The slight univallate hillfort and motte and bailey castle at Merdon Castle survives well, and geophysical survey has shown that the monument retains archaeological remains relating to its multiple episodes of use. Further archaeological remains and environmental evidence relating to the original construction of the monument and the landscape in which it was constructed can also be expected to survive.
The monument includes a substantial motte and bailey castle constructed within the ramparts of an earlier slight univallate hillfort on a prominent, south facing chalk spur near the village of Hursley. The hillfort is of probable Late Bronze Age to Early Iron Age date (eighth to fifth centuries BC). The later castle was built by Henry de Blois, the Bishop of Winchester between 1129 and 1138, although there is tentative documentary evidence to suggest that it may originally have been the site of a Saxon defended manorial residence dating from the eighth century. It was partly demolished in 1155 on the accession of Henry II but was used as a bishop's palace until at least the 14th century. The polygonal hillfort defences enclose an area of approximately 3.7ha within a single rampart and outer ditch which are reinforced across the neck of the spur by a slight counterscarp bank. They have been modified or augmented around much of the circumference by the later construction of the motte and bailey defences, but are best preserved to the north east where the rampart stands 2m above the interior and 4m above the outer ditch. They have been further disturbed to the south by modern ploughing, but here there is evidence either of an earlier phase of construction or an attached enclosure, represented by a semicircular section of bank extending from the south eastern defences. There is no clear trace of an original entrance, although there is a slight out-turning and overlapping of the ramparts on the western side, indicating a possible hornwork. The later castle fits tightly within the hillfort defences and includes a massive rampart and correspondingly large outer ditch which encompass an oval shaped motte and a semicircular bailey to the south. They are most substantial around the motte, where the rampart stands up to 5m above the interior and 12m above the ditch. Here the rampart is capped to the east by a number of subrectangular platforms and mounds and, to the west, is terraced into the interior and partially revetted by two lengths of flint walling standing up to 3m high. The bailey's defences are comparatively simple, but are only slightly less substantial. The castle's ramparts have been disturbed between the motte and the bailey by the construction of a modern farm track across the monument. The interior of the motte stands some 2m-3m higher than the bailey and includes the remains of a flint lined well and a substantial, two storeyed flint and stone rubble tower situated within a break in the rampart on the northern side. It has an archway facing the ditch and appears to form a gatehouse, although its use as an entrance is flawed by the absence of an approach through the hillfort rampart opposite, which has been bolstered by a series of earthen buttresses. A more definite entrance is formed by a simple gap in the rampart on the south side of the bailey. A geophysical survey of the monument in 1994 indicated the presence of the buried foundations of walls and buildings within the castle, particularly within the motte where a polygonal arrangement of buildings around an inner courtyard is indicated. Some of these buildings, however, may relate to the castle's later use as a bishop's palace for which there is documentary evidence of substantial structures within the motte, including a bishop's chamber, chapel and hall, and of wooden structures within the bailey, including stables and other farm buildings. Further buried remains associated with the earlier use of the monument as a hillfort, including traces of round houses, granaries and pits, can also be expected to survive. The use of the monument during the medieval and post-medieval period is also indicated by a series of hollow ways and banks to the north west which are depicted in a map of 1588 as forming part of an old road curving around the northern side of the castle. There are also documentary records of the site as the location of a medieval village, although there is no visible archaeological evidence to support this. Later, more recent use of the monument as a military camp during World War I and World War II is represented by a comprehensive series of earthworks and building foundations situated between the castle and hillfort ramparts to the south east and west. (Scheduling Report)
Williams-Freeman considers that at Merdon Castle a Norman keep and bailey utilizes the banks of a hillfort. This is suggested by the plan, and the name, taken together with its early importance as the centre of a Saxon hundred.
Probable I.A. Hillfort. This would seem to have been of the plateau type, such as Winklebury or Tidbury. Starting from the park gate and going in a clockwise direction, there is a single rampart and outer ditch, which makes an almost right-angled turn and after 50 yds. is interrupted by the Norman castle. On the W. side of the bailey the rampart and ditch is reinforced by a counterscarp bank. This is prominent around much of the northern part of the earthwork and at the termination by the park gate. The double rampart is across the neck of the ridge. No orginal entrance can be traced.
The Castle. The foundation is ascribed to Henry de Blois, Bishop of Winchester, who built and fortified it in 1138. The castle was doubtless of importance during the struggle between Stephen and Maud but thereafter seems to have been a bishop's palace rather than a military castle.
Though by the 14th century some parts had fallen into decay, Bishop Edendon seems to have resided there as late as 1365.
Both inner ward and bailey are of the normal rampart and outer ditch construction; from the rampart of the inner ward to the bottom of the ditch measures 49 ft., and similar measurement of the bailey is 37 ft.
The ruin on the northern side with flint walls 8 ft. thick is doubtless the gatehouse and has an archway facing the moat. Across the moat on this side, in the interspace before the I.A. rampart, a level space is flanked on each side by a low broad bank and ditch and may represent the position of the barbican.
The present entrances between the inner ward and bailey "are probably due to mutilation".
The well is of great depth and diameter (Williams-Freeman; VCH).
The site comprises two main features, the hillfort earthwork and the mediaeval castle. Beyond these, on the N. and W., are a confusion of banks and ditches representing hollow ways and the remains of the park pale (for which see Hants 42 NW 5
The I.A. Hillfort. Williams-Freeman's contention that this earthwork is the remains of an I.A. fortification is clearly upheld by its plan and condition in relation to the medieval work.
It appears to have been a polygonal enclosure formed by a strong rampart and outer ditch and possibly supplemented on the north by a counterscarp bank. The later, however, may also represent the remains of the mediaeval park pale. Most of the southern side and part of the western side of the earthwork has been destroyed or overlaid. On the western side, at SU 41922647 there is a gap with inturning of the bank which is just possibly the remnants of an original entrance: all other breaks in the rampart are certainly modern.
The situation affords no natural defence from its plateau-like position, and northwards from the area the ground rises slightly.
The Castle. The Norman fortification is eccentrically placed within the hillfort. The usual plan has been adhered to and the I.A. earthworks are merely an additional obstruction. The inner ward and the bailey to the south are encompassed by a massive rampart with a correspondingly large outer ditch. The breaks at the east and west sides of the castle are not original features.
The entrance to Merdon Castle was evidently on the south side of the bailey at SU 4207 2637 where there is a break in the rampart. The ditch is not causewayed, indicating that it was bridged.
Williams-Freeman suggests that the structure at the north of the inner ward is the gateway but this is very doubtful. An entrance here would ify the whole concept of the fortification, and there is no evidence of an approach route to the interspace beyond the wide ditch. The bank and ditches in this area which he supposes to be the site of the barbican are merely three quarry pits, not original features.
The structure, at SU 4208 2654, is of flint and stone rubble, up to 8.0m. high, and overgrown with ivy. Most of the facing stone has been robbed. It comprises two two-storeyed chambers with walls from 1.0 m two 2.3 m thick.
On the west side of the inner ward is one long and one short length of revetment walling; in both cases 2.0m. high.
The well of Merdon Castle at SU 4209 2650 is now only 3.3 m deep, due to infilling. The diameter is 2.5 m (F1 NVQ 08-AUG-61).
The site was surveyed by staff of RCHME between December 1993 and January 1994. The recognition of a ploughed-down remains of a substantial rampart in the field to the S of the earthworks has led to a reinterpretation of the shape of the outer prehistoric enclosure. A rise in the medieval ditch in the SW corner coincides with the prehistoric rampart's projected alignment and may represent where it has been cut through by the medieval ditch. On the E, the alignment of the substantial bank which turns nearly at right angles from the prehistoric rampart, closely matches the E apex of Merdon Park pale as depicted on a C16 estate map (Hughes). The survey has added a number of important points of detail concerning the morphology of ringwork construction. Results from a geophysical survey by AML suggest the curtain wall continued at least as far as the 'gatehouse' (Cole). A full field survey report is available in NMR Archives (JD Donachie/June 1994/RCHME Field Survey Report: Merdon Castle, Hursley).
The geophysical survey carried out in 1994 indicated the existence of buried foundations within the castle, particularly within the motte, where the evidence suggests a polygonal arrangement of buildings around an inner courtyard. It is possible that some of these may relate to the castle's later use as a bishop's palace. A series of earthworks and buried foundations lying between the castle and the hillfort ramparts represent more recent use of the castle as a military camp during World War I and World War II (Scheduling Report). (PastScape)