An isotopic study shows that the projectiles found in Switzerland’s Surses Valley came from Hispanic mines, while those discovered farther north, at Lake Constance, have a different origin, calling into question the route taken by Augustus’ legions.
In the summer of 15 BC, Emperor Augustus’ two stepsons, Tiberius and Drusus, brought the Central Alps and the Swiss Plateau under Roman control. That campaign concluded decades of efforts to link the Gallic provinces with Illyricum and secure the routes between northern Italy and the Danube. However, while the western and eastern routes of that offensive are well documented in ancient sources, the central route, which would have crossed the Grisons passes, is barely mentioned in the texts.
Until 2002, there was no archaeological evidence of a major military operation in that central corridor. But over the past twenty-five years, archaeologists have uncovered a genuine conflict landscape in the Surses Valley, in the Swiss canton of Graubünden. There, near the Crap Ses gorge, they have found the remains of a marching camp, a battlefield, an observation post, and a small fort, along with thousands of military objects: swords, shields, hobnails, coins, and, above all, nearly five hundred lead sling bullets.
These projectiles, known in Latin as glandes, have an exceptional feature: most of them bear the name of one of three Roman legions, stamped with a punch or cast into the mold: the Third, the Tenth, or the Twelfth. Until now, the practice of stamping the name of a legion onto lead projectiles was almost unknown in the Roman Empire, highlighting the uniqueness of these discoveries.
The Twelfth Legion, identified by the symbol of a bundle of lightning bolts, is the same one that had fought against the Parthians in the Near East in 20 BC. Meanwhile, the coins recovered from the battlefield, more than eighty in total, date the events to after 28 BC, which fits Augustus’ Alpine campaign, although the researchers acknowledge that they could also be linked to a later punitive expedition not mentioned in the sources.
A Supply Problem
The lead in those projectiles was not only used to kill, but has now become a tool for understanding Roman logistics. The authors of the study, led by Christopher D. Standish and colleagues from the Universities of Southampton and Basel, posed several questions: where did the lead used to manufacture those bullets come from?
Map showing the routes of the Alpine Campaigns, with the two main thrust axes of Tiberius (red arrows) and Drusus (turquoise arrows) alongside the central thrust axis (yellow arrows). Credit: Stefan Scherrer, Archäologischer Dienst Graubünden
Did the projectiles found at Crap Ses and those recovered farther north, at the Roman settlements of Brigantium (modern Bregenz, Austria) and Tasgetium (Eschenz, Switzerland), both on the shores of Lake Constance, share the same origin? If the lead was the same, that would support the hypothesis that the same troops advanced northward from the Surses Valley. If it was different, it could indicate that the northern projectiles belonged to another unit or another period.
To answer this, the scientists turned to lead isotope analysis. This technique works like a geochemical fingerprint: each mining region has a characteristic ratio of the different lead isotopes (atomic variants), which is preserved in the extracted metal. By analyzing those ratios in artifacts, it is possible to trace their origin.
The team sampled the cores, rather than the corroded surfaces, of 47 sling bullets, two lead tent weights, and two pieces of shapeless lead. Forty of the projectiles came from Crap Ses, while seven came from Brigantium and Tasgetium. After dissolving small amounts of lead in acid and measuring their isotopes with a high-precision mass spectrometer, the results were clear.
Two Groups, Two Origins
The data reveal two clearly distinct groups. On the one hand, the projectiles from Brigantium and Tasgetium (Group 1) have a homogeneous isotopic composition with less radiogenic values. On the other hand, the projectiles from Crap Ses (Group 2) display a more variable but still clustered composition, with more radiogenic values. Statistical analysis confirms that the difference between the two groups is highly significant: 95.7% of the variability is due to differences between the groups, and only 4.3% to variation within them.
Beyond the isotopes, there are other physical differences. The Group 1 projectiles are heavier, averaging 80 grams, and lack legionary inscriptions. Those in Group 2 are lighter, averaging about 50 grams, and bear the name of the Third, Tenth, or Twelfth Legion.
This indicates that they were manufactured using different strategies: one workshop produced large, unmarked projectiles for general supply, while another produced smaller projectiles stamped with the legion’s emblem for a specific contingent. More importantly, within each group, the lead composition is consistent, pointing to centralized production from a homogeneous batch of metal.
Where Did the Lead Come From?
To determine the source of the lead, the researchers compared the isotopic signatures with those of the principal mines known to have been active in the Roman world during the first century BC. They quickly ruled out Italian sources, such as Tuscany, because there is no archaeological evidence that lead mines were being worked there at the time. Moreover, Pliny the Elder mentions a Senate decree prohibiting mining within Italy itself. The deposits of the Aegean, Turkey, Egypt, Britain, Germania, and the Balkans also do not fit, either because they were not being exploited at the time or because their isotopic signatures do not match.
The most likely origin lies in the Iberian Peninsula and, to a lesser extent, in Gaul (modern France). The Group 2 projectiles from Crap Ses have a signature that matches the Almería-Cartagena volcanic belt in southeastern Spain. That region had been an important mining center since the third century BC, and the Cabezó del Pino mine, in particular, provides an almost exact match.
But the data also show that this lead was not pure: it contains a mixture with another type of lead, probably originating in the Sierra Morena region of south-central Spain or from the Iberian Pyrite Belt in Huelva.
As for Group 1, from Brigantium and Tasgetium, its signature matches perfectly with lead from Sierra Morena, the Iberian Pyrite Belt, or, alternatively, with certain deposits in France’s Massif Central, such as those in the Rodde area of Haute-Loire. The authors note that although Gallic mines were being worked at the time, their isotopic signatures overlap with those of Iberian mines, so they cannot completely rule out a Gallic origin for Group 1. Nevertheless, the historical and archaeological evidence favors Iberia, since the Spanish mines were the most active and productive in the western Mediterranean during that period.
What Does This Mean for History?
The most striking conclusion is that the projectiles from Crap Ses and those from Brigantium/Tasgetium do not share the same lead source. This has direct implications for the march route hypothesis. If the same legions that fought in the Surses Valley had continued northward to Lake Constance, it would be logical to expect them to have carried projectiles made from the same batch of lead.
Examples of the lead objects investigated in this study. Credit: Amt für Archäologie Thurgau / Jakob Näf / Julia Rabitsch / Peter Thomas / Stefan Scherrer, Archäologischer Dienst Graubünden
But that is not the case. As the authors state in the article: The contrasting isotopic compositions of projectiles from Brigantium/Tasgetium and Crap Ses do not provide clear evidence in support of the hypothesis that Roman military action at Brigantium and Tasgetium relates to later activity by the same vexillation involved in events at Crap Ses (and the Septimer Pass).
In other words, the data do not support the idea that the northern projectiles belonged to the same advance force as those from the south. What explanations do the researchers consider? One possibility is that a different vexillation operated at Brigantium and Tasgetium, perhaps the one advancing along the western route under Tiberius’ command.
Another is that the same unit, after exhausting its projectiles in the battle at Crap Ses, received a new shipment from a different supplier. Whatever the case, the isotopes do not allow a continuous route to be traced.
Other Objects and Supporting Finds
The study also analyzed two lead tent weights and two shapeless pieces of lead. The tent weights, which served as counterweights for Roman military tents, have the same isotopic signature as the Group 2 projectiles, reinforcing the idea that all of that unit’s lead equipment came from the same centralized supply.
By contrast, one of the shapeless lead fragments, found at Filzbach, north of the Alps, has a signature similar to Group 1, suggesting that it may be related to Roman activity in the Lake Constance region, although it cannot be confirmed that it belonged to the exact same batch.
The authors also compare their results with previous analyses carried out in the same area. Data from Klingler (2008) and from Rageth and colleagues (2010) match those of Group 2 perfectly, confirming that the bullets from Crap Ses and the Septimer Pass came from the same production.
They also note that other projectiles from the same period found in Belgium, Germany, and Iberia display similar signatures, suggesting that Spanish and Gallic lead supplied much of the Roman army during the final third of the first century BC.
A Window into Roman Logistics
This work opens a new avenue for understanding how the supply of combat equipment was organized in the Roman Empire. The rarity of projectiles stamped with the legion’s name, together with the isotopic uniformity of the batches, points to centralized, carefully planned production for a specific campaign.
The authors point out that all the lead required to manufacture the 500 projectiles discovered so far weighs less than a single 33-kilogram Roman ingot, suggesting that they were cast specifically for that operation, while each legion used its own mold and punches on the same metal.
Questions remain unanswered. Why are the northern projectiles larger and uninscribed? Were they manufactured in a different workshop, perhaps for another unit, or simply for a different tactical purpose? Could future analysis of projectiles associated with Tiberius’ western route resolve the mystery?
The research team encourages further work along these lines: comparing the signatures of bullets found in the Rhône Valley or at the Great St. Bernard Pass with those from this study could confirm whether there were indeed two distinct supply networks serving the two main routes.
For now, what is clear is that Roman sling lead tells a story that the ancient texts left untold. Geochemistry thus becomes an ally of archaeology in unraveling the movements of the legions, their supply chains, and the logistical decisions made by Augustus’ generals during the conquest of the Alps.
Above all, it demonstrates that even a projectile lost in a mountain pass can contain precise information about the origin of its metal, the organization of the army, and the routes traveled by soldiers two thousand years ago.
SOURCES
Christopher D. Standish, Hannes Flück, et al., Supplying the Augustean conquest of the Alps — investigating the origins of Roman slingshot projectiles by lead isotope analysis. Journal of Archaeological Science: Reports, vol.74, October 2026. doi.org/10.1016/j.jasrep.2026.105939