Wall Mile 28, with the ditch and Military Road

Having admired the milecastle, we can now move on, heading briefly northwards and then east to circumnavigate a small plantation. Next we cross a road and the Trail rejoins the Wall once we enter the field. We are now closer to the Military Road than the wall and ditch, but no matter. The very fact that we are walking between the 18th-century road and Hadrian’s Wall reminds us that not all of the wall was destroyed by the road being placed on top of it. Continuing into the next field we then come up against a major diversion which sends us to the north, across the line of the wall and ditch (at a point where a modern quarry has removed them) and then along the northern rim of the ditch for a while, before being sent off even further to the north-east (by a route which seems to change each time you walk it and is seldom clearly signposted) before we stumble onto the minor lane that leads us back to Walwick and the course of the Wall.

I think this translates as 'Welcome to Walwick'... but I could be wrong

As we pass through the hamlet, we can admire the rather aggressive notices warning walkers against using the farm buildings as lavatories (and of course begging the question of why appropriate facilities have still not been provided if there is a demonstrable demand). Finally we are back at the Military Road and the line of the Wall, which we rejoin by walking a few metres downhill to the bend, placing us near the site of Milecastle 28.

Milecastle 28 (extreme left) and Wall Mile 28

Milecastle 28 (Walwick) was a long-axis milecastle on the other side of the road, but there is, you will be unsurprised to learn, now nothing to see.

Wall Mile 29

Heading eastwards from the dramatic and slightly forlorn pinnacle of rock, we find that the ditch gets ever shallower, evidently having been little more than marked out, rather than fully excavated. The Trail currently passes to the south of the wall next to the trig point, then heads south until a break in the newly emerged consolidated curtain wall, where we cross the barrier again onto the berm and make our way down to the end of the field. Note that the Military Road is not on the line of the wall, here, but rather sits on the north mound of the Vallum (where it stays for nearly four Wall miles).

Black Carts and Turret 29a

Crossing a small lane and entering the next gently sloping field, we find a splendid stretch of curtain wall and ditch at Black Carts. Two-thirds of the way down are the remains of Turret 29a.

Turret 29a (Black Carts), famously depicted in one of the woodcuts in Collingwood Bruce’s Handbook, survives to eleven courses within its recess, and has the familiar wing walls of a broad-gauge turret wed to the narrow-gauge curtain wall. It was first excavated by John Clayton in 1873 and subsequently re-examined in 1971. The threshold block in the doorway is of particular interest, since it retains the settings for the monolithic uprights that formed the door jambs, and the socket on the eastern side shows which way the door opened (remembering defensive doorways and gateways always opened inwards).

Turret 29a

Black Carts centurial building stone

If you are feeling adventurous, you can nip round to the north side of the wall and hunt for a building inscription. One lies 55m from the west end (or 90m from the east) of this stretch of wall and records construction work under a centurion from the first cohort of a legion by the name of Nas(…) Ba(ssus). It has been suggested that legio XX was responsible for the initial construction of this section of wall. Bassus crops up elsewhere and an almost identical stone can be seen in Chesters museum which, although unprovenanced, may well be the pair to this stone. It has long been thought that building inscriptions were only placed on the south face of the curtain wall and that those on the north side were a result of rebuilding work. The fact that this stone is in the second course may give pause to question this argument for their placing, but it may equally hint at a very thorough rebuilding of this bit of the curtain wall (and such major reconstruction work is known elsewhere).

Leaving Black Carts behind us, we cross a lane and then head up a gently sloping field towards a plantation near the top, the ditch still visible to our left, but the curtain wall now hidden. Near the top of the field is the site of Milecastle 29.

Milecastle 29 from the air

Milecastle 29 (Tower Tye), like Milecastle 38 (remember those bankers?), exists now solely as an earthwork, but is nevertheless an extremely interesting example. Excavated by John Clayton, the robber trenches for its walls are still sharply defined. However, there is an additional detail that the keen-of-eye may be able to make out and that is the fact that the milecastle is one of the few known to have had a ditch around it. It shows up now as a shallow depression around the west, south, and east sides.

Quarrying east of Carrawburgh

Departing Carrawburgh, we head eastwards up the field, past some recent quarrying next to the Military Road, and before long the Vallum re-emerges, now with the Military Way perched on its north mound. We now have to negotiate another crossing of the Military Road (so, once again, take care as there are maniacs in cars along here), backtracking slightly to get into the far field and resume the Trail to the north of the ditch. Before heading on, look back towards Carrawburgh and you will see a view that has changed little from when J. P. Gibson took a photograph of it before the First World War, although his model was then able to sit down in the road (which did not have a tarmac surface and was much lower than the current ‘blacktop’).

The Military Road at Carrawburgh before tarmac and cars

We plod on to the north of the ditch, now being a good time to admire those roadside drystone walls (they are easier to see from either side than from the road, due to the changes in level since the road was built). The larger blocks, curiously familiar from our perambulations next to the curtain wall, are interrupted by regular lines of throughstones. This pattern of construction becomes obvious when poor quality repairs are attempted in places, although this stretch is in good order.

The construction of the Military Road began in 1749 with a survey from west to east, undertaken by military engineers Dugal Campbell and Hugh Debbeig (the latter serving later at Wolfe’s side at Quebec), the actual work of building the road and walls being contracted out to civilian companies (unlike the other military roads in the Highlands, which the Hannoverian army built).

The unfinished ditch at Limestone Corner

Finally, we cross a stile and reach one of the great enigmas of Hadrian’s Wall: Limestone Corner. Why an enigma? Well, there’s no limestone there for starters; it starts further to the east – no matter. More perplexingly, this is the point at which the Roman army got fed up digging their ditch and gave up. It is perplexing as the vallum diggers had no such problems and ploughed through the whin stone along here regardless, so it seems the ditch-diggers may have had a bad Friday afternoon. The most exciting thing about Limestone Corner is that the course of the ditch (and curtain wall) moves offline in order to stay at the top of the scarp, thereby (once again) enhancing the defensive effect of the ditch (although not quite as much as if they had actually finished it, of course). To the south of us, large chunks of whin have been discarded down the slope, the largest of which (subsequently split into two) has been estimated as weighing around twelve tonnes (naturally, there is no record of anybody having actually weighed it; this is a guesstimate). Such pieces probably had to be removed with sheer legs, a technology with which the Roman army were not unfamiliar.

Wedge holes in the top of the living rock

If we move a little further on from the point where the Trail crosses the ditch, we can stop by a large rock protruding from the base of the ditch. Wedge holes can still be seen in the top of it where the attempts to split it were given up; it must have been a bad day.

Milecastle 30 (Limestone Corner) survives as an earthwork to the south of the field wall (which is set back slightly from the line of the curtain wall), excavation showing that the narrow wall butted against broad wing walls.

Milecastle 30 and the Limestone Corner ditch debris

Carrawburgh from Wall Mile 31

Our gentle climb continues until we reach Carraw Farm, where we are diverted off to the north (this bit can get a bit plodgy) and then back round and onto the upcast mound again, north of the ditch. Now, however, since we have passed the crest, Carrawburgh fort heaves into view south of the road, after we have negotiated the dip. You will have noted that, now that the Wall has come down from the crags, it is being extremely well-behaved and traversing the landscape with nice straight stretches, just like the Vallum has been doing all along. Now there is nothing to stop the two of them running along, hand-in-hand, for a few miles.

Our second crossing of the Military Road is about to occur, but here there is no pedestrian crossing as there was in Stanwix so keep a sharp lookout as they drive fast around these parts and the crossing is at the bottom of the dip.

The slightly moist site of Coventina's Well

Once over the road, there is an extremely boggy portion off to the right of the path (between us and the imposing ramparts of the fort) and the section with a pond in the middle of it is in fact the site of the shrine of Coventina, a local water nymph. When it was excavated in 1876, it produced vast amounts of coins (more than 13,000; some were melted down and cast into a bronze eagle – must have seemed like a good idea at the time) as well as other votive material, some of which we can see in Chesters museum very soon, and is characteristic of the Celto-Roman veneration of water deities.

Carrawburgh fort (BROCOLITIA) (pronounced Carra-Bruff) is 7.6km (4.75 miles) from Housesteads and is one of the forts that sits astride the Wall, rather than attached to the rear or even detached. Occupying 1.6ha (3.9 acres), it was constructed after the Vallum, the course of which runs under it. It was garrisoned by the cohors I Aquitanorum in the 2nd century and other units attested include the cohortes I Cugernorum, I Frixiavonum, and I Tungrorum (the last of which, as we know from Housesteads, was milliary, so only a detachment would have fitted in). Cohors I Batavorum was recorded in the 3rd and 4th centuries.

The Carrawburgh mithraeum

There is little to see of the fort, although its platform is still prominent, but the mithraeum outwith the fort in the civil settlement is on display. When excavated, the waterlogged conditions preserved many organic remains that enabled a detailed reconstruction to be built in the former Museum of Antiquities, now recreated as a rather-less-successful video display in the Great North Museum, both in Newcastle. On site, the organic components have been cast in concrete, which is also the medium employed for the replica statuary and altars.

Altar to Mithras from Carrawburgh

Designed to mimic a cave and produce what excitable marketing types would probably call ‘an immersive experience’ these days, devotees entered at the south end of this small quasi-apsidal building, encountering a diminutive lobby or vestibule, separated from the rest of the interior by a wooden screen. Beyond the screen were two wicker-lined benches, one on either side, attended by Mithras’ familiar torch-bearing companions Cautes and Cautopates (the former with his torch held upwards, the latter downwards). Cautes has lost his head, but of poor old Cautopates, only the feet remain. At the northern end, there are three altars, dedicated by commanders of the cohors I Batavorum. The one on the left incorporates a nice little effect, whereby the radiate crown of Mithras has been pierced, enabling a lamp to be placed behind it for some minimalistic visual trickery. Evidence of what went on in here includes burnt pine cones, a chicken’s head, and bones from pork, lamb, and more chickens: obviously somebody’s idea of a fun night out in the vicus. The whole thing was thoroughly trashed in the 4th century AD and it is speculated that Christians may have been responsible.

Mithraism was an elitist cult (the temple could only accommodate twelve so it was obviously not meant for the common soldiery), with a strict hierarchy that mimicked the army’s rank structure, and a series of ordeals beloved of such institutions

Immediately outside the entrance at the western end of the mithraeum was another small shrine, dedicated to the nymphs (unsurprising, given the presence of so much water in the vicinity) and the genius loci (literally ‘spirit of the place’).

As we have already seen, the remains of Coventina’s shrine lie nearby, but there was also a bath-house on this side of the fort, excavated by Clayton but not now visible… well, in fact, to be brutally honest, it is currently ‘lost’.

The car park, beyond that the fort at Carrawburgh, with Carraw Farm on the horizon

Milecastle 31 (Carrawburgh) lay just beyond the eastern end of the car park, part of one of the robbed walls having been found. Needless to say, there is nothing to be seen now.

The National Trail crosses the ditch

As we depart Milecastle 33, we are still on the approximate line of the curtain wall, but shortly we cross over the ditch and find ourselves firmly in Barbaricum for a while. This occurs at the point where the Military Road shuffles in and adopts a position on top of the curtain wall which it will retain, for the most part, until we reach the outskirts of Newcastle. Now we are walking along the line of the upcast mound and there is a little to see, beyond the ditch, hard up against the north wall of the Military Road.

Wall Mile 32, with the Military Road to the right and Carraw Farm in the distance

These walls are interesting (as indeed is the Military Road, about which we shall say more in due course). Before the road was built, Hadrian’s Wall itself was the major land division, so destroying it and laying a road required that walls be built on either side of it, to placate the landowners. Therefore, the road-laying gangs also had to undertake wall construction, and there are still some fine examples of the drystone-wallers’ art to be seen as we progress towards our destination.

Milecastle 32 and the Military Road

After an interminable distance (or so it seems) we cross a drystone wall by means of a ladder stile and then it is only some 220m to the location of Milecastle 32 (Carraw). Don’t bother looking for it, as it is on the other side of the road (remember: the Military Road sits on top of the curtain wall, now, so the milecastle is south of that), but it survives as a low earthwork, was of the long-axis type, and was excavated in 1971.

Wall Mile 33 from Milecastle 34

The landscape now takes on a more gentle aspect for the walker. We are also soon going to encounter the Military Road again after a long interval and it will be a close companion until we reach Newcastle. Meanwhile, the remains of the curtain wall are visible as a low mound with occasional blocks of stone poking out. Treat it gently and tread carefully. It continues like this until we reach the remains of Turret 33b (Coesike).

Turret 34b, with blocked doorway

Another of the short-lived turrets, with broad gauge footings cut away by the narrow gauge wall. Once again, the recess-filling wall is present, albeit not of the best quality workmanship, probably to enable a wall walk to cross it safely. The doorway still retains its blocking, so the turret was evidently not completely reduced upon abandonment.

The Military Road crosses the Vallum

To the south of us, the Military Road emerges from a softwood plantation to swoop across the Vallum and now keep pace with us, although still not actually on the curtain wall. We cross from walking behind the wall to walking along the berm and a field wall now sits just to the south of the curtain wall’s remains, beginning a gentle climb up towards Carraw Farm.

Milecastle 33

We proceed in this fashion until we come to Milecastle 33 (Shield-on-the-Wall), the side walls of which we have to cross, as the curtain wall is to our left and the field wall to our right.

The north gate and parts of the side walls of this long-axis milecastle are still exposed, perhaps a bit too much for those who worry about potential damage to the monument. One interesting detail to note is how excavation has changed its flora and made it stand out. Excavated in 1935–6, it usually shows as a patch of bracken, with an old spoil heap standing proud at its south-east corner (it is not generally thought good practice for archaeologists to leave their spoil heaps lying around, but it sometimes happens).

The north gate of Milecastle 33

The curtain wall and the cist

About 50m east of Milecastle 35, there is a small stone box next to the south face of the curtain wall. This is the remains of a cist burial, presumed to date to the post-Roman period. The curtain wall along here is very obviously narrow gauge on broad foundations (some of which were of whin) and, noting a narrow cleft which William Hutton was told was a tunnel dug by adventurous Picts in order to sneak under the curtain wall, we follow it until we reach the plantation around Sewing Shields farm. Emerging on the far side, it is now clear that we are nearing the end of the crags. We have another turret to inspect before we get too carried away, Turret 34a (the site of 34b was in the plantation).

Turret 34a

Turret 34a (Grindon West) was furnished with exceedingly small wing walls and this was another of those turrets which was only occupied in the 2nd century and, after abandonment, had its northern recess filled in. In the doorway, the curious will note the settings for the stone jambs of the door as well as the socket for the door pivot on the east side.

Sewingshields Castle and Milecastle 34

To our north are the earthworks that are all that remains of Sewingshields Castle, still visible from the air, but our principal concern is the proximity of the next milecastle, marked by a walled plantation on top of it. As we reach it, we note that the ditch has rejoined us to the north, since the crags are now behind us and a man-made obstacle is once more needed.

Milecastle 34 with the ditch to the right

Although the position of Milecastle 34 (Grindon) is conveniently marked by the plantation, there is nothing to see of the milecastle itself, beyond an information panel, but it makes a fine observation point from which to observe the ditch in either direction.

Busy Gap from Sewingshields Crags, looking towards Housesteads

The curtain wall, beneath the by-now-familiar field wall, descends into Busy Gap and is breached by a modern gateway called the King’s Wicket which seems to have a history. Busy Gap was a traditional route through the wall in the medieval and post-medieval period, ne’er-do-wells who used it for their nefarious activities earning the nickname Busy Gap Rogues (a term of abuse that remained in use into the 19th century). It has an even older significance, however, as an earthwork dyke that may date as far back as the Bronze Age runs through the gap and on towards what is now Scotland. Once again, the Roman Wall merrily slices across a traditional landscape. The angle between the wall and the dyke is adapted into a triangular enclosure by the earthwork known as Black Dyke, here thought to be used as a post-medieval stock enclosure associated with the passage through the wall. The Wall ditch reappears across Busy Gap, recognising its tactical vulnerability but terminates again once it begins to ascend Sewingshields Crags.

Curtain wall atop Sewingshields Crags

We follow the wall up, passing the site of Turret 35b and, once we achieve the summit, can pause to look back to the west, where we can see Broomlee Lough, Greenlee Lough beyond it, and Housesteads Crags, with Crag Lough and Peel Crags in the distance. At the top, a short length of curtain wall emerges from underneath its guardian field wall, just to remind you of its existence.

Turret 35a and the trig point

Before long we stumble unexpectedly on Turret 35a (Sewingshields). Constructed on a broad gauge foundation but with a narrow gauge curtain wall, this turret, with its entrance at the eastern end of the south wall, was only briefly occupied before being demolished and its recess filled in.

The next stretch of curtain wall we find has a rather nicely consolidated expansion near its eastern end, confirming that these were not just a product of the imagination of Clayton’s workman but were a genuine feature of the south face of the curtain wall, along the Central Sector at least.

Length of curtain wall with a reduction point

Moving on we encounter further spasmodic sections of curtain wall bursting out of the turf and before too long we reach another trig point, which is a good place to consider King Arthur. Who? Why? Well, remember King Arthur’s Well in Wall Mile 44? Tradition (although not a very old one, truth be told) has it that he and his sleeping knights lie nearby, waiting for the call to defend Britain once again. Having dozed through sundry national threats (the Armada, Napoleon, for example) he was supposedly disturbed from his slumbers by a Northumbrian shepherd in pursuit of a ball of twine (string-related mishaps being common among northern stocksmen). This has little relevance to Hadrian’s Wall, other than to show how it has acted as a focus for myth formation as much as any other human activity (but most notably stone robbing), and Arthur is both ubiquitous and ‘sticky’, as well as beloved of tourist authorities the length and breadth of the land.

More stretches of curtain wall lead us to the site of Milecastle 35, clinging on to the edge of Sewingshields Crags by its fingertips.

Milecastle 35

Milecastle 35 (Sewingshields) was excavated in 1978–82 and the first thing the visitor notes is that this long-axis milecastle has no north gate. This is one of those few instances where it would be truly superfluous. The interior of the fortification is occupied by several phases of Roman building on either side of the central roadway, culminating in its re-use as a medieval farmstead. The later Roman phases included evidence of metalworking on the site.

Before long, we exit the plantation and a gate on the right takes you down past the fort to pay for a ticket at the small museum to examine Housesteads itself, whilst the Trail itself rather grumpily lurches to the left and would take you along the northern wall of the fort and down a rather steep and badly eroded slope without a chance of a peek at this most impressive of forts. The museum has a small display about the site and some of the finds, as well as a bijou shopportunity.

Housesteads from the air

Housesteads fort (VERCOVICIVM)

By dint of visitor numbers alone, Housesteads is the best-known and most popular Roman fort in Britain. For many, it is Hadrian’s Wall. It lies 9.7km (6 miles) beyond its neighbour, Great Chesters, but only 3.2km (2 miles) from the Stanegate fort of Vindolanda (the older name for which, Chesterholm, is seldom used now). Housesteads is another fort that is oriented east to west, in this case in order to fit it into the limited available space at the end of the dolerite ridge above the gap through which the Knag Burn flows. As it is, it occupies 2ha (5 acres) and still slopes quite considerably inside.

A water tank

There was no ready source of fresh water within the fort, although there is the Knag Burn down the hill which supplied the bath-house down there. Therefore, quite unusually for a British fort and more in keeping with its cousins in Jordan or Syria, Housesteads was dependent upon the collection of run-off from roofs and road surfaces, so great attention was paid to the provision of water tanks. This in turn allowed the flushing of the latrine building in the south-east corner of the fort.

The garrison was the cohors I Tungrorum milliaria (about 800 infantrymen), which moved there from Vindolanda (the Tungri originated west of the Rhine, around the Ardennes). It was later supplemented by a cuneus Frisiorum and the numerus Hnaudifridi, both quite clearly Germanic in origin. The depiction of an archer on a sculpted panel from Housesteads (now in the Great North Museum, so we can see it later) has led to the suggestion that a detachment of cohors I Hamiorum (who we know were based in the Stanegate fort of Carvoran) may have been based there at some point (and they came from Hamah in Syria).

The south-west corner of the fort

From the museum, we can make our way to the entrance to the fort (which will soon revert to the south gate, after many years of being through a gap in the south wall near the south-west corner), but should pause briefly on the way to look at the exterior of the south-west corner, noting the Crunchie-bar-shaped blocks of stone that were used to repair it in the late Roman period.

The south gate

Duly enlightened, we can now proceed to the south gate. The first thing to note about this double-portalled gateway is that it has been adapted and the east tower has acquired an additional structure, since this was a fortified medieval farm, notorious in its day for the unruly nature of its inhabitants, as well as a corn dryer in its eastern tower. The south gate itself, you will not be surprised to learn, had its east portal blocked (and the blocking removed by 19th-century excavators). The central pier (or spina) between the portals contains two fine examples of pivot holes for the gates, each with channels to allow the gate leaf to be fitted. Looking up the hill from the gate, we are looking along the via principalis (the main short-axis street) from the porta principalis recta (or south gate) towards the porta principalis sinistra (or north gate, which we can’t actually see because of the shape of the hill). So now it is time to go exploring the central range, since this is the first fort we have encountered that will let us examine all three principal components.

The commanding officer's house

First, on our left-hand side, we see the commanding officer’s house (praetorium). Although it conforms to the Mediterranean-style, high-status courtyard dwelling, it is a radically unusual example: the awkward terrain has forced its builders to terrace it into the hillside. The south-eastern corner contained a stable, whilst the east and west wings climb up the slope to the north wing, which is considerably higher than the south and has been taken to imply a second storey at the lower level (thereby pre-empting the medieval Borderers’ habit of living above their animals). The courtyard in the centre is worth a look, as it has been paved in a late phase with Crunchie-bar-shaped blocks and even bits of window head (monolithic blocks with a semi-circular cut-out, imitating an arch, that acted as lintels above window openings): heritage hardcore as crazy paving. The north range has a series of rooms with underfloor heating, something of a prerequisite for this area in a winter, but actually a common feature in commanding officers’ houses everywhere. The commander and his familia (his slaves being included within that term) lived and worked within those four ranges of rooms, socially delineated, functionally adapted, and decidedly terraced.

The headquarters building

Now we move uphill to the principia or headquarters building, a fine example of its kind. Entering from the main north to south street (the via principalis, if you recall), we encounter the first of the three components of this building: the courtyard. Open to the elements, but surrounded by a peristyle, it harked back to the days when the centre of a Roman camp was its forum, where the soldiers could assemble. The Romans had no gutters on roofs, so there was an eavesdrip round the courtyard which channelled the run-off into the drainage system. Moving westwards, we proceed into the cross-hall, a high covered structure with additional entrances at each end (to our left and right). To our right is the raised podium or tribunal (yes, that’s where we get the English word from) from which the commander could address his troops or, more likely, his centurions at the daily morning briefing (for which we have documentary evidence from other sites), when the daily password was set and unit statistics passed on. In front of us is the rear range of offices, with the shrine of the standards, the aedes principiorum, directly in front of us. Remember, this is placed so that it is visible from the porta praetoria, in this case behind us, to the east. Offices on either side of the aedes contained the clerks who handled unit administration and looked after the records.

The hospital

Immediately to the west of the headquarters building is an additional courtyard structure. Since another house seemed unlikely, the excavators decided this must have been a valetudinarium or hospital. All forts had them (the example at Vindolanda is mentioned in the famous writing tablets) so it is not an unreasonable deduction, although the evidence (similarity with other, larger, such structures) is circumstantial, rather than conclusive (like, say, a lopped-off limb or two or a set of medical implements). Rooms were arranged on four sides around a courtyard, one of them suggested as an operating theatre (it is bigger than the others), and with a latrine incorporated in the south-west corner of the structure.

The fort had two granaries, to the north of the HQ, but they are rather unusual, since it has been suggested they may originally have constituted one large structure. A central row of column bases, subsequently concealed between the north wall of the south granary, and the south one of its northern neighbour, are one clue, whilst those two butted partitions, inserted between the end walls, are another. If we stand at the west end of the northern granary, in its doorway, there are a number of useful things we can note. To the east is the interior, with its floor (long gone now) raised on small stone columns or pilae; these, together with the ventilator slots in the side walls, allowed air to circulate beneath the floor to keep the contents cool and (it is always said) discourage (but not necessarily completely defeat) vermin. The threshold upon which we are standing is in fact a loading platform, against which carts could be backed up, so that gives us a good idea of street level here in the Roman period. Looking down, you can see that sockets and openings reveal that there were once two inward-opening doors here. If feeling energetic, we can nip round to the south side of the south granary to see more evidence of the adaptation of this fort building into a medieval farm: another corn dryer in the middle of it.

Turret 36b

Just north of the granaries, beyond another store building squeezed into the available space, we find the remains of Turret 37b, demolished (along with a stretch of Hadrian’s Wall) once it was decided to construct Housesteads. The north wall of the fort was pushed further north, right to the lip of the slope, to gain as much room as possible, hence the need to level the existing curtain wall.

The north gate from outside

The north gate now has an imposing drop below its external threshold, but this is a result of a causeway having been removed during the 19th-century excavations. We can stand near the edge and look at the usual attributes of a gateway, most notably marking-out lines which facilitate the placing of the massive opus quadratum blocks of the gate piers and jambs.

Between the north gate and the north-east corner, the Romans suffered repeated problems with the stability of the wall now that it was placed nearer the edge of the slope. This was compounded by the habit of removing the rampart to insert rampart-back buildings like workshops, then demolishing them and putting the rampart back! Anyway, several collapses later, they started reinforcing the back of the much-abused rampart, the back of which (marked now by lines of kerb stones) crept ever further across the road towards the barrack. What started out with room to drive a cart along ended up barely wide enough for a single person to get through. Excavations showed those workshops were busy working with leather (mainly cobbling, since the water tanks ended up full of old shoes, betrayed by their hobnails), blacksmithing, and casting copper-alloy equipment.

Things had also gone a bit wrong at the north-east corner of the fort at a very early stage. Placing the angle tower in the correct position meant, inconveniently, that it was not at the junction with Hadrian’s Wall itself (one wonders at which point this was noticed!), so that was demolished and a new tower placed slightly to the west of it. This speaks volumes about how the whole story of the Wall was one of adaptation (or fudge, the unkind might observe).

Barrack XIII

Two barrack buildings, XIII and XIV, have been consolidated in their late ‘chalet’ form, with each contubernium in the form of a separate hut, but more standard long barracks were located beneath them and some of the walls were reused in the later versions. The officers’ buildings were at the east end of the buildings and one of them contained a piece of a hackamore from a horse harness. Next to Barrack XIV, to the south, was Building XV, originally a storehouse and later adapted to contain a small bath-house. Note those Crunchie-bar-shaped stones used in its reconstruction. East of the two barracks is a late interval tower, but that is perched (rather precariously) on top of a larger Hadrianic bakehouse that was found to contain two circular bread ovens (which you can no longer see), presumably one for each barrack.

East gate blocking

The east gate has a potent piece of folk mythology associated with it, the ruts in the threshold block supposedly influencing the Standard Gauge of 4ft 8½in (1.435m) adopted by Stephenson for his railways. The debunking of this myth is done with the aid of horses’ bottoms (naturally). The axle width of a cart is dictated by the need to comfortably fit a horse into the poles; hence modern carts resemble Roman carts in a lot of details, including axle width. Since railways evolved out of the horse-drawn waggonway carts that hauled coal along Tyneside from mine to staithe, also one horse’s width, we have our equally interesting, but less romantic, answer: it’s a coincidence.

Moving on downhill, we see the remains of an ascensus or stairway to the south of the east gate. This is one of the means by which soldiers got onto the rampart walkway. Next there is another interval tower before we reach the heavily modified south-east corner tower and its attendant facilities.

The latrine

This corner provides everybody’s favourite bit of Housesteads: the latrine. This much-sniggered over piece of functional engineering was flushed by water held on the large header tank with the scalloped edges, the water passing clockwise around the inner channel so that soldiers’ sponge sticks could be rinsed, and then anti-clockwise around the sewer beneath the seating space, finally debouching through an arched outlet straight into the civil settlement. Property prices in that area were probably rock-bottom.

The sewer outfall

The header tank for the latrine

The scalloping around the tank has caused some comment, it often being suggested that this was caused by soldiers sharpening blades (unlikely, since hones, found by the dozen in the fort, were a much more efficient way of doing that). It may instead have been caused by washing clothes, the slight downward trend of the ‘scallops’ being a possible indication of this. Like much of life, if you have to sum up Hadrian’s Wall (and certainly Housesteads), you can probably do it with this latrine.

The west gate

Now, how keen are you on gates? For the sake of completeness, you should see the splendid west gate on the far side of the fort but we shall be understanding if you decided to skip this part of the tour; we can leave you sitting morosely on part of the site (assuming it’s not raining, which it does quite a lot at Housesteads). Two portals again, both ultimately blocked (the northern first, it is suggested), and the north pier surviving to the height of the arch springer. Slots can still be found to secure the gates, as can the usual sockets, threshold blocks, and more marking-out lines.

Murder House

Outside the fort are some of the vicus buildings. Next to the south gate is the House of the Beneficarius [sic] (yes, there were even spelling errors on old Ministry of Works signs), and further down the hill is the inspiringly named Murder House (you’ll never guess what happened there: two Roman bodies under a newly laid floor, one with the tip of a blade between the ribs). Don’t make the mistake of thinking the circular sheep stell is Roman; that belongs to Housesteads’ long history as a farmstead, rather than a Roman fascination with building circular structures with no apparent entrance.

We can usefully resume our journey by passing round the outside of the south-east corner of the fort and making our way diagonally down the slope towards where the curtain wall crosses the Knag Burn. As we go, examine the outer face of the fort wall and see more long blocks typical of late rebuilding: this is one heavily patched fort!

The valley of the Knag Burn was not only the site of the bath-house for Housesteads (no longer visible) but also a gateway through the Wall.

Knag Burn Gateway

Knag Burn Gateway, thought to have been built during the 4th century and examined in the mid-19th century, consists of two towers, one on either side of a single portal. As such, it is not particularly noteworthy, but it does give us a clue what the gateways on Roman roads at Carlisle, Portgate, and (possibly) Newcastle looked like. This, however, is not on a major road, but rather a minor route, perhaps a pre-existing transhumance route. Interestingly, there were two sets of pivot holes and it has been suggested that two sets of gates were in use at the same time. Clearly, there may have been other gates along the Wall which have not as yet been found.

Wall Mile 36 from King's Hill

The Trail now leads us on into an angle in two field walls where a stile takes us over and into another plantation. A short length of ditch survives to the north-west of the plantation but has not been identified over the rest of the Knag Burn valley, although it might have been anticipated. Out the other side and we are now yet again following a field wall to our left that is on top of the curtain wall. The ditch stops as we ascend Kennel Crags but the Military Way is still with us to the south. Further down the dip slope is the Vallum, die-straight as ever. We climb a small hill, descend into another gap, then climb higher to a small plateau, King’s Hill, where we find our next milecastle.

Milecastle 36 (King’s Hill) was a long-axis example perched on a hill, overlooking Busy Gap to the north-east. The identity of the king in question will become clearer once we get up onto Sewingshields Crags in a short while.

Hotbank Crags, looking towards Housesteads

As soon as we have finished with Milecastle 38 we have an invigorating climb up to Hotbank Crags. There are almost permanent problems with erosion up here so try not to tread on eroded areas (remember that grass is very resilient until it is worn down to its roots and then it becomes vulnerable). It is one of those hills that just keeps on giving (or taking, depending upon your point of view) and whilst the hardy will stick to the curtain wall, the less resolute can branch off just past the hexagonal plantation and rejoin the less-demanding Military Way. The ditch is rendered unnecessary again along the crags, whilst further down the dip slope the Vallum is gradually converging with the course of the curtain wall.

Housesteads plantation from Cuddy's Crags

Once we reach the top and have caught our breath, the going is easier again and we barely notice as the Claytonized wall gives way to a modern drystone wall, which then descends into Rapishaw Gap, this time being pierced by a gateway that marks the line of the Pennine Way. Crossing a stile, we are confronted with a prominent outcrop, on top of which is the neatly terminated end of another stretch of Claytonized curtain wall. The easiest way round this is to head south for a short way and then almost double back to get up to the line of the wall itself, so that we are only climbing easy inclines and not risking life and limb scrambling up rocks. The timid can take the opportunity to carry on along the line of the Military Way, but to do so will involve missing another of those iconic views. Back on the crags, the Trail takes us 180m along Cuddy’s Crags to another re-entrant (with ditch, naturally) that suddenly presents us with the most famous prospect of Hadrian’s Wall: the gap between Cuddy’s Crags and Housesteads Crags with Housesteads Plantation perched on the edge of the precipice. It is highly unlikely that you have never seen this view somewhere, whether it be on a poster, postcard, or book cover. One of the earliest versions was a postcard produced at the beginning of the 20th century by J. P. Gibson, Hexham pharmacist and photographer, and himself no mean excavator of the Wall. Having duly recorded the view for posterity in an appropriate fashion (camera, watercolours, charcoal, Etch A Sketch…), we press on down into the gap and up the other side where, before long, we encounter Milecastle 37 in all its Claytonized magnificence.

Milecastle 37

Milecastle 37 (Housesteads) is perhaps the most visited, by dint of the fact it is closest to Housesteads (which enjoys the highest visitor numbers for Hadrian’s Wall), and is within staggering distance for the more adventurous car-bound visitor. It is presented in the same Claytonized form as the curtain wall on either side, facing stones reconstructed up to a regular height and topped with turf. It has been excavated four times between the middle of the 19th and end of the 20th centuries and, quite apart from offering an excellent sheltered location for a walker’s lunch, provides more insights into the nature of the milecastle.

The north gate of Milecastle 37 from the air

Beginning with the north gate, we can see that the reduction in width to pedestrian access is still in place. Comments are occasionally made that it is daft to provide gateways for some of the milecastle along the crags, but access would have been needed along the front of the curtain wall and ditch for the purposes of maintenance and many afforded some sort of rudimentary route to the north, the pedestrian blocking being a recognition of the fact that this was probably usually not by wheeled vehicle. In fact, the most recent excavation showed how partial collapse of the north gate led to its being blocked soon after construction and only opened up for pedestrian access at a later date. The lowest two voussoirs of the southern arch of the north gate are still in place on either side, but the others have been replaced in recent times for effect (a drawing of 1879 by James Irwin Coates shows those two springers, as they are known, in situ).

There is one internal building, east of the central north–south roadway, recalling the arrangement we have already seen at other milecastles (although the only excavated sign of a western structure here was a couple of hearths).

The gates at Milecastle 37

The south gate is less well-preserved than its northern companion but still stands to an impressive height and shows the use of responds on either side of the gate itself. ‘What’s a respond?’ you cry; it’s the sticky-out bit on either side of the jamb (upright) of the doorway. Why would you care about responds? Because they are one of the identifying factors that distinguishes the three (or four) types of milecastle gateway (which scholars think mark construction work by different legions). What do they do? They carry the archway over the gate; so now you know.