Welsh Mines Society, 10/11 June 2000, Lampeter

field notes

Dolaucothi - Ogofau Gold Mine

Today the central part of the mine, the Ogofau Pit, is dominated by 20th century structures which played no part in the working of this mine. The head frame, winder, compressor house and associated structures were brought into the site in the late 1980s from Olwyn Goch near Halkyn in north Wales when that mine closed. Their use at Ogofau was to provide a focus for visitors in the interpretation of gold mining on the site in the 20th century. Whilst late 19th and 20th century working did have an impact, the Ogofau site is primarily a Roman or earlier gold mining site of considerable significance. Its standing is comparable with those in Romania and the late Iron Age gold sites of Limousin in central France.

Despite its importance surprisingly little is known about the operation of the mine in the Roman period. In fact it was only in recent years that the workings were positively identified as being from the period of Roman occupation. Work by Lewis and Jones in the 1960s highlighted the magnitude of the workings and mapped the extensive watercourses associated with the site but presumptions were made on dating based on artifacts found close to the site, pottery finds and the proximity of the fort at Pumsaint.(1) One of the watercourses is now known to have been long abandoned, and largely filled with debris, by c.AD 800. The ore processing area adjoining Carreg Pumsaint has also now been dated to the early Roman period and is possibly contemporary with the drainage wheel found under the Ogofau Pit in the 1930s, the timber of which has a date of 90 ± 70 BC.(2) However the purpose of the watercourses and the way that water was employed on the site is not yet fully understood.(3)

It is however clear from recent work carried out by French archaeologists, including Béatrice Cauuet (Research director with CRNS), that water, from the tanks fed by the leats on Allt Cwmhenog, was not used for extractive purposes. An initial survey of the site suggests that when water was brought into the mine the opencast workings were already in place. Comparison with similar features in the gold mines of Limousin, abandoned on the arrival of the Romans, also suggests that some of the opencast working at Ogofau might be preRoman.(4) This is supported by a circa 6th century BC dating for material found beneath spoil near Carreg Pumpsaint.(5)

The gold mineralisation exploited at Dolaucothi is found within a network of thin discontinuous veins with occasional lenses of gold bearing quartz and no dominant vein or lode is in evidence.(6) Along the north-western facing slopes of Allt Ogofau and Allt Cwmhenog are a series of pits and openworks extending for 500 metres on either side of the Ogofau Pit at NGR SN663402 all of which are pre 19th century. Some of these pits and openworks have been continued underground. In the case of the Ogofau Pit, working was continued below the water table under the openworks. On Allt Ogofau are two large cross section adits, driven south-east with little vertical or lateral separation. These are commonly referred to as the Upper and Lower Roman Adits although all that can be said at present is that they are pre 19th century. During the late 19th and early 20th century some of the earlier workings were extended, and at least three new adits were driven under those workings. In the mid 1930s the sinking of New Shaft was resumed to a depth of 480 feet (144 metres) and levels driven at 100, 160, 260, 360 and 460 feet below collar to work the mineralisation under the Ogofau Pit. The mine ceased production in 1938 having milled 16,862 tons of ore yielding £11,106 of gold in its final brief, but most productive, period of modern working.(7)

Ogofau is part of the Dolaucothi Estate owned by the National Trust and they have developed the site as a visitor attraction. Access to the underground workings is subject to a Crown lease, as the minerals worked are a royal prerogative, and that lease is now held by the Trust. Prior to 1999 the lease was held by the University of Wales Cardiff who had rehabilitated sections of the mine as a field training centre. The university will continue to use the site for educational purposes under the new leasing arrangement although priority will be given to the site's archaeological importance.

Archaeological work carried on outside the immediate area of the mine also has potential for improving understanding of mining during the Roman period. Dr Barry Burnham, of the University of Wales Lampeter, has already excavated sections of the fort at Pumsaint and future work in that direction may provide positive dated links with the organisation of gold mining.

The Llanddewibrefi mines: a brief outline of their history.

From the late medieval period through to the 19th century Llanddewibrefi formed part of the estates of the bishop of St Davids. Lead was being worked there by at least the 14th century although it is not yet possible to identify which of the small mines in the area was active at that period. Lead mines within the bishop's forest were documented in 1326, at which time the forest area probably included all the high ground in the parish east of the Teifi.(8) By the 17th century Fforest yr Escob, Crynnenyth or the four forests of Treminits, was confined to the eastern townships of the parish between the Pysgotwr and the Camddwr in the Towy catchment.(9) Grants of mining rights on the bishop's estate, including the forest, are documented from the mid 18th century onwards. Rhydtalog Mine, on the eastern boundary of the forest, was probably worked from 1754 until around 1765 in conjunction with the minerals under Rhywhalog, in Caron parish, to the east of the Camddwr.(10) That mine was certainly abandoned by about 1770 when a waterwheel was removed to Rhysgog, although there was renewed interest in 1785/6 and the site was tried again in the 19th century.(11)


On the north bank of the Pysgotwr at NGR SN745509 but only accessible by road from the Teifi valley, this is a remote mine site located within the post 17th century bishop's forest.

In terms of its working history and production this mine is in no way spectacular. It is reputed to have been an old mine in 1860 when reopened by the Brynambor Lead Mining Company. The outcrop of the lode is cut at the confluence of two streams and mineralisation might therefore have been noted at an early date but Hall notes no reference to the 'old man's workings' in the Mining Journal reports.(12) A series of limited liability companies attempted to work the mine until circa 1874 with a recorded production of only 10 tons of dressed lead ore.(13)

What is of interest at Brynambor is the remains of the pumping and winding machinery lying on site as abandoned in the mid 1870s. The mine is so remote that the scrap merchants cannot have found it economic to remove the metalwork. Time has taken its toll and virtually all the woodwork has gone and the winding drum housing has collapsed, leaving the drum frame in the stream. Some of the iron flat rods have been robbed for fencing posts but a substantial section remains close to the engine shaft.

Rhysgog, or Rhyscog, Mine

This is certainly an ancient working, how ancient is unclear but probably pre-gunpowder in its origins as adits are reported 'which have been wrought with extreme care, and at a costly rate, being apparently cut down with chisels' and the mine has been linked to the 16th century entrepreneur Hugh Myddelton.(14) However, claims that it was the Bronflwyd (Bronfloyd) Mine are rather far fetched.(15) The mine was active from at least 1752, probably into the 1770s, and in about 1810 it was described as 'an old work' although 'nothing has been done at it these thirty years and then it was worked by Thomas Johnes Esq.'(16) At this period, Rhysgog and other mines in the area, Esgair Gadfach and Cwm Trinant, lay within the commonland of the bishop's manor. One or more of them, or a mine within the forest, was still producing a small amount of lead ore in 1784 when the bishop received £2 11s. 2½d. in duty but interest in the mines was not really renewed until the 1830s; no doubt stimulated by rises in the price of lead.(17)

A partnership led by Oliver Lloyd (probably the solicitor of Cardigan who held a lease of the Llanfyrnach Mine in Pembrokeshire) worked Rhysgog from around 1830 and are credited with the construction of a leat, 9 miles in length, from Llyn Berwyn.(18) The leat enters the site along the south side of Cwm Brefi to feed a small dam high above, and to the east of, the mine. Operations lasted ten to fifteen years. In May 1847 the bishop's land agent served a notice to quit on the mine. By this time one partner was already bankrupt and Lloyd had died but his brother-in-law maintain some interest in the mines until the following year. By that time John Taylor & Co. had also taken an interest in Llanddewibrefi and were then granted a licence to work the mines. Taylor's tenancy did not last long and they subsequently removed all the machinery from Rhysgog.(19) Thereafter a succession of limited liability companies took the Llanddewibrefi mines on lease.(20) Some did a small amount of work at Rhysgog and the new machinery was installed by 1855 but there was no recorded production. The lease Rhysgog and all other mines in the the parish of Landdewibrefi, excluding those in the bishop's forest, was held from December 1860 by Messrs Bevan and Marsden. They were no doubt behind the companies launch to work the mines after that date. By 1878 so little work had been done and little or no ore produced that the bishop's agent had to take action to recover the mine as the lessees were in breach of their agreement.(21)

The workings at Rhysgog are centred on the point where the productve vein crosses Nant Rhysgog, one kilometre south-east of Rhysgog Isaf, at NGR SN680537. The adit, referred to as deep adit in 1855, commences as a crosscut south from the stream to Engine Shaft where it turns east along the vein. East of the stream is another working shaft and two air shafts on the adit. New Shaft at the eastern extremity of the workings does not appear to connect with adit. About 150 metres west of Engine Shaft is a trial crosscut driven south. There is also a new deep adit, commenced in 1861, in Cwm Brefi 600 metres north-east of the main workings, at NGR SN685541.


Other workings around Rhysgog

To the west of Rhysgog there are two groups of workings which, in the 19th century, were generally included in the lease for that mine. Esgair Gadfach, with workings on either side of the mountain road to Farmers at NGR SN667538, is reported to have been discovered in the 1770s but is probably much older.(22) By the 1860s the workings here comprised an adit and at least four shafts, one of which was sunk on the underlie of a lode. Nothing much appears to have done here after that date and at least one shaft had been filled in by 1878.(23)

At Cwm Robert, also known as Cwm Trinant, on the hillside immediately east of the Teifi at about SN648533 there were three adits by the second half of the 19th century, with a winze connecting the lower two. This mine is reported to have been discovered around 1800 and worked by Saunders of Porth y Berllan and Humphreys of Ivybush Inn in Carmarthen who abandoned it without a fair trial. Oliver Lloyd and his partner, John Davies, attempted to launch the mine as a separate enterprise in the 1830s or 40s. Work was carried out on one of the adits here after 1861 but had fallen in by 1878.(24)

In addition to these there are 19th century workings at 'Cwm Brefi Mine' (at about SN695541) including prospects on two or three 'lodes' which cross the valley in that area.(25) Trials were made at Cwm Dewlas, a short distance to the east, between circa 1861 and 1878.(26)

Prospects for future research

There is scope for some detailed research into the history of mining in this part of mid-Wales. Our knowledge of mining in the area is currently weighted towards the latter part of the 19th century and there is undoubtedly a wealth of documentary material available for earlier periods should researchers wish to investigate further. During the late medieval and early modern periods the small mines of Llanddewibrefi were but part of a wider industry producing lead, and probably silver, both for local consumption and for export to a wider market. They did not have the high output of mines in the modern industrial period but produced only what was required to compensate for losses in a society geared to the recycling of metal resources.

One aspect of metal production not touched upon above is the processing of lead ores, in particular lead smelting. Prior to the 17th century ores would be processed close to the mining sites. Smelting sites undoubtedly exist in the area but, as far as the author is aware, none have been identified. In this, and other aspects of early mining, archaeological investigation is likely to provide a significant contribution. A survey of part of the area, the post 17th forest, is already in hand and hopefully others will follow.

References and notes.

  1. Lewis, P R. and Jones, G D B. 'The Dolaucothi Gold Mines I: the surface evidence', The Antiquaries Journal, XLIX, 1969, 244-272.
  2. Burnham, Barry C. 'Roman Mining at Dolaucothi: the implications of the 1991-3 excavations near Carreg Pumsaint', Britannia, XXVIII, (1997), 323-336.
  3. By the time WMS visit the site in June more information may be available on the sequence and purpose of the application of water to gold working on this site. A team of French archaeologists with considerable expertise on gold working in Limousin and Romania are currently carrying out a two weeks evaluation of the site (9 April 2000).
  4. For details of the Limousin workings, see Cauuet, B. 'Celtic Gold Mines in West Central Gaul', in Morteani and Northover (eds.) Prehistoric Gold in Europe, (1995), pp. 219-240.
  5. Burnham, op cit., p. 332.
  6. Annels, Alwyn E. and Burnham, Barry C. The Dolaucothi Gold Mines, 3rd. edn., 1995, 16.
  7. For details of the workings see Annels and Burnham, op cit. and Hall, G W. Metal Mines of Southern Wales, 2nd edn., 1993, xii-xiii and 39-48.
  8. Willis-Bund, J W. (ed.) The Black Book of St David's, London, 1902. Given the presence in the forest of the mill at which the tenants made suit, its western boundary in 1326 must have been close to the vill of Llanddewibrefi. The area was also large enough to support the members of the eight gwelyau and the 240 cows within the forest.
  9. Speed, Map of Cardiganshire, 1640. National Library of Wales (NLW), Morgan Richardson 2043.
  10. NLW Lucas 2810-2. NLW, Calendar of Crosswood Deeds, 214.
  11. Hall, op cit., 71, quoting Mining Journal (MJ) 1870, 100. Pembrokeshire Record Office DX/99/13.
  12. Hall, op cit., 71-73.
  13. Burt et al. The Mines of Cardiganshire, 1985, 8.
  14. Cambrian Journal 3 (1856), 103. Smyth,Warrington. 'On the Mining Districts of Cardiganshire and Montgomeryshire', Mems. Geol. Surv., Vol. 2, Pt. 2 (1848), 681.
  15. MJ 1843, 353.
  16. NLW Tyllwyd 243; BRS 2196. Meyrick, S R. History and Antiquities of the County of Cardigan, 1907, 108.
  17. NLW Gilbertson schedule, 2. A payment in August 1783 of £10 19s. 9d. in respect of duty on 14 tons 13 cwts of ore, receipted by John Warren, bishop of Bangor, to Rev. Isaac Williams (as in 1784) may also relate to the Llanddewibrefi mines but they are not specified.
  18. NLW Druid Inn 57.
  19. NLW Lucas 1992, Harvey to Burder (the bishop's secretary) 10 May 1847; Harvey to Taylor 10 Sept. 1847; Harvey to Messrs. Bolton and Co. 24 Jan. 1849. Lucas 2118.
  20. See Hall, op cit., 70-71 for details of those companies.
  21. NLW Lucas 2117
  22. Meyrick, op cit., 108.
  23. NLW Lucas 2341 and 2117.
  24. Meyrick, op cit., 108. NLW Lucas 2341 and 2117. Hall, op cit., share certificate illustrated on page 76.
  25. Hall, op cit., 71.
  26. NLW Lucas 2341 and 2117.


Peter Claughton / Dept. of History
Last modified 10 December 2000