View along Offa’s Dyke near Craignant as it descends towards the Morlas Brook with Selattyn Hill behind, view from north

Offa’s Dyke is a late 8th-century linear earthwork (Britain’s longest), attributed to the Mercian ruler Offa (r. 757-796). To understand it, one has to have a copy of Sir Cyril Fox’s 1955 book, the last and only time the monument has been surveyed from one end to the other. This post reflects on my own observations and what Fox has to say.

Last weekend, in cold and wet weather (but not as wet as this weekend), I went to explore a part of Offa’s Dyke I hadn’t visited before. It was a rare opportunity to get out into hitherto unexplored territory without kids in tow, so I decided to get out relatively early and explore some rather steep and muddy hillsides, in places still covered in a very thin layer of snow.

Offa’s Dyke descending the southern (north-facing) slope of the Ceiriog valley, as seen from the west

The section of Offa’s Dyke I walked runs south from the Ceiriog valley at Chirk Mill (Wrexham) to Craignant (west-north-west of Selattyn, Shropshire). This section is particularly interesting because of the dyke’s relationship with the Ceiriog Valley at Castle Mill – the Gate of the Dead – and the fact that it traverses a steep ravine of the Nant Eris and crosses the Morlas Brook on its route to Selattyn Hill.

Offa’s Dyke walk signposts

This was also intriguing for me since for almost all of this distance, the dyke had unusually restricted views westwards over the territory it was built to defend against and dominate. Fox (1955: 79) refers to the general tendency of the dyke to follow western edges of plateaus, ridges and knolls, but also there are ‘remarkable examples of indifference to tactical position’. However, Fox did not see examples of this south of the Dee, where he regards that ‘practically all advantages offered by the terrain were seized’ (Fox 1955: 79) so that the power ‘occupying the lowlands desired to include within its territory key positions giving visual control of this part of the mountainous borderland’ (Fox 1955: 80).

I wanted to get a sense of why Fox felt this applied to the Chirk-Craignant section when views were so restricted westwards from the line of the earthwork.

This relates to my research interests in the linear earthworks, which I have discussed most recently when I visited both Offa’s and Wat’s dykes near Ruabon. The trip was also useful to reconnoitre ahead of potential future student trips.

Looking north along Offa’s Dyke from the bank with the V-shaped ditch well-preserved on the left (west)

I walked from south to north (the direction of description preferred by Hill and Worthington but counter Fox) and photographed over 2km of the dyke as it rose up from crossing the Morlas Brook at Craignant northwards uphill to Plas-crogan farm and then downhill to cross another steep ravine of Nant Eris. Then it rose again over a second hill before beginning its final descent, first gradual, then steep, down to the Ceiriog at Bronygarth.

Fox (1955)

In terms of Sir Cyril Fox’s 1955 survey Offa’s Dyke, this stretch corresponds to his not-so-succinctly titled sections:

Fig. 23  The southern end of Chrik Park to the Crogen wood-mound wood area; through the parish of Chirk, Denbighshire, and on the boundary between Denbighshire (Glyn Traian parish) and Shropshire (Weston Rhyn parish). pages 56-58

and his:

Fig. 24 The N. flank of the Morlas brook valley to the plateau S. of Orseddwen farm: on the boundary between Denbighshire (parish of Glyn Traian), Shropshire (Weston Rhyn and Selattyn parishes) and through Selattyn parish.

The last point that Offa’s Dyke is visible descending the northern side of the Ceiriog valley: the ditch to the left (west) is enhanced to by the line of a stream, giving it today an impressive profile. However, downslope of that point, towards the photographer, the dyke disappears towards the oak at the Gate of the Dead

Fox’s notes how the dyke disappears on the steep scarp descending to the Ceiriog valey bottom. It remains lost in the Ceiriog valley bottom but then identified it for the first time in gardens and closes of Pen-y-bryn hamlet. Rising up the southern valley-sode, Fox describes how increasingly well-preserved this section is. From here to the Nant Eris stream, he describes it as having a ‘striking profile’ in places.  Fox also notes an ’embayment’ to the west of the dyke, created by its diagonal ascent up the slope, affording ‘magnificent’ views up the Ceiriog valley.

Offa’s Dyke views from west , just north of the Nant Eris

Further up, onto moorland pasture, the bank often retains its steep slope and its ditch retains angular contours. In places damaged by rabbits and foxes, it has been lowered by cultivation between the road and the Nant Eris ravine.

Well-preserved ditch and bank of the dyke south of the Ceiriog valley

Fox also discusses the ‘remarkable appearance’ of the dyke as it rises steeply from the stream of the Nant Eris up the wooded scarp of the ravine with an ‘immense V-shaped ditch on the W. side silhouetted against the southern skyline’.

Offa’s Dyke descends upon the steep north-facing slope of the Nant Eris, view from the north-north-east

Further south, around Plas-crogen, the bank is less impressive since it has been built of soil that has slumped but the ditch has remained intact, in some places over-deepened by water action as it descends to the Morlas Brook.

Offa’s Dyke from east: seen crossing the Morlas Brook at Craignant

Hill and Worthington (2003)

Hill and Worthington note a specifically important point about the choice to cross the Morlas Brook where it does: to be downstream of a confluence with another stream. For this section, their commentary does not contradict that of Fox.

In addition, they conducted one excavation in this section, when Fox had conducted none (his nearest was north of Chirk). This is Hill and Worthington’s Site 63, Woodside, Selattyn – where they recorded a partial section of bank that had become exposed in the rebuilding work near a cottage cutting the east edge of the dyke’s bank.  Half the width of the bank was exposed and a layer of large stones was noted at the base of the bank, with a layer of hard-packed clay with some overlying stones , all covered by looser clayey material (Hill and Worthington 2003).

My Perspective

Based on these sources alone, I think I gain a sense of the narrow line of the earthwork itself and, in crude terms, a sense of its varying levels of survival. However, I don’t get a sense of how it was surveyed, organised and maintained. In particular, I think there are addition things to say about views from and of the dyke, and gaps in it. To do this, we must begin to look out from the dyke; considering the wider topography both west and east.


I have now personally witnessed the gaps in the dyke required by the passage of streams at Plas Power (Nant Mill) and in this section at Morlas Brook and Nant Eris and can confirm that Fox was right to say that they were only a few yards across.

The absence of the dyke at the base of the Ceiriog valley therefore must be, as Fox suspected, the result of alluvial action over the millennium since. I’m not convinced it was never there, seeing the nature of the earthwork coming down the sides of the valley and the breadth of the space left open were it to have not existed.

Views East

One point of important that I haven’t seen observed by others is views to the east. While Offa’s Dyke held views west, it also afforded views east over the Cheshire and Shropshire plains. In the key stretches on this section, views were afforded north-east to the Mid Cheshire Ridge and south-east towards the Wrekin. Was the dyke positioned as much to communicate east – forewarning these territories of unexpected movements of people, as much as dominate the west?

Views east-south-east from the dyke near Plas-crogen. You can see the Wrekin, partially in cloud

Views West

Having walked this section of dyke, I am convinced that it cannot have made any dominant presence in the landscape to its west apart from its bisection of key valleys/ravines where it does, of course, block and dominate those approaching from the west.

I would argue that this stretch of dyke is therefore an area of compromise between the need to retain the dyke’s long-distance orientation and trajectory and simultaneously maintaining long-distance views to the west. To retain longer distance views, the dyke would have had to have detoured significantly westwards to take in higher ground.

Fox identified other sections like this to the north of the Dee. Yet I am surprised he didn’t discuss this stretch in a similar light as a part of the dyke that fails to achieve westerly views.

Views Along the Dyke

These sections, do, however, enjoy views along the length of the dyke itself, allowing communication with possible lookout points just behind its line. From parts of these upland sections, the dyke was observable, and communicable from any lookout point or fortification on the sites that were to later become Chirk Castle. Likewise, were there any station upon Selattyn Hill, approaches to the dyke at points where its view was restricted westwards would be flanked.

I think that more attention is required to the importance of viewing along and behind the dyke in communication systems that helped defend it.

Chirk Castle from the dyke
Views of Chirk Castle from the dyke near Ty’n-y-coed Farm


Viewing from the West

As Keith Ray has recently discussed in lectures I have attended, and discussed in the imminently published book by Ray and Bapty, when approaching river crossings, the dyke bends to show off a united front to those approaching from the west. This is very clear on the slopes down to the Ceiriog near Bronygarth and must have been the case for the Morlas Brook, south of which the earthwork bends westwards to navigate Selattyn Hill and therefore overlooks those approaching along the valley from the west. In both instances, Fox does not fully recognise, as Ray and Bapty are now suggesting, the importance of display and reconaissance in the line of the earthwork when it enters valleys.

Offa’s Dyke descending in a clear curve into the Ceiriog valley with Chirk Castle behind, thus showing itself off to anyone approaching along the valley bottom and valley sides
Offa’s Dyke on the slopes of the Ceiriog valley, view from south-east

In summary, I’m looking forward to exploring more of Offa’s Dyke. Offa’s Dyke has been destroyed or depleted over large sections. Without trespassing, there are large sections that have no public access and public footpaths run far from it. Where it is accessible to a footpath, on this section in particular, for large parts it runs behind the dyke or on it, thus facilitating only particular senses of the monument and how it operates. There are relatively few places I could easy get to see it from the west and appreciate how it was intended to look when approached by those wishing to move eastwards. I think these are some of the reasons why it still remains under-appreciated by the public and scholars, and its character and function, even if it is a unified frontier project as Fox says, are a challenge for us all.


Fox, C. 1955. Offa’s Dyke: A Field Survey of the Western Frontier-Works of Mercia in the Seventh and Eighth Centuries AD, London: British Academy.

Hill, D. and Worthington, M. 2003. Offa’s Dyke: History and Guide, Stroud: Sutton.

Ray, K. and Bapty, I. 2015. Offa’s Dyke: Landscape and Hegemony in Eighth Century Britain, Oxford: Windgather.