AMGUEDDFA GENEDLAETHOL CYMRU

 

NATIONAL MUSEUM OF WALES

 

 

Mines, Mills and Furnaces

 

  

by

 

D. MORGAN REES

 

ED. Excellent history of the Industrialization of Wales. The chapter on Ironmaking is excerpted here. (J. McV April 2008)

 

PREFACE

 

This short volume has been written by the Keeper of the Department of Industry during a period which has coincided with the re‑establishment of the Department in the Museum's new, west wing. The fieldwork entailed in its preparation has already resulted in the making of exhibits relating to the sites of old metalliferous mines in Wales. It also provided an opportunity to record, in slides and photographs, many important industrial sites before they disappeared, and to acquire for the collections original objects connected with them.

 

The Department of Industry, now only in its eighth year, has cause to be grateful to many people who are interested in industrial archaeology for their practical support. The author, in preparing this introductory volume on the subject, wishes to acknowledge gratefully the great assistance and advice he has received from Dr. F. J. North, formerly Keeper of Geology at the National Museum of Wales. He is also especially indebted to Professor Melville Richards, University College of North Wales, for his advice on the translation of Welsh place‑names, to Mr. W. A. M. Jones, Cardiff, for his work on maps and figures and to Mr. G. A. Hall, Gloucester, for making available, so readily, his collection of references to metalliferous mining in the Mining ‑Journal. The kindness of the editors of Archaeologia Cambrensis, Ceredigion and of the journal of the Merioneth Historical and Record Society in allowing the use of material contained in articles on industrial archaeology contributed by the author to their journals is also gratefully acknowledged.

 

D. DILWYNJOHN

Director

July 1967

 

CONTENTS

 

List of illustrations  xi

 

Introduction  xiii

 

1. Gold Mining  I

 

II. Lead and Zinc Mining . . . . . 16

 

III. Copper Mining  43

 

IV. Ironmaking . . . . . . . 52

(i) The sixteenth century 52

(ii) The seventeenth century 56

(iii) The eighteenth and nineteenth centuries  6o

 

V. Steel and Tinplate  88

 

VI. Forging and Casting . . . . . 96

 

Bibliography  99

 

Index 107


 

Excerpt – Pages 51 - 87

 

IV. IRONMAKING

 

THE SIXTEENTH CENTURY

 

It may be true to say that the iron industry which was established in Glamorgan and Monmouthshire during the second half of the sixteenth mtury, was the first in Wales to be financed from capital, as we know today, with the profit motive in mind. Ironmasters, who had already een concerned with ironmaking for this reason in Sussex, moved to the alleys of the Taff and the Cynon in Glamorgan after certain official ‑strictions were placed on their operations, and those of others, in the Veald of south eastern England; these restrictions were really directed t preventing the consumption of timber for making charcoal, the fuel ien used in the smelting of iron. It has long been said that the timber ,as needed for shipbuilding in naval yards, in view of the sea battles gainst Spain, but it has also been argued that the cordwood used in charcoal‑making was not suitable for shipbuilding.

 

The valleys, which the Sussex ironmasters chose to exploit in Glamorgan, were well‑wooded and promised ample supplies of charcoal iel; they were also well‑endowed with rivers and streams, which produced the water for driving the waterwheels, which operated the crude ellows providing the blast for heating the fuel used in smelting. In addition, iron ore was easily worked from outcrops and shallow pits.

 

Between 1564 and 1600113 blast furnaces and forges appeared in the Taff Valley near Tongwynlais (sward near the white brook) at Pentyrch (boar head), Pontygwaith (work bridge), Pont‑y‑Rhun (Rhun's bridge), Dyffryn (vale), and Blaencanaid (head water of river Canaid, white); remains of the furnace at Pont‑y‑Rhun were to be seen on the western bank of the Taff in 1874.96  In the Cynon Valley a furnace was established at Cwmaman (Aman valley), at another site called  Dyffryn, with a forge a little way downstream and another forge at Llanwonno church of Gwynno) to the south west. A second Pontygwaith in the Rhondda Fach (little Rhondda, noisy one) has also been mentioned as a last furnace location. The furnace ‑ the Taff furnace ‑ near Tongwyn,is supplied iron to a forge at Rhyd‑y‑Gwern (alder ford). Sir Robert Sidney, afterwards the Earl of Leicester, 'an experienced industrialist ad ironmasterl’ 82 set up ironworks at Llanhari (church of St. Hari), Llantrisant (church of the three saints) and at Angelton and Coity to the north of Bridgend.

 

Similar conditions favourable to ironmaking, obtained in Monmouthshire, in the valleys of the Afon Lwyd (grey river), the Rivers Ebbw and Usk. Ironmaking, on an extensive scale, in the county, started a few years later than in Glamorgan and became concentrated in and around Pontypool. The importance of water power may be seen in the names of some of the locations which developed furnaces; for example, Cwmffrwdoer (valley of the cold stream) 1570, Trosnant (cross brook) 1576 and Pontypool itself (bridge of the pool).

 

Foremost among the works developed in Monmouthshire at this time, were the Tintern Wireworks, and the leading personality was Richard Hanbury, a native of Worcestershire. He was sufficiently astute to acquire much of the woodlands of the river valleys, which gave him control over fuel resources, and led to his becoming a partner in the Tintern works and the proprietor of other ironworks in Monmouthshire, in addition to those at Abercarn (mouth of the river Carn) and Monkswood.

 

The picture, which has emerged, is of an industry composed of a number of small units, scattered throughout the two counties. Its dependence upon charcoal, and the difficulty in maintaining adequate supplies of this fuel, made it necessary to locate the works at considerable distances apart and in well‑wooded districts. The works were committed to their chosen locations by the 'tyranny of wood and water."80

 

Despite this, the iron industry in Glamorgan and Monmouthshire flourished sufficiently to develop an export trade, the finished iron being sent from the ports of Cardiff and Newport to the midland counties of England, to Bristol and its hinterland, to the western counties of Wales, to Ireland and to the Low Countries. Records in the Welsh Port Books 99 show that consignments of 'Welsh iron' were exported from Cardiff on a number of occasions between 1586 and 1601, to Bridgwater, Bristol and Flushing and that Edmund Mathew, the proprietor of the Pentyrch Ironworks was exporting from Cardiff to London, 'pieces of ordinances', called 'sakers' and 'mynions', a trade which was not strictly legal at the time.

 

The ruins of 16th century ironmaking blast furnaces may be seen at Cwmaman, Blaencanaid and Angelton. A blast furnace is a tall structure fed at the top with raw materials and continuously producing metal in a liquid form, which, at the bottom, is tapped from time to time. It has been used, according to this principle, but in developing forms, from the 16th century until the present day.

 

The furnace at Cwmaman (ST 004992) was described by William Llewellin as a sixteenth century blast furnace built of sandstone and lined with the same material.101 He wrote of evidence of the existence of a waterwheel, the furnace having been built near the confluence of two streams, and gave details of the dimensions of the furnace in plan and section.

 

An interesting detail on the Cwmaman furnace is contained in a history published in 187597; it was, supposedly, built by three brothers who followed respectively, the trades of stonemason, blacksmith and wood turner. The blast for the furnace was provided by two men, each operating 'a blacksmith's bellows', but, 'the works did not respond to the cost' and the venture was abandoned.

 

The stonemason and blacksmith left Cwmaman, but the wood turner remained to carry on with his original trade and to make chairs of distinctive designs, which achieved local fame. A number of the chairs survived into the nineteenth century, the owner of one of them, in 1830, composing an 'englyn', an alliterative stanza, to it, which contained a line claiming that the chair was more than 300 years old. The deduction then reached was that the Cwmaman furnace was built about 1520.

 

The present-day remains are deteriorating rapidly, but illustrate the principle adopted in the early days of blast furnace building, that of building the structure into the slope of the ground, so that the mouth of the furnace and the charging platform were at a higher ground level, thus making it easier to put in the raw materials. The sandstone lining is still to be seen in a right angled fragment of the interior of the furnace about 3 feet high.

 

The blast furnace which operated at Blaencanaid (so 035042) has been described as one which was 'simple to a degree'.118 It was situated in a natural hollow at 900o feet on the western side of the Taff Valley. It provides an admirable example of using sloping ground, on the west side of the hollow, to make for easy charging of raw materials. A stream flows into the hollow, and at a point behind the furnace and to the north of it, forms a waterfall 8 feet high. It passes the side of the furnace at a distance of 20 feet and long, flat, inclined stones lying on the bed of the stream suggest that they were placed there to increase the flow of water, possibly to drive a waterwheel, which operated the bellows providing the furnace with its air-blast.

 

It was a very small furnace, sandstone-built, probably from the stone of a nearby outcrop; the stones, in the lower levels, or near the base of the furnace, are nearly 3 feet long and generally 3 inches thick. Most of the stonework has collapsed; only a height of about 3 feet 6 inches remains at the front of the furnace, which measures 9 feet across (Plate 32). The present height of the remains of the back wall of the furnace is 8 feet and it is probable that the original overall height of the structure was no more than 12 feet. The length of each side wall is calculated at about 12 feet, but excavation is necessary to determine the exact penetration into the slope of the bank. The front aperture of the furnace is 3 feet high and 1 foot 6 inches wide and 3 feet long, the side walls of the aperture tapering inwards towards the tap-hole and the hearth stone. The roof of this aperture, which sloped downwards towards the tap hole, has collapsed and the opening into the furnace at this point, is now very irregular in shape. The bottom of the furnace hearth has a diameter of 1 foot 9 inches the lining tapering outwards to a diameter of 4 feet at a height of 6 feet from the bottom; this was the widest point of the furnace.

 

Plate 32 Blaencanaid  Blast Furnace Remains

 

 

A limited excavation of the site in July 1965, by D. M. Evans of the Department of Archaeology, University College, Cardiff, revealed the existence of a charging platform (Plate 33) and a short length of side wall built outwards from the northern side of the furnace. The existence of this wall, and the discovery of a number of flagstones in the ground fronting on to the furnace, suggest that a roofed building of some kind had been built in front of the furnace. The area immediately in front of a furnace of this period, was the casting floor, which was often housed in a simple building.

 

Plate 33 Blaencanaid blast furnace charging platform.

 

 

A little way downstream from the furnace, there are the remains of two stone buildings, one on each bank of the stream, and on the southern edge of the hollow there is an entrance to a small coal level. Unfortunately, the thickness of the natural growth makes it impossible to form any conclusion on the significance of the ruins of these buildings.

 

Nothing is known of the ownership of this furnace. It may have been worked in association with the furnace established at Pont-y-Rhun, a mile farther down the Taff Valley.

 

Another sixteenth century site which merits attention is that of the Angelton or Coity blast furnace (SS 905821), near Bridgend in Glamorgan.

 

‘Near the left bank (of the river Ogmore) on the windward, westerly, side of the hill, denoting its early date, prior to utilisation of the artificial air blast, stand the serried ruins of a blast furnace. Entwined in the interstices of its sandstone blocks grow the fibrous roots of a sycamore tree, rearing its head skywards in all the beauty of its brilliant foliage, but year by year bringing by disintegration of its masonry, destruction to this interesting memento of an ancient industry.’111

 

In the article which contained these words, the height of the furnace remains at the time, was given as 12 feet with a hearth 61 feet square, the furnace stack tapering to a measurement of 3 feet by 3 feet at the top. An accompanying drawing suggests an original height of 18 feet the outside measurements being 18 feet square at the base, and 16 feet square at the top, inward ledges at heights of 9 feet and 13 feet effecting the reduction.

 

A contemporary contributor to a newspaper, is also quoted as saying that when he visited the site many years previously, the hearth of the Angelton furnace was intact - 'it was a square hearth, not round as they are now made ... in the slag heap I found pieces of charcoal but no coke or coal from which fact I concluded they used as fuel charcoal only'.

 

The present remains (Plate 34) show very clearly the shape of the inside of the furnace, but it has disintegrated considerably since 1896.

Plate 34 Angelton blast furnace remains.

 

 

An interesting feature is the furnace interior, or lining, which was built of tiled stone blocks if inches thick. For some reason this kind of structure led to the conclusion that the Angelton furnace was Roman in origin, but this cannot be sustained.

 

It has been claimed82 that the furnace was built by Sir Robert Sidney, a son of Sir William Sidney of Sussex and brother to Sir Phillip Sidney, after he had leased land in the area for, '. . . the power to build a work for melting, making and casting iron sows, to make iron by forge and furnace or other means'. His interests in ironmaking in Glamorgan were extended to other districts and a site near Coity village (ss 928827) may well have been established by him. The excavation of this site could reveal the remains of the hearth of an early blast furnace.

 

An ironworks, which was probably established late in the sixteenth century, and with others survived until its partial demolition by Oliver Cromwell's troops about the middle of the seventeenth century,101 was located at Pontygwaith (ST 079979) in the valley of the river Taff. There are no obvious surface remains on the site but a water-colour, said to be a copy of the original, reputedly of the Pontygwaith furnace is preserved in the borough library at Merthyr Tydfil.

 

A photograph (Plate 35) of this painting provides a very good example of the integration of buildings achieved in an early ironworks. Towards the top (right) is seen a tram, loaded with raw materials, about to enter the charge house, at which level the materials would have been charged into the furnace. To the right of the small, square furnace stands a building which probably housed the bellows, which provided the the blast for the furnace; the bellows were usually operated by a waterwheel. The cast house, into which the molten iron flowed, abuts on to the front wall of the furnace; it has roof and wall ventilation. The piled material at the front may have been pig iron.

 

Plate 35 Pontygwaith Iron Works.

 

(ii) THE SEVENTEENTH CENTURY

 

During the years after 1600 the export of iron was considerably curtailed, following competitive production in other regions, notably the Forest of Dean, and a number of the sixteenth century ironworks succumbed. Some of them, because of ownership, became involved in the politics of the seventeenth century and were partially destroyed by Cromwell's troops. It was also to be expected that some would cease to function following the scarcity of charcoal fuel, due to the rapid deforestation which had taken place, and to the fact that no attempts were made to replenish the natural supply by planting coppices. As the seventeenth century advanced, therefore, the ironworks of South Wales became fewer in number and were, accordingly, more scattered.

 

In Glamorgan a furnace was established at Caerphilly in 1680 and the works at Pentyrch continued to operate; in Carmarthenshire the Cydweli (territory of Cadwal) furnace came into being and in Breconshire ironmaking was carried on during the seventeenth century at Ynyscedwyn (Cedwyn's watermeadow) and Llanelly (church of St. Elli). In Monmouthshire the works at Tintern were operating and two furnaces, at least, one near Llandogo (church of St. Euddogwy) the other near Trellech (great stone), were built to provide the works with iron.

 

The sites at Llandogo, Trellech, Llanelly and Caerphilly still provide physical evidence of iron making during the seventeenth century when charcoal was still used as the fuel.

 

The Coed Ithel (Ithel’s wood) blast furnace, near Llandogo (ST 527027) in the Wye Valley, is sited on a hillside about 30 feet above the bed of a stream and on a level area, contained by stout retaining walls, one of which runs parallel with the roadway. The stream flows regularly throughout the year.

 

Excavation has disclosed a furnace structure about 24 feet square.116 The present maximum height from the foundation level is about 20 feet; the furnace never exceeded this height by more than a foot or two.

 

The square shaft was made up of 3 inches thick grey sandstone and the circular hearth was built up from the bottom with 6 inches of white sandstone. It joined the shaft at a point more than half way up the interior of the furnace. The hole which took the nozzle of the bellows, that provided the blast, was about 18 inches from the bottom of the hearth of the furnace.

 

Among the finds were three cast-iron runners which were probably used in the sandbeds or pig beds, that is, the casting area, to connect the main channels, along which the molten iron ran, into parallel furrows wherein the pigs were formed.

 

It may be helpful at this stage to explain the origin of the terms 'sows' and 'pigs' in relation to iron-making.

 

A sand-bed would be made to slope gently away from the furnace; it was made up of a long main channel, running from the furnace tap hole, a number of transverse channels in the sand, and a series of parallel furrows with their long axis directed towards the furnace, sufficient walls of sand being left between the furrows to form barriers strong enough to resist the pressure of the molten metal.108 At first the lowermost of the furrows would be connected with the main channel, communication between it and the other rows being then made in succession, by the removal of sand barriers previously erected. The transverse feeding channels formed the 'sows' and the parallel furrows the 'pigs'. These names were fancifully suggested by a sow feeding her litter of pigs.

 

Dr. Tylecote has concluded, that on the evidence which he obtained, the lines of the Coed Ithel furnace are unique, although its operation appears to have been typical of that of a mid-seventeenth century furnace in other areas. It used iron ore from the Forest of Dean, and charcoal and bloomery cinder (slag containing iron left on sites of early forges where iron was shaped into a bloom or ingot), but limestone was not included in the charge.

 

The furnace was in being in 1651 and between 1672 and 1676 it had an average weekly output of 18 tons. It lasted until the beginning of the eighteenth century but does not appear in the list of furnaces for 1717.95

 

The Trellech furnace, at Woolpitch Wood (ST 487048), is in an overgrown area with trees growing out of the furnace ruins, but enough of the exterior remains for it to be seen as square in shape. The furnace structure is 26 feet square at the base and diminishes to about 20 feet square at a height of to feet from the ground (Plate 36). It stands alongside a stream which could have provided the power for driving a waterwheel.

 

Plate 36 Trellech  Blast Furnace.

 

The ground rises steeply from the working level of the furnace on the west side, and there is evidence of the foundations of a bridge arch which connected the charging platform with a roadway on level ground above. Beyond the roadway are the remains of the walls of a building, measuring about 6o feet by 20 feet, which was connected with operations at the furnace. It is likely that this furnace also depended upon the Forest of Dean for iron ore, but found an abundance of local timber, for converting into charcoal fuel for smelting.

 

The third of the seventeenth century sites, where furnace remains may still be seen, is in the southeast corner of Breconshire on the northern side of the river Clydach in the parish of Llanelly (So 235142); there are also the remains of an interesting building near the furnace.

 

The blast furnace at Llanelly was in production during the 1600s, when it was associated with the Hanbury family. This site must not be confused with a site at Clydach, on the slope of a hill, to the south of the river (SO 232 128) where, it is claimed, that a small furnace about 12 feet high was erected in 1590 and that a forge operated nearby in I600, both establishments ceasing to exist in 16O7.113a

 

It has been said that the Llanelly furnace came into being in 16o6 and a forge to serve the furnace, within easy distance, in 1615.113a There is, however, more direct evidence in support of its existence at the end of the seventeenth century rather than at the beginning.

 

The Llanelly furnace continued to produce until the 1860s and in its present condition can be readily identified as a stone-built blast furnace, by the arched passage 5 feet high and 3 feet 6 inches wide, running between the main furnace building and the retaining wall built into the rising ground behind. This passage can be entered from the west opening only; the opposite end of the passage has collapsed thus limiting its length.

 

This is yet another existing illustration of the practice of building a blast furnace at the bottom of a slope, to bring the mouth of the furnace level with the higher ground for easy charging. At Llanelly the arched roof of the passage was in effect the under-part of a bridge arch, which connected the furnace charging floor with the upper ground level. In this passage, in the back wall of the furnace 2 feet from the floor there is a hole, possibly the tuyre, which took the nozzle of the bellows which provided the furnace with its blast.

 

The Llanelly furnace was built of sandstone and was 26 feet square, the retaining wall, which contained the hillside at the back, continuing for a further 12 feet on each side of the furnace. The length of the retaining wall, which runs from the open end of the passage, stands, but the other side has collapsed along with the front and second side of the furnace. Extensive excavation would be necessary to uncover details of the hearth.

 

On the higher ground behind the furnace remains, there is a two storey stone building, part ruinous and part occupied (Plate 37). This could have been the charcoal store at one time, the 'colehous' referred to by Major John Hanbury (who inherited the works at Llanelly and Pontypool from his father Capel Hanbury) in his 'Observations on the Making of Iron at Pontypool and Llanelly' dated 1704.113b The ruinous part of the building presents some puzzling features. At the end, there is a passage with an arched roof, 8 feet high extending from the front to the back, where it is met by another passage which runs along the back of the building at a height of 3 feet from the ground. This transverse passage has a similar roof to the first and is 3 feet from its floor to its ceiling: the remains of a similar passage are also to be seen at first floor level.

Plate 37. Ruined buildings behind Llanelly (Brecks.) furnace.

 

Within easy reach of the site there is a large dwelling-house, which may have been built or re-built by Francis Lewis, who was the Clerk of the Llanelly Furnace at the end of the seventeenth century and during the early years of the eighteenth. The initials FL and the date 1693 are clearly to be seen along the bottom of the coat-of-arms above the entrance to the house. A boundary wall, made entirely of fairly large lumps of iron slag, runs along an adjacent roadway for some distance.

 

Hanbury's dependence upon, and respect for Lewis, are made quite clear in his 'Observations' some of which were based on Lewis's experience in making cast iron. It is interesting to note that the production programme of the furnace depended upon the collection and cutting of wood and its 'coling' into charcoal.

 

'For the future I think of making about 300 tun of pigs yearly at Llanelthy & I am of opinion the best way would be to blow every year.

 

I think the proper time would be to begin constantly in the beginning of September and 300 tun would be cast in January at Farthest, so there would be all the spring to fill the house (the charcoal store) to keep till winter & from Sept. to the later end of Novbr. to bring the present stock.'

 

The layout of the furnace area, with its coal-houses and tenements and its proximity to Llanelly Forge, is to be seen in a survey of 1779 by John Aran.79 The furnace with its attendant charging house, cast house, and waterwheel, is admirably illustrated in Wood's, 'Rivers of Wales'. Nothing remains in the forge area, apart from some retaining walls and ramps but there is obvious evidence of the existence of the forge pond.

 

A tinplate mill was established on this site about 1800, the plates being made from charcoal iron from the forge. A present-day dwelling house at the end of Forge Row, which leads to this old industrial site, was the original tin‑house. The sheets were liberally coated with tin, for they The works,carried a coating of 6 pounds and upwards of tin per box. 83  The works was closed in 1884

 

The site of the Caerphilly furnace is to the north west of the town at T 142878, but there are no physical remains to point to a stone blast furnace; there are, however, slag heaps and slag is to be found in the nearby stream.

 

The furnace was built in 1680,113c and was under the ownership of the Tredegar family. 96a In 1694 the furnace was owned by John Morgan (of Tredegar), Roger Williams and Roger Powell, the last of the three living at Energlyn (originally 'Geneu'r Glyn') House within easy reach of the furnace. In 1787 a bigger furnace was built to replace the original and 'a powerful blowing engine'96a was provided in place of the old bellows which were actuated by a waterwheel.

 

Before the re‑building, about 200 tons of iron were produced each year, but afterwards 503 tons were produced in 1796. This was an appreciable production figure at the time.

 

The Caerphilly furnace was under several different ownerships – the names of Pratt and Harford being associated with it and a well‑known lcal figure Edward Lewis, had a share in it in 1747. The furnace remained in production until 1819. 96a It is included in John Fuller's list of 1747 ‑ producing 200 tons; some of this iron was, during the eighteenth century, supplied to the two forges at Machen.113c

 

iii) THE EIGHTEENTH AND NINETEENTH CENTURIES

 

In the preface to his book on the South Wales ironworks of the 1760 - 1840 period, John Lloyd wrote of'. . . the introduction of a great chain of Iron Works into South Wales in barely more than a generation, and their rapid development to a great height of prosperity . . .'102 This was the period which gave to South Wales its first concentration of industry within a defined area - the northern rim of the South Wales coalfield; it was a compulsive concentration which sprang from the easy availability of all the raw materials necessary for the making of iron.

 

At the heads of the valleys, the 'blaenau', and on the uplands beyond, iron ore was easy to come by, '. . . worked at slight expense in patches by turning over, like a garden, the open mountain slopes’.101a

 

The work of Abraham Darby, at Coalbrookdale, had culminated in 1709 in the use of coke as a fuel, in place of charcoal; the coal for coking was available in abundance in the South Wales coalfield and there were adequate supplies of limestone, used as a flux in the smelting process, and of stone for furnace building.

 

The blast furnace of this period was, in essence, a vertical, circular chamber, often called, the shaft, which 'widened from the bottom upwards and from the top downwards'.108a The inner lining was made of re‑brick, common brick and a rubble of stone as a filling. Externally the furnace was a solid, square structure massive in appearance. The internal chambers varied in shapes as the period advanced, each depending upon the theories of the furnace designers and builders.

 

The furnace top was a cylindrical chimney surrounded by a platform; around the lower part of this chimney there were, usually, four openings of regular shape through which the furnace was charged from the platform.

 

The stone-built furnaces were gradually replaced by rounded structures clad in wrought iron plate, and with the introduction of hoists for bringing up the raw materials to the charging platforms, the need for building furnaces into sloping ground ceased.

 

At the beginning of the period, waterwheels were used to drive the bellows which provided the blast, but steam‑driven blast engines were in use at the beginning of the nineteenth century. This resulted in a great increase in efficiency and brought increased production. The blast traveled through the blast main, a cast‑iron pipe which passed round the furnace, injecting the blast into the furnace through pipes connected with the openings in the furnace walls, the tuyeres.

 

In attendance upon the blast furnace were kilns in which the limestone and iron ore were calcined, or roasted to eliminate moisture and impurities, before the smelting process in the blast furnace. At the front of the furnaces there were cast houses, and always nearby, the engine houses which contained the blast‑producing engines.

 

The process of conversion of the cast iron into malleable or wrought iron, was carried on in buildings which housed the puddling furnaces in which the conversion was carried out. As the period advanced, rolling mills for the production of bar iron were developed and each ironworks' site was clustered with many buildings.

 

Many of the works, which were established on the strip of land extending from Blaenavon (headwater of the river Afan) to Hirwaun (long moor), began to run down about one hundred years ago. It is surprising that so much structural evidence of the works remains at the end of that period. Amongst it there are a number of interesting features ‑ one would expect it to be a rich field of study ‑ some of which lead to related features and sub ects, of which mention must also be made.

 

The approach to the industrial archaeology of the period of the ironworks must be selective as far as possible, to avoid needless repetition, and a geographical approach, starting on the eastern side, may be better than one based upon the chronology of the ironworks, which could lead to bewildering movements to and from different parts of the area.

 

MONMOUTHSHIRE

 

The ironworks at Blaenavon dated from 1789 in which year a lease was granted by the Earl of Abergavenny to Thomas Hill of Stafford and his associates Thomas Hopkins and Benjamin Pratt. There is, today, clear evidence in the towering, disintegrating masonry, of the original blast furnaces clinging to the hillside. Of all the blast furnace remains still to be seen on the sites of old works, these at Blacnavon present the classic example of using the upper levels of the ground for charging the furnaces (Plate 38)

Plate 38 Remains of furnaces Blaenavon.

 

The furnaces were described by Archdeacon Coxe in 180184 as. . the buildings which are constructed in the excavations of the rocks . . .' In this book there is an engraving of the Blaenavon Ironworks in 1799 which indicated that three furnaces were in operation and that each one was connected with a charging house, above and behind, and a cast house in front and on ground level. It shows also, some tramroads and the horses and trams which operated on them.

 

A plan of the Blacnavon works dated 1814 and preserved at the National Coal Board survey office on the site, gives a number of interesting details. It shows the five furnaces as two pairs and a single furnace this is seen in the present ruins. There were coke yards and 'the Mine Kilns' behind the furnaces on the charging level. The tram roads to 'mine works' and 'limestone rock' are shown and other features included, are 'mouths of levels for getting coals', 'mouths of levels for getting Mine or Iron Ore', a new and an old blowing engine, two melting fineries and cinder heaps.

 

A map of 1819, preserved at the Monmouthshire County Record Office, shows that there were five furnaces in blast; they were still in production in 1839.105 Among the remains, the round shape of a furnace stack is prominent, revealing a number of brick courses, and the stone masonry of the outside walls of another furnace is also to be seen. The disintegration has also revealed the use made of wrought iron bands and keys, as reinforcement during the building of the furnaces.

 

The outside walls and gable ends of a number of cast houses have survived, as have the remains of an engine house stack. Another interesting feature on the site, is what remains of a water balance tower erected near a coal level which was driven to the south cast of the line of furnaces.

 

In 1817 the Blaenavon works became associated with a works built at Garnddyrys (bramble cairn) on the western slopes of the Blorenge mountain. Cast iron was sent to this works from Blaenavon for finishing; the sites were connected by a tramroad. The Blaenavon works was connected by tramroad to the terminal of the Monmouthshire Canal at Crumlin (originally 'Crymlyn', curved lake) and the Garneldyrys works was similarly joined to the Brecon and Abergavenny Canal at Llanfoist (church of St. Ffwyst); from the two works, between 1802 and 1840, 447,392 tons of iron were transported to Newport.114 This was in the form of pig iron and cast iron and finished iron such as puddled bars, rods and rails.

 

The Garnddyrys works was closed in i86o and by 187o Blaenavon had ceased to produce iron rails -this applied to all ironworks which could not compete with the Bessemer steel rails which were then being produced at Ebbw Vale and Rhymney.

 

At the head of the Sirhowy (Howell's land) river, stand the remains of the Sirhowy Ironworks, which, after an indifferent start, following the granting of the first lease of the land in 1778, was from 1794 operated jointIy102b by William Barrow, an 'ironmaster' and the Revd. Matthew Monkhouse, Clerk, of Sirhowy and Richard Fothergill - a name which became well known in the South Wales iron trade - of Surrey, who had a small ironworks in the Forest of Dean.118a In 1800 the partners were joined by Samuel Homfray of the Penydarren (top of the hill) Ironworks and a new works, the Tredegar (Tegyr's farm) Ironworks, came into being less than a mile downstream from the Sirhyow works, the two works coming under the joint ownership of the new partnership. The Tredegar works, in time, became the bigger of the two having five blast furnaces to Sirhowy's four.

 

In 1818 the Sirhowy lease was up and Richard Fothergill's belief that he would secure its renewal, led him to confide in James Harford, who had become a partner of the Honifrays in their enterprise at Ebbw Vale in 1816. It seems that Harford acted unethically by acquiring the lease for himself, and as his acquisition of the Sirhowy works happened soon after the Homfray family had given up their interest in Ebbw Vale, he became one of the most powerful ironmasters in Monmouthshire. This led to the discontinuation of the joint management of Sirhowy and Tredegar and Fothergill ordered the removal of all moveable plant and rolling stock, such as engines, trams, tram-plates and barrows to within the Tredegar area, so 'that there should be no connection, however trifling, between the two works in future."118b

 

The remains at Sirhowy are fairly typical of those of the ironworks of the period - lofty arches, collapsed furnaces and walls gradually disappearing beneath coal rubble. It is known that a steam engine was introduced in 1797 'to assist the water power'118a but it seems that a waterwheel survived until the late 1870s at Sirhowy (Plate 39).

Plate 39 Sirhowy ironworks late I870's

 

The evidence, which is available in old photographs, is of considerable assistance in giving an idea of layout, the juxtapositions of furnaces, engine houses and other buildings and of the techniques employed in ironmaking. They also illustrate the changes in the design and appearance of blast furnaces and adjacent housings as the ironworks developed.

 

The history of three sites at Rhymney, which were developed as ironworks, reveals the usual agreements on leases, partnerships and the subsequent developments in production when commercial bargaining had subsided. There is a great deal that is interesting on the three sites today, that of the first development at Upper Furnace, of the Union Works and of the Bute Iron Works, all three eventually being merged into the Rhymney Iron Company. ,

 

The site at Upper Furnace was first developed in 1800 as the Union Iron Company by some Bristol merchants who were attracted to the area. A stone furnace was built on the left bank of the River Rhymney, inside the county boundary of Brecon. By 1802 'a considerable iron furnace had been erected and Iron Works of considerable extent established upon the premises, and a Dwelling-House built on the leasehold land adjoining to the furnace as a residence for the Manager, Mr. Richard Cunningham.102c

 

The ruins of the base of the furnace only remain but the topography of the site is such that it can be recognised, by comparison with others, as one which accommodated a small nineteenth century ironworks. The dwelling house of the manager still stands and alongside it a warehouse-like building which bears the date 1802 above one of its doorways.

 

The Union Iron Company became a target for Richard Crawshay, Cyfarthfa Ironworks, Merthyr Tydfil, who wished to establish a son and son-in-law in an ironworks. He succeeded, and in 1803 the Union Iron Works Company came into being, two of the previous proprietors being entrusted with its management. One of these, Thomas Williams, may have been the first Welshman to have participated, as a proprietor, in an ironmaking enterprise. This works was developed on the same side of the river, at a point downstream where the bottom of the valley was wider.

 

In 1804 it was taken over by Richard Crawshay and Company and was carried on under the ownership of Benjamin Hall, Crawshay's son-in-law until 1820.

 

On the opposite side of the river, in 1825, the Bute Iron Works was established; three blast furnaces were erected and they 'were of a somewhat pretentious style of architecture, having a front of Egyptian design.'102d

 

Within a few years the two works were amalgamated into the Rhymney Iron Company, a concern which for many years afterwards was a prominent industrial name, first in ironmaking and subsequently in coal-mining and marketing.

 

The site of the Rhymney Iron Company has yielded some interesting industrial remains. Until a few years ago there was some evidence of the fanciful style of the blast furnaces on the tight bank of the river. It is of interest to refer to the furnace built by John Bedford at Cefn Cribwr (comb ridge) about forty years earlier, described on Page 78, and his inclusion of a 'Ballustrated. Battlement Level with the Bridgehouse floor'.

 

The interior of one of the Rhymney furnaces may still be viewed from above, and part of one of the bridge arches - these spanned the distance between the charging platforms and the level ground beyond remains. The bridges themselves, however, are all in a collapsed state.

 

At the northern end of the works site, there are the remains of batteries of coke ovens, each oven being of the long rectangular type, which had no provision for the utilisation of the gas given off, for heating purposes in smelting and other processes in the works. An ironstone level, on this site was penetrated for a distance of about seventy-five yards in February 1966 by members of the mining and metallurgy section of the South East Wales Industrial Archaeology Society and a small, wooden tram  was found about thirty yards from the level mouth, completely submerged in water but standing on rails. The level is about six feet high and six feet wide with a semicircular arched roof and masonry lining. The rail track was on a raised portion with drainage channels on either side, the rails themselves of cast iron right-angle plates, the tram wheels being without flanges. A small amount of ironstone pins and nodules, were removed from the tram before its recovery.

 

The body of the tram measures 4 feet 3 inches long by 1 foot 6 inches high by 1 foot 3 inches wide and it carries strengthening straps of wrought iron around the sides and bottom. One end was open for loading and unloading, the inside of the panel at the other end bearing an indentation, consistent with contact from a shovel during the removal of ironstone. The wood of the tram seems to be birch.

 

NORTH GLAMORGAN

 

It could be claimed that the story of the iron industry in Merthyr Tydfil (church of St. Tudful) during the years under review, epitomises the story of the iron industry throughout the region. Four ironworks of great magnitude - Dowlais, Penydarren, Plymouth and Cyfarthfa grew out of small enterprises, dependent upon waterwheels and bellows for air blast. Physical evidence of these ironworks in remains of furnaces and buildings, is not extensive, but a brief look at the historical development of the works, and the height of production achieved, is permissible if only to emphasize how quickly the evidence of a vast industry disappears, and how limited is the time still available to the industrial archaeologist.

 

The story of Dowlais (black brook) Ironworks begins with the acquisition by Thomas Lewis, Llanishen (church of St. Isien), of a lease of land in 1757 in Dowlass and Tory Van and the formation in 1759 of a partnership of nine persons with a capital of £4,000, which agreed to build a furnace or furnaces 'for Smelting of Iron Ore or Iron Mine or Stone, into Pig Iron'. 90

 

Dowlais furnace was the second coke furnace in South Wales, the first being at Hirwaun. It has been generally accepted that charcoal was the original fuel used in firing this furnace, but this theory has been confidently refuted.104 Merthyr Tydfil's elevation above sea-level was not conducive to tree growth and the original lease (which reserved all timber trees for the lessor) included an agreement to share 'all Metals, Castings, Iron, Timber, Wood, Coal, Coak, Charcoal, Braises, Tools, Utensils.'

 

The fact that the enterprise 'was a direct result of the Horseha furnaces, Shropshire which used coke, and in 1757 were each making 15 tons of iron per week, a total of nearly 1600 tons per annum’104 lends considerable weight to this argument. 'Any statement that, after 1757, any South Wales furnace started on charcoal is mythological', ‑ Dr. R. A. Mott, one time president, the Sheffield Trades Historical Society and of the Coke Ovens' Managers' Association, in a letter to the author.

 

John Guest, of Broseley, near Coalbrookdale, where coke was first used successfully in the smelting of iron by Abraham Darby in 1709, was appointed manager of the Dowlais furnace in 1767 after a number of years at the Plymouth furnace, named after the land owner, the Earl of Plymouth. In his early days at Dowlais the furnace made 18 tons of iron per week using 8 tons I cwt. of coal per ton of iron produced. 78

 

The general use of coke was soon followed by other improvements in the technique of ironmaking; blowing cylinders replaced the wood and leather bellows for providing the blast, and the steam engine took the place of the waterwheel. Due emphasis should be given to these technical developments, which enabled the emergent ironmasters to take full advantage of the easy availability of the raw materials they required; the rapid progress made would not have been possible without the use of coke and the power of the steam engine.

 

In 1786 the management of the Dowlais Ironworks was taken over by Thomas Guest, son of John Guest and by this time an annual output of 1,500 tons in 1763 had been increased to 5,500 tons. The export of Dowlais iron to America started in 1780

 

In 1800 three blast furnaces were in operation and the works was developing, but it must not be assumed that the development of the Dowlais Iron Company and other ironmaking concerns went unimpeded.

 

Among the original lessees was Isaac Wilkinson of Wrexham, who held a 1/16th share in the venture, and who had in 1757 taken out a patent for a cylinder blower, in which a piston was operated by a waterwheel.104 A cylinder blower of this kind was installed on the Dowlais Brook and was referred to as an 'Engine' on a map of July 8th, 1769, which is with the Dowlais Papers at the Glamorgan County Record Office; the length of the pipe carrying the air from the cylinder to the furnace appears to have been considerable. Unfortunately the Dowlais brook could not provide a constant, adequate supply of water and a high rate of production was not reached until a Boulton and Watt double‑acting engine and the John Wilkinson, son of Isaac Wilkinson, cylinder blower were introduced in 1803 ‑

 

In addition to technical deficiencies, the wars of the period 1795 to 1815 brought depressions, as the problems of over production had to be solved, and the re‑adapting of ironmaking plants for products of peace became necessary, only to be followed swiftly by a re‑conversion, to meet the needs of war as another outbreak occurred. In such circumstances, a stable development was difficult and the final magnitude reached by this and other ironworks is all the more creditable.

 

John Josiah Guest assumed control of the works in 1807. By 1815 five blast furnaces produced 15,600 tons per year and in 1823 ten furnaces produced just over 22,000 tons. In 1845 eighteen blast furnaces provided a weekly production of 101 tons of iron per furnace; six of these furnaces were included in the new Ivor Works built to the north of the original works.

 

The Dowlais Company mined coal and iron ore and produced pig iron, much of which was sold to companies which used the iron for manufacturing purposes; for example, Brown Lenox & Co, the chain-makers of Pontypridd and the Neath Abbey Iron Works, where various kinds of engines were made. The Company also refined its own iron, and iron bars were forged and rolled into rails after the coming of the railways ‑wrought iron rails for tramroads were made in earlier times. The rails for the Stockton and Darlington Railway, the first passenger railway, came from Dowlais in 1821.

 

Wrought iron was produced in puddling furnaces of which there were many in an ironworks such as Dowlais. The process was invented by Henry Cort and was patented by him in 1784

 

A puddling furnace (Plate 40) had a shallow hearth, with a fire grate separated from it by a firebrick bridge; it was formed externally of cast iron plates and provided with suitable openings in front for the fire hole and the working door, and lined internally with firebrick. The crown of the furnace was also lined with firebrick, and at the end opposite to the fire grate, there was a flue connected to a simple, rectangular stack provided with an iron damper.

Plate 40. Iron Puddling Furnace

 

Before the puddling process was started, the furnace bottom was specially prepared and it was important that it should be constantly kept in good working order.

 

When the furnace was red-hot, pig iron, usually in half pigs, was fed into it with a quantity of hammer slag. The heating of the iron went on for some twenty minutes and the pigs were then turned to heat them more uniformly. The mass was then stirred up with an iron bar to bring up any pieces of iron not completely melted.

 

The next stage was the puddling or rabbling of the charge with long iron bars (rabbles), bent at the end at right angles, through a hole in the side‑wall, exposing it evenly to the action of the flames. The iron was thus 'cleared' or purified and then brought to the boil and became pasty, or in the language of the puddler 'came to nature'. The pasty iron was then worked by the puddler with his rabble, into a number of balls which were lifted from the hearth with tongs and transferred to a hammer to be rid of the slag. This was then rolled into 'puddled bars', the name given to crude wrought iron.

 

The whole operation lasted about two hours, the work being extremely arduous and the physical effort required to manipulate the metal in the heat, was 'the severest kind of labour ever undertaken by man'. 87

 

In the mid‑forties Dowlais Ironworks was the largest works in the world, employing 10,000 and having in addition to the furnaces, rolling mills, forges and foundries.

 

The great expanse of land once occupied by the Dowlais Ironworks is now fairly clear of buildings. There is much rubble but the remains of the brick works are still to be seen and those of the side walls of blast furnaces which jut out from higher ground. All this is in marked contrast to the spectacular evidence of the works in full production, contained in three watercolours painted by G. Childs in 1840; these are at Guest Keen Iron and Steel Works, Cardiff. One shows at least 14 of the huge blast furnaces in operation at the time; another (Plate 41) shows in close‑up the charging platform of a furnace and the preparation of the charge for a furnace indicating quite clearly the participation of women in this work.

 

Plate 41. Preparation of Raw Materials Dowlais Ironworks, 1840.

 

Among the remains on the site are the ruins of early coke producing ovens (Plate 42). They suggest that originally each oven was rectangular in shape and in its horizontal section had the form of a rectangular chamber, covered with a flattened arch. Ovens of this type were separated from each other by a comparatively narrow brick wall.115 The width of the oven varied from 7 feet to 8 feet and the dividing wall was between 18 inches and 3 feet in thickness. A feature of these walls was that they were sufficiently thin to transnmit heat from one oven to the next, so that when an oven had been discharged and was being re‑started, the ovens on each side helped it along by the transmission of heat.

Plate 42. Dowlais Ironworks – Ruins of Early Coke Oven

 

The ovens were charged by hand from the front and from floor level, and the gases formed, escaped into the outside air through a hole in the roof of each oven. At the charging end there was a lifting door made of cast iron.

 

The remains of the Dowlais ovens, reveal that the thickness of the coal charged was about 12 inches, there being a lack of carbon deposit on the lower three courses of the brickwork interior. The present inside measurements suggest that each coke oven was about 6 feet 6 inches wide, by 6 feet high and 12 feet 6 inches deep. They were stop‑ended or single door ovens.

 

The Penydarren Ironworks was established on the left‑hand side of the Dowlais Brook, adjoining the Dowlais Ironworks to the south. Its present day remains are negligible, comprising only the broken remains of furnaces, clinging to the rising ground of the small hill which gave the works its name. In Tram Road Side, a short distance from the main site, there are also the ruins of small, stone re‑heating furnaces, which appear to have been built in pairs, with a chimney common to each pair.

 

This ironworks cannot be dismissed summarily in view of its association with the Penydarren Tramroad and the Cornish engineer, Richard Trevithick.

 

The works came into being in 1784 under the ownership of three sons of Francis Homfray ‑ Jeremiah, Thomas and Samuel and George Forman of London. Francis Hornfray had been invited by Anthony Bacon, some two years previously, to establish a refinery and a forge at Cyfarthfa, the iron to be supplied by Bacon from his own furnace there. This arrangement was discontinued when Bacon, then Member of Parliament for Aylesbury, withheld the iron supplies, after he had negotiated with the Government to supply cannon for the army and the navy, which were then engaged in the American War of Independence.96b

 

It is assumed that the Penydarren site was chosen by Francis Hornfray and that the lay-out of the works was based on his experience as an ironmaster in the Midlands; the management of the works, however, was undertaken by two of the brothers, Jeremiah and Samuel Homfray. 'This was in 1786, or thereabouts, and in the course of the next six or eight years a pretty little Iron‑Works, complete in every respect, had been established close to the left bank of the Dowlais Brook."Ole One readily accepts that a nineteenth century ironworks merited such a description.

 

In common with the other ironmasters in Merthyr, the Hornfrays were hampered by the lack of efficient communication with the port of Cardiff. Their joint efforts resulted in the passing of the Glamorganshire Canal Act in 1790, and by 1792 the canal was navigable from Merthyr - Cyfarthfa Ironworks - to Pontypridd, and to the sea-lock at Cardiff on February 10, 1794. The canal has, by now, been filled in for much of its length but at various points along its route the ruins of locks still remain and some of the original bridges are still standing.

 

The Glamorganshire Canal did not prove to be the complete answer to the transport problems of the ironmasters; there was constant congestion, due to the heavy demands made upon the waterway, particularly between Merthyr and Abercynon, where numerous locks had been built to compensate for the steep descent. This, and a deeplying disagreement with Richard Crawshay of Cyfarthfa Ironworks, led the partners of the Dowlais, Penydarren and Plymouth Ironworks to build a tramroad from Merthyr to the canal basin at Abercynon. It became known as the Penydarren Tramroad103 and it achieved fame as the tramroad on which Richard Trevithick's locomotive drew a load on rails - the first steam locomotive to have done so - on Tuesday, February 21st 1804, 'a date forever memorable in the history of the locomotive.'88

 

In 1803 Trevithick became associated with Samuel Hornfray and the ironmaster interested himself in Trevithick's work on high‑pressure steam engines, providing him with the facilities to build his 'Tram Waggon' at Penydarren.

 

The tramroad is not without its interest today; appreciable lengths of it reveal parts of the stone sleepers, which carried the 3 foot lengths of cast‑iron plate rails, which can still be recovered. The passing place at Pontygwaith (ST 081976) remains in being; a relic of the tramroad, a cast‑iron wheel, was discovered in June 1966 by two telephone line workers, about half‑a‑mile from this point, in the direction of Abercynon, partially buried in the side of the embankment which drops steeply to the river Taff. This cast‑iron wheel, 31 inches in diameter, contains eleven spokes and it is reasonable to assume that it was part of a tram which traveled along the tramroad at one time.

 

Trevithick's locomotive was not a complete success because its great weight caused the iron tram-plates to break; it was ultimately converted into a stationary engine and used as a prime mover at the ironworks.

 

The Plymouth Ironworks, the second to have been established in the Merthyr area, followed upon the granting of a lease of land by the Earl of Plymouth to Isaac Wilkinson of Wrexham, and John Guest of Broseley (associated with Dowlais from 1767) in December 1763, who, to quote from the lease, intended 'to erect ... certain ffurnaces, fforges, Mills, pothouses, or other Works for the making and Manufacturing of Iron'. From the outset the works bore the name of the landowner, the first occupiers styling themselves as 'Messrs. Wilkinson and Guest, Plimouth Company',102f but for some time the works bore the name of 'Ffwrnes Isaf', the lower furnace, its original location being given as Coedcae Glynmil (quickset hedge of mill). 96c

 

In common with many of the ironworks of South Wales, the story of the beginnings of the Plymouth Ironworks has a strange fascination, but it will be enough to mention that Anthony Bacon already had an interest in 1766 ‑ he was thinking of buying more shares in the venture in June of that year122 and his Cyfarthfa Company became the owners by the end of 1766 ‑ and that his brother‑in‑law Richard Hill took over the management in 1784 and subsequently acquired the works which remained in the possession of the Hill family until 1862.

 

This works, although possibly overshadowed by the other works at Merthyr Tydfil, was not unimportant. On the original site two more furnaces were producing by 1800 and a fourth in 1815 ‑ the remains of one of these may still be seen at SO 059049, much of the surrounding area being recognisable as the site of the ironworks.

 

In 1807 the Plymouth Forge Company established a forge at Pentrebach (little village) (SO 06204I), which was followed by a rolling mill in 1841. A short distance to the south, at Dyffryn, a blast furnace was erected in 1819 and followed by two others and a blowing engine and a battery of coke ovens by 1824. A plan of the Dyffryn Furnaces drawn in 1861 (a copy of which has been deposited in the Department of Industry, National Museum of Wales) shows five blast furnaces, numbered 5, 6, 7, 8 and 10, kilns and limestone sheds, a refinery, cast house, engine houses, a double‑waterwheel and a 'fitting‑up' shop.

 

The remains of the blast furnaces have been slowly covered by coal refuse tipping, a rough half section of one only still standing, the wall of which is slowly disintegrating.

 

Anthony Hill, son of Richard Hill, emerged as an important member of this Company. Regarded as one of the leading chemists and metallurgists of the day he was responsible for a number of experiments aimed at improvements in ironmaking. During his time the iron produced by the Plymouth Iron Company was ten shillings higher in price than any other iron 'because it was of better quality.’96d

 

The fourth of Merthyr's ironworks was the Cyfarthfa (barking place, where animals stand at bay) Ironworks inevitably linked with the Crawshay family. The main works covered an extensive area on each side of the river Taff, on the north‑western outskirts of Merthyr and a subsidiary, the Ynysfach (little island) Ironworks was developed on the right hand bank of the river, west of the centre of the town.

 

The accounts of the origins of this vast enterprise - 'Cyfarthfa works were by 1807 the largest ironworks in the world, with six furnaces making over 10,000 tons per annum' - are fairly consistent. In a flourish of hyperbole, Charles Wilkins118c states that the first furnace at Cyfarthfa was built in 1765; John Lloyd102g accepts this date and locates it as being 'some little distance higher up the river than the present main works. This No. 5 has a plate on it dated 1765, W.C (William Crawshay) 1827, and is one of the sights of Merthyr'; in 'Hanes Morgannwg'96C the date is also given as 1765 and the location as near the confluence of the two Taff rivers, Taf Fawr (great river Taff) and Taf Fechan (little river Taff).

 

The Charles Wood Diary122, however, reveals that the first furnace was built in 1766/67 by Wood himself for Anthony Bacon and Company, in which Wood was a partner. Charles Wood was the son of William Wood, an ironmaster of Wolverhampton and entries in his diary in 1766 and 1767 give numerous details, relating to the building of the Cyfarthfa furnace and forge building, and of the association with the Plymouth Ironworks.

 

A forge, taking pig iron from the Plymouth furnace, had already been built and Wood, whilst the furnace was being built, was adding other buildings on the forging site. By this time, the Plymouth furnace had been acquired by Bacon and many of the castings for the Cyfarthfa furnace were being made there.

 

Unfortunately the Diary does not give the exact location of the furnace; there are constant references to the farms of Llwyncelyn (holly grove) and Rhyd‑y‑Car (vehicle ford) at present districts of Merthyr Tydfil, but there is nothing conclusive about either of these. It may be that the location given in 'Hanes Morgannwg' is the right one, because Wood refers to the farm called 'Tai Mawr' (big houses), known to be immediately to the west of Taf Fawr and a little above its confluence with Taf Fechan. On the other hand, immediately below the confluence of the two rivers, on the right bank of the Taff, are to be found the remains of the circular hearth of an early furnace (Plate 43) and there is a substantial depth of water in the river at this point, but it cannot be argued with certainty that this was the original site.

Plate 43. Cyfarthfa Ironworks Early Furnace Remains

 

It must be accepted that the site was fairly close to the main river, in the light of the entry in Wood's Diary for September 9th 1766. 'Leveled the Bank, against which, the furnace is proposed to be built; it is 47 feet high to the flat part of the field, & from thence to the surface of water (about 6 inches running over the Wear) 19 feet more, in the whole 66 feet. The stack may be 50 feet high, the foundation 6 feet and then there will remain 10 feet for the Cistern & cut, or back race, for a flat bottom boat to convey the metal to the flourishing furnace etc. It may be contrived, for the metal to be put into the Boat, out of the Cistern, which will save room for binns & Labour, in taking it out of the Binns. This may be considered of.'

 

The reference to the 'Cistern' and the conveyance of the metal to the 'flourishing furnace' is mystifying.

 

This entry also contains a reference to an agreement for raising of imine' (ironstone) from Penywain (end of the moorland) which was due west of Tai Mawr and not far distant.

 

Wood also mentioned the building of a furnace stack, 36 feet square and 50 feet high and 'the Holme where the Blast‑furnace is proposed to be erected' is described as a 'very convenient place, a fine bank for an high one and if there should not be found room for a bridge-house (this could mean a charging house) at the back of the Stack, an Arch may be sprung to the Rock upon the Bank . . .' In this entry, which was for June 22nd 1766, there follows the significant phrase, 'The bank for coking the coal will be inconvenient . . ‘, which makes it fairly certain that the Cyfarthfa furnace used coke from its first operations. This view is strengthened by a further entry, dated June 25th, when Wood showed the site of the proposed blast furnace to Isaac Wilkinson (of Dowlais and Plymouth) 'which he much approves of but advises to take the field on the other side of the road, for a Bank to burn Stone and Coke Coal.'

 

The widespread remains of kilns, blast furnaces and various kinds of buildings, long lengths of retaining walls and enormous slag heaps, suggest that the Cyfarthfa Ironworks at the height of its production, covered a large area of land to the west of the Taff. There are photographs and paintings which show this to be so.102h The most interesting of the remains are those of the Ynysfach Furnaces a short distance to the south of the main works. There are four furnace structures, but all the front arches have been bricked in and the stacks have collapsed inwards at a height of about 20 feet. It is still possible, however, to walk the length of the arched passage, which ran beneath the bridge house and which stood on the same level as the charging platforms of the furnaces; it is fairly certain that this passage carried a blast pipe from one of the two engine houses (Plate 44) ‑ even in its ruined state an impressive building ‑ which stood at the end of the line of furnaces and is seen in the background of a photograph of the site, taken in 1905.102g This engine house, and the second, had stacks in attendance and there were cast houses in front of the furnaces; beyond the cast houses stood the refinery.

Plate 44. Engine House Serving Ynysfach Furnaces

 

One of the photographs of these furnaces shows a cast iron key plate bearing the initials 'C & G' and the date 1801.102g Two furnaces were built as a pair in 1801 by Richard Crawshay, who bought the Works from Anthony Bacon's heir in 1794, and Watkin George, 'the mechanical genius of Cyfarthfa' 118d who became a partner in the company in 1792. The cast iron bridge (Plate 45) which spanned the Taff to the east of Ynysfach, was designed and built by Watkin George in 1800 of an iron reputed to be rustless; this bridge was unnecessarily removed in 1964.

Plate 45. Cast Iron Bridge Merthyr Tydfil

 

Another key plate still remains at Ynysfach, in the arch of the third furnace; it bears the date 1836 and the initials of William Crawshay II, who became the sole owner of the Cyfarthfa Ironworks in 1834, and probably indicates the date of the building of the remaining two of the four furnaces.

 

Ironworks were established at Abernant (mouth of the brook) immediately to the north‑west of Aberdare, in 1802, by Jeremiah Homfray and his partners and came under the direction of Rowland Fothergill in 1819, when they were grouped with the Aberdare Ironworks under the title of the Aberdare Iron Company. Chancery proceedings in 1846 brought forth a list of Particulars referring to a proposed sale of both works; a copy of this list, dated 11th June 1846, is available at the Cardiff Central Reference Library.

 

On the site at Abernant there are at present only the remains of a blast furnace, which has crumbled almost to the top of the arch above the fore‑hearth, and a chimney stack which served an adjacent enginehouse. A photograph of the site about 50 years ago, which has happily been preserved, shows a blast furnace standing to three‑quarters of its original size, yet in a ruined state, with a three‑storeyed engine house, and its attendant chimney stack, a very short distance away from it. Between the furnace and the engine house, there is an air reservoir or regulator, which regulated the blast provided for the furnace by the engine. The air reservoirs in use at ironworks in South Wales were reputed to be spheres made of iron plates ‑ the one in use at Penydarren Ironworks, Merthyr between 1825 and 1850 is known to be spherical ‑ but the reservoir at Abernant is seen (Plate 46) to be ellipsoidal and located in an upright position. The blast main leading from the base of the reservoir into the side of the furnace is plainly to be seen. This photograph provides rare evidence of the use of a regulator at a South Wales ironworks during the first half of the nineteenth century.

 

Plate 46. Abernant Ironworks

 

The sale particulars of 1846 indicate that there were at Aberriant three blast furnaces, fineries, three blowing‑engines, water‑wheels, forges and rolling mills 'with engine power complete' and 'blast and other pipes in use and regulators in use'. The particulars also include the important information that one of the blast furnaces 'has hot air apparatus in full work', and another the same apparatus, 'but not in a state quite fit for present use'. This indicates that hot blast was used in smelting iron at Abernant at least before 1846. The introduction of hot blast is credited to James Beaumont Neilson in 1828, when it was used at the Clyde Ironworks; James Budd first used it in South Wales ‑at the Ystalyfera Ironworks in 1844

 

The first blast furnace in South Wales to use coke as fuel,104 was the Hirwaun furnace established in 1757 by John Maybery who had operated a forge at Powicke in Worcestershire. The coal used was anthracite.

 

The ironworks which was developed on this site suffered many reverses; it was leased for a time (1780‑86) to Anthony Bacon and was, afterwards, under different ownerships until 1819, when it was acquired by William Crawshay I and developed into an ironworks with four blast furnaces and a rolling mill, on the left bank of the river Cynon, which took the works into the parish of Penderyn and the county of Brecon. The ruins of four blast furnaces, a collapsed bridge arch and other building remains, provide evidence of the disintegration which happens over a long period of time ‑ the works was disposed of by the Crawshays in 1864.

 

During one of its thin periods, the sale of the works was considered and a list of particulars was included in a printed document, drawn up for a proposed auction sale in 1813, a copy of which, dated 26th January 1813, is in the Department of Industry at the National Museum of Wales.

 

The preservation of documents and papers relating to works, which have long ceased to function, and to their technical contents, and to works which are disappearing even currently after only a limited existence, is important in the field of industrial archaeology. These documents often point to different manufacturing techniques, and the rapidity of changes in these directions emphasises the importance of such documents in recording industrial development and progress.

 

The Hirwaun document lists in detail the capital equipment and resources of a medium‑sized ironworks in South Wales at the beginning o,f the nineteenth century. Under Lot I it lists:

 

TWO WELL CONSTRUCTED FURNACES

each about 4o Feet high and of proportionable Diameter;

Two Cast Houses, one about 45 Feet by 4o, and the other

36 by 33.

 

AN AIR FURNACE, TWO FINERIESA CAPITAL BLAST ENGINE

 

On BOLTON and WATT'S improved Principle,

now Blowing the Two Furnaces and Two Fineries

with 78 Inch Blowing Tube and 38 Inch Steam Cylinder, working

a 6 Feet 8 Inch Stroke and Water Regulator.

 

A FORGE

 

One Hundred and Fifty‑Seven Feet in Length, 44 Feet in Width

at one End, and 34 Feet at the other, with io Pudling

and 5 Ball Furnaces.

 

TREVITHICK'S STEAM ENGINE

 

Working by a 6 Feet Stroke, Two Pair of Pudling and One

Pair of finish Rollers, capable of Rolling from

80 to 100 Tons Weekly.

Forge Counting House, Pattern Room, Drying Sheds,

Carpenters' and Smiths' Work Shops

Water Wheel, Turning a Lathes for the Rollers, Grinding

Clay &c.

Brick Furnace Kiln, of sufficient Size to burn 13,000 common

Bricks.

 

FOUR KILNS for CALCINING the IRON STONE,

 

Mineral Yard, Coke Banks, Two Counting Houses, Three Lime Kilns, which supply Lime to the surrounding Neighbourhood to a considerable annual Profit: and every Requisite for conducting the Business.

 

Under "The Mines of Iron Stone", it says that:

 

"Several Veins of Ore of an excellent quality, in various thicknesses, from 8j inches down to 3 inches, are worked by Levels, the distance from the mouth or opening of the farthest does not exceed 23 miles from the Furnaces, and the nearest within 1200 yards.

 

"The Collieries comprise Four Levels, called the Old Lime Kiln, Old and New Glovers and the Gothlyn. Having several veins of Coal, some of good Coking Quality, other for the Furnace, in almost inexhaustible supply, from 4 feet to 9 feet in thickness, these Levels are all within Two Miles of the Works."

 

The need for meeting the housing needs of the workers is reflected in frequent references to tenements; sixty‑four near the works site, five at Penhow, thirty‑eight at Coedcafellin (quickset hedge of mill), two (one unfinished) at Rhydia (fords) Mill, others at outlying farms and in a reference to 'Two Pieces of Land part of Hirwain Common ... for Building Houses for the Workmen near the Collieries.'

 

 

WEST GLAMORGAN, CARMARTHENSHIRE AND PEMBROKESHIRE

 

At Neath Abbey stand the ivy‑clad remains of two stone blast furnaces, which tower above the ground. The original drawing of the furnaces has survived and has been deposited at the Glamorgan County Record Office; it shows Number I furnace to have been some 51-1/2 feet high from the furnace bottom and Number 2, 63-1/2 feet. In 1798 they were referred to as, ‘…Two immense blast furnaces belonging to Messrs. Fox & Co. . . constantly at work, each of them producing upwards of thirty tons of pig‑iron every week. They are blown by iron bellows, worked by a double engine, constructed on the plan of Messrs. Boulton and Watts, with a steam cylinder of forty inches in diameter.’117

 

The site was leased in 1792 to the Cornish Quaker family, Fox, the names of Peter Price, Samuel Tregelles and John Gould being also associated with the operation of the ironworks in subsequent years.110 The firm's letters were variously signed by the different partners102, their recipients often being addressed 'Esteemed Friend', the letters ending, 'Your friends' or 'Thine Truly'. The works remained under the management of the Quakers until 1875

 

In their 'engine Manufactory', this firm established for itself a very high reputation for the building of locomotives, stationary engines, marine engines and many kinds of machinery. It is very gratifying to be able to say that plans and drawings of these machines, numbering many hundreds, have been preserved and are available for study; they were deposited at the Glamorgan County Record Office in 1964 by Mr. A. W. Taylor of Taylor & Sons Ltd, Briton Ferry, Glamorgan.

 

There are, still in existence, a number of examples of the finished work of the Neath Abbey Works. In 1964 the Vivian Tinplate Works of the Briton Ferry Steel Co. Ltd, were demolished; this works was established in 1926 on the site of the Margam Copper Works. Parts of the roof structure of the building was supported on two cast iron pillars, 16l inches in diameter, each bearing the date 1800 and the name 'N. Abbey'; a short length of one of these pillars including the inscriptions is preserved in the Department of Industry, National Museum of Wales.

 

Two further examples of Neath Abbey engines, which have survived, are to be seen at Glyn Pits, near Pontypool (ST 265999). One is a beam engine, together with pump, installed in 1845 in a building bearing the date and the initials C.H.L., Capel Hanbury Leigh, who inherited the industrial enterprises of Major John Hanbury of Pontypool. The other is a winding engine which drove two reel drums, each 15 feet in diameter, carrying flat winding ropes.

 

The remains of the ironworks at Ystradgynlais (strand of Cynlais) and Ystalyfera (water meadow at end of short share) are important because both works were closely connected with developments in the technique


of ironmaking and were situated in the Tawe Valley, which benefited from the opening of the Swansea Canal in 1798.

 

The ironworks at Ystradgynlais, the Ynyscedwyn Ironworks, developed from a blast furnace built in 1696, there being a succession of owners but no great success, until the works was taken over in 1823 by George Crane of Bromsgrove, Worcestershire.

 

A number of attempts had been made at smelting iron using anthracite coal as fuel, but without success. 'It was reserved for Mr. George Crane, of the Ynyscedwyn Ironworks, to solve this difficult problem, and which he has effectually done by making strong and excellent pig iron in respectable quantities for 2 years together, by means of hot blast, and with raw anthracite coal.'105a

 

The manager of the works at the time, David Thomas, trained at the Neath Abbey Ironworks, has also been credited with the discovery, but it is likely that the two men collaborated to the same end. Crane reported on his discovery in a paper read before the British Association in 1838. He had taken out a patent for the use of hot blast with anthracite coal in 1836; a patent for melting iron by hot blast using coke was granted to James Beaumont Neilson of the Glasgow Gas Works in 1828.

 

In 1837 there were three blast furnaces in production at Ynyscedwyn, increased to six by 1853. The remains of blast furnaces are readily recognizable on the site of the former works, but it is now remarkable for a chimney stack, dated 1872, and two fairly high walls, each containing six archways, made of yellow‑faced bricks, ruined monuments of an unfinished building.

 

A tinplate works of three mills was built on the ironworks site in 1889. It was operative for sixty years and its plant was finally dismantled in, 1946 under the Tinplate Redundancy Scheme.13 The building, although in a dilapidated state and used as an iron foundry for a number of years, still stands, as do the chimneys of the heating furnaces and the engine house stack.

 

At Ystalyfera a blast furnace was built in 1838 by Benjamin Treacher and Evan James of Swansea; the site was acquired by a company consisting of Sir Thomas Brancker and J. J. Hogan of Liverpool and Edward Budd of Swansea in 1840‑ In a short time, Edward Budd's son, James Palmer Budd became the general manager and ultimately the proprietor of the works, being responsible for its development and success over a period of forty years.119

 

In 1846 there were six blast furnaces at work and eleven in 1851, a new forge being in operation by 1853‑ Within a few years, forty puddling and balling furnaces were operating and a sixteen‑mill tinplate works had been established. In 1865, 30,000 tons of iron were produced and in 1872, 182,000 boxes of tinplate were sold. At this time it was regarded as being among the biggest tinplate works in the country. This integrated works now employed 3,000, and the company also had a further 1,000 workers employed in coal and ironstone mines.

 

References & Bibliography

 

x

IV.  IRON, STEEL AND TINPLATE

 

77. ADDIS, J. P. (1957). The Crawshay Dynasty, Cardiff, p. 138

 

78. ANON (1937). Guest Keen Baldwins Iron and Steel Co Ltd, History and Development, Cardiff, p.16.

 

79. ARAN J. (1779) Survey of several estates situated in Brecon and Monmouth Counties and belonging to John Capel Hanbury, 47, Llanelly Estate and Works. Preserved at Pontypool Park Estate, Pontymoile.

 

80. ASHTON, T. S. (1963), Iron and Steel in the Industrial Revolution Manchester, third edition, p. 22.

 

81. BEDFORD PAPERS, 1767-1793, N. L. W., Aberystwyth; other MSS in library of Royal Society of Arts, John Adam Street, London, W.C.2.

 

82. BEVAN, T. (1956). Sussex Ironmasters in Glamorgan, T.C.N.S., Vol. LXXXVI, Cardiff, p.5

 

83. BROOKE, E. H. (1944). Chronology of the Tinplate Works of Great Britain, Cardiff, p. 146.

(a) P- 73   (b) p. 69   (c) p. 143

 

84. Coxe, W. (1801). An Historical Tour in Monmouthshire, London, p. 227

 

85- CHAPPELL, E. L. (1840). Historic Melingriffith, Cardiff, p. 30

(a) p. 33

 

86. DAVIES, W. (GWALLTER MECHAIN) (1815). General View of the Agriculture and Domestic Economy of South Wales, London, Vol. II, p. 456.

 

87. DENNIS, W. H. (1967). Foundations of Iron and Steel Metallurgy, London, P-15

 

88. DICKINSON, H. W. and TITLEY, A. (1934). Richard Trevithick, the Engineer and the Man, Cambridge, p. 64.

 

89. DODD, A. H. (1951). The Industrial Revolution in North Wales, Cardiff, pp. 133-152.

(a) p. 134

 

90. ELSAS, M. (1860). Iron in the Making, Dowlais Iron Company Letters 1782-I860, Glamorgan County Record Office, p. vii.

 

91. GREEN, F. (1915) - Carmarthen Tinworks and its Founder. West Wales Historical Records, Vol. V, p. 250

(a) p. 262   (b) p. 245

 

92. GREEN, F. (1806-7). Carmarthenshire Tokens. Trans. Carmarthenshire Antiquarian Soc., Vol. II, p.199.

 

93. GRIFFITHS, L. (1953). From making iron to making saints in the Llynfi Valley. MS. in the Department of Industry, National Museum of Wales.

(a) p. 8.

 

94. HOWELL, J. M. (1926). The Birth and Growth of Aberayron. Cardiganshire Antiquarian Soc. Trans., Vol. IV, p.10.

 

95. HULME, E. W. (1930). The Statistical History of the British Iron Trade from 1717 to I 750. Trans. Newcomen Soc., Vol. IX, p. 2 1.

 

96. JONES, D. W. (1874). Hanes Morgannwg-History of Glamorgan, Aberdare, p. 408

(a) p. 255   (b) p. 416   (c) p. 419   (d) p. 420   (e) p.415

 

97. JONES, O. (1875) ed., Cymru - Wales, historical, topographical and biographical, London, Vol. I, p. 30

 

98. LOUTHERBOURG, de P. J., Iron Foundry near Machynlleth, pen and ink drawing at National Library of Wales, Aberystwyth.

 

99. LEWIS, E. A. (1927), The Welsh Port Books 1550-1603.Cymmrodorion Record Series, No. XII, London.

 

100. LEWIS, W. J. (1950-5 1). Aspects of lead raining in Cardiganshire in the sixteenth and seventeenth centuries. Ceredigion Journal Cardigan Antiquarian Soc., Vol. 1, p. 178.

 

101. LLEWELLIN, W. (1863). Sussex Ironmasters in Glamorganshire. Arch. Camb., Vol. IX, pp. 86-88.

 

102. LLOYD, J. (1906). The Early History of the Old South Wales Iron Works, London, p. iii.

(a) p. v   (b) p. 147   (c) p. 131   (d) p. 129   (e) p. 87   (f) p.72   (g) p. 63   (h) pp. 51, 63, 71   (i) pp. 102, 103

 

103- MERCER, S. (1953). Trevithick and theMerthyrTramroad, Trans. Newcomen Society, Vol. XXVI, pp. 89-103.

 

104. MOTT, R. A. (1961). Coke-making in Wales, Coke Oven Managers Association Bulletin, Vol. XXX, No. 3, p. 6

 

105. MUSHET, D. (1840). Papers on Iran and Steel, London, P. 415.

(a) p.424

 

106. MINCHINTON, W. E. (1957). The British Tinplate Industry, Oxford, p. 38

(a) p. 32

 

107. NORTH, F. J. (1962). Mining for Metals in Wales, Cardiff, p. 42.

 

108. PERCY, J (1864). Metallurgy: Iron and Steel, London, p. 493

(a) p. 350

 

109. PHILLIPS, M. (1932-33). Early Development of Iron and Tinplate Industries in Port Talbot District. Trans. Aberafan and Margam Dist. Hist. Soc, Vol. V, p. 25

(a) p. 30.

 

110. PHILLIPS, D. R. (1925). The History of the Vale of Neath, Swansea, p. 289.

 

111. RILEY, W. (1894-95). Intrenchments and Camps on Mynydd Baidan and Mynydd Margam. T.C.N.S., Vol. XXVII, Pt. II, Cardiff, p. 77

 

112. ROYAL SOCIETY OF ARTS, Guard Books, Vol. 3, No. 119, p. 1.

 

113. SCHUBERT, H. R. (1957). History of the British Iron and Steel Industry from 450 B.C. to A.D. 1775, London, pp. 176-179

(a) p. 371   (b) p. 423  (c) p. 369.

 

114. SCRIVENOR, H. (1841). A Comprehensive History of the Iron Trade, London, pp. 126-127, 294-295.

 

115. SOLDENHOF, R. de (1892-93). Progress made in Coking during the last few years. S.W.I.E. Trans., Vol. XVIII, p. 303

 

116. TYLECOTE, R. F. (1966). Blast furnace at Coed Ithel, Llandogo, Mon. Journal of the Iron and Steel Institute, Vol. 204, pp. 314-319

 

117. WARNER, R. (1813). Second Walk through Wales in August and September, 1798. London, 4th ed., pp. 96‑97­

 

118. WILKINS, C. (Igo

Wales, Merthyr3). History of the Iron, Steel and Tinplate Trades Merthyr Tydfil, p. 14.(a) p. 176 (b) p. 179 (c) P‑ 49 (d) p. 68.

 

119. WILLIAMS, D. G. (c‑ 1930). Hanes Plwyf Llan‑giwg ‑ History of the  parish of the church of Ciwg. MS. in library of University Collef Swansea.

 

120. WILLIAMS, J. (19 15). How to improve Welsh Tinplate Rolling Mill Practice. T.S. W.I.E., Cardiff, Vol. XXI , pp.. 151‑170.

 

121. WILLIAMS, J. G. (1866). A Short Account of the British Encampmntse; lying between the Rivers Rheidol and Llyfnant in the County of Cardigt and their Connection with the Mines, Aberystwyth, p. 18.

 

122. WOOD, C. (1766‑1777) ‑ MS. diary written at Cyfarthfa, microfilm in possession of M. W. Flinn, University of Edinburgh.

 

123  WOOD, J. S. (1813). The Principal Rivers of Wales.