Hod Hill, Dorset.

Hod Hill

Scheduled Ancient Monument #1002678; NGR 38571106

Surveyed between 2006 – 2008

TEXT by Dave Stewart and Miles Russell

Full details of this survey and the interpretation of the results can be found in Stewart, D and Russell, M ‘Hillforts and the Durotriges’ published by Archaeopress 2017.

The Site

Situated in the parish of Hanford, the hillfort of Hod Hill, also known locally as Lydsbury Rings, is located 5.5km to the north-west of Blandford Forum and less than 2km from Hambledon Hill.

The hillfort sits on an outcrop of the Upper Chalk where the Stour Valley opens out into the fertile pastures of the Blackmore Vale. It is separated from the main scarp on the east by the Iwerne valley and bounded by the river Stour to the west, a few metres upstream of an ancient crossing place.

Spettisbury and Dudsbury occupy similar riverine locations but were not expanded and developed. Hod is not visible from the sea like Chilcombe, Eggardon, or Pilsdon Pen but shares with these other developed hillforts the access to large areas of productive farmland, making its position on the waterway of the Stour an ideal place to control trade, or indeed to defend it. It is designated as a Site of Special Scientific Interest for its population of butterflies and orchids and is owned and managed by the National Trust.

Hod Hill is a multivallate hillfort enclosing an area of 22 ha and has a Roman fort situated in its north-western corner. Double ramparts surround the hillfort, except on the western side where the steep slope to the river Stour is protected by a single bank. There is an inner scarp below which quarry pits form an almost continuous line.

Two of the five entrances that presently pierce the ramparts have external hornworks. These, the Steepleton and Southwest Gates are deemed to be original, Home Gate is medieval or later while the Ashton Gate and Northwest (or ‘Water’) Gate are both thought to be of Roman origin.

The Roman fort, which reused the north and western ramparts of the hillfort, enclosed an area of around 2.6 ha, and was defended on its southern and eastern edge by a single rampart fronted by three ditches. Two wedge-shaped gates, at the south and east, were faced with small banks, known as tituli. Projecting internal mounds to the north of the east gate and east of the south gate may originally have been designed as artillery platforms.

The hillfort has been the subject of three published excavations, Boyd-Dawkins (in 1900), Brailsford (in 1949) and Richmond (in 1968) as well as the less well-organised artefact ‘scavenging’ activities of Henry Durden. Geophysical surveys were carried out by Dave Stewart between 2005 and 2007.


The scars of plough damage are evident across the western half of the hillfort. The most striking feature of the plot is the number of circular anomalies, 10 to 15m in diameter with a gap to the south-east side, which seem typical of ring ditches that surround prehistoric roundhouses. The structures on Hod are densely packed and their footprints often overlap.

Each appears to contain one or more strong pit-like anomaly: comparison of magnetometry and resistivity plots indicate a mixture of hearths and pits, some probably combining both. A line of similar pit-like features without ring ditches appear near the eastern rampart and a scattering of the same can be seen in the south-west.

One roundhouse structure, possibly the site of pit 4 from the 1897 excavations, was re-surveyed at finer resolution in order to test the preservation of internal features. The results clearly show the inner ring of postholes and a compact magnetic scatter that could possibly represent the remains of a hearth. The wall debris has been ploughed away in the south-east quadrant giving a glimpse of the original wall slot or stakeholes beneath. To the southeast of this structure and intruding on it, is a rabbit warren covered by nettles, brambles and hawthorn where the intruders have taken advantage of the loose backfill.

A dendritic road system, extending from the Steepleton Gate is evident from lack of cut features rather than as a positive feature in its own right. Various branches terminate at areas of magnetic scatter below the ramparts or divide groups of huts. Some sections of both main and minor trackways are bounded by ditches, where the gullies surrounding structures have been linked to form a continuous drainage channel. A higher resolution plot of one such section near a junction showed it to be also bordered by rows of four-post structures, generally about 2.5m square with 4, or in two instances 5 postholes.

A line of responses commensurate with stakeholes appears to extend between the post-built structures and the nearest ring-ditch. The trackways divide the internal area into blocks of features and within each block examination of the magnetometry reveals at least one structure surrounded by its own enclosure. In the south-east corner, there are two areas abutting the quarry pits that appear to be screened with palisades.

A Royal Commission survey conducted in 1970, notes the existence of a slight bank and ditch running some 200ft from the southern rampart and disappearing into the ploughed area. The magnetometry shows that this linear feature continues before turning north-west towards the angle of the Roman and prehistoric ramparts. There is a shorter linear running east – west near the northern rampart. The double ditch and rampart that surround the Roman fort can be seen, as can the two titulus ditches that served to better protect the entranceways. The blank areas clearly evident in the plot are where the slope proved too steep to survey.

The ‘target area’ between the ditches is devoid of magnetic features. Within the Roman camp the outlines of rectangular buildings are evident, with particularly strong responses and internal divisions showing in one feature in the south of the camp. These buildings clearly overlie the faint traces of circular structures which can also be seen at the northern edge. One very clear circular feature near the centre of the camp marks the structure currently identified and scheduled as a Bronze Age barrow and five strong magnetic responses occur around the outside of the Roman rampart.



The ring ditches almost certainly represent Pimperne style Iron Age dwellings, comprising a central ring of post-holes for roof supports surrounded by a ring of smaller holes for stakes that supported the daub walls. In many cases the walls have collapsed to form a low bank over the stake holes and the whole structure is enclosed by a shallow ditch or drip gully with an entrance to the south-east side. Two of these were excavated by Boyd-Dawkins who described them as ‘cooking huts’, believing that there would be no room to live inside with a fire lit. It is worthy of note that Boyd-Dawkins had not excavated the bank of wall material that encroaches on the interior.

Richmond identified just one round house within an enclosure, basing his interpretation of the Roman artillery Blitzkrieg (sic) on this, evidenced by the number of iron-shod ballista bolts, as being the ‘chieftain’s hut’, the destruction of which was followed by rapid capitulation of the hillfort inhabitants. However, if some or all of the enclosures revealed by magnetometry are contemporary it would suggest that there was more than one chieftain and therefore, perhaps, a federal system or oligarchy. Alternatively, Richmond’s interpretation of the excavated house as the main structure belonging to the ‘chief’ was simply incorrect.

The western portion of the survey area has few magnetic features, but resistivity survey of this zone in 2008 detected circular platforms that appear to have been terraced into the slope, further identifying several ring ditches that are invisible to magnetometry. Given that the magnetic susceptibility analysis indicates significant enhancement of soil in the occupied area, natural silting of cut features could be expected to produce a higher contrast. The supposition is, therefore, that the features were intentionally back-filled.

Unlike those from the Steepleton Gate, there is no clear evidence for the route of any tracks to or from the Southwest Gate, although some may be inferred from the overall distribution of pits. When the National Trust excavated a narrow pipe trench to the cattle trough visible as magnetic distortion in the south-west corner, only three postholes were encountered. This was attributed to plough damage but possibly the trench happened to follow the southern branch of a dendritic system from the Southwest Gate.

Within the Iron Age structures excavated, Boyd-Dawkins found artefacts relating to spinning, weaving iron smelting and other forms of metalworking. Soil samples taken from an area of magnetic scatter during the initial survey of Hod Hill proved to contain iron slag and hammerscale, confirming the existence of an on-site smithy. Hence the extensive magnetic scatters in several quarry pits may be interpreted as evidence for iron smelting away from occupied areas in the shelter of the ramparts rather than discarded modern rubbish.

The linear feature dividing the hillfort interior could be a field boundary although it does not conform to the known episodes of ploughing. It does however align with the eastern side of the enigmatic area of earthworks outside the north-west corner, considered to be an unfinished extension to the defences. It is therefore probable that the bank and ditch represented by this linear predate not only the ploughing but also the raising of the Iron Age rampart, forming a prehistoric territorial boundary.

The ramparts of Hod Hill were refurbished, the original box rampart replaced by the later glacis style and the style of roundhouses may also have changed over time but the interior was densely populated over an extended period. We have lost the pattern of trackways or streets in the western section but it is clear that the dendritic layout was long established with drainage and post-built structures lining the tracks.

The distribution of higher status enclosed dwellings within each housing block is, if contemporary, suggestive of a federal system, an interpretation which can only be verified by future excavation. Overall there is nothing in the evidence to contradict the view that Hod Hill was effectively the Iron Age version of a bustling market town commanding a busy trade route.

The Roman fort on Hod Hill is, in its position, form and nature, unique and having been the main focus of his excavation, is well described in Richmond’s 1968 report. The magnetometry results conform to his plan of the layout, the strong response from the southernmost barrack block attests that it was, as he suggested, destroyed by fire.

The feature recorded as a Bronze Age bowl barrow which lies inside the Roman fort, occupies the highest point of the interior which would make it invisible from any low-lying areas around. More significantly it sits in the middle of the via praetoria of the fort, just in front of the headquarters building. Boyd-Dawkins put a trench across the ditch surrounding this feature which he thought ‘might be the base of a tumulus’ and found it full of ‘Roman rubbish’. It does however seem rather odd that the barrow and ditch should have been left standing by the Romans in such an inconvenient position. A detailed plot of the so-called ‘barrow‘ bears a striking resemblance to the round house and an alternative interpretation would be that this structure actually post-dates the Roman camp and represents a partial reoccupation of the hillfort, utilising the vacant cleared space of the via praetoria.


OS map reference: ST 8565 1066. Nearest town/village: Stourpaine.

You may also like...