Trial digs at Crymych point to a Roman auxiliary fort in north Pembrokeshire, its finder says

The case is cumulative, and each piece of it is circumstantial. The earthworks have the playing-card outline of an auxiliary fort and cover about three hectares (7.4 acres), a size the author describes as typical for this type of installation ^[5]. A leat enters the south of the enclosure through a masonry channel, and the author notes that "Every Roman fort needed a water supply" ^[6]. The A478 runs straight beside the site, though the line is lost in places ^[6].

The author is open about the resistance. "Like Fenton, my observations have attracted a good deal of skepticism," the author wrote ^[14]. Richard Fenton first identified the nearby villa in the early 19th century, and his reliability has been questioned, mostly because of a long-held perception that the Romans did not reach west Wales ^[1]. Two earlier results cut against that perception without depending on the author's road theory. A stretch of Roman road in the vale of Whitland was excavated in 1995 ahead of a bypass ^[3], and a fort turned up at Wiston near Haverfordwest in 2012-2013 ^[4]. It took 22 years to get from the 2002 villa rediscovery to the 2024 earthworks ^[11].

The digs are small so far. The October 2025 survey was limited, and it placed two rectilinear buildings in the right place ^[7]. Trial excavations in March and July 2026 turned up coins, a gaming counter, Roman pottery and a spindle whorl ^[8]. The author wrote that "A more comprehensive evaluation excavation was needed, however, to establish for certain whether the site was a Roman auxiliary fort" ^[8]. In early September, five trenches went in: three across the suspected principal buildings, one where a perimeter road would be expected, and one upslope on a suspected track ^[9]. In the first two, the team found the bases of floor structures built of clay and packed with thin stone blocks ^[10].

The author does not report a date for the floors or the finds, and the date matters for the legend.

The Mabinogion tale places the late fourth-century usurping emperor Magnus Maximus, Macsen Wledig in Welsh, on Frenni Fawr, the easternmost peak of the Preseli Hills ^[15]. That peak is a little over three kilometers (1.9 miles) northeast of the site ^[15]. There are no visible remains of a Roman marching camp on it ^[16]. The author began to think the fort may relate to Cadair Maxen, the emperor's camp ^[17]. The association is drawn from position.

In our view the fort claim is the strongest part, because shape, size, water supply and survey results point the same way, and the September trenches are the first to look inside the enclosure. The road network is a hypothesis with one more anchor point if the fort holds, and the author believes the site belongs to a major arterial road linking Pembrokeshire with Ceredigion ^[13]. The legend link would need evidence of occupation in the late fourth century.