Otford & District Archaeological Group (ODAG)

The Romano-British Cremation Cemetery at Frog Farm, Otford, Kent, in the context
of contemporary funerary practices in South-East England by Clifford P. Ward 1990

Frog Farm Site, Otford

The importance of the Pilgrims’ Way/North Downs Ridgeway is underlined by Margary (1965) who commented that ‘this route is certainly one of the most important prehistoric thoroughfares in the south-east of England, for it was from the earliest times the main link between the Continent and that central downland area of Wiltshire which played such an important part in the settlements of early man in this island’. The Darent crossing was one of only five river crossings required between Folkestone, Farnham and Wiltshire. It comprised a ridgeway and terraceway below the steep escarpment of the chalk downs, and ‘formed a valuable link between various Roman sites at the foot of the downs’. Thus, in the absence of any known east-west Roman road through Kent south of Watling Street, it may be assumed that the Pilgrims' Way formed a part of the main communications network of southern Britain.
   The geology of Frog Farm is dictated by the river Darent which in its course to the Thames broke through the chalk North Downs at this point. Underlying the chalk is a stratum of gault clay forming the flat floor of the Vale of Holmesdale, while superimposed on this is the alluvium of the river flood plain, considerably greater in extent than befits the present water flow. The actual cemetery site is on a spread of Second Terrace river gravel, and stands at O.D. 220 feet (Dines et al 1969).
   At this point the Pilgrims' Way utilises a slight ridge to climb westwards from the Darent crossing, and the cemetery would have been visible from all directions. (map.)
   Apart from a small group of Roman pots unearthed north-east of the farm in 1906, and presented to Maidstone Museum (Acc. not traced.), the first indication of Roman occupation of the area came in 1927 when a contractor was laying a gas pipe beneath the Pilgrims Way. A ‘very hard surface’ was encountered, which was identified, on very dubious grounds, as a Roman road. The fact that the Sevenoaks Council granted an additional £35 12s 6d to cut a trench through it gives some indication of its strength and/or extent (G. Ward c.1930), but unfortunately nothing further is known, despite a watching brief kept on subsequent holes in the vicinity. Its massiveness suggests a structure rather than a road, just possibly a shrine or further mausoleum, or even the elusive villa long-sought to the west of the river.
   At about the same time a burial group, dated to the second century A.D., was unearthed at Frog Farm, vessels from which are now deposited in Sevenoaks Museum. (Acc. K1394).
   The site of the settlement at Wickham Field, the former Isolation Hospital, was subjected to trial excavation in the 1920s & 30s when evidence of a settlement was found (Clarke & Stoyel 1975), and a series of investigatory trenches was opened there by the newly-formed Otford & District Historical Society Archaeological Group (ODAG) in 1965.
   At the end of that year, during construction of a shallow drainage trench for a potato-clamp immediately outside the back garden of Frog Farmhouse, a group of Roman pots was discovered by the farmers, and ODAG was invited to investigate. This group was identified as a cremation burial group (Grave 1), and in another part of the trench a substantial ragstone wall was noted which exhibited a 45° corner and was pierced by a somewhat enigmatic flue. This subsequently proved to be the best-preserved portion of an octagonal mausoleum (see below).
   In view of the shallowness of the grave bottom, about 16 inches below the surface, there was a strong possibility that any further disturbance of the site would cause serious damage to any other artefacts nearby, and, with the encouragement of the Booker family, the farmers, a decision was taken to excavate the site in order to rescue any other remains which might have survived, and to endeavour to trace the angular building which disappeared beneath a manure heap.
   A 10 feet by 10 feet grid was constructed aligned on the adjacent garden boundary (a gooseberry hedge) and work proceeded over the next two years. Due to farming needs it was not possible to extend the grid northwards or westwards apart from in one area (E 6), and the garden precluded any extension to the south, although the hedge was curtailed slightly (See Plate 1).

Plate 1. Frog Farm Cremation Cemetery and Mausoleum - Schematic Plan

   Over the area excavated a series of grave-groups was located immediately beneath the plough level, mainly in shallow, ill-defined scoops in the underlying brickearth, and from the damage occurring to all of the larger vessels, it appeared that the ground surface had been lowered by up to 1½ feet since the cemetery was in use.
   Although no regular pattern of deposition could be ascertained, unlike the "lines" suggested at e.g. Borough Green (G. Payne 1899) and other places, the groups were generally spaced at 3-5 feet between centres, with, in only two instances (Groups 7 & 57), suggestion of two urns buried together.
   The urns had all been placed in holes dug into the gravel sub-soil, at least three being lined with flints. They were set in an upright position with ancillary vessels, where present, grouped round them, having in 8 instances a platter used as a lid to seal the mouth of the urn. In one case only (Group 18) two fragments of roof-tile (tegula) had been added to the plate, presumably to enhance the effect.
   The damage to the urns in particular, indicated that they had not been buried deeply, as many of the vessels had been crushed, with their rims having fallen on to the bones, leaving the upper portion of the walls upstanding. In many cases these had been ploughed off subsequently, rendering complete restoration impossible, but retaining the most significant identification features i.e. the rim types together with reasonable indications of the overall size and character of the vessels. Some of the urns, however, had been virtually ploughed out, leaving only the base in situ. All were invaded by grass and other roots, and many were badly distorted.
   Some attempt at analysis of the composition of the grave-groups has been made (Figs 3-9), but there is little indication of strong trends in either pottery associations or dating in different areas of the cemetery, although Patchgrove urns preponderate, and of ancillary vessels, Samian appears to form a higher than usual proportion.
   The majority of the samian ware has been identified and dated from 150-200 A.D. In common with most others the cemetery does not contain any decorated samian ware, which suggests that it was deemed unsuitable for funerary purposes. The predominant form is the platter (Dr 18/31 etc), of which there are 26, with cups (Dr 33 only) numbering 17 (see analyses below).
   The piece de resistance came from Grave 45 where a two-handled samian cup of rare form imitating a metal vessel was placed within a greyware pie-dish, and had a small samian dish (Dr 36) over it. The cup was submitted to Dr Grace Simpson who appraised it as Dr 34 having both body and slip of clays containing very fine mica, as found at Lezoux. The slip is black, through a reduction process in firing. The vessel is dated A.D. 138-180. Of the original two handles, one had become detached prior to burial and the scar appears to have been deliberately smoothed off to some extent, end it is tempting to think of the cup as perhaps rejected after accidental damage in a nearby villa, and rescued for use by one of the local peasants, for whom it became a favourite drinking vessel, and ultimately accompanied him to the grave.
   Towards the north-east, the burials were further apart, shallower and more fragmentary. It appears that the land surface had been subject to greater erosion, presumably through ploughing, although at slightly lower elevation than the mausoleum. The absence of any remains whatsoever in Grid B6 may indicate the extent of the cemetery on the eastern side, although no trace of any boundary was encountered.
    In view of impending changes in agricultural operations it was decided to conclude the excavation by attempting to recover the plan of the building which until then lay beneath the manure heap. This was accomplished over the last weekend of April 1968 (plan) and immediately backfilled. Due to the arrival at that time of the writer’s baby daughter, the overall appearance of the mausoleum is known only from photographs (Plates 3, 4 ) and thus some of the details, e.g. the existence of two small pieces of lead, suggestive of coffin sheeting of Keston, (not preserved among the finds), said found in a slight depression at its centre cannot be attested by the writer.

Plate 3. Frog Farm, Otford. Mausoleum from the south-south-east with entrance on right.

Plate 4. Frog Farm, Otford. Mausoleum from the west-south-west showing badly robbed
                 foundations and earlier cremation graves in situ in foreground and to the right.

  The structure was found to be badly robbed over most of its circumference, with only that portion adjacent to the entrance standing above the foundation which was composed of 2 courses of flint nodules bedded in gravel 36 inches wide. It was octagonal in plan with an entrance only 20 inches wide in the centre of the east-south-east face. Above the flints was one course of neatly-laid faced ragstone blocks set in white mortar. These formed walls 34 inches wide. The wall/foundation did not continue below the entrance, which had been used at some time, apparently late in the Roman period, as a flue.
   The comprehensive robbing had removed all traces of floor and superstructure beyond a few small opus signinum and tile, both tegula and imbrex, fragments. No indications of postholes or sleeper beam slots were noted in the ragstone wall, but the insubstantial nature of the foundations suggests that the superstructure was either timber-framed, or, if stone built, limited to one storey. Nothing can be adduced about the roof, but the few tile remains point to a tile cladding rather than thatch, and the ragstone for at least the lower courses of the walls must have been transported a minimum of two miles from the lower greensand ridge to the South, suggesting a building aspiring to some degree of ostentation. ODAG's identification of the structure as a mausoleum was confirmed by the Museum of London (N.C. Cook pers. discussion).

[NB by adding in the additional groups found in 2006, and the subsequent Nil groups west of the N-S track to the fields, it should be possible to make a better guestionmate of the likely size of the cemetery - perhaps as a Supplement to include the commercial archaeological contractors work and ODAG's own. We came to the conclusion we did not want any more, provided they were not under threat!]

Funerary Practices       To Frog Farm Cemetery contents page      Discussion