Skip to main content Accessibility help

Roman Gold-Mining in North-West Spain, II: Workings on the Rio Duerna

  • R. F. J. Jones (a1) and D. G. Bird (a1)


Recent studies of Roman gold-mining, in Britain at Dolaucothi, and in Spain at las Médulas, Montefurado and Puerto del Palo, have shown that many valuable results can be obtained at such sites by a careful programme of field-work. The present article is the result of field-work in the region of the Rio Duerna in 1970. It was also possible to undertake a special in-depth survey of one of the mines, the working known as los Castellones. This mine was chosen as one of the most important and interesting in the valley.

Previously attention has centred on the hard-rock mines of the south of Spain and Portugal, and there are few sources referring to the gold-mines of the north in any but very general terms. O. Davies has compiled a list of most of them and thus rendered a valuable service, in spite of some errors. Strangely he does not give the most comprehensive source as his reference for the mines of the Duerna area (which he calls ‘Quintanilla’ on his map IIIa), although he makes use of it elsewhere.



Hide All

1 Lewis, P. R. and Jones, G. D. B., ‘Dolaucothi Gold Mines I: the Surface Evidence’, Ant. J. XLIX (1970), 244 ff., henceforth ‘Lewis and Jones, Dolaucothi’.

2 Lewis, P. R. and Jones, G. D. B., JRS LX (1970), 169 ff., henceforth ‘Lewis and Jones, N.W. Spain I’. Our attention has been drawn to a paper by Domergue, C., ‘Introduction à l'étude des mines d'or du nord-ouest de la péninsule ibérique dans l'antiquité’, in Legio VII Gemina (León, 1970), 255 ff. We were unable to obtain this until after the completion of the present article, but fortunately have found that it has nothing to add.

3 The authors wish to express their gratitude to the following organizations and people: the Spanish Government for the grant of a scholarship to D.G.B. in 1969–70; the Department of Education and Science for extending a grant to allow a trip to Spain in October-November, 1970; Hulme Hall, Manchester University for the grant which made possible the special survey of los Castellones led by R.F.J.J. in April, 1970, and the members of that expedition, Messrs. P. Barber, W. S. Hanson, N. C. Meadmere and J. M. Slack; Dr. P. R. Lewis and Professor G. D. B. Jones for much useful help and comment; and the people of Luyego, especially Sr. D. Generoso Turienzo Alonso and his wife, for their hospitality.

4 Davies, O., Roman Mines in Europe (1935), 102–3.

5 ibid.; the brief mention by Longridge, Mining Journal (henceforth M.J.), 29 Jan., 1898, 139, cannot be compared with the survey by Oriol, Revista Minera, Metalurgica y de Ingenieria (henceforth R.M.M.I.), XLVII, serie C, 14 (1896), 197 ff.; or the information given by Jones, J. A., Trans. Fed. Instit. Mining Eng. XX (19001901), 426–7. Mention should also be made of Gómez-Moreno, M., Catálogo Monumental de España: Provincia de León (Madrid, 1925), texto 98, and of an article in R.M.M.I. (1 Dec, 1850), 387 ff.

6 Roman Asturica Augusta. See Richmond, I. A., ‘Five Town Walls in Hispania Citerior’, JRS XXI (1931), 90–1. For a map of the area, see fig. 2.

7 See below, pp. 73–4.

8 op. cit. 427.

9 Such as a military inspection by Su Alteza Real el Principe Don Juan Carlos de Borbon, which robbed the los Castellones survey of the first three days of its planned work schedule.

10 Mining terms: Lewis and Jones, ‘Dolaucothi’, app. 1, 268–9. Pliny translation: Lewis and Jones, ‘N.W. Spain I’, app. 1, 181–4.

11 ibid.; cf. Pliny, Nat. Hist. XXXIII.

12 See for example Paul, R. W., California Gold: the Beginning of Mining in the Far West (University of Nebraska, 1947), especially 61–2.

13 Luengo, J. M., ‘Explotaciones auriferas romanas en Rabanal del Camino, León’, Archivo Español de Arte y Arqueologia (1935), 287 ff., and personal observation in 1970.

14 See fig. 2.

15 Oriol and Longridge, opp. citt.

16 Throughout this section reference should be made to fig. 3 for the positions of the sites.

17 Arroyo is the Spanish term for a stream.

18 Personal observation in 1970.

19 A small margin of error must be allowed in the measurements of the tanks because of the difficulty of distinguishing on a gentle slope where the floor ended and the bank began.

20 There is a fairly certain tank on the mesa of site VII, and this must add considerably to the probability of G being a tank.

21 At the confluence of the Llamas and the Duerna, and below site XVI.

22 There are similar rock cuts below the workings at site XVI.

23 See below, pp. 70–1.

24 There are other examples of aqueducts discharging directly into a gully after passing through a ninety-degree turn, including one at Puerto del Palo: Lewis and Jones, ‘N.W. Spain I’, 180.

25 Such a system can be seen in the Duerna valley at site X, and also at Dolaucothi in Area I, Lewis and Jones, ‘Dolaucothi’, 266.

26 Other examples in the Duerna valley lend support to this theory. Modern canals from the right bank of the Llamas through site II, and to site XII near Boisan both pass through ancient tanks.

27 See below, pp. 70–1.

28 Forster, W., A Treatise on a Section of the Strata from Newcastle-upon-Tyne to Cross Fell (3rd ed., Newcastle, 1883), 161 ff.

29 There is another example on the Valdedillo, between the Prado and Boisan.

30 Although it is impossible to make a precise calculation of the amount of water that dam 2 could hold, it has been roughly estimated at 2,500,000 litres. This compares with rough figures of 1,380,000 litres for tank A, 2,000,000 for tank B, 1,350,000 for tank C, at least 2,000,000 each for tanks D and F, and 400,000 for tank E. These figures are based on average water depths of 1·50 metres in tanks A, B and C, 2·00 metres in D, E and F, and a maximum depth of 2·50 metres behind dam 2. Too much of dam 1 has been lost for any estimate to be made for it. If the modern tank was filled only to a depth of 1 metre, it would still contain more than 5,000,000 litres.

31 J. A. Jones, op. cit. 426, refers to the Rio Sil and León Mining Company Ltd. (see below, n. 36); and there is also a local tradition of French working, which is not otherwise substantiated.

32 See fig. 9.

33 op. cit.; the other was, of course, los Castellones.

34 ibid.; he probably meant visible from, rather than at, Molinaferrera, as Fucochicos is on the other side of the river from that village.

35 A similar method was used in nineteenth-century California; see R. W. Paul, op. cit. (n. 12), 151–2.

36 op. cit. 426: ‘In 1887 the Rio Sil and León Mining Company Ltd. worked on the Duerna on one of these beds. They found it to consist of 180 feet of poor gravel overlying a rich layer that rested on the bedrock. It was found that the Romans had already worked some of this layer in patches.’

37 Maclaren, J. M., Gold: its Geological Occurrence and Geographical Distribution (London, 1908), 89. See also R. W. Paul, op. cit. 99.

38 Probable aqueducts are known tapping the Llamas, the Espino, the arroyo through Boisan (see pl. III, 3) and the Prado. Three aqueducts from the Prado are proved certain by rock-cut sections (see pl. III, 2). Each of these streams is tapped by modern agricultural canals which run even throughout the summer.

39 See below, p. 74.

40 R.M.M.I. (1 Dec. 1850), 387 ff.

41 Mezquiriz de Catalan, M. A., Terra Sigillata Hispanica (Valencia, 1961), 11, plate 92, especially 1496; note also 1495, 1506 and 1510. Our thanks are due to Joanna Morris for this reference.

42 L'Année Épigraphique (henceforth AE), 1910, 325–6, numbers 1, 2, 3, 4, 5 and 6. The last four had been previously published in CIL ii, 2552, 2553, 2554 and 2556 respectively. AE 1910, 2 was given minor corrections in AE 1966, 188.

43 Ptolemy, Geography ii, 6, 28; Itinerario de Barro: see Diego Santos, F., Epigrafia Romana de Asturias (Oviedo, 1959), 244 ff., and the references given there, or Schulten, Adolf, Los Cántabros y Astures y su Guerra con Roma (Madrid, 1943), 191.

44 AE 1910, 2 = AE, 1966, 188.

45 AE 1967, 229, 230 and 231 (= AE, 1963, 28).

46 For the campaigns see Magie, D., ‘Augustus' war in Spain’, Classical Philology xv (1920), 323 ff.; Syme, R., ‘The Spanish war of Augustus’, American Journal of Philology LV (1934), 293 ff.; A. Schulten, op. cit.; and now also Syme, R., ‘The conquest of north-west Spain’, in Legio VII Gemina (León, 1970), 83 ff., noting and summarizing all the earlier work on the subject, which is undoubtedly the best account of the present state of our knowledge. What has not been adequately emphasized is that the real need now is for intensive fieldwork coupled with aerial surveys, in an area which has received very little effective archaeological work of any sort.

47 See for example Davies, op. cit. 101; Charlesworth, M. P., Trade Routes of the Roman Empire (Cambridge, 1926), 153; and Blásquez, J. M., Emerita xxx (1962), 117, n. 4, and the references given there. Syme, Am. J. Phil, cit., 295, n. 5, is doubtless correct to say that Augustus was concerned with ‘tidying up’ the peninsula, but surely he should not exclude completely the possibility that the mineral profits of conquest were also taken into consideration.

48 Similar deposits to those on the Duerna and further west were worked at some stage by the Romans in what is now the province of Salamanca, at Cavenes del Cabaco (Gómez-Moreno, op. cit. 97). These may have served as a pointer.

49 Florus, II, xxxiii (Loeb translation). Syme, R., ‘Pliny the Procurator’, Harvard Studies in Classical Philology, 73 (1969), 218, notes that the figures given by Pliny for gold production ‘might well be Augustan’, although they could presumably refer to the time when Pliny himself was procurator.

Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The Journal of Roman Studies
  • ISSN: 0075-4358
  • EISSN: 1753-528X
  • URL: /core/journals/journal-of-roman-studies
Please enter your name
Please enter a valid email address
Who would you like to send this to? *


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed