, 23, 25 

Howland, Isabel de, 10 

Hoyle, Little, Remains Unearthed, 

Hoyle's Mouth, 373 

Hughes, Edward, High Sheriff of 
Denbigh-hire, 331 

Hughes, Harold, Report on Exca- 
vations at Tre'r Ceiri, 38-62 

Hutchins, Randolph, 389, 416 

lolyn, Edward ap David ap, 24 ; 

Morgan ap, 25 
Irish Language in Dyfed, I^ersist- 

ence o^ 91 
Tsycoed, Five Townships of, 3 
Ithel, David ap, 23, 25 
Itinerary in Jrales, Leland's, Edited' 

by L. T. Smith— Reviewed, 

254, 255 

John, Richard ap, Vicar of St. 
Peter's, Carmarthen, 338 

Kidwelly Castle, 240 ; Church, 241 
Kitchen, William, 318 
Knyght, Thomas, 24 

Lake Dwelling, Tally, 387 

Late-Celtic Objects, 387, 388 

*'Latymer," 26 

Laud, Bishop, Preaches at St. 
Peter's, Carmarthen, 343 

Laughame, General, 235 

Church, 235 : Pre - Norman 

Cross, 235 
Castle, 234 ; Winchester Mea- 
sure, TaUy Sticks, 235 

Lawgam, Robert ^Courtemain), 290 

Laws, E., on Prehistoric Man, 125, 
126, 374 

Leech, John, of Holt, 332 

Legh, Geoffrey, 23 

L'Estrange, Roger, 7 

Digitized by VjOOQ IC 



Leland's Itinerary in Wales, Edited 
L. T. Smith, Reviewed, 264, 

Lewis, Anthony, of Burton, Bequest 
to Gresford, 353 

Llan and Eglwys, 258, 259 

Llandaff, Oudocetis, Bishop of, 279 

Llandawke Church, 233, 276; 
Effigy Legends, 277 ; Font, 
278 ; Origin of Name, 278, 
279 ; Stone, 76 

Llanddowror = Llanddyfrwyr, 365 ; 
Church, 225 ; Pilgrim Stones, 
226 ; Rev. Griffith Jones, 
226 ; Madam Be van's Chair, 

Llandtulyddog and Carmarthen 
Identified, 281, 282 ; Given 
by Henry I to Battle Abbey, 

Llandyssilio Church, 243-246 -In- 
scribed Stones, Chalice, Vi- 
car's Diary, Communion 
Plate, 1651, Circulating Li- 

Llanfihangel Abercowin, Pilgrim's 
1 hurch, 216, 217 ; Tombs, 
217 ; Chalice, Font, 218 ; 
Pilgrim Stones, 222 ; Road, 

Llansaint, Rev. George Eyre Evanj>, 
63-65, 66-76 ; Inscribed 
Stones, 239, 240 

Llanstephan Church, 213 ; Castle, 
108-118, 214, 215 ; Maurice 
Fitzgerald, Lord of, 289 

** Lletfer," meaning, 309 

Lletrach= Irish Leitrioch, 365 

Lleweljm apGruffydd Breaks Truce, 
Storms Aberystwith and Ha- 
warden Castles, Attacks 
Rhuddlan and Flint Castles, 

Lloyd, Professor, J. E. — Carmar- 
then in Early Norman Times, 

Lloyd, Sir Richard of Esclus, De- 
fends Holt Castle, 331 

Llwyn Gwyddel, 365 

Llystyn Gwyn Stone, 96-102 

Lothar, Lancelot, 23, 24, 328, 329 ; 
Thomas, 318, 328 

Maddock, Edward, 416 
Madoc, Einion ap, 4 ; Fychan, P 

Maen Prenvol, 368 

Maes y Prior, 289 

Mainweyring, William, 24 

Marford (Merford) and Hoseley 
Annexed to Flintshire, ^ 
Henry VIH, 19 

Marios, St Margaret^ 266, 267 ; 
PhUip, 264 

Mollington, John, Forester, of 
Bromfield, 12 

Morgan, Col. W. LI.— Report on 
Excavations at Coelbren, 129- 
174 ; On Llanstephan Castle, 
214, 215; Bastion Traces, 238 

Moulton, Robert, Auditor of Wales, 

Mural Paintings, 266 

M>ddleton, Edward, Mayor of Car- 
marthen, 340, 341 

Neolithic Man : Conditions of Life, 
Weapons, Clothing, Pottery, 
Food, 376, 377 

Nevern Stone, 81 92 

Norden, John, 323-325 ; Sur\'ey of, 
1620, 389-407 

Norfolk, Thomas Duke of, Justice 
of Brooi field, 17 

Norman Arch (St. Clear's), 224; 
Font, 218 

Norman Supi-emacy secured in Car- 
marthenshire, 284 

Norman Times, Carmarthen in 
Early, 281-292 

Notes on East Window, Gresford 
Church— Rev. E. A. Fiah- 
boume, 352-357 

Nott, General Sir William : Monu- 
ment, 349 

Ogam (Barrivendi), 232, 233; 

Gwarmacwydd, 243 ; Eglwys 

Cymmyn, 260 
Llandawke, 279, 280; Deccaibar 

Vuglob, 297-306 
Oudoceus, Bishop of Llandaff, 279 
Owain ap Cadwgan, 285 ; Ap Cara- 

dog Defends Carmarthen 

Castle, 285 

Palmer, A. Neobard — ^The Town of 
Holt : Castle, Church, Fran- 
chise, 1-34, 311-334, 389-434 

Pant y Gwyddel, 365 

Pare Owning, 316 

Digitized by VjOOQ IC 



Pare y Ceryg Sanctaidd, 232, 233, 
258, 267-276 

Pare y Parsonage, 219 

Pate, Thomas, 389, 416, 417 

Pedigree— Bostocke of Holt, 330 ; 
PoweU of Holt, 420, 421 ; 
Powys Fadog, 8 

Pelunyawc CPeulini<^, Pelumawc), 
from Peulin, Paalinus, 362 

Pembrokeshire Archaeological Sur- 
vey, Treasurer's Statement, 

Pembrokeshire Association for the 
Preservation of Ancient 
Monuments, 437 

Pembrokeshire, Encroachments by 
Carmarthenshire on, 362, 
365; Attributed to Sir 
Thomas Jones, M.P., 363 

Perrott, Sir John, Tablet, 266 

Pickering, John, 22, 25, 319, 418 ; 
Francis, 389, 418 ; Sir Ed- 
ward, 418 ; Sir Christopher, 

Pilgrim Stones, 222, 226 ; Road, 220 

Pont y Cim, 310 

Porter, Hugh, 23, 26 

Powell, Thomas (of Hoseley), 317, 
318, 330 ; Pedigree, 420, 421 

Powys Fadog, Pedigree of Later 
Princes, 8 

Prehistoric Hearths in Carmarthen- 
shire, 380-383 

Prestland, Thomas, 24, 419 ; John, 
Lancelot, and Richard, 419 

Priceof Yale, David, 330; Robert, 

Prichard, Thomas, 348 

Puleston, John, 23, 25 ; John, jun., 
23, 26, 321 ; Roger, 317 
Sir Roger Opposes Grant to 
William Bentinok, Earl of 
Portland, 326 

Pulford of Barton, Thomas, 423 ; 
Of Holt, 389, 423; John, 
Lancelot, William, 409, 423 

Quaker's Yard (Holt), 430 

Reviews and Notices of Books, 264- 

Rhyd y Gors Castle, 237 ; Caer, 284 
Rhys, Professor John — Epigraphio 
Notes, 66-76 ; The Capel 
Mair Stone, 293-310 

6th 8BB., VOL. VII. 

Richard 11 at Holt Castle, 13 

Roberts, Clement, 323 

Robyn, David, 338 

Roden, Sir William, Rector of 

Gresford, 324 
Roydon, John, 23 ; Richard, 24, 

317 ; R<ier, 424, 425 

Salisbury, Ro^er, Bishop of, 291, 292 

Salusbury, John, 318 

Sam Helen wholly Roman, much 
used till Norman Times, then 
disused, 130 

Sayce, Professor, on Scientific Ex- 
cavation, 126 

St. David's, Bishop of, Bernard, 
Institutes House of Augus- 
tinian Canons, 287, 288; En- 
dows St. John's Priory, Car- 
marthen, 337 
Thomas Wallensis, 261 ; Wil- 
frid, 292 ; Peter de Leia, 337 

Scrope, William le. Earl of Wilt- 
shire, Justice of Chester, 13 

Seymour, Thomas, 320 ; Thomas, 
Lord Admiral, 320 

Speed, David, 389, 425 ; John (An- 
tiquary), 425 

Stanley, Thomas, Earl of Derby, 
352, 356 ; Sir WiUiam, Chief 
Justice of Wales, Chamber- 
lain of Chester — Wealth, 
Execution, 21, 22 ; Has Grant 
of Bromfield Lordship, 17 

Stansty Abbatis, 19 

Steele, Sir Richard, Buried in St. 
Peter's, Carmartlien, 344 ; 
Monument, 349 

Stone Circles, Relics of Bronze Age, 

Stone Coffin Lid, 345 ; Implements, 

Stones, Cursing, 271 ; Resting, 275 

Stones, Inscribed, 65-68 ; Llan- 
dawke, 76 ; Nevem, 81-92 ; 
Llystin Gwyn, 96-102 ; Tref- 
lys, 92-96 ; Penymynydd, 
134; Llansaint, 239, 240; 
Parcau, 243 ; Llandyssilio, 
243-246 ; Traws Mawr, 248. 
See also "Ogam." 

Survey of Bromfield and Yale, 
Tidderley's, 311 ; Norden's 
(Holt), 389 

Symkins, Richard, 318 


Digitized by VjOOQ IC 



TachlowmonsTelich Clowmon, 364 

Tafam Diflas, 268 

Talacharn (Laugharne), 364, 365 

Tally Sticks (Laugharae), 235 

Tarleton, Thomas, 23 

•* Taylor's Cushion," 362 

**Te Deum" Window, 355 

Teulyddoff, 282 

Tidderley s Survey of Bromfield and 
Yale, 311 

"Tir y Grarreg," corrupted into 
** Tithe Garret," 407 

Toch Castle, 273 

Tomb, Sir Rhys ap Thomas's, 347 

Toye, Robert, Mayor of Carmar- 
then, 340 

Trahem, John, Warden of Grey 
Friars, Carmarthen, 348 

Tranmoll, John, Forester of Brom- 
field, 12 

Treasurer's Accounts, 1906, C. A. A„ 
251 ; Pembrokeshire Survey, 

Trefenty — Passage, 219 ; Encamp- 
ment, 220 

Treflys, Note by Professor Rhys, 92 

Treheme, G. G. T., Notes on 
Eglwys Cymmyn, Parc-y- 
Ceryg Sanctaidd, and Llan- 
dawke, 257-280 

Tre'r Coiri — Introductory Note by 
Professor Boyd Dawkins, 36- 
37, 124, 125 
Relation to other Fortifica- 
tions, 35 ; Cyttiau Gwyddel- 
od, 35 
Compared with Primitive 

Houses in Lreland, 36 
Report on Excavations at — 

Harold Hughes, 38-62 
Details of Finds, 40-57 ; Bronze 

Torque, 40, 52 ; Pin, 47 
Entrances, 57-60 

Treth Dan (** Fyre Silver"), 21 

Trevor, John, 321 ; Sir John, Kt., 
426; Sir Richard 426 

Tumulus, Castle Hill, Kiffig, 386, 
387 ; Banc Benisel, 384 

Turbridge, Robert, 321 

Turstin, Prior, 292 

Twlc y Viliast, Cromlech, Ffynnon 
Newydd, 369 

Ty Cwrdd (Meeting House), 75 
Ty Gwyn ar Daf, 241-243 

Vernon, Richard, 389 

Walter of Glo'ster, 285 
Warburton, Sir Peter, Burgess of 

Holt, 403 
Ward, John — Roman Remains at 

Cwmbrwyn, 175-209, 226 
Warrenne, John de, Bromfield and 

Castle of Dinas Bran granted 

to, 8 ; Dies 1304, 9 
Warrenne, William de, 9 
Webbe, Sir David, Vicar of St. 

Peter's, Carmarthen, 338 
Welsh Court House, Holt, 391, 

Wesley, John, at Carmarthen, 344 
Whitland Abbey, 241-243 
Wilde (Weld, Wells, Welles, Wild), 

389, 426, 427, 428 
Wilkinson, John and Thomas, 389, 

Williams, Sir John, Bart., Presi- 
dent's Address, 108-118 
Sir W., Opposes Grant to Wm. 

Bentinck, Karl of Portland, 

Lancelot, 409 
Peter, 264 
Winchester Measure, Laugharne, 

Wodehay, Richard de, 24 
Womack, Lawrence, Bishop of St. 

David's, 343 
Wren, John, 25 ; Thomas, 25 
Wrexham, Abbot, 19 
Wrexham (Wrightesham, Wristles- 

ham), 2, 11 
Wrighte, John, 323, 389 ; George, 

Wristlesham (Wrexham), Castellum 

de, 2 
Wysham , John de, 9 

Yardley, John, 389, 429 
Ystrad House, 313 ; Pillar Stones, 

Digitized by VjOOQ IC 




Report on the ExcaTations carried out at Tre'r Ceiri in 1906 : 

Plan (Plate) ...... 38 

Fragments of Lance-head, Bronze Plates, Earthenware 

Vessels, etc. ..... 41-51 

Plan of South- West and North-West Entrance . . * 68 

Inscribed. Stones at Llansaint, Carmarthenshire (Two Plates) 64 

Epigraphic Notes : 

Inscribed Stone at Llansaint, Carmarthenshire . .67 

Inscribed Stone at Llandawke, Carmarthenshire . . 77 

Lintel Stone with Interlaced Work at Nevern . . 82 

Inscribed Stone No. 2 at Nevorn, Pembroke . . 84 

Inscribed Stone at Trefljs, Carnarvonshire . . 92 

Inscription on Stone at Treflys, Carnarvonshire . . 94 

Inscribed Stone at Lljstyn Gwjn, Carnarvonshire . 98 

Latin and Ogam Inscnption on the Llystyn Gwyn Stone 99 

Excavations at Coelbren : 

Map showing Position of Camp .... 131 
Plan of Camp (Plate) . . . . .136 

Sections of Camp .... 138-150 

Pottery and Glass found at Coelbren . . 160-1 (>6 

Roman Remains at Cwmbrwyn, Carmarthenshire : 

Section of 6-in. Ordnance Map of Carmarthenshire 

(Plate) 176 

Plan before Excavation (Plate) . . . . 1 78 

Plan after Excavation (Plate) . . . .180 

Plan of Building and Sections of Rampart and Ditch 

(Plate) . . . . . .182 

View of the Gateway looking to the North (Plate) . 182 

Various Articles found, etc. . . . 184-205 

Digitized by VjOOQ IC 



Norman Font in Llanfihangel Abercowin Church . . 223 

Inscribed Sbone at Eglwys Cymmjn, Carmarthenshire 232 

Pre-Norman Cross at Langhame, Carmarthenshire . . 234 

Inscribed Stone at Parcan ..... 242 

Inscribed Stone from Castell Dwyran (Plate) . . 244 

Inscribed Stones at Llandjssilio, Pembrokeshire . -. 244-245 

Inscribed Stone at Traws Mawr, Carmarthenshire (Plate) . 246 

Ende Pillar Stone at Traws Mawr . . . .248 

Capel Mair Inscribed Stone : 

Sketch of (Two Plates) .... 296, 297 

Existing Fragments .... 299, 301 

Ogam Inscriptions ..... 304 

St. Peter's Church, Carmarthen : 

Interior View, looking East .... 336 

Coat-of-Arms from St. John's Priory . . . 338 

Sepnlchral Slab of Richard .... 346 

Tomb of Sir Rhys ap Thomas . . . .347 

Remains of Ancient Bridge, Carnarvon : Plan and Elevation . 359 

The Town of Holt, in County Denbigh : 

Early Sketch of Holt Castle . . . .393 

Early Ground Plan of Holt Castle . . .395 

Elevation of Holt Castle in 1620 . . . , 397 

Ground Plan of HoU Castle in 1620 . . .398 


Digitized by VjOOQ IC 



9atron0. , 


The Right Hon. the EiiBL of Powis 

The Right Hon. the Earl of Cawdor 

The Right Hon. Earl op Plymouth (President, 1898 and 1899) 

The Right Hon. Viscount Trkdboar (President, 1885) 

The Right Rev. the Lord Bishop of St. David's 

The Right Rev, the Lord Bishop of Llandaff 

The Right Hon. Lord Dynevor 

The Right Hon. Lord Kenyon 

The Right Hon. Lord Mostyn ( President y\H90) 

The Right Hon. Lord Penrhyn (President, 1894) 

The Right Hon. Lord Aberdare (President^ 1900) 

The Right Hon. Lord Harlech 

The Right Hon. Lord Llamoattock 

The Right Hon. Lord Swansea 

Sir John Williams, Bart. 

Sir Richard H. Williams Bulkelsy, Bart. 

H. R. Hughes. Esq., Lord Lieutenant of Flintshire 

Sir John Evans, D.C.L., F.R.S., V.P.S.A. 

Sir C. E. G. Philipps, Bart. {President, 1880 and 1883), Lord Lieutenant 

of the Town and County of HaverJPordwest 
R. H. Wood, Esq., F.S.A., F.R.G.S. (President, 1903) 
Sir John Talbot Dillwyn Llewelyn, Bart., M.A., F.L.S. (President, 

Lient.-Col. C. S. Mainwaring (President, 1887) 

Digitized by VjOOQ IC 


M. k Dr. DE Olosmadeuc (Pretidmt, 1889), Prudent de la Soci^t^ 

Polymathique da Morbihan 
JohN Rhys, Esq., M.A., LL.D. (President, 1891), Professor of Celtic, 

and Principal of Jesas College, Oxford 
W. Boyd Dawkins, Esq., M.A., F.R.S., F.S.A., Professor of Geology, 

Owens College, Manchester 
The Rev. A. H. Sayce, M.A., LL.D., Professor of Assyriology, Oxford 
The Rev. Hugh Pbiohabd, M.A. 

The Yen. Archdeacon Thomas, M.A., F.S.A. (President, 1906) 
Sir Jambr Williams Drummond, Bart, (/'reairfen^, 1892), Lord Lieutenant 

of Carmarthenshire 
Sir Owen H. P. Scourfiild, Bart. (President, 1897) 
Edward Laws, Esq., F.S.A. 
The Rev. Canon Rupert Morris, D.D., F.S.A. 
J. W. WiLUs-BuND, Esq., F.S.A. (President, 1905) 
Henry Owen, D.C.L., F.S.A. 
The Rev. Prebendary Garnons- Williams, M.A. 
Tlie Rev. 8. Baring-Gould, M.A. 

W. R. M. Wynne, Esq. (Lord Lieutenant of Merionethshire). 
Sir John Williams, Bart. 


The President, with all those who have held that office ; the Yioe-Presi- 
dents ; the Treasurer ; the General and Local Secretaries ; and the 
Editorial Sub-Committee, with the following : 

Yen. Archdeacon Thomas, M.A., F.S.A., Chairman, 

Thos. Mansel Franklen, Esq. 

Rev. John Fisher. 

Rev. E. J. Newell. 

Professor Anwyl. 

Professor Lloyd. 

Professor Powel. 

Iltyd NichoU, Esq., F.S.A. 

H.Harold Hughes, Esq.. A.R.LB.A. 
J. Romilly AUen, Esq., F.S.A. 
J. Ward, Esq., F.S.A. 
Mrs. Allen. 
Edward Owen, Esq. 
A. N. Palmer, Esq. 

J. Romilly Allen, Esq., F.S.A., 28, Great Ormond Street, W.C. 

^Iritotial J^ttb^Committre. 

Professor Rhys, M.A., LL.D. 

The Rev. Canon R. Trevor Owen, M.A., F.S.A. 

Worthington G. Smith, Esq., F.L.S. 

Col. W. L. Morgan, R.E., Brynbriallu, Swansea. 


R. H. Wood, Esq., F.S.A. 
W. R. M. Wynne, Esq. 
Colonel W. Gwynne- Hughes 

Digitized by VjOOQ IC 



ffenetal 3^ttt$tsti$§. 

Rev. Canon R Trevor Owen, M.A., F.S.A., Bodelwjddan Vicarage, 

Rhaddlan (Flintshire), S.O. 
Bey. Gharles Chidlow, M.X, Llawhadon Vicarage, Narberth 

Corre»9onlriitg ^tttttavi$%, 

France — Mons. Oharles Mettier, F.S.A., Oaen 

Brittany — M. de Keranfleo'h Kernezne, Oh&tean de Qn^l^neo, Mar de 

Bretagne, Cdtes du Nord, France 
Scotland — Joseph Anderson, Esq., LL.D., Mosenm of Antiquities, 

Ireland— R. Cochrane, Esq., I.S.O., F.S.A., 17, Highfield Road, Rathgar, 

Cornwall — Edwyn Parkyn, Esq., Royal Institute, Truro 

Rev. S. Baring-Gould, Lew Trenchard Rectory, N. Devon 

Aonotars MtmUn. 

M. Alexandre de Bertrand, Paris 

Mons. Charles Hettier, F.S.A., Caen, France 

floral ^tn$tati$%, 

Anglesey .... J. E. Griffith, Esq., F.R.C.S., P.L.S., Bryn Dinas, 
Upper Bangor 

Professor J. Morris Jones, Tycoch, Llanfair, P.G. 
CamarvoTishire . . Edw. Roberts, Esq., M.A., H.M.I.S., Carnarvon 

T. E. Morris, Esq., LL.M., Lombard St., Portmadoo 
Denbighshire . . A. Foulkes-Roberts, Esq., 34, Vale Street, Denbigh 
Flintshire . . . Rev. W. LI. Nicholas, M.A., Rectory, Flint 

L. J. Roberts, Esq., H.M.I.S., Rhyl 
Merionethshire . . Rev. J. E. Davies, M.A., The Rectory, Llwyngwril 
Montgomeryshire . 
Brecknockshire . . Lieut. -Col. R. D. Garnona- Williams, Ty Mawr, 

Cardiganshire . . Professor Anwyl, M.A., Univ. College of Wales, 

J. H. Davies, Esq., M.A., Cwrtmawr, Aberystwyth 

H. M. Vaughan, Esq., Plas, Llangoedmore, Cardigan 
Carmarthenshire . Alan Stepney-Gulston, Esq., Derwydd, Llandebie 

Rev. D. D. Evans, B.D., Llangunnor Vicarage, 
Olamorganshire . Thos. Powel, Esq., M.A., University College, Cardiff 

C.Wiikin8,Esq.,F.G.S., Springfield, Merthyr Tydfil 

C. H. Glascodine, Esq., Cae Pare, Swansea 

D. Lleufer Thomas, Esol, Hendre, Swansea 
Pembrokeshire . . Herbert J. Allen, Esq., Norton, Tenby 

Rev. James Phillips, Haverfordwest 
Radnorshire . . . Rev. E. H. Day, D.D., Abbey Cwm Hir, Penybont 
Monmouthshire . . A. E. Bowen, Esq. , Town Hall, Pontypool 
Hie Marches . . JamesDavies, Esq., Gwynfa,Broomy Hill, Hereford 

Rev. C. H. Drinkwater, M.A., St. George's Vicarage, 

Henry Taylor, Esq. F.S.A., Curzon Park, Chester 

Digitized by VjOOQ IC 




His Majesty the Kino . 
Swansea, The Rt. Hon. Lord 
Allen, Mrs. Thomas . 
Allen, W. Bird, Esq., M.A. 
Allen, J. Romilly, Esq., F.S.A. 
Alien, Rev. W. Osborn, M.A. . 
Asher, Messrs., and Co. 
Baring-Gould, Rev. S., M.A. . 

Baynes, E. Neil, Esq. 
Biblioth^ue Nationale 

Birmingham Free Library . 
Blundell, Joseph Hight, Esq. 
Bond, F. Bligh, Esq. 
Boston Public Library, U.S.A. . 

Bridger, E. K., Esq. 
Brownrigg, The Very Rev. Dean 
Bumard, R., Esq., F.S.A. . 
Obambres, Rev. G. C, M.A. . 
Chetham Library 
Cochrane, REsq., F.S.A.,I.S.O., 

Hon. Sec. Royal Society of 

Antiquaries, Ireland 
Columbia University . 

Cnnliffe, Major J. Williams 
Cunnington, B. Howard, Esq., 


Cunnington, Mrs. B. Howard 
Davies, Timothy, Esq., M.P. . 
Dawkins, W. Boyd, Esq., M.A., 

F.R.S., F.S.A. 
Dawson, Mrs. 

D'Arbois de Jubainville, M. 
De Keranflec'h Kemezne, M. 

Detroit Public Library 

Buckingham Palace, S.W. 

24, Motcombe St., Belgrave Square, W. 

42, Connanght Square, W. 

158, Portsdown Road, Maida Vale, W. 

28, Great Ormond Street, W.C. 

83, St. George's Road, S.W. 

13, Bedford St., Covent Garden, W.C. 
Lew Trenchard Rectory, Lew Down, 

N. Devon 
120, Warwick St., Eccleston Sq., S.W. 
Paris (c/o Continental Export Co., 
4, High Street, Bloomsbury, W.C.) 
Birmingham (c/o A. Capel Shaw, Esq.) 
157, Cheapside, E.G. 
Star Buildings, Bristol 
(c/o Messrs. Stechert & Co., Star Yard, 
Carey Street, Chancery Lane, W.C. 
Berkeley House, Hampton-on -Thames 
Booking, Braintree 
Huccaby House, Princetown, S. Devon 
Grammar School, Wigan 
Manchester (c/o W. T. Browne, Esq.) 

17, Highfield Road, Rathgar, Dublin 

New York, U.S.A. (c/o Messrs. 

Stechert & Co., Star Yard, Carey 

Street, Chancery Lane, W.C.) 

17, Inverness Terrace, Hyde Park, W. 



Pant y Celyn, Putney 

Fallowfield House, Fallowfield, 

Hartlington Hall, Skipton 

84, Boulevard Mont Pamasse, Paris 
Ch&teau de Qu^l^nec, Mur de Bre- 

tagne, Cdtes du Nord, France 
(c/o Mr. B. F. Stevens, 4, Trafalgar 
Square, W.C.) 

* Members admitted since the Annual Meeting. 1906, have an asterisk prefixed to 
heir names. 

Digitized by VjOOQ IC 


Edwards, J. Watkin, Esq. . 
Evans, Sir John, F.R.S., K.C.B. 
Evans, E. Vincent, Esq. (Hon. 

Sec, Honourable Society of 

Cymmrodorion) * . 
Fryer, Alfred, Esq., Ph.D., 

M.A., F.S.A. 
Gabriel, J. R., Esq., M.A. 

Gordon, Mrs 

Griffiths, Joseph, Esq., M.D. . 
Guildhall Library, E.G. . 
Harford, Mi&s .... 
Hartland, Ernest, Esq., M.A., 


Hartland, E. Sidney, Esq., F.S.A. 
Harvard College Library . 

Hereford Free Library 

Ho worth. Sir Henry H., 

K.C.LE., F.R.S., F.S.A. . 
Humberston, Mrs. — 
Jackson, J., Esq. 
James, 0. R., Esq., . 
Jesus College Library 
Jones, E. Alfred, Esq. 
Jones, Rev. G. Hartwell, M.A. . 
Jones, Lawrence, Esq. 
Jones, Robert, M.D., F.R.C.S. . 
Jones, W. E. T., Esq. 
*Jones, J. Prichard, Esq. . 
Joseph-Watkin, T. M., Esq. . 

King's Inns' Library 
Lewis, William F., Esq. . 

Liverpool Free Public Library . 
Lloyd, Alfred, E8q.,F.C.S.,F.E.S. 
London Library 
Manchester Free Library . 
'"'Masterman, John Story, Esq. . 
Melbourne Public Library . 

Morris, The Rev. Canon Rupert 

H., D.D., F.S.A. . 
Morris, T. E., Esq., LL.M. 
McClure, Rev. Edmund, M.A. . 
New York Library . • • 

46, Albert Terrace, Middlesbrough 
Brittwall, Berkhampstead, Herts 

64, Chancery Lane, W.C. 

13, Eaton Crescent, Clifton, Bristol 
Technical College, Swindon 
9, St. German's Place, Blackheath,S.E. 
1, St. Peter's Terrace, Cambridge 
(c/o Edward M. Borrajo, Esq.) 
Blaise Castle, Henbury, Bristol 

Hardwick Court, Chepstow 

Highgarth, Gloucester 

Cambridge, Mass., U.S.A. (c/o Messrs. 
Kegan Paul, Triibner A Co., 
43, Gerrard Street, Soho, W.C.) 


30, CoUingham Place, S.W. 
ll,Pelham Crescent, Kensington, S.W. 
25,Leazas Terrace, Newcastle-on-Tyne 

5, Raymond's Bldgs., Gray's Inn, W.C. 

Hampden House, Phoenix Street, N.W. 
Nutfield Rectory, Redhill 

6, Water Street, Liverpool 
Claybury, Woodford, Essex 

6, Stone Buildings, Lincoln's Inn, W.C. 

76, Canfield Gardens, Hampstead, N. 

Herald's College, Queen Victoria Street, 

Dublin (J. Carter, Esq.) 

2109, Walnut Street, Philadelphia, 

Liverpool (c/o Peter Cowell, Esq.) 
The Dome, Upper Bognor, Sussex 
St. James's Square, S.W. 

St. Margaret's, Dorking 
c/o Agent-General for Victoria, 142, 

Queen Victoria Street, E.C. 
St. Gabriel's Vicarage, 4, Warwick 

Square, S.W. 
8, Fig-Tree Court, Temple, E.C. 
80, Eccleston Square, S.W. 
New York (c/o Mr. B. F. Stevens, 

4, Trafalgar Square, W.C.) 

Digitized by VjOOQ IC 



Nonnan, George, Esq., M.D. 
Owen, Edward, Esq. . 
PenQsylvania Historical Society 

Peter, Tharstan C, Esq. . 
Preece, Sir W. Henry, K.C.B., 


Price, Hamlyn, Esq. . 

Price. — , Esq 

Pritchard, John Jones, Esq. 

Pritchard, L. Jones, Esq. . 

Prichard, Rev. R. W., M.A. 

Prichard-Morgan, W., Esq. 

Rennes, Biblioth^qae Universi- 

Rhys, John, Esq., M.A., LL.D., 
Professor of Celtic and Princi- 
pal of Jesns CoUege 

Roberts, Sir Owen 

Rock, J. Dennis, Esq. 

Sayce, Rev. A. H., LL.D., Prof. 

of Assyriology 
Smith, Worthington G., Esq., 


Sydney Free Public Library 

Taylor, His Honor Judge, K.C. . 
Thomas, Lieut. -Col. G. T. 
Thomas, Rev. W. Mathew, M.A. 

Toronto Public Library 

Vaughan, H. F. J., Esq. . 

Vaughan- Williams, F., Esq. 

Willis-Bund, J. W., Esq., F.S.A. 
Williams, Miss M. C. L. . 
♦Wynne, 0. M. E., Esq. . 

12, Brock Street, Bath 

India Office, Whitehall, 8.W. 

(c/o Messrs. B. F. Stevens & Brown, 

4, Trafalgar Square, W.C.) 
Redruth, Cornwall 

Gothic Lodge, Wimbledon 

lo. King Street, St. James's Square, 

43, Pall Mall, London, S.W. 
6, Stanley Road, Waterloo, Liverpool 
Menai Lodge, Chiswick, W. 
Stoke Vicarage, Chester 
1, Queen Victoria Street, E.G. 

Rennes, Ille-et-VilaiDe, France 

Jesus College, Oxford 
Horley Park, Guildford 
Oharlemont, Eliot Park, Blackheath, 

Queen's College, Oxford 

121, High Street North, Dunstable 
(c/o Messrs. Truslove, Hanson & Co., 

163, Oxford Street, W.) 
4, Harcourt Buildings, Temple, E.C. 
The Bush, Walton-on-Thames 
Billingboro' Vicarage, Folkingham, 

S.O., Lincolnshire 
(c/o Messrs. C. D. Cazenove & Son, 26, 

Henrietta St., Covent Garden, W.C.) 
30, Edwardes Sq., Kensington, W., 
and Humphreston Hall, Salop. 

HoUyhurst, Barton under Needwood, 

15, Old Square, Lincoln's Inn, W.C. 
6, Sloane Gardens, S.W. 
67, Torrington Square, W.C. 

Digitized by VjOOQ IC 



Beade, Lady .... 
Bulkeley, Sir Biohard H. 

Williams, Bart., Lord Lieu- 

tenaot of Anglesey. 
Meyrick, Sir George, Bart. 
Vemey, Sir Edmund, Bart. 

Adeane, Miss .... 
Evans, Rev. Evan 
Griffith, Rev. Ellis Hughes 
Hall, Rev. W. E. Scott . 
Hampton-Lewis, Mrs. 
Jones, Professor J. Morris 
Lloyd Theakstone, Mrs. . 
^Massey, Miss .... 
^Massey, Miss Gwendolin . 
Prichard, Rev. Hugh, M.A. 
Prichard, Thomas, Esq. 
''^ThomaB, Fleet - Surgeon J. 

Lloyd, R.N 

Williams, Rice R., Esq. . 

Carreg-lwyd, The Valley, S.O. 

Baron Hill, Beaumaris, S.O. 

Bodorgan, Llangefni, S.O. 

Claydon House, Winslow, Bucks ; and 

Rhianva, Menai Bridge 
Plas Llanfawr, Holyhead [S.0 

Llansadwm Rectory, Menai Bridge, 
Llangadwaladr Vicarage, Llangefni, 
Plas, Llanfaelog [S.O. 

Henllys, Beaumaris, S.O. 
Tycoch, Llanfair P.G., S.O. 
Fir Grove, Menai Bridge, S.O. 
Comely, Beaumaris, S.O. 
Comely, Beaumaris, S.O. 
Dinam, Gaerwen, S.O. [S.O. 

Llwydiarth Esgob, Llanerchymedd, 
Denmor, Penmon Beaumaris ; and 

Royal Naval Barracks, Chatham 
Stanley House, Holyhead 


Greaves, J. E., Esq., Lord 
Lieutenant of Carnarvonshire 
Mostyn, The Lady Augusta 
Penrbyn, Rt. Hon. Lord . 
Arnold, Professor E. V., M.A. . 
Breese, Charles E., Esq. . 
Davids, Miss Rose 

Davies, John Issard, Esq., M.A. 

Davies, J. R., Esq. 

Dodson, William M., Esq. . 

Evans, Colonel O. LI. G. . 

Evans, Mrs. Lloyd 

Foster, W. A., Esq. 

Gardner, Willoughby, Esq., 

F.L.S., F.R.G.8. . 
Griffith, J. E., Esq., F.R.A.S., 


Hughes, H. Harold, Esq., 


Job, Rev. J. T 

Jones, C. A., Esq. 

Bron EijQon, Criccieth 

Gloddaeth, Llandudno 

Penrhyn Castle, Bangor 

Bryn Seiriol, Bangor 

4, Marine Terrace, Portmadoc 

Greenhall, High Blantyre, N.B. ; and 

Plas Llanwnda, Carnarvon 
Llysmeirion, Carnarvon 
Ceris, Bangor 
Bettws-y-coed, S.O. 
Broom Hall, Chwilog, S.O. 
Broom Hall, Chwilog, S.O. 
Glyn Menai, Bangor 

Y Berlfa, Deganwy 

Bryn Dinan, Upper Bangor 

Aelwyd, Bangor 
Bethesda, Bangor 

Digitized by VjOOQ IC 


List O^ MBMBfiRS. 

Jones, L. D., Esq. . 
Jones, Bey. Canon, M.A. . 
■^Lewifl, ReT, J. P. . 
Lloyd- Jones, Miss 
Lloyd, Prof. John Edward, M.A. 
Morrice, Rev. J. C, B.A. 
North, Herbert L., Esq. 
Owen, Cledwyn, Esq. 
Parry, R. Ivor, Esq. 
*Pany, W. H., Esq. . 
Roberts, E.,Esq., H.M.I.S.,M.A. 
Roberts, T. E., Esq. 
Thomas, Owen Edward, Esq. 
University College Library 
Watts-Jones, Mrs. H. 
^Williams, Mrs. Ignatius . 
Williams, W. P., Esq. 
Williams, J. A. A., Esq. . 
Waiiams, Rev. J. E. 
Wyatt, J. W., Esq. . 

3, Edge Hill, Garth, Ban((or 

The Vicarage, Llaudegai, Bangor 

Llanystumdwy Rectory, Criocieth 

Pennillt, Penmaenmawr, S.O. 

Waen'Deg, Bangor 

5, Brynteg Terrace, Bangor 

Llanfairfechan, S.O. 

PwllheH, S.O. 

Pwllheli, 8.0. 

Pabo Isaf, Conway 

Plas Maesincla, Carnarvon 

Plas-y-Bryn, Carnarvon 

301, High Street, Bangor 


Glyn, Penmaenmawr 

Hendregadredd, Pentrefelin, Port- 

Cae'r Onnen, Bangor [madoc 

Aberglaslyn, Beddgelert, Carnarvon 

The Vicarage, Portmadoc 

Bryn Gwynan, Beddgelert, Carnarvon. 


^Dundonald, The Countess of, . 
Williams- Wynn. Lady 
Williams -Wynn, Sir Watkin, 

Bart., C.B., Lord Lieut, of 

McLaren, Sir Chas. B. B., Bart., 

K.O., M.P 

Barnes, Mrs 

Behrens, George B., Esq. . 
Berkeley, A. E. M., Esq. 
Darlington, James, Esq. . 
Davies, D. S.. Esq. . 
Davies, John, Esq. . 
Ellis, Rev. E. Lodwick . 
Fisher, Rev. John, B.D. . 
Foulkes-Roberts, A., Esq. . 
Foulkes, E. A., Esq. 
Halhed, Wm. B., Esq. 
Harrison, 8. H.,Esq., F.R.G.S., 


Hughes, Edward, Esq. 
Hughes, J. 0., Esq. 
Hughes, Rev. Meredith J. 
Jones, T. E., Esq., M.D. . 
Jones, A. Seymour, Esq. . 

Gwyrch Castle, Abergale, S.O. 
Llangedwyn, Oswestry 

Wynnstay, Rhnabon 

Bddnant, Eglwysfach, S.O. 
The Quinta, Chirk, Rhuabon 
Caerfedwen, Trefnant, S.O. 
Gredington, Whitchurch, Salop 
Black Park, Rhnabon 
Castle House, Denbigh 
Bryn-y-Parc, Denbigh 
Bettws Vicarage, Abergele 
Cefn Rectory, St. Asaph 
34, Vale Street, Denbigh 
Eriviatt, Denbigh 
Brynderwen, Llanrwst 

Cartrefle, Abergele, S.O. 
Glyndwr, Bersham Road, Wrexham 
Estate Office, Llangedwyn, Oswestry 
Brynymaen Vicarage, Colwyn Bay 
Henar, Llanrwst 
Pendwr, Wrexham 

Digitized by VjOOQ IC 



Jones-Bateman, Bey. B., M.A. 
Kyrke, B. V., Esq. . 
Lloyd, Humphry, Esq. 
Lynch, Francis, Esq. 
Mainwaring, Lieot.-Gol. 
Morris, £., Esq., M.A. 
Morris, John, Esq. 
Palmer, A. N., Esq. . 
Boberts, Bev. C. F., M.A. 
Sandbaoh, Colonel 
Trevor- Parkins, The Wor. 

Williams, Thomas, Esq. 
Wynne, Mrs. F. 

Pentre Mawr, Abergele 
Nant-y-firidd, Wrexham 
Morannedd, Llanddulas, Abergele, S.O. 
Glasooed, Wrexham 
Galltfaenan, Trefnant, S.O. 
Walmer Villa, Wrexham 
Lletty Llansannan, Abergele, ,3.0. 
1 7, Bersham Boad, Wrexham 
Llanddnlas Bee tory,. Abergele, S.O. 
Hafodunos, Abergele, S.O. 

Glasfryn, Gresford, Wrexham 
Llywesog, Denbigh 
Ystrad Cottage, Denbigh 

Hughes, Hugh B., Esq., Lord Kinmel Park, Abergele, S.O. (Den- 

Lieutenant of Flintshire 
Kenyon, Bight Hon. Lord 
Mostyn, Lady . 
Mostyn, Bight Hon. Lord 
St. Asaph, Very Bev. Dean of 
Davies, Bev. W. J., B.A. 
Felix, Bev. J. . . . 
Godsal, Philip T., Esq. . 
Hook, Bev. Paul 
Jones, Bev. D., MA. 
Lewis, W. A., Esq., M.A. 

Gredington, Whitchurch, Salop 
Talacre, Bhyl ^ 

Mostyn Hall, Mostyn 
Deanery, St. Asaph 
Bryntirion, St. Asaph 
Cilcain Vicarage, Mold 
Isooed Park, Whitchurch, Salop 
St. Mary's College, Holywell 
The Vicarage, Gorsedd, Holywell 
Glangwynedd, Bhyl 

♦Meredith, W. F., Esq. . . St. Mary's College, Holywell Mi/e) 

Mesham, Colonel 
Morris, Dr. — • • . 

Nicholas, Bev. W. LI., M.A. . 
Owen, Bev. Canon B. Trevor, 

M.A., F.S.A 

Pennant, Philip P., Esq., M.A. 
Powell, Bev. E. W., M.A. 
♦Price, W. Fred., Esq. . 
Bawlina, F. L., Esq. 
Bichardson, Mrs. 
Boberts, L. J., Esq., H.M.LS. . 
St. Beuno*s College Library 
St. DeinioUs Library, 
Storey, W. J. P., Esq. 
Tayleur, C. Bichard, Esq. 
Vaughan-Jones, Bev. W., M.A. 
Williams, Bev. B. 0., M.A. 
Williams, P. Mostyn, Esq. 

Pontruffydd, Trefnant, S.O. (Denbigh- 
Bodowen, Holywell 
The Bectory, Flint 

Bodelwyddan Vicarage, Bhuddlan, S.O. 

Nantllys, St. Asaph 

St. Asaph 

Fron Haul, Bodfari, Trefnant, S.O. 

Bhyl {Denbigh$hire) 


Tegf an, Bussell Boad, Bhyl 

St. Asaph 

Ha warden, Chester 

Preswylfa, Bhyl 

Brynllithrig, St. Asaph, S.O. 

Mostyn Vicarage, Holywell 

Bose Hill, St. Asaph, S.O. 


Digitized by VjOOQ IC 




Wynne, W. R. M., Esq., Lord 
Ideutenant of Merionethshire 
Owen, Lady 
Davies, Rev. J. E., M. A. 
Davies, R. O., Esq. . 
Dodd, F. R., Esq. . 

Griffith, Edward, Esq. 
Griffith, Miss Lucy . 
Leigh-Taylor, John, Esq. 
Morris, R. Jones, Esq. 
Owen, Rev. William 
Owen, Owen, Esq. 
Vaughan, Rev. T. H., B.A., . 
Wynne- Williams, Kor O., Ebq. . 
Wood, R. H., Esq., F.S.A., 

Peniarth, Towyn, S.O. 
Cae'r Ffynnon, Talsamau, S.O. 
The Rectory, Llwyngwril, S.O. 
The Square, Blaenau Festiniog 
Intermediate School, Blaenau Festi- 
Ooedcymmer, Dolgelly 
Arianfryn, Dolgelly 
Penmaen Uchaf , Dolgelly 
Tycerrig, Talsamau, S.O. 
Llanelltyd Vicar*ige, Dolgelly 
Llys Dorvil, Blaenau Festiniog 
Glyndyfrdwy Vicarage, Llangollen 
Bronwylfa, Llanderfel, Corwen 
Belmont, Sidmouth, S. Devon ; and 
Pant-glas, Trawsfynydd 


Lord Lieutenant of Shropshire 

Pryce-Jones, Lady . 

Dugdale, J. Marshall, Esq., M. A. 

Jones, Pryce Wilson, Esq. 

Jones, R. E., Esq. 

Leslie, Mrs. 

Lewis, Hugh, Esq , M.A. 

Lloyd Vemey, Mrs. . 

Lomax, J., Esq. 

^Macnair, Mrs. 

Mytton, Captain 

Phillimore, Egerton, Esq. 

Pughe, Mrs. Arthur 

Rees, Dr. . 

Temple, Miss . 

Thomas, Yen. Archdeacon, M.^., 

Turner, E. R. Horsfall, Esq . 

Willans, J. Bancroft, Esq. 

Powis Castle, Welshpool 
Dolerw, Newtown, Mont. 
Llwyn, Llanfyllin, Oswestry 
Gwynfa, Newtown, Mont 
Cefn Bryntalch, Abermule, S.O. 
Bryntanat, Llansantfraid, Oswestry 
Glan Hafren, Newtown, Mont. 
Clochfaen, Llangurig, Llanidloes 
Bodfach, Llanfyllin, Oswestry 
Pennal ToWer, Machynlleth 
Garth, Welshpool 
Penrhos Arms, Cemmaes, S.O. 
Gwyndy, Llanfyllin, Oswestry 
Caersws, S.O., Mont. 
Llandysilio, Oswestry 
Llandrinio Rectory, Llanymynech, Os- 
westry; and The Canonry, St. Asaph 
Llys Efrog, Llanidloes, S.O. 
Dolforgan, Kerry, Newtown, Mont. 

Digitized by VjOOQ IC 



Bradley, Mrs Cefn Pare, Brecon 

Dawson, Mrs Scethrog, Bwlch, S.O 

Evans, David, Esq. 

Oamons- Williams, Lt.-Col. R.D. 

Jebb, J. A., Esq. 

Jones, Rev. M. H., B.A. . 

Price, Rev. John, M.A. . 

Williams, Rev. Preb. G., M.A. 

Wood, Thomas, Esq. 

Ffrwdgrech, Brecon 
Tyraawr, Brecon 
Watton Mount, Brecon 
Trevecca OoUege, Talgarth, Brecon 
Uanf eigan Rectory, Brecon 
Abercamlais, Brecon 
Gwemyf ed Park, Three Cocks Junc- 
tion, S,0. 


Davies-Evans, Lieut.-Col. H., 
Lord Lieut, of Cardiganshire 

Lloyd, Right Rev. John, D.D., 
Bishop Suffragan of Swansea 

Anwyl, Prof essor, M.A. 

Bebb, R«v. J. M. LI., M.A. . 

DavieB, J. H., Esq., M.A. . 

Davies, John, Esq. . 

Ellis, Thomas, Esq. 

Evans, Captain E. W. W. 

Evans, Rev. George Eyre . 
Francis, J., Esq. 
Hughes, Joshua, Esq. 
Lewes, Miss Evelyn . 
Lloyd, Charles, Esq., M.A. 
Phillips, Rev. T. 
Pritchard, Dr. ... 

Pritchard, Mrs. 
Rees, Rev. R. J., M.A. . 
Roberts, T. F.,Esq., M.A., Prin- 
cipal of Univ. ColL of Wales . 
Rogers, J. E., Esq. . 
St. David's Coll., ThoLibrarian of 
Samuel, David, Esq., M.A. 
Vaughan, Herbert M., Esq. 
Waddingham, T. J., Esq. . 
Williams, Ven. Archdeacon, M,A. 
Yerward- James, W. E., Esq. . 

Highmead, Llanybyther, S.O. 

The Vicarage, Lampeter 

Univ. Coll. of Wales, Aberystwyth 

St. David's CoUege, Lampeter 

Cwrtmawr, Aberystwyth 

Bridge Street, Lampeter 

Glascoed, Aberystwyth 

Blenheim Club, 12, St. James* Square, 

S.W. ; and Camnant, Llandyssul 
Ty Tringad, Aberystwyth 
Wallog, Borth, S.O. 
Rhosygadair Newydd, Cardigan 
Ty-Glyn Aeron, Ciliau Aeron, Cardigan 
Waunifor, Maes y Crugiau, S.O. 
Aberporth Rectory, Cardigan 
The Priory, Cardigan 
The Priory, Cardigan 
Rhos, Aberystwyth 


Abermeurig, Felinfach, Cardiganshire 



Plas Llangoedmore, Cardigan 

Havod, Devil's Bridge, S.O. 

Abergeldie House, Aberystwyth 

Cae Morgan, Cardigan 

Digitized by VjOOQ IC 



WilliamA-Drammond^Sir J.,Bart., 

Lord Lieat. of Carmarthenshire 
St David's, The Lord Bishop of 
Dyneyor, The Bight Hon. Lord 
Williams, Sir John, Bart., M.D. 
Barker, T. W., Esq. . 
Bishop, His Honour Judge 
Brigstocke, T. £., Esq. 
Bnokley, Captain James 
Camber- Williams, Rev. Canon 

R., M.A 

Collier, Ernest, Esq., M.S.A. . 
Davies, A. Llewelyn, Esq. 
Davies, Rev. D. H. . 
Davies, Rev. Wm. 


Edwinsford, Llandeilo, S.O. 

The Palace, Abergwilly, Carmarthen 

Dynevor Castle, Llandeilo, S.O. 

Plas Llanstephan, Carmarthenshire 

Diocesan Registry, Carmarthen 

Dolygarreg, Llandovery 

54, King Street, Carmarthen 

Castell Gorfod, St. Clears, S.O. 

Evans, Mr^. Colby 
Evans, Rev. D. D., B.D. . 
Gwynne-Hughes, Colonel W. 
Hughes, John, Esq. . 
James, Daniel, Esq. . 
Jones, John D., Esq* 

The Parade, Carmarthen 

4, Quay Street, Carmarthen 

Brynderw, Carmarthen 

Ffinant, Newcastle Emlyn 

Vicarage, Llanfihangel Abercowin,. 

St. Clears 
Guildhall Square, Carmarthen 
Llangunnor Vicarage, Carmarthen 
Glancothy, Nantgaredig, S.O. 
Belle Vne, Llandeilo 
Vrondeg, lilandeilo 
Post Office* Carmarthen 

Johnes, Mrs Oolancothy, Llanwrda, S.O. 

Lloyd, H. Mearic, Esq.» M.A. 
Morgan, J. B., Esq. . 
Morris, Rev. J., M.A. 
Poole-Hughes, Rev. W. W.,M.A. 
Spurrell, Walter,Esq. 
Stepney*Gulston, Alan J., Esq. 
Thomas, Rev. Griffith 
Thomas, Rev. John, M.A. . 
Thomas, Rev. O. Jones 
Thomas, Mrs. R. M. 
Treheme, G. G. T., Esq. . 
Wheldon, J. P., Esq. 
Williams, Rev. J. A. . 
WUliams, Rev. R. , M. A. . 
Williams, W. Llewelyn, Esq., 

M.A., M.P 

Williams, Mrs. W. J. 

Delfryn, Llanwrda, S.O. 

50, New Road, Llanelly 

Vicarage, Llanybyther, S.O. 

The College* Llandovery 


Derwydd, Llandebie, S.O. 

Troedybryn, Carmarthen 

Laugharne Vicarage, St. Clears, 8.0. 

Llandyssilio Vicarage, Clynderwwn 

Llanddowror, St. Clears, S.O. 

7, Bloomsbury Square, London, W.O. 

National Provincial Bank, Carmarthen 

Llangathen Vicarage, Golden Grove, 

Vicarage, Llandeilo [S.O. 

Lamb Buildings, Temple, E.C. ; and 

Glansawdde, Llanipeuiock 
21, Picton Terrace, Oarmarthea 


. Plymouth, The Rt. Hon. Earl of. 
Lord Lient. of Glamorganshire 

«Bute, The Most Noble the 
Marquis of ... . 

Llandaff, The Lord Bishop of . 

St. Pagan's Castle, Cardiff 

The Castle, Cardiff 
The Palace, Llandaff 

Digitized by VjOOQ IC 

List Olr ItiEHBEttS. 


Aberdare,The Right Hon. Lord . 
Lewis, Sir W. T., Bart. . 
Llewelyn, Sir John Talbot 

Dillwyn, Bart., M.A. . 
Llandaff, Very Rev. the Dean of 
Benthall, Ernest, Esq. 
Bloflse, E. F. Lynch, Esq. 
Cardiff Free Library . 
Cathedral Library 
Clark, Godfrey L., Esq. 
Corbett, E. W. M., Esq. . 
Corbett, J. Stuart, Esq. . 
David, W. W., Esq., M.D. 
Da vies. Rev. David, M.A. 
Davies, Dr. .... 

Davies, C. Morgan, Esq. . 
Davies, Rt-v. H. C, M.A. 
Edward8,W.,E8q.,M.A.,H.M.I 8. 
Edmondes, Yen. Arch., M.A. . 

Edmondes, Mrs 

Evans, Rev. A. F., M.A. 
Evans, Pepyat W., Esq. . 
Evans, Rev. W. F., M.A. 
Evans, W. H., Esq. . 
Evanson, Rev. Morgan, B.Sc. . 
Franklen, Thos. Mansel, Esq. . 
George, Isaac, Esq. . 
Gibbins, Wm., Esq. . 
Glascodine, C. H., Esq. . 
Gordon, Mrs. R. W. . 
Gray, Thomas, Esq. . 
Griffith, Rev. John . 
Griffiths, W., Esq. . 
Hybart, F. W., Esq. . 
James, C. H., E^sq. . 
James, Frank T., Escj. 

Jenkins, Mrs 

Jones, D. W., Esq. . 

Jones, Edmund, Esq. 

Jones, Evan, Esq. 

Jones, Miss Ada 

Jones, Oliver Henry, Esq., M.A. 

Jones, Edgar, Esq., M.A. . 

Jones, Rhys, Esq. 

Jones, J. Amallt, Esq., M.D. . 

Kirkhouse, Mrs. Herbert 

Dyffryn, Aberdare 
Mardy, Aberdare 

Penllergare, Swansea 

Deanery, Llandaff 

Glantwrcb, Ystalyfera, S.O. 

Coytrehen, Aberkenfig, S.O. 



Talygarn, Llantrisant, Glam., S.O. 

PwU-y-pant, Cardiff 

Bute Estate Office, Cardiff 

The Glog, Pontypridd 

Canton Rectory, Cardiff 

Bryn Golwg, Aberdare 

112, High Street, Merthyr Tydfil 

St. Hilary Rectory, Cow bridge 

Courtland House, Merthyr Tydfil 

Nolton Court, Bridgend 

Old Hall, Cowbridge 

Vicarage, Neath 

33, Newport Road, Cardiff 

The School, Cowbridge [Cardiff 

Llanmaes House, Llantwit Major. 

Merthyr Mawr Rectory, Bridgend 

St. Hilary, Cowbridge 

The Grove, Mountain Ash 

Gktrthmor, Neath 

Abingdon Gardens, Kensington, W. 

Nottage Court, Bridgend, Glam. 

Underbill, Port Talbot, Glam. 

Llangynwyd, Glamorgan * 

Pencaemawr, Merthyr Tydfil 

19, Castle Street, Cardiff 

Conway Road, Canton, Cardiff 

64, Park Place, Cardiff 

Penydarreu House, Merthyr Tydfil 

Gellystone, Llandaff 

Galon-uchaf, Merthyr Tydfil 

The Forest, Glyn Neath, Glam. 

Ty-mawr, Aberdare 

Maindy, Ynyshir, Pontypridd 

Fonmon Castle, Cardiff 

County School, Barry 

Godrecoed, Neath 

Heathmont, Aberavon, Port Talbot 

Hazelwood, Cathedral Road, Cardiff 

Digitized by VjOOQ IC 



Knight, R. L., Esq. . 
Lawrence, Arthur, Esq. 
Leigh, Dr. 
Lewis, Rev. Canon 

Lewis, Arthur, Esq. . 
Lewis, Rev. Daniel 
Lewis, Lieut.-Colonel D. R. 
♦Lewis, Mrs. 
Linton, H. P., Esq. . 
Llewellyn, R. W., Esq. 
Metford, Miss . 
Moore, G. W., Esq. 
Moore-Gwyn, J., Esq. 
Morgan, Colonel W. L., R. 
Morgan, J. Llewellyn, Esq 
Morgan, Taliesin, Esq, 


NichoU, Iltyd, Esq., F.S.A. 
Nicholl, J. L D., Esq. 
Owen, J. Trevor, Esq., M.. 
Phillips, Rev. T. 0. . 
Powel, Professor Thomas, M.A. 
*Pughe- Jones, J. E., Esq. 
Rees, Hoirell, Esq., M.D. 
Reynolds, Llywarch, Esq., M.A. 
Richards, J. E., Esq. 
RUey, W., Esq. 

Royal Institution of S. Wales . 
Ryland, C. J., Esq. . 

Salmon, Principal David . 
Seaborne, George, Esq. 
Stockwood, 8. H., Esq. 
Swansea Free Library 
Talbot, Miss . . . . 
Thomas, A. C, Esq. . 
Thomas, D. Lleufer, Esq., M.A. 
Thomas, J. Lynn, Esf^., C.B. . 
Thomas, Rev. J. L., M.A. 
Thomas, Trevor F., Esq. . 
♦Thomas, Lewis D., Esq. . 
Thompson, Herbert M., Esq. . 
Traherne, L. E., Esq. 
Tyler, Mrs. Trevor . 
University College Library 
Vachell, C. T., Esq., M.D. 

Tythegston Court, Bridgend, Glam. 
6, Park Place, Cardiff 
Glynbargoed, Treharris, Glam. 
Ystrad - y - f odwg Vicarage, Pentre, 

Tynewydd, Llandaff 
Rectory, Merthyr Tydfil 
Phw Penydarren, Merthyr Tydfil 
Greenmeadow, Tongwynlais, Cardift 
Llandaff Place, Llandaff 
Baglan, Briton Ferry 
Glasfryn, Dinas-Fowys, Cardiff 
Pen lUtyd, Palace Road, LUndaff 
Dyffryn, Neath 
Brynbriallu, Swansea 
Bryn Teilo, LUndaff 
12, Queen's Chambers, Queen Street, 

The Ham, Cowbridge 
Merthyr Mawr, Bridgend, Glam. 
Grammar School, Swansea 
Skewen Vicarage, Neath 
University College, Cardiff 
43, Trafalgar Terrace, Swansea 
Glyndwr, 190, Newport Road, Cardiff 
48, Glebeland Street, Merthyr Tydfil 
Woodlands, Neath 
Newcastle House, Bridgend 
Cardwell Chambers, Marsh Street, 

Bristol; and Southemdown 
Training College, Swansea 
Brynheulog, Hengoed, Cardiff 
Bridgend, Glam. 

Margam Park, Taibaoh 
103, Cathedral Road, Cardiff 
Hendre, Swansea 
21, Windsor Place, Cardiff 
Pont-neath-Vaughan, Neath 
Ely Rise, Cardiff 
27, Rope Walk, Neath 
Whitley Batch, Llandaff 
Coedriglan Park, Cardifl 
Llantrythid, Cowbridge 
11, Park Place, Cardiff 

Digitized by VjOOQ IC 

LtSt Ot M1SMBSR6. 


Ward, John, Esq., F.S.A. . 
WheatJey, J. L., Eaq. 
♦Wniiams, W., Esq., M.D. 
WiUiamB, Mrs. 
Wilking, Charles, Esq., F.G.S. 

Public Maseum, Cardiff 
174, Newport Road, Cardiff 
Ponarth, Glam. 
Cartrefle, Hirwain, Aberdare 
Springfield, Merthyr Tydfil 


Cawdor, The Right Hon. Earl, 
Lord Lientenant of Pem- 
brokeshire .... 

Philipps, Sir C. E. G., Bart., 
Lord Lieutenant of the Town 
and County of Haverfordwest . 

Scourfield,SirOwenH. P.,Bart. 

Alien, Miss Mary 

AUen, Herbert, Esq. 

Bancroft, J., Esq., H.M.I.S. 

Bowen, Rev. Preb. . 

Bushell, Rev. W. Done, M.A. . 

Cathedral Library 
Chidlow, Rev. C, M.A. . 
Green, Francis, Esq. . 
Hilbers, Yen. Archdeacon, M. A . 
Jones, E. D., Esq. . 

Laws, Edward, Esq., F.S.A. 
Leach, A. L., Esq. . 

Lewis, Rev. T 

♦Lloyd, Richard, LI., Esq. 

Phillips, Rev. James 
PhiUips, J. W., Esq. 
Phillips, Rev. J. . . . 
Samson, Louis, Esq., F.S.A. 
Thomas, A. H., Esq., A.R.LB.A. 

Thomas, Rev. O. Jones 
Thomas, Mrs. James 
Wade-Evans, Rev. A. W. . 

Stackpole Court, Pembroke 

Picton Castle, Haverfordwest 

Williamston, Neyland 

c/o C. F. Ecerton Allen, Esq., Hill 
Cottage, Tenby 

10, The Norton, Tenby, and Winton 
House, Leamington 

Somerset House, Tenby 

Monkton Priory, Pembroke 

The Hermitage, Harrow ; and Caldy, 

St. David's, Pembroke 

Llawhaden Vicarage, Nar berth 

Glanymor, St. David's 

St. Thomas Rectory, Haverfordwest 

6, Addison Road, Kensington, W., 
and Fishguard 

Brython Place, Tenby 

Giltar, Shrewsbury Lane, Plumstead, 
S.E. ; (Tenby and Co., News Office, 

Lampeter Velfrey Rectory, Narberth 

Pen ty park, Clarbeston Road, S.O. 

44, Oxford Terrace, Hyde Park, W. ; 
and Poyston, Pembroke 



Uzmaston Vicarage, Haverfordwest 

Scotch weU, Haverfordwest 

County Surveyor's Office, Haverford- 

Llandyssilio Vicarage, Clyndtrwen 

Rook House, Haverfordwest 

41, Goldsmith Avenue, Acton, W., 
and Fishguard, Pembroke 

Digitized by VjOOQ IC 




Day, Rev. E. Hermitage, D.D. . Bryn Moel, Abbey Cwm Hir, Peny- 
Jones, John, Esq. . . . Ash Villa, Rhayader (bont, R.S.O. 
Sladen, Mrs Rhydoldog, Rhayader 

Thomas, Rev. J. J., B.A. 
Thomas, R. Wellings, Esq. 
Venables-Lle welyn, Charles, Esq. 
Williams, Mrs. 
Williams, Rev. Preb. T., M.A. 

The Manse, Rhayader 

County Surveyor's Office, Uandrindod 

Llyadinam, Newbridge-on-Wye 

Penralley, Rhayader 

Llowes Vicarage, Hereford 


Tredegar, Right Hon. Viscount, 
Lord Lieut, of Monmouthshire 
Llangattock, The Kt. Hon. Lord 
Jackson, Sir U. M., Bart. . 
Anthony, Miss .... 
Bowen, A. E., Esq. 
Bradney, Joseph A., Esq. 
Hanbury, J. Capol, Esq. 
Martin, E. P., Esq. . . The Hill, Abergavenny 

Rickards, R., Esq. . . The Priory, Usk 

Secretary, The .... Public Library, Newport, Mon. 
Williains, Albert A., Esq. Penyparc, Llangibby, Newport^ Mon. 

Tredegar Park, Newport 
The Hendre, Monmouth 
Llantilio Court, Abergavenny 
The Grove, Caerphilly, Mon. 
The Town Hall, Pontypool 
Tal-y-Coed, Abergavenny 
Pontypool Park, Mon. 


Harlech, The Right Hon. Lord . 
Banks, W. H., Esq., B.A. . 
Corrie, A. Wynne, Esq. 
Davies, James, Esq. . 
Drinkwater, Rev. C. H., M.A. . 
Gleadowe, T. S., Esq., H.M.LS. 
Lloyd, Edward, Esq. 

Longley, Mrs 

Newell, Rev. E. J., M.A. 
Nicholson, A. C, Esq. 
Owen, John, Esq. 
Parry- Jones, J., Esq. 
Pilley, Walter, Esq. . 
Sitwell, F. Hurst, Esq. 
Summers, H. H. C, Esq. . 
Swainsun, Rev. J. G. 

Taylor, Henry, Esq., F.S.A. 
Woodall, Edward, Esq. 

Brogyntyn, Oswestry 

Hergest Croft, Kington, Herefordshire 

Tedsmore Hall, West Felton, R.S.O. 

Park Hall, Oswestry 

Gwynva, Broomy Hill, Hereford 

St. George's Vicarage, Shrewsbury 

11, Stanley Place, Chester 
Meillionen, Hoole, Chester 
Dinhara House, Ludlow [timer, Salop 
Neen Solars Vicarage, Cleobury Mor- 

12, Salop Road, Oswestry 
Tawelan, Newton Lane, Chester 
Plas, Glyn, Rhuabon 

The Barton, Hereford 

Ferney Hall^ Craven ArmF, Shropshire 

Picton Villa, Oswestry 

Wistaiistow Rectory, Craven Arms, 

S.O., Salop 
12, Curion Park, Chester 
Wingthorpe, Oswestry 

Digitized by VjOOQ IC 



The Society of Antiqaaries, Burlington Hoase, London (c/o W. H. 

St. John Hope, Esq.) 
The Society of Antiquaries of Scotland, Queen Street Museum, Edin- 
burgh (c/o Joseph Anderson, Esq., LL.D.) 
The Royal Society of Antiquaries, Ii-eland (c/o R. H. Cochrane, Esq., 

F.S.A., 6, St. Stephen's Green, Dublin) 
The British Archseological Association, Brooklyn Lodge, Mill Hill, 

Barnes, S.W. (c/o R. H. Forster, Esq.) 
The ArchsBological Institute of Great Britain and Ireland, 20, Hanover 

Square, W. (c/o The Secretary) 
The Royal Society of Northern Antiquaries, Copenhagen 
The Royal Institution of Cornwall, Truro (c/o Major T. Parkyn) 
The Cambridge Antiquarian Society, Cambridge 
The Bristol and Gloucestershire Archasological Society (The Society's 

Library, Eastgate, Gloucester) 
The Chester Archaeological and Historical Society (c/o I. E. Ewen, Epq., 

Grosvenor Museum, Chester) 
The Shropshire Archaeological and Natural History Society (c/o F. 

Goyne, Esq., Shrewsbury) 
The Cumberland and Westmorland Antiquarian Society, Kendal 
Society of Antiquaries of Newcastle-on-Tyne (R. Blair, Esq., F.S.A.) 
La Soci^t^ d'Arch^ologie de Bruxelles, Rue Ravenstein 11, Bruxelles 
The Smithsonian Institution, Washington, D. C, U.S.A. 
The Library, Bureau of Ethnology, Washington, D.C., U.S.A. 
Kongl. Vitterhets Historic och Antiquitets Akademien, Stockholm 

(c/o Dr. Anton Blomberg, Librarian). 

All Members residing in South Wales and Monmouthshire are 
requested to forward their subscriptions to the Rev. Charles Chidlow, 
M.A., Llawhaden Vicarage, Narberth. All other Members to the Rev. 
Canon R. Trevor Owen, F.S.A., Bodelwyddan Vicarage, Rhuddlan, 
Flintshire, S.O. 

As it is not impossible that omissions or errors may exist in the above 
list, corrections will be thankfully received by the General Secretaries. 

The Annual Subscription is One Cruinea^ payable in advance on the first 
day of the year. 

Members wishing to retire must give six months* notice previous to the 
first day of the following year, at the same time paying all arrears. 

All communications with regard to the Archceologia Ca/mhrensis should 
be addressed to the Editor, J. Romilly Allen, F.S. A., 28, Great Ormond 
Street, London, W.C. 

Digitized by VjOOQ IC 

18 lAWS. 



Cambrian ^Ircbaeologtcal ^tssociation. 


In order to Examiney Preserve, and Illustrate the Ancient Monuments and 

Remains of the History, Language, Manners, Customs, 

and Arts of Wales and the Marches. 


1. The AsBOoiation shall consist of Subsoribing, Corresponding, and Hono- 

rary Members, of whom the Honorary Members must not be British 


2. New members may be enrolled by the Chairman of the Committee, or by 

either of the General Secretaries ; but their election is not complete 
nntil it shall have been confirmed by a General Meeting of the A^ocia- 


3. The Goyemment of the Association is vested in a Committee consisting 

of a President, Vice-Presidents, a Treasurer, a Chairman of Committee, 
the General and Local Secretaries, and not less than twelve, nor more 
than fifteen, ordinary subscribing members, three of whom shall retire 
annually according to seniority. 


4. The Vice-Presidents shall be chosen for life, or as long as they remain 

members of the Association. The President and all oti^er officers shall 
be chosen for one year, but shall be re-eligible. The officers and new 
members of Committee shall be elected at the Annual General Meet- 
ing. The Committee shall recommend candidates ; but it shall be 
open to any subscribing member to propose other candidates, and to 
demand a poll. All officers and members of the Committee shall be 
chosen from the subscribing members. 


5. At all meetings of the Committee the chair shall be taken by the Presi- 

dent, or, in his absence, by the Chairman of the Committee. 


6. The Chairman of the Committee shall superintend the business of the 

Association daring the intervals between the Annual Meetings ; and 
he shall have power, with the concurrence of one of the G^nerid Secre- 
taries, to authorise proceedings not specially provided for by the laws. 
A report of his proceedings sl^ll be laid before the Committee for their 
approval at the Annual General Meeting. 

Digitized by VjOOQ IC 

LAWS. 19 


7. There shall be an Editorial Snb-Oommittee, oonsiBting of at least three 

members, who sliall saperintend the publications of the Association, and 
■hall report their proceedings annually to the Committee. 



8. All Subsoribing Members shall pay one guinea in advance, on the 1st of 

January in eaoh year, to the Treasurer or his banker (or to either of 
the General Secretaries). 


9. Members wishing to withdraw from the Aissociation must g^ive six 

months' notice to one of the General Secretaries, and must pay all 
arrears of subscriptions. 


10. All Subscribing and Honorary Members shall be entitled to receive all 

the publications of the Association issued after their election (except 
any special publication issued under its auspices), together with a 
ticket giving free admission to the Annual Meeting. 


11. The Secretaries shall forward, once a month, all subscriptions received 

by them to the Treasurer. 


12. The accounts of the Treasurer shall be made up annually, to December 

Slst; and as soon afterwards as may be convenient, they shall be 
audited by two subscribing members of the Association, to be appointed 
at the Annual General Meeting. A balance-sheet of the said accounts, 
certified by the Auditors, shall be printed and issued to the members. 


1 3. The funds of the Association shall be deposited in a bank in the name 

of the Treasurer of the Association for the time being ; and all bills 
due from the Association shall be countersigned by one of the General 
Secretaries, or by the Chairman of the Committee, before they are paid 
by the Treasurer. 


14. The Committee shall meet at least once a year for the purpose of nomi- 

nating officers, framing rules for the government of the Association, 
and transacting any other business that may be brought before it. 


15. A General Meeting shall be held annually for the transaction of the 

business of the Ajssociation, of which due notice shall be given to the 
members by one of the General Secretaries. 


16. The Chairman of the Committee, with the concurrence of one of the 

General Secretaries, shall have power to call a Special Meeting, of 
which at least three weeks* notice shall be given to each member by 
one of the G^eral Secretaries. 


17. At all meetings of the Committee five shall form a quorum. 

Digitized by VjOOQ IC 

20 LAWS. 


18. At the Annual Meeting the President, or, in his absence, one of the 

Vioe-Presidents, or the Chairman of the Ck>mmittee, shall take the 
chair ; or, in their absence, the Committee may appoint a chairman. 


19. At all meetings of the AjBsociation or its Committee, the Chairman shall 

have an independent as well as a casting vote. 

20. The Treasurer and other officers shall report their proceedings to the 

General Committee for approval, and the General Committee shall 
report to the Annual General Meeting of Subscribing Members. 


21. At the Annual Meeting, tickets admitting to excursions, exhibitions, 

and evening meetings, shall be issued to Subscribing and Honorarj 
Members g^tuitously, and to corresponding Members at such rates as 
may be fixed by the officers. 


22. The superintendence of the arrangements for the Ajinual Meeting shall 

be under the direction of one of the General Secretaries in conjunction 
with one of the Local Secretaries of the Association for the district, 
and a Local Committee to be approved of by such General Secretary. 


2.S. All funds subscribed towards the local expenses of an Annual Meeting 
shall be paid to the Joint account of the General Secretary acting for 
that Meeting and a Local Secretary ; and the Association shall not be 
liable for any expense incurred without the sanction of such General 


24. The accounts of each Annual Meeting shall be audited by the Chairman 

of the Local Committee, and the balance of receipts and expenses on 
each occasion be received, or paid, by the Treasurer of the Ajssociation, 
such audited accounts being sent to him as soon after the meeting as 


25. Any Subscribing Member may propose alterations in the Rules of the 

Association ; but such alteration must be notified to one of the G^eral 
Secretaries at least one month before the Annual Meeting, and he shall 
lay it before the Committee ; and if approved by the Committee, it 
shall be submitted for confirmation at the next Meeting. 

(Signed) C. C. Babington, 
Augfust 17tb, 1876. Chairman of the Committee, 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

Digitized by VjOOQ IC 

"32101 063966^1 

Digitized by VjOOQ IC