Tresminas and neighboring Jales comprise one of the largest and best-preserved Roman goldmining districts of the early Imperial period. Nestled in a rural corner of northern Portugal, just above the Douro River, the archaeological zone includes underground mining galleries and expansive opencast pits, traces of a settlement and amphitheater, and quarries and aqueducts that sustained large-scale gold mining in the 1st and 2nd c. CE. The spectacular site was designated an Imóvel de Interesse Público, a cultural property of public interest, in 1997 under Portuguese heritage law. With the recent opening of a small interpretive center, the area has become more accessible as a regional tourist attraction. Although the mine is well known in specialized circles, it has not been published extensively until now. This new volume, Roman Gold from Tresminas (Portugal) – Prospection — Mining — Treatment, by Regula Wahl-Clerici (W.-C.), provides a detailed, well-illustrated resource that makes this mining complex accessible to the wider scholarly community. This is a significant contribution because of the site's remarkable historical and archaeological value. Further, the techniques described in the volume differ substantially from those used at other, better-known gold mines and contribute substantially to our knowledge of ancient mining technologies.
The volume joins a robust and growing bibliography on gold mining in northwest Iberia, which took place on a large scale from the time of the Augustan conquest of the territory through the late 2nd or early 3rd c. CE, primarily to supply ore for gold coinage.Footnote 1 The UNESCO World Heritage Site of Las Médulas in León, Spain is the best known of these Iberian mines; it has often been referenced as the type site for gold mining in northwest Iberia because of its unusual preservation and long history of publication.Footnote 2 It can be misleading to consider Las Médulas as typical, however, because the considerable geological diversity of the region necessitated different mining technologies and organizational strategies. Recent advances in archaeological prospection techniques, particularly the use of LiDAR, have expanded our knowledge of gold mining in northwest Iberia well beyond Las Médulas.Footnote 3 Farther south, the scale of gold mining along the Tagus River is also becoming much better understood.Footnote 4 Outside of this archaeological research, the Roman mines of Iberia are increasingly incorporated into wider scholarship on Roman economic history.Footnote 5 Alfred Michael Hirt, for instance, makes ample use of Iberian epigraphic and archaeological evidence, including several mentions of Tresminas, in his monograph on the organization of Roman imperial mines and quarries.Footnote 6
This is the first volume in an anticipated series covering different aspects of the territorium metallorum Tresminas/Jales, the term W.-C. gives to the territory encompassed by the Roman mining district or metalla. The book is the product of decades-long engagement with the site: the author and the late Jürgen Wahl first began fieldwork at Tresminas in 1985, sponsored by the German Archaeological Institute and the German Mining Museum in Bochum. The project has continued into the present in various iterations (18–19). Jürgen Wahl's Reference Wahl1988 preliminary report and subsequent guidebook have remained the most useful site-specific resources for the mines themselves, but these can now be replaced with this comprehensive volume.Footnote 7 Following the introduction (Chapter 1), which provides a historical overview and a summary of the site's research history, the book is laid out in five chapters. The core chapters systematically take the reader through the chaîne opératoire of mining from prospection (Chapter 2), to extraction (Chapter 3), to treatment (Chapter 4), while Chapters 5 and 6 deal with supporting infrastructure.
In Chapter 2, W.-C. outlines Roman methods for locating and determining the extent of ore deposits. Most mines were opened in places with pre-existing mining traditions, which left behind traces that the Romans used to find ore deposits. The Roman-era techniques employed at Tresminas, however, have no clear Iron Age precedent. While the author concedes that inhabitants of the nearby Iron Age castro, Cidadelhã, were almost certainly aware of the locally available ore, their exploitation of it was probably as technologically simple as panning for gold along the riverbanks (22). Additional Roman prospecting was therefore necessary to locate the deposits, though W.-C. makes the important point that prospecting is only preserved or easily identified in places deemed unsuitable for working. She documents, for instance, small-scale surface sampling, or “pocket-like recesses,” cut into the rock across the territory (22). These interventions represent prospecting with unsatisfactory yields. She suggests that the larger mining systems at Corta de Covas and Corta da Ribeirinha, by contrast, were initially opened for prospecting at multiple subterranean elevations, and continued to be mined when found profitable. This chapter is useful, particularly considering discussions of prospection elsewhere are usually cursory or heavily reliant on Roman textual descriptions.Footnote 8
The second part of the chapter details evidence for a disaster in an underground passageway, Galeria dos Alargamentos, when a stairwell was opened to prospect below the original gallery. This intervention destabilized natural rock fissures, and the stairway collapsed, blocking the passage below. The fallen rockface was subsequently modified with holes for placing oil lamps and wooden beams, which happened, W.-C. convincingly argues, following the incident, during efforts to rescue miners and/or retrieve equipment. The author's careful recording, which traces the relationship of the prospection efforts to the original gallery as well as each stage of this disaster in detail, is extremely rare in mining archaeology, where attempts to date mines within less than a century are uncommon. This section is a prime example of micro-histories that can be told through the detailed phasing and documentation efforts of the author and her collaborators.
The heart of the book is Chapter 3, which traces and illustrates the evidence for gold exploitation at the major sites of the mining district in an expansive 130 pages. Tresminas is so named for its three distinct mining zones: Corta de Covas, Corta da Ribeirinha, and Lagoinhos. All were once thought to be opencast – or open air – mines. W.-C. shows, however, that Lagoinhos was actually an underground mine and only appears to be a pit mine because of a later collapse – an important demonstration of how a mixture of extraction methods were used simultaneously. Roman exploitation of the nearby Campo de Jales and Gralheira deposits is no longer intact because of later 20th-c. exploitation, but each is described briefly. A section on extraction techniques is particularly useful because it points out differences between those used at Tresminas and at Las Médulas, where secondary gold deposits were mined using hydraulic methods. Larger underground galleries at Tresminas preserve cart tracks, for instance, attesting to the use of wheeled vehicles, and perhaps pack animals, alongside human labor. Some galleries also retain their dome shape, characteristic of fire-setting, which involved heating a rockface with open flames and dousing it with water to make it more friable prior to manual extraction of the ore.Footnote 9
Next, Chapter 3 details the organization of production in the largest of the extraction sites, Corta de Covas. W.-C. describes the vast opencast mine with four underground galleries incorporated into its lateral edges, once used for prospection, water evacuation, and ore removal. As with the section about the disaster, the detail included here is considerable and gives a remarkable sense of the gradual development of the mine and the foresight necessary for efforts of this scale. Precise construction time estimates are given for each gallery (e.g., the earliest, Galeria Esteves Pinto, took ca. 5.5 years to complete). Especially helpful are elevation drawings that reconstruct the mine in each phase, as the opencast pit grew and deeper galleries were added. The chapter also draws attention to technical experimentation: in the Galeria do Pilar, two wide platforms for capstans were excavated, only one of which was completed and used. W.-C. suggests that the device helped to transport water or other material out of the gallery. If true, this adds a unique example of mechanical water removal in Iberian mines, alongside waterwheels operated by treadmills, Archimedean screws, and bucket chains.Footnote 10
This section is followed by detailed descriptions of each of the underground galleries associated with Corta de Covas and Corta da Ribeirinha. The range of different construction techniques that W.-C. demonstrates is remarkable. In some cases, a qanat system – the same as that used for underground water transport in other contexts (e.g., 6th-c. BCE Samos) – was used to advance galleries at their intended orientation. This involved sinking a series of vertical shafts and then connecting their deepest levels with a horizontal gallery. The Galeria dos Alargamentos, by contrast, was excavated from both ends to meet precisely in the middle – an impressive engineering feat. Sometimes small galleries, or “pilot drift[s]” (160), were opened first to mark out their intended extension and only widened later, while at other times galleries were opened from the outset at their full height and width. Finally, the chapter also documents waste-rock dumps near extraction zones, which W.-C. uses to trace prospecting and processing as well as to estimate the volume of material excavated in places where the extent of the mines is unknown.
Chapter 4 describes the next phase of mining: treatment of the ore to isolate and purify the gold. At Las Médulas, gold was sorted from waste material and collected as nuggets without further processing. The deposits at Tresminas, Gralheira, and Jales all required more complex treatment, because the gold was intermixed with other materials into hard rock deposits. The various stages of treatment, including dry mechanical processing, wet processing, and smelting, are all attested at the site. Dry processing involved breaking up the extracted rock, either by tossing it down a shaft or by using a mechanical stamp mill. The next step consisted of grinding the ore with a two-piece round mill. The stone elements of these mills are preserved in astonishing numbers across the landscape and were often reused (e.g., in the Roman town, to construct a support in one of the galleries [Galeria do Pilar], and in construction of later periods). W.-C. touches upon this reuse activity, but this topic deserves further integration with scholarship on reuse and recycling.Footnote 11 Wet processing involved washing ground material to separate out gold from waste. This is attested on a hillside at Tresminas with two parallel rows of platforms, which the author hypothesizes were fitted with wooden washing tables. Finally, the gold was smelted to remove impurities. All three parts of the process are attested at Forno dos Mouros, associated with the Gralheira deposit. W.-C. details this spot and identifies other processing areas in the Tresminas extraction zone by tracing concentrations of stone mills, waste, and slag.
The final section of the chapter is dedicated to tracing the gold contents in the three deposits, with the goal of explaining why the mines were abandoned. Here, W-C. cites decreasing gold yields or shifts in production at the end of the 2nd c. CE as possible causes for abandonment (230). She assembles previous studies that calculate the gold content per ton of material in each of the major deposits, the fineness of the gold itself, and the volume of extraction. From this data, she produces a new estimate that the mining district could have yielded 10 million tons of extracted material with a ratio of ore to overburden of 1:10 for a total of 10,000 kg of raw gold during its two centuries of operation. She hypothesizes an annual production of between 6,100 and 8,200 aurei (233). These estimates are substantially lower than previous calculations based on incomplete data from the mines. For instance, earlier estimates of 20 million tons of extracted material misunderstood areas of rockfall to be opencast mining.Footnote 12 These new numbers must therefore be more accurate, as more is known about the size of the mines. While believable, the math is impossible to verify given that all samples of gold were taken from areas of the mine that were abandoned and the deposits deemed valuable enough for Roman mining are now gone. In the end, the specific historical question of why mining ceased at Tresminas is not (and cannot be) answered definitively.
Finally, Chapters 5 and 6 deal with water infrastructure and quarries, providing detail about the diverse interventions necessary to support both mining and the town. The water infrastructure around Las Médulas has been extensively published,Footnote 13 so a chapter on this subject in the volume allows for detailed comparison across these two gold mines. At Tresminas, there is no evidence for ruina montium, the hydraulic mining technique described by Pliny the Elder (HN 33.21.66) and attested at Las Médulas. Other types of water management and hydraulic techniques were necessary, including water transport for fire-setting and for the ore washeries, to power mills, and to supply the community's needs. Tresminas had a basin associated with the settlement and two dams, which were linked to the nearby streams of Tinhela, Curros, and Peliteira. Additionally, the chapter traces 12 aqueducts that carried water from these sources to the mining zone. This section is of particular value, as it adds a set of rural case studies to the well-published urban aqueducts of Hispania.Footnote 14 W.-C. meticulously documents the basins and aqueducts, as well as infrastructure (e.g., channels cut into mining galleries, runoff using natural slopes, and sewage channels in the settlement) for disposing of wastewater produced through mining, crafts, and human use.
In Chapter 6, W.-C. turns to quarrying, which provided local granite, schist, and loamy earth for mining needs. She documents granite exploitation throughout the zone and differentiates between the specific uses of stone from each geological source. Biotite granite, the hardest local stone, was used for stamp-mill bases and ore mills. The softer two-mica granite was purposed for grain mills, epitaphs and other inscriptions, and construction. Shist was also quarried near the settlement of Tresminas, in two distinct areas, and employed in dwellings, the amphitheater, the water tank for domestic use, and the ore-washing facilities. The chapter also documents extraction pits for soil, which was needed for dam construction and to seal water pipes. By examining the exploitation areas in concert with construction, W.-C. convincingly argues that local needs were met through on-site exploitation.
One of the most exceptional parts of the book is the diversity and quality of its illustration program. The written chapters are enhanced by maps, full-color traditional and aerial photography, line-drawings, artistic reconstructions, and images produced through 3D-laser scanning. The 3D images, by Markus Helfert, Klaus Mechelke, and Maren Lindstaedt, are innovative to-scale black-and-white section images that detail connections between multi-level mining galleries. Underground mines of this complexity are difficult to visualize and understand, but the variety of illustrations brings the reader as close as possible to doing so without an in-person visit.
One of the difficulties of studying Roman mining is that site reports tend to be published in local and regional journals or books without wide international distribution. Mines of the Roman world also have not been integrated well with studies of mining from a global perspective, both because of these accessibility issues and because of disciplinary divisions separating classical and anthropological archaeology.Footnote 15 In light of this, a great asset of this volume is its accessibility. The book is downloadable for free on the publisher's website as a PDF, which makes the contents searchable for research use. (As there is no index, the digital version is best for scholarly research.) Nothing is lost in quality in the print version. The large-format hardback book has the same spectacular, full-color image program as the digital version. Jargon is kept to a minimum and the text is precise, which is important given the site's complexity and its potential interest to non-specialists. All ancient texts are provided in the original Latin or Greek and accompanied by the English Loeb translations. Given these features, the book will be approachable for both the serious Roman mining scholar and the generalist with interests in ancient technology or Roman provincial archaeology.
Still, the book is the beginning rather than the culmination of what can be done with a site this rich. Chapter 1 provides only cursory historical background, and there is no dedicated conclusion, so there is ample room for contextualization and analysis of the data presented. W.-C. poses intriguing historical questions but does not evaluate them in detail. For instance, in describing the mining disaster, she wonders whether the miners, the procurator, or other government officials oversaw disaster relief in imperial mines and quarries (40). This fascinating topic could lead to a much larger discussion. In another instance, the author suggests that the quarries, like the gold mines, belonged to the Roman state. She hypothesizes that once stone had served its use (i.e., when mills became worn), its ownership was transferred for private use. This would explain the extensive use of mill stones as spolia at Tresminas (274–75). Ownership of refuse from imperial mines and quarries is another topic deserving of solid treatment in the future. Overall, the text is largely descriptive and focused on the technological aspects of mining. Future desiderata include detailed presentation of the habitation, funerary, and epigraphic evidence, as well as a catalogue of the objects from the site to help highlight social and economic aspects of life at Tresminas. These are hardly critiques, though, as the book provides what it promises: a meticulous site report covering all aspects of prospection, mining, and treatment in rare detail. The volume will become a staple for anyone working on Roman mining.
Competing interests
The author declares none.
