Vol 1, No 1: Geology, Archaeology and Pre-Norman Roath

From Project Newsletter Vol 1. No 1. Oct 1983:-



The terrain of the East Moors, Pengam Moors and the flood areas of the River Rhymney and the Roath Brook consist of alluvium and estuarine marls. Further inland in the area of Splott, Roath and Cathays the subsoil is mainly river gravels. North of the old village of Roath the land rises gently to a height of just over 200 feet on the summit of Pen-y-lan Hill where the reservoir is sited.

Pen-y-lan has a two-fold geological significance. Firstly, it is here that are found the oldest rocks in the neighbourhood – an outcrop of rocks of the Silurian age.  Professor F.J.North tells how hundreds of millions years ago the great masses of strata of coal measures, carboniferous limestone and Old Red Sandstone were gradually eroded away leaving exposed the ancient Silurian rocks. The thickness of missing strata he estimates to have been at least 8000 feet. Secondly, Pen-y-lan is significant as marking the southwest boundary of the massive ice sheet of the last Great Ice Age (c. 200,000 B.C.) the line of which extended N.N.E. from Pen-y-lan more or less following the line of the present Welsh border, continuing northward and then eastward to the East coast of England.  Tree trunks rooted in black clay resting on Triassic Marl at a depth of 17 feet were found in 1901 during dock excavations at Roath.


One of the most significant of all archaeological finds in Wales is the Old Stone Age axe found about 50 years ago on Pen-y-lan Hill; it is on exhibit in the National Museum of Wales with the following caption:

” Hand axe of Acheulian type made of quartzite from Pen-y-lan. Date: Middle Acheulian (c. 200,000 B.C.) The implement is the first of its kind to have been recorded in South Wales, though some flint hand axes, apparently of Acheulian type have been found in Pontnewydd cave near St. Asaph in North Wales. It was probably made locally from a subangular fragment derived from one of the beds of gannister which occur in the coal measures. It was picked up by the donor, Miss M. Remington, from among stones cleared from an allotment on Pen-y-lan Hill; many unworked cobbles of the same material were found with it and probably reached the spot by solifluxion during the last Ice Age, long after the great interglacial period during which Middle Acheulian man flourished.”

Penylan handaxe

Quartzite handaxe found in an allotment at Penylan.

Also held at the National Museum is an end scraper on ridge backed flake of white patinated grey flint found at Roath Park. The bold working is described as of Pre-Neolithic character (Catalogue No. 193 (91.445)). A polished flint axe with thin  butt and slightly flattened oval section was found in a peat bed “some feet below” the surface on the site of the Dowlais Steel Works East Moore in 1896.

In 1902 during work on preparing the site of the Roath Power Station some prehistoric bones were excavated and are described in the Public Library Journal Vol. III pt. 6, March 1902:


Not the Penylan Urus skeleton but one found near Copenhagen (photograph: Wikipedia)

“Excavations on the site of the Roath Electric Power Station have revealed interesting relics in the shape of ancient human bones and those of hoofed animals. The deposits in which the discoveries were made are alluvial, consisting of 20 or 30 feet of clay used for brick-making and underlain by gravel and shingle. Foundations are being laid upon these last beds and in every case the bones were found at this level resting upon the sand and pebble. Our readers will remember the press notice of the chief find, the upper part of the skull of the URUS, with horn-cores complete. This ancient animal is believed to be the direct ancestor of the British wild oxen, the nearest approach to which are the well known “Chillingham Cattle” preserved in the park of that name in Northumberland. A rather larger pair of horn-cores of the same animal was also found but in fragments. When mended this will prove a fine museum specimen.”

The bones were identified by Mr. John Storrie, curator of the Cardiff Museum.

Roman pottery was found in the proximity of Pen-y-lan Hill and the Cardiff inner by-pass. It was discovered in a disused quarry near the recently demolished Edward Nicholl home. O.S. 171 (154) 1978.788.

Sources, etc.:

National Museum of Wales – various publications.

NORTH; F.J. Transactions of the Cardiff Naturalists’ Society, Vol. XLVIII (1915) pp 36 – 49 article on water borings in Roath.

NORTH, P.J., “The Evolution of the Bristol Channel.”; National Museum of Wales. 3rd ed. 1964. Cardiff.

ANDERSON,J.G.G., Article on Geology in “The Cardiff Region; a Survey” (prepared for the meeting of the British Association in Cardiff), Cardiff. pp 22 – 24.

Public Library Journal, Cardiff, as described and acknowledged in the text.


From Project Newsletter Vol 1. No 1. Oct 1983:-


Many of the buildings that stand in Roath today are the product of suburban development undertaken during the last quarter of the nineteenth century and the first two decades of the twentieth. The area was viewed as being ripe for development by speculative landowners primarily to accommodate Cardiff‘s increasing population following new dock and railway construction. The landowners largely responsible for the suburban infilling were John, 5rd Marquess of Bute (1847 – 1900), Sir Charles Morgan (created 1st Baron Tredegar in 1859) and Godfrey, 2nd Baron Tredegar (1851 – 1913),  Alfred Donald Mackintosh (sometimes referred to as the Mackintosh of Mackintosh), and the Crofts-Williams family of Roath Court.

If a survey were carried out amongst current Roath inhabitants (particularly the younger element) it would not be very surprising if many would find it difficult to imagine the landscape of Roath as being anything but the serried rows of terraces and shop fronts that are familiar today. Such a conception would be fallacious however since for the greater-part of its evolution, the Roath landscape has been predominantly agricultural, and characterised by dispersed farmsteads.This is clearly evident from the Roath tithe map of 1840.

The area known as Roath is also much older than is generally appreciated. In its time Roath has served as a manor parish village and suburb, and it is imperative that the local historian define clearly which of these units he is describing. It should also be borne in mind that, as well as some of these units existing at the same time, other complications arise in Roath’s manorial and parochial contexts which have to be clarified.

Manorially, Roath was originally one manor, established by the first Norman lord of Glamorgan and Cardiff in the 1090s, and known as the Manor of Roath. During the twelfth and thirteenth centuries the manor became fragmented so that three manors were forged out of the original unit. This process will be described in more detail in a future Project letter.

Parochially, we need to-distinguish between ecclesiastical parishes and civil parishes (although the latter never applied to Roath), The ecclesiastical parish of Roath is reputed to have been created in the sixteenth century, although its antecedent, namely the chapelry of Roath, originated in the early twelfth century. The ecclesiastical parish was the unit which served the spiritual needs of the community although it also came to perform administrative (i.e. secular) functions such as the organisation of Poor Relief. As a consequence of population growth and urbanisation in the second half of the nineteenth century, the boundaries of ecclesiastical parishes lost their significance. New ecclesiastical districts emerged to serve new parish churches that were built within the confines of an original parish, to cater for the increasing population. This was the case in Roath. (This is why modern parochial boundaries bear no resemblance to the original ecclesiastical ones). Although many civil parishes did follow the boundaries of their ecclesiastical namesakes they had a fundamentally different purpose. They were the product of late nineteenth century legislation, being the creation of the Local Government Act of 1894.  They were exclusively civil and administrative units, and formed a third tier of local government until they in turn were abolished by the Local Government Act of 1972.  (The 1894 Act in terms of civil parishes had no effect on Roath which had since 1875 come under the administration of the extended borough of Cardiff).

In its manorial context, Roath has a wealth of source material available. This will become clearer when we study such records in future Project Newsletters.

It is appropriate that this first Project Newsletter should contain a reference to the origins and characteristics of pre-manorial Roath. In order to do this however it is necessary to place the latter in a larger context for the reason that describing pre-manorial Roath without reference to wider issues (regional and “national”) would be virtually meaningless, with the ensuing description extremely superficial. As it is, what follows is based entirely on secondary source material and interpretation, because written source material is virtually non-existant, whilst non-written sources for Roath are largely based on legend.

Professor William Rees in his history of Cardiff suggests that the present Roath Court Funeral Home (at the junction of Albany Road with Newport Road) may have been one site of royal centre of the many independent Welsh rulers who filled the vacuum following the departure of the Roman garrison from Wales in c400, but when it served as a royal centre is uncertain. He bases this view mainly on the observations of the sixteenth century antiquarian Rice Merrick (Rhys Meurig) who in his “Booke of Glamorganshire Antiquities” published in 1578 stated that within the lordship of manor of ‘Rothe Dogfilde stood an old Pyle, compassed with a Mote, which is called ‘The Court’ but now in ruyne”.  Rees believed this mound to be suggestive of Pre-Norman origins.  Moreover the ‘Pyle’ and ‘Mote’ are also seen as authenticating the claim that the place name Roath (in Welsh, Y Rhath) is of Gaelic extraction, being derived from ‘Rath’ the old Irish word for an ancient earthwork, fortress or enclosure.

It should be pointed out however that in an article entitled ‘The Problem of the Raths of Pembrokeshire‘, Professor W.F.Grimes is cautious in ascribing archaeological derivations to place names which have ‘rath’ connotations. He writes:

‘Rath is a term commonly used in Ireland to define a particular type of earthwork a circular enclosure defended by a bank-and-ditch system which is normally very strong in relation to  the very small area enclosed. Apart from its presence in certain place names (where in any case its origins is probably different) the word occurs in Wales only in one quite limited area:  the central part of Pembrokeshire, here taken to be a piece of country with a radius of about twelve miles around Haverfordwest.

(See the article in Donald Moore (ed) The Land of Dyfed in Early Times, Cardiff 1964;

In the absence of any contrary information as to the origin of the place name we may take it that the derivation from an earthwork or fortress is the most we can go on for the present.

Finally, John Hobson Matthews hypothesised that there was reason to believe that the name of Roath predates the Norman foundation of Cardiff, and that the area of land between the Taff and Rhymney rivers, incorporating the site of Cardiff castle, equated with ’the District of the Rath’.

What follows broadly describes the main territorial and administrative features of pre-Norman ‘Wales’ with special reference to the ‘Cardiff’ region.


After the withdrawal of the Roman forces from Britain (in the fifth century) native dynasties arose in South East Wales and elsewhere. Tradition has it that somewhere between the fifth and seventh centuries peety kings of the line of the legendary king GLYWYS ruled an area which extended from the river Tawe in the west to the river Usk in the east, known as GLYWYSING. Glywys was reputed to be the great grandson of Eugenius, one of the sons of Magnus Maximus (Macsen Wledig), the late Roman ruler. In the course of the seventh century a new dynasty under MEURIG ap TEWDRIG emerged. It is from his grandson MORGAN ab ATHRWYS who ruled in the eighth century that the name MORGANNWG is said to be derived. The kingdom of Morgannwg was larger than Glywysing and included the ancient Kingdom of Gwent. It covered an area from the Tawe in the west to the Wye in the east.


The Gwlad

In pre-Norman times, Wales was divided into GWLEDYDD or states, so that the kingdom of Morgannwg was a Gwlad. The latter represented a region subject to single rule and a single dynasty.  In each gwlad the king or prince exercised certain privileges and prerogatives, maintaining at his court or LLYS a royal household of high-ranking officers and retainers such as the Disdain or steward, the Judge, the Court priest and the Bard.

Cantrefs and Commotes

The most important administrative unit of the gwlad was the CANTREF, which was itself derived from a unit comprising one hundred TREFYDD (or townships). In turn cantrefs were divided into two or more CYMYDAU or Commotes in order to facilitate the collection of dues and the holding of local courts. These ancient administrative units reveal a highly developed system of local government. In each cantref of commote the king or prince (the tywys og) had a lesser court-house which served as a centre for local government.

The sites of these court-houses were known as MAERDREFI or royal villages, and they were the largest centres in each commote. It was here that the king or prince’s local officer, the Rhaglaw (deputy or viceroy) held the court of the commote, the Maer collected the dues of the Bondment while the Rhingyll (sergeant or beadle) collected those of the Freemen.


The old Welsh System was based on kinship. Society was composed of tribes with all members united by real or supposed descent from a common ancestor. Within the tribe there were smaller groupings or clans based upon a closer blood relationship. Members of the clan were responsible for the payment of a variety of dues to the Chieftain, and the Gwestfa (a food rent) to the king or prince when he and his officials were in the area. (Interestingly, mention is made of a guest rent in a Minister’s Account of 1301 for the Manor of Rumney. It would appear that this was also claimed by Norman lords.)  Members of the clan were Freemen who occupied scattered farmsteads. Gavelkind was the method of inheritance, where property was divided equally among a man’s sons.

In addition to the freemen were the, Bondmen who lived in BOND VILLAGES – the only villages to be found in Wales before the coming of the Normans. These bondmen too were subject to dues and services.


The kingdom of Morgannwg consisted of seven constituent cantrefi, whose boundaries ran along the lines of the river valleys that run from the uplands to the sea. They included GWRINYDD (GORFYNYDD) which lay between the Tawe and the Thaw, PENYCHEN between the Thaw and the Taff, GWYNLLWG between the Rhymney and the Usk, GWENT ISCOED and GWENT UCHOED.

Historical evidence of the medieval period also points clearly to the fact that SENGHENYDD or ‘y Cantref Breiniol’ had been an administrative entity in the Welsh period.



(Drawing: Caerphilly Local History Society)

The pre-Norman cantref of particular interest to us is  Senghenydd, lying between the rivers Taff and Rhymney, and extending from the boundary with Brycheiniog in the north (near Merthyr) to the sea at Cardiff. The cantref was divided into three commotes:

(1) UWCH-CAIACH (above Caiach) to the north of Caiachbrook near Llancaiach.

(2) IS-CAIACH (below Caiach) south of the Caiach brook as far as the ridge of the Caerphilly mountain.

(3) CIBWR (Kibbor) lying to the south of Caerphilly mountain, as far as the Severn shore or Bristol channel.

The Mediaeval Lordship of Cardiff - Copy

The Mediaeval Lordship of Cardiff (Drawing: W F Grimes)

It is the commote of Kibbor, whose Llys or Court-house  was reputedly at Llysfaen (Lisvane – meaning ‘Stone Court‘) which is of relevance here, as it was this area which was adopted in its entirety by the Normans as their Lordship of Cardiff – an area which also incorporated the MANOR of Roath and which later encompassed the ecclesiastical parishes of Roath, Llanedeyrn, Lisvane, St John’s and St Mary’s in the ‘towne’ of Cardiff (east of the Taff) and part of Whitchurch.

The manor of Roath will form the basis of a future Project Newsletter.


The passages devoted to the origins of Roath are based on MATTHEWS, J. HOBSON,(ed.), Records of the County Borough of Cardiff (6 Vols. Cardiff 1898-1911, especially Vols I, II & V).

REES, William, Cardiff. A History of the City (Cardiff 1962, revised ed. 1964.

MERRICK, Rice, A Booke of Glamorganshires Antiquities (1578) (ed. J.A. Corbett 1882; reprinted Barry 1972) (The first complete edition of Merrick’s study is scheduled to be the first publication issued under the auspices of the recently formed South Wales Record Society. Entitled Morganiae Archaigoraphia: a Book of the Antiquities of Glamorgan, it will be edited by Brian L1. James.

The background to pre-Norman society was based on the chapter by J. Beverly Smith entitled “The Kingdom of Morgannwg and the Norman Conquest of Glamorgan”, in Vol. III of the Glamorgan County History Series (ed. T.B. Pugh, Cardiff 1971) and the same author‘s article in Morgannwg entitled “The Lordship of Glamorgan”.


There is nothing for Roath comparable to Edgar Chappell‘s history of Whitchurch. It is the intention of the Roath Local History Society to publish a history of the ecclesiastical parish some time in the future, using original source material.

Much of the material for Roath is subsumed within more general histories, e.g. of Cardiff or Glamorgan. Of these by far the most important is the Matthews compilation of Cardiff records detailed above. This can be consulted at the Cardiff Central Library; University College, Cardiff, Arts and Social Studies Library (Salisbury Collection) and the Glamorgan Record Office.

Apart from those already listed the following works will also be found useful:-

ANDREWS, D.A., Cardiff – a Sketch Book. London, 1920.

BALLINGER, J. Guide to Cardiff CIty and Port.  1908.

BALLINCER, J., (ed.) , Cardiff. An illustrated handbook. 1996.

CAERDYDDIAN (pseud.);”Reminiscences of Cardiff“, South Wales Echo, June 1892′

CARDIFF CITY COUNCIL. Cardiff 1889-1974. 1974.

”           ”          ”           The Cardiff Guide. 1976.

Various Authors. The Cardiff Region. 1960.

CLARK, J.H., Cardiff and its neighbourhood 1853

DAUNTON, M.J., Coal metropolis, Cardiff, 1870-1914, Leicester.1977

DAVIES, J., Cardiff and the Marquesses of Bute. Cardiff. 1981,

EVANS: C.J. (ed.), The Book of Cardiff, London and Oxford 1937

EVANS, C.J.O., Glamorgan: its history and topography. Cardiff 1938.

HALLETT, H.M., (ed.), Handbook for Cardiff and the Neighbourhood. 19

HILLING, J.B., Cardiff and the Valleys. London. 1973.

HODGES, T.M., “The peopling of the Hinterland and Port of Cardiff” in W.E.Minchinton (ed). Industrial South Wales 1750-1914

JENKINS, J.A., The Book of Cardiff, l912

”            ”     Cardiff and its history, 1905.

JENKINS, W.L., A history of the town and castle of Cardiff. 1854.

MORGAN, C., A brief guide to Cardiff and the neighbourhood; 1891.

PATERSON, D.R., Early Cardiff. Exeter, 1926.

THOMPSON, H.H., Cardiff. 1930.

VACHELL, C.T., The history of Cardiff. 1913.

WARNER, Marmaduke, History of Roath St Germans.

WALKER, D.G, “Cardiff“ in R.A. Griffiths (ed.) Boroughs of Medieval Wales. 1978.

WILLIAMS, G, (ed), The Glamorgan County History Vol 4. 1974.

WILLIAMS, Moelwyn, “Cardiff  – its people and its trade 1660 – 1720“ in Morgannwg , Vol. VII. 1963.

WILLIAMS, S.(ed), The Cardiff Book. Vols 1,2 &3. Barry, 1973,74,77.

”         ”     The Glamorgan Historian. Vols 1 – 15,

”         ”    Cardiff yesterday, 7 Vols. 1979 – 1983.

”          ”     South Glamorgan : a county history   Barry. 1971

NATIONAL ASSOCIATION OF HEADTEACHERS, The book of Cardiff.  Oxford. 1937.

The Parish of Roath –  publication celebrating the centenary of the rebuilding of Roath Church. 1970.

REES, W., The placenames of Cardiff. South Wales and Monmouthshire Record Society. Vol. 2. 1955..

Most, if not all the above, should be held at the Cardiff Central Library. Many more works on Cardiff and its hinterland are also listed in the Library’s extensive Local History collection.