NGRef: SH7770
OSMap: LR115, OL17
Type: Minor Settlement, Fort
Canovium Reconstruction

Reconstruction of the Canovium fort
and its associated vicus settlement
Drawing by David Swarbrick
S (10) to Brynygefeiliav (Gwynedd)
WSW (15) to Caemetta
Iter XI: ENE (19) to St Asaph
WSW (20) to Segontivm

The Caerhun Fort

776704 c.410 x 410 ft
(c.125 x 125 m)
c.3¾ acres
(c.1.5 ha)

The Platform of the Canovium Roman Fort
Viewed from Tanrallt across the Afon Conwy to the east beside the A470(T)
(Picture taken in May 2005)

About the year 1650 the antiquarian Samuel Lee unearthed a hypocaust and tiles stamped LEG XX V, and Gale in 1719 reported others, recently unearthed, bearing the legend LEG X, which may have been broken. In 1801 Samuel Lysons uncovered a bath-house, 128 feet (39m) long, outside the north-east defences of the fort, along with tile-stamps marked LEG XX VV.

This fort is contemporary with the forts at Cicucium (Brecon Gaer/Y-Gaer) and Segontium (Caernarfon), being built around 75AD. This is a square fort, each side measuring 410 feet within the ramparts, giving an occupation area of 3¾ acres. Defenses consisted of a 20 foot wide clay bank, fronted by two ditches. The gateways and internal buildings were of timber construction.

The size of the fort and the arrangement of its interior buildings suggest that Caerhun housed a Cohors Peditata Quingenaria, a regiment of foot-soldiers nominally five-hundred strong. The names of none of the garrison units stationed at Canovium are known.

Plan of Canovium - Click to Enlarge
Plan of the fort at Canovium
Adapted from Webster's The Roman Imperial Army (fig.45)
"Additions in stone were made in the first quarter of the second century, and early rather than late in that period. The outer margin of the clay rampart was cut off to a width of 2 feet, and a stone wall 6 feet thick at its base built between the rampart and ditch. The inner ditch was filled up soon afterwards in order to strengthen the foundation of this wall. ... The gateways also were rebuilt in stone. The east gate (porta praetoria) was a double opening with guard-rooms, singular in having its two arches of different widths (15 feet and 5 feet respectively). The new south gate was a double opening with no guard-rooms; but one of the arches seems to have been blocked up during construction for use as a guard-room. At the same time the internal buildings were all reconstructed in stone." (Collingwood, p.37)

Excavation has revealed two timber periods in the early history of this fort, rebuilding being carried out sometime during Flavian times. The sacellum in the centre of the camp was the first building to be replaced in stone during the reign of Trajan, followed by the rampart-wall in Hadrian's reign. Hadrianic and Antonine samian ware shows continued occupation through these times, but the well in the principia was filled around 196/7AD, which may indicate either destruction or desertion at this time. Occupation at the fort was soon resumed, however, as attested by the building of a new cook-house behind the rampart around 235, and continued occupation throughout the third and fourth centuries is proven by pottery and coins dateable to both these periods. The last coin recovered from the site is one of Gratian (367-383).

After the fort was destroyed in c.200AD, the civilian settlement or vicus outside the defences was only sporadically occupied until the 4th century when it was finally abandoned. There were Roman copper mines at Pen-y-Gogarth (Great Orme's Head), eight miles north of the settlement near Llandudno at the mouth of the River Conwy.

Canovium - The Evidence

The Antonine Itinerary was a list of routes and posting-stations used by the Roman army of the late-second century, the British section of this document has fifteen such itineraries, and the Caerhun fort is included in Iter XI - Item a Segontio Devam 'Itinerary Eleven - The route from Segontium to Deva'. The route is listed as 74 (Roman) miles long, starting from SEGONTIVM (Caernarfon, Gwynedd) the next station Conovio is 24 miles away, which can only be Caerhun. From here Iter XI proceeds another 18 miles to VARIS (St. Asaph, Clwyd?) then on to its eastern terminus at DEVA (Chester, Cheshire), a further 32 miles. The seventh-century Ravenna Cosmology (R&C#83) lists Caerhun as Canubio, where it appears between the entries for SEGONTIVM and MEDIOLANVM (Whitchurch, Shropshire).

The only texts reported in the Roman Inscriptions of Britain are not very helpful; a building stone inscribed ...OA... (RIB 437) which is undecipherable, and also a lead letter D (RIB 438), which had originally been inset into a stone inscription, now lost.

Milestones From The Canovium Area

IMP(erator) CAES(ar) TRAIANVS HADRIANVS AVG(ustus) P(ontifex) M(aximus) TR(ibuniciae) P(otestatis) V P(ater) P(atriae) CO(n)S(ul) III A KANOVIO M(ilia) P(assuum) VIII

"The emperor Trajan Hadrian Augustus, high priest, holding tribunician power for the fifth time,¹ father of his country,² consul three times.³ Eight thousand paces from Kanovium."

(RIB 2265; milestone; Rhiwiau Farm, 7 miles west of Caerhun; 120-1AD)

  1. Hadrian was made Pontifex Maximus and first held tribunicia potestas following his accession on 11th August 117AD. Tribunician power was renewed on 10th December 117, and on that date annually thereafter; ergo on this basis, the stone must date to 120AD or 121.
  2. Hadrian was titled Patria Patriae 128AD; there is an obvious disparity here. It is not likely that the letters P P were added at a later date, though it is possible that the mason who inscribed the stone made the simple mistake of assuming that Hadrian already posessed the Pater Patriae title.
  3. Hadrian was consule suffectus 108AD, and consule ordinaris in 118 with Gnaeus Pedianus Fuscus Salinator, and in 119 with Publius Dasumius Rusticus. He held the consulship only in these three years.

Four Roman milestones have been found in the countryside around Caerhun, all can be reasonably dated; the most important one was found on Rhiwiau Farm, 7 miles west of Caerhun and mentions the name of the Canovium Fort (vide RIB 2265 supra), another (RIB 2266 vide infra) was found on the opposite side of the road on the same farm, and one more along the same Roman road but only 4 miles from Caerhun (RIB 2267 infra) and another was recently discovered nearby at Bwlch-y-Ddeufaen, 3½ miles west of the fort (JRS xlvi 1956, p.148, no.11; vide infra).

Inscription Togo-Translation RIB
IMP CAES L SEP SEVERVS P P ET M AVR ANTONINVS AVG ET P... "Emperor Caesar Lucius Septimius Severus, Father of the Fatherland, and Marcus Aurelius Antoninus Augustus and ?P[...]." 2266;

"[...]i? the spirit of the Augustus the divine Constantinus Pius, son of the Augustus."

INV AG DIVI CONSTANTI PIA G FILIO "To the unconquered Augustus, the divine Constantinus Pius, son of the Augustus." JRS 1956.11;

The find-spot of the last Roman milestone mentioned is nearby a visible section of the old Roman road from Caerhun to Caernarfon (NGRef: SH 725 714), and this entire stretch of the road north of the Afon Tafalog abounds with ancient hut-circles, cairns, burial chambers and standing stones.

Click here for a map of Canovium from

See: The Roman Imperial Army by Graham Webster (Constable, London, 1979);
Air Reconnaissance in Britain, 1973-76 by J.K. St. Joseph in J.R.S. lxvii (1977) p.151;
The Roman Inscriptions of Britain by R.G. Collingwood and R.P. Wright (Oxford 1965);
The Archaeology of Roman Britain by R.G. Collingwood (Methuen, London, 1930);
The Military Aspects of Roman Wales by Prof. F. Haverfield (London 1910; pp.28-31);
All English translations, including any inherent mistakes, are my own.

For a brill ground plan and loads of photo's of the Canovium fort (among other things)

Caerhun Roman Fort - by Dave Alexander

If I rated other dude's pages, this site would be ranked somewhere in the stratosphere.

Roman Name

Ravenna Cosmography: Canubio; Antonine itinerary: 11-2 Conovio

Page Citation: Kevan White (2018) "Roman Britain: CANOVIVM"