The final agreed order of work was:
Clearing the remains of the cascade (IP2) on dam 1 for photographs
Recording the revetment on the poolside of dam 1
Observing and recording the dismantling of spillway, cascade and
Recording of the pump house (IP16)
Excavation of the grotto (IP17) and immediate surroundings
Back-filling of selected trenches from last year
Examination/excavation of possible observation platform (IP18)
Additional Suggested Options
Assessment of terraced walkways/track ways with some small scale
excavation (1m test pits?) to search for metalling (IP22
onwards depending on number of sites examined).
Complete the excavation of the sunken feature on dam 3 (IP8) down
to floor level to try and really nail what's going on there
Survey plus possible sample excavation of summerhouse (IP20)
Examination of building debris behind the 20th. C. pump house
(IP19) up from dam 7, site marked as conduit on estate map 1790s
Clearance of growth around 20th. grave plot and recording of
reused 18th. century masonry (IP21)
Here's how it all worked out.
An initial chat with representatives from Birch Brothers
established what needed to be done urgently. Afterwards Janine,
David ( for the volunteers) and Colin ( the contractors ) and I
walked up the valley agreeing priorities and discussing methods.
Some time was spent in the pump house trying to assess the scale
of the archaeological work needed. It was generally agreed that
this was a much bigger set of tasks than envisaged at first and
that further talks were needed to agree on methods and timing. The
grotto was considered next with advice being taken about the
safety of the roof. A small trench was marked out within the
grotto to examine the flooring plus further trenches to the west
and north to look at the wider setting. Further up the valley side
the observation platform/ turning circle was considered. The
location of a trench to look at the perimeter was marked out and
the possibility of adding a further trench looked at should the
use of a mini-digger prove possible. It was agreed that a
preliminary earthwork survey and a metal detector survey be done
once the volunteers had cleared the undergrowth and some fallen
branches. Further along the same path a set of stone steps was
noted heading towards a slightly sunken area flanked by possible
historic planting. It was agreed that some of the undergrowth
would be cleared and then a decision would be made about
additional survey and possibly excavation. Finally a visit was
made to the twentieth century grave site and the options for
clearing and recording the earlier masonry on the site considered.
In the afternoon the volunteers helped with clearing the lower
portions of the cascade below dam one. This uncovered a section of
walling previously buried below a fallen tree. Work was then
transferred to the revetment wall on the poolside which was due to
be removed for the new spillway. This was cleaned up then recorded
with a drawn plan and photographs.
Remains of the cascade and retaining wall below dam 1 brilliantly
cleaned by the team...
who were them moved up to the pool side revetment for further
cleaning and here it is ready for planning and photography.
An early start to meet our metal detecting specialist and lead him
down to the valley. Initially the area around the grotto was
examined and signals marked with yellow flags. He then moved on to
the possible observation platform where a team of volunteers
cleared the ground. We had hoped for considerable evidence of
either vehicle movements or clues to the location of built
structures or seating. Unfortunately signals were thin on the
ground so no conclusions could be reached.
Later on the strimming crew moved onto the summer house site and
cleared around 8 square metres of dense vegetation whilst leaving
in place the rhododendrons which flanked the site. A small test
pit was then dug through an overburden of roughly 20 cm. of leaf
mould derived mainly from bracken. This revealed a small exposure
of what appeared to be a well laid paved stone surface. As there
was no archaeological features seen within this build up of plant
material we propose tomorrow to clear more of the surface and
clean the approaching steps under archaeological supervision.
At the grotto site a base camp was set up and an initial
trench marked out taking in an area immediately in front of the
larger opening and reaching across the terrace; only0 a little way
down the slope. About half of this was excavated to a depth of
around 10cm revealing the front edge of the brick paving from the
inside of the grotto and a surface of loose brick and stone rubble
beyond. This will be examined further tomorrow. Once the
volunteers had finished we began to plan the layout of the grotto
by offsets. We also visited dam 1 and noted that one of the large
stone slabs from the upstream end of the spillway had been lifted.
We’ll pop back and photograph this tomorrow.
The newly discovered steps going up.. to where?
To here, the newly opened test pit with the first sight of the
stone flags, base for a summer house?
The area in front of the grotto strung up and the 'topsoil', if
that is what you call a hard dry mixture of dust and rubble,
The day began with a visit to dam 1 to photograph the large stone
from the spillway and the lower part of the cascade cleared on
Monday. Back at the grotto the time before our volunteers arrived
was taken up with completing the ground plan of the grotto. As
there were several new faces time was taken to give a briefing on
the key aspects of the project including health and safety.
One group began on the grotto where we continued clearing the
area in front of the main opening. It became clear that there was
a distinct area of disturbance with traces of burning, potentially
a hearth or fire pit. Other indications of possible extensions to
walling also emerged. Later in the day work was started on the
alignment of stones to the north to establish whether they
represented a wall or a pathway.
The sheet piling inserted so that the draw down of the water can
begin once construction starts.
The second group began up on the site provisionally named ‘the
summer house ‘. Unfortunately we under estimated the scale of
finds that were to be made and we found we had to undertake
considerable reorganization of resources to cope with what became
a collection of 21 pieces of ironwork, primarily large iron nails
and spikes. Work began on clearing a deposit of leaf mould largely
derived from bracken from a series of stone slabs. One of these
proved to be inscribed with rather coarse lettering which bore
traces of a blue pigment. More work needs to be done on cleaning
the stone but it appears to include three sets of initials ending
in the letter K above the line B G. GUARDIAN 1838 ( the last two
digits are uncertain) below that are what may be a couple of lines
of verse. This is obviously an important find and may require a
broadening out of the investigation of this site.
In front of the grotto a distinct patch of dark earth begins to
The setting for the 'summer house' and hand excavation begins as
the density of finds -particularly ironwork- becomes evident.
Plotting of finds underway when suddenly we have lettering and a
very careful clean up begins.
After checking in with the contractors we drove up the valley and
completed the removal of the topsoil from the extended area in
front of the grotto. The volunteers finished cleaning this area
before turning their attention to the deposit of charcoal rich
loam. A large part of this was removed to uncover a dense layer of
small rubble packed with mortar which may be construction debris
for the grotto. Inside the grotto a sizable portion of the floor
was cleaned showing that there was a well preserved brick floor
flanked by an edging of small rubble in lime mortar. As this floor
level is only a couple of centimetres down we plan to uncover most
of it so as to better understand the internal layout of the
grotto. Work clarifying the line of the path to the north is
Conditions up on the summer house site remain quite
challenging with water breaks being taken every hour. We continued
to recover large quantities of iron work, mainly nails and spikes.
The inscription was covered over for most of the day whilst the
trench was extended to the east. We now appear to have defined the
sides of the paved area to east and west but there are still a
number of large slabs disappearing under the baulk to the north.
The decision will have to be made for next week as to how far we
A trench was started on the south west perimeter of the
observation platform/ turning circle. This revealed, down to a
depth of about 30 cm, a loose mix of dry silty loam and small
angular rubble with virtually no covering of topsoil. It certainly
has the appearance of a fairly recent leveling operation with no
evidence of earlier structures or deposits.
A major effort to clear and record the remarkably well preserved
floor to the grotto whilst at the same time coming close to
completing the excavation of the area immediately outside which
has given us a few small finds, notably some early glass which may
help with dating. I suspect the path may be an early effort (
1960s? ) by the Trust to put in some hard landscaping when first
opening for the public. The trench in front of the scour tunnel to
dam 7 was backfilled. Recording of the trench in the turning
circle was completed.
Up at the 'summer house ' the trench was cut back to the
currently defined limits to uncover more slabs and metalwork.
There are some hints of preserved timber which we will have to pay
special attention to next week. We took the decision to recover
the slabs with the loose topsoil/leaf mould generated by the
bracken to provide some security for the site over the weekend.
(We tried a plastic tarpaulin but that generated too much
Continuing excavation as we climb the steps.
The end of the session on Friday cleaning up with stone slabs
disappearing under the bank to the north.
...and then it was the weekend of stifling temperatures and World
Cup football... except...
The inscribed slab, slightly modified image.
... for the following correspondence:
"In 1838 Croft Castle estate
was being held in trust for the three surviving children of
Somerset Davies' daughter Ann and her husband James Kevill. Ann
had died in 1826 and James in 1831. So the two boys (Edward
Hammond and William Trevelyan) and daughter Anne Isabella ( just
found there was a surviving daughter so no dates so far) were in
the care of Somerset's widow Ann and two Guardians ( Rev John
Moore Stevens and Langham Rokeby). I would guess that Ann moved
back into Croft but not yet whether the boys went away for
school. Ann lived into her nineties and present when William
came of age and inherited the Davies' estates in 1847 (Edward
died in 1844 just before his majority)
In 1838 Edward (the heir) was
about fifteen and William about twelve (no age for sister) so
the K could refer to something all three had set up or was about
them. . 1838 was also significant because it was the year Ann
and the guardians successfully petitioned for Edward to add
Davies to the Kevill name as per Somerset's will. They had to
petition again in 1847 for William so the Davies only got added
to the heir's name.
The estate appears to have been
managed by trustees and presumably an estate manager. I have not
found any records about what activity was going on in FPV
at that time but from what I have gathered the picturesque had
gone out of fashion and with the expansion of the empire plants
were being brought back from the far east and gardens
concentrated on herbaceous borders and show of flowers. I
can imagine the children going to explore FPV - more suited to
boisterous youngsters than neat borders!"
Many thanks for this and yes how interesting/exciting. I
must emphasize that the reading of the date is only
provisional but this information could support a date of
1838 and suggests the initials could be I.K.
(Isabella), E.K. (Edward) and W.K. (William) plus also
giving us a context for 'B.G. Guardian' although I'm not
sure what 'B.G.' stands for. It will be fascinating to
unravel the rest of the inscription and establish a context
for this highly unusual find. We have yet to give the stone
its final careful ... very careful clean and then we
have the option for further advanced imaging tasks.
Arriving on site late morning we began by calling in at dam 1 to
review progress with the contractors. The stones on the west side
of the former spillway had been removed by hand revealing a
section through the made up ground beyond the spillway to the
west. This uncovered bands of silty rubble alternating with
cleaner sandy spreads. The section was cleaned and photographed.
After lunch work continued on the grotto. A new area was opened up
below the small bay on the north side. Access was restricted
because of the presence of timber props but an area of around 0.5
m by 0.75m was examined. Below a large dump of rubble and modern
mortar was an area of compact stone fragments but there was no
evidence of paving. A post pit was excavated which may have been a
setting for a prop or possibly support for a bench. In front of
the main grotto a small test pit was dug into the underlying
deposit of clay and rubble so that the sequence of construction
could be examined. Planning of all features associated with the
grotto was completed and detailed photographic record of all
The volunteers were sent up to the 'summer house' site where they
did a marvelous job in difficult conditions of clearing the
undergrowth for an additional metre to the sides and to two metres
heading up the slope revealing some interesting changes in
gradient. We also cleared a new access path to relieve of the need
to use the existing stone steps which are in rather fragile
An extraordinary effort goes into clearing more vegetation 'up
Work began continuing with the recording of the grotto by taking
levels on to all significant layers and structures. Attention was
transferred to the dig down on dam 3 where the upper two trenches
were back-filled and leveled with the assistance of the volunteer
team. Everyone then adjourned to the upper section of the valley
where the extension to the trench on the summer house ( IP20 ) was
set out. Work then concentrated on removing the upper layer of
leaf mould from the whole area. Once this was complete a start was
made on excavating the underlying hill wash, a loose grayish clay.
This uncovered the beginnings of two flanking walls set at a
slight angle so as to frame the central slabbed area. It seems
likely that these will meet up with a retaining wall which will
define the northern boundary of the site. More architectural iron
work was recovered but also of significance was the presence of
traces of timber boarding which may represent preserved remnants
of a wooden structure or possibly benching.
The final photographic efforts on the grotto recording elevations
The second main focus for the day was the clearance of undergrowth
from around the grave site on the summit of the adjacent hill.
Centred on the grave of James Croft, the eleventh baronet, two
concentric roughly positioned stone rectangles were cleared along
with the surrounding area. This showed that two separate sets of
moulded stones had been used in a fairly rough and ready fashion
to create these 'enclosures'. It seems likely that these stones
represent two different stages of a stepped plinth which had been
dismantled for reuse. The fact that the layout has been positioned
with a degree of carelessness suggests that no great effort has
been put into this construction and therefore that these heavy
stones have not been brought from far away. Other stones
positioned as part of the layout include two blocks of old red
sandstone and a remarkable moulded block which appears to be the
base of a Gothic screen with attached column. The quality of the
stonework and the level of preservation indicate that this may be
a nineteenth century piece. Once the area was cleared preparations
were made to start planning the complex.
An initial visit was made to dam 1 and an area on the south side
of the spillway where a large tree root had been removed was
examined. Some additional details of the existing walling were
noted and photographed.
Planning was started on both the summerhouse (IP20) using a
planning frame, and the grave site (IP21) by offsets. Once the
volunteers arrived one person was sent to help with the planning
at the grave site whilst the rest made a start on removing the
remainder of the overburden from the north side of the paved area.
This resulted in the discovery of further architectural iron work
including a large hinge socket. Most remarkable however was the
exposure of the end wall which not only proved to be apsidal but
was also capped for almost its full perimeter by heavily decayed
On the grave site further stone cleaning was done, particular
attention was paid to the rough dressing on the inner faces of
some of the moulded blocks again suggesting their location within
an enclosed plinth.
Before the volunteers arrived recording was finished at the grotto
and context records completed for the ‘summer house ‘
afterwards planning was continued both here and on the grave site.
The volunteers helped complete the final stages of excavation
on the ‘summer house ‘ and then began the big clean up for
photography. This was a challenging task especially after the
early morning rain but brilliantly carried out by the volunteers.
Having thought about the layout of this site it seems to me that a
better appellation for the site would be an arbor. I envisage an
over arching framework of curved timber members to create a shell
like cover above the paved area. In the afternoon both sites
were photographed in detail. At the grave site each stone was
numbered and photographed individually. At the same time a group
of volunteers completed the back filling on dam 7 and then moved
on to clear some undergrowth in the area indicated on the late
eighteenth-century estate map as being the location of a
‘conduit’. Extensive spreads of building debris were uncovered.
The last bit of excavation round the edges of the
in the jungle nobody goes except those searching for the conduit.
After taking down the gazebo at ‘base camp’ we paid a final visit
to the contractors on dam 1 to examine the area cleared below the
cascade where the substrate of loose rubble was recorded. This was
the final visit as part of the current watching brief. A portion
of the retaining wall to the pool has yet to be removed but is
unlikely to provide any new information. We also had a brief
conversation with the contractor working on the repairs to the
The last of dam 1, the area of the cascade is covered over.
The remainder of the morning was given over to recording . The
stone by stone planning of the grave site was completed and
measured profiles of the key mouldings drawn. Similarly on the
‘summer house’ planning of the paved surface was finished and a
profile of the walling and timber work drawn. There was
insufficient time to plan in detail the steps so these were
recorded with a series of scaled vertical and oblique photos.
Finally a comprehensive set of photos were taken of the building
debris revealed by yesterday’s clearance of undergrowth from the
site of the conduit.
The volunteers got on with back-filling the trenches next to the
grotto whilst leaving the paved interior clear. We had discussed
options for covering and protecting the ‘summer house ‘ site until
such time as decisions can be made regarding its future and we
left them as they started to work this out.