Belfry Bulletin

Search Our Site

Article Index


A Map of the Three Streams which enter St. Cuthbert's Swallet.

Surveyed By Roger And Frances Stenner In 1996.


Frankie and I were in the final stages of making the survey, in head-high rushes near the tramway. After this time there wasn't much need to talk much, we were just getting on with the work.  A family - couple, two small boys and a big dog - came by, and stopped to watch.  Of course the adults and the older boy were far too reserved to say anything, but the smaller boy's curiosity got the better of him, and he asked us what we were doing.  We told him, and the father couldn't hold his tongue.  "Why don't you just buy a map?"  Why indeed!  Up-to-date large scale maps of the area are available in theory, but the price is high enough to make them effectively unavailable.  The 1:2500 ("25 inch" series) O.S. map of ST 5450 and ST 5550, published in 1963, showed only two of the three streams connected with the cave.  The 6" to 1 mile O.S. map, ST 55 SW, published in 1961, showed only one stream, which looked like the last part of Ladywell stream tagged onto a piece of creative fiction.  So even in the days when maps were still available, the situation was unsatisfactory.

In 1967, when I needed to show the location of the three streams, the only option was to survey them myself. Since I only needed to illustrate the streams to the south of the line through Plantation Swallet and Maypole Overflow Corner, the surveying was limited to this area.  The map was included in a description of the hydrology of the cave (Stenner, 1968).  A later drawing, extended to include the streams as far as the Mineries Pool, was made after re-plotting the 1967 data at the scale of 1:2500.  This map was included in the draft of Caving Report No.13 Part L, which was never published.  To illustrate the surface hydrology, the St. Cuthbert's Report contained a map which included incorrect features from the 1961 6":1 m O.S. map (Irwin, 1991, Fig.7, p.64).  At some time between 1985 and 1991, Plantation Stream changed its course. From being a mess, the map situation had become an obsolete mess.

The description of the surface hydrology, also written up in the draft caving report, had likewise become obsolete. Stream studies needed to describe the current situation began in 1994, and once again a map was needed.


In 1996, Frankie and I began surveying the streams, using Suunto gear belonging to the club, and a Fibron tape (a genuine Cuthbert's veteran!).  Closed traverses created a framework on which to hang the data.   As to calibration, something new was tried. I took one wall corner near the Belfry, and a second near Fair Lady Well, both of which were clearly identified on the 1963 1:2500 map.  The justification for using these two wall corners will be examined later.  The plan co-ordinates of the two wall corners were measured on the map to 1/4 m.  The angle needed to yield the best closure from the first corner to the second was the compass correction for the day.  Spreadsheets by Lotus were used for all the computations.  Four different routes from one corner to the second were surveyed, each one in a single day.  One of them contained a mistake, but the other three gave mutually consistent data.

Early attempts by Ellis to survey in the St. Cuthbert’s Lead-works in Priddy Minery failed because of magnetic disturbances.  In the 1967 survey and in the present survey, the method described for detecting and avoiding magnetic anomalies (Irwin and Stenner, 1977) was used successfully.


25” Map of the Cuthbert’s area


V St. Cuthbert’s Swallet Entrance Shaft

1.                  Sample site, Mineries Pool outflow stream

2.                  Sample sitte, St. Cuthbert’s stream main sink.

3.                  Sample site, Fair lady Well.

4.                  Maypole Sink.

5.                  Intermittent.

6.                  Maypole overflow corner.

7.                  Intermittent tributary complex.

8.                  Aqueduct.

9.                  Source of former (pre 1985) Cuthbert’s stream.

6” Map of the Cuthbert’s area


8 St. Cuthbert’s Swallet Entrance Shaft

1.                  Sample site, Mineries Pool outflow stream

2.                  Sample sitte, St. Cuthbert’s stream main sink.

3.                  Sample site, Fair lady Well.

4.                  Maypole Sink.

5.                  Intermittent.

6.                  Maypole overflow corner.

7.                  Intermittent tributary complex.

8.                  Aqueduct.

9.                  Source of former (pre 1985) Cuthbert’s stream.

10.              Former Pool, now breached.


Seven closed traverses were examined, and the results are shown in the following table.  Lengths are in metres.




Horiz. Misc.

Vert. misc.

Slope Misc.



















































The method developed for the St. Cuthbert's Survey for closing networks (Irwin and Stenner, 1975) was used.  Seven survey routes from the first corner to the second were calculated and tabulated. The resulting co-ordinate changes, given in the following table, show that route 7 contained a substantial error. Details of the seven routes led to the identification of the section of route 7 which contained the error, and the misclosure error was distributed along this section.  Spot heights were calculated using the height of the top of the Entrance Shaft.  The 1967 survey data were recalculated, and were consistent with the new data.





Mean change of  6 routes (metres)

Standard deviation of  6 routes

Change given by route 7

Difference between route 7 and Mean

Change from 1963 1:2500 O.S. map

Difference from Mean



















While the survey was being made, the positions of present and former stream courses were measured. The intention was to try to reconcile the map being made with the 1961 and 1963 maps, and with earlier maps, and hence to establish the sequence of the stream changes.  Copies of 6":1 m O.S. maps of 1883 and 1903, and the 1:2500 O.S. map of 1903 were also available.  During this stage of the work, Chris Richards of Weston-super-Mare (and Axbridge C.G.) gave me a photo-copy of a map which was probably made in about 1860, in connection with the Ennor/Barwell legal dispute which came to court in 1860.  The map shows, at a scale of 1:2500, the hydrology of the Chewton and Priddy Mineries. It included these features: Priddy Pot (=Potable?) Water Stream, Wheel Pit (at Maypole Overflow corner) Swallet (at Plantation Swallet) and another Swallet at Five BuddIes Sink.  The Mineries Pool did not exist, and upstream of its present position were two Reservoir Ponds, supplying water to the Washing Works near Five Buddles Sink.  The map showed the route of the stream in the valley under the Reservoir Ponds, so it is possible either that the ponds were relatively new in 1860, or that they were drained from time to time.  Efforts to reconcile the information in the various maps will now be discussed.


Fair Lady Well and the Ladywell Stream were shown (as Spring and Priddy Pot Water Stream) on the 1860 map. From the spring to close to Plantation Swallet, the stream followed a bowed route, slightly to the east of the present route.  Beyond Plantation Swallet, the map of the sinuous course can be superimposed perfectly on the present map of the stream, drawn to the same scale.  When, in about 1850, Ennor took on the lease of the Priddy Minery (later renamed St. Cuthbert's Leadworks) one of the conditions stated that the Priddy Pot stream must not be used for mineral purposes (Tilly, 1967, p3). Before being diverted, Ladywell Stream would have flowed into the depression and into the cave; so in about 1850 the Ladywell stream was already an artificial feature.  Close to Plantation Swallet the 1860 map showed Priddy Pot stream crossing a stream to Plantation Swallet, so a Ladywell aqueduct must have been in place then.  The date when the Ladywell stream course was constructed is not known.  Perhaps parish records may contain relevant information. The reason why the stream followed a sinuous course from Plantation Swallet to the comer near Mr. Foxwell's Drive is straightforward; it followed the hydrological contour around the depression. This shows that the fundamental geography of the depression has not changed since before 1860.  The impression given by the 1860 and 1883 maps, that the area between the Belfry and St. Cuthbert's Lead-works was featureless, is misleading.

In 1860, the stream from Fair Lady Well was shown to have been augmented by drainage from Chewton Minery. It would have been a larger stream than it is now.  Annual maintenance would have sealed leaks such as those which currently allow the stream to shrink so much between the source and the Belfry.  As late as 1954 the stream still reached Eastwater Lane, and vanished in a marsh on the far side of the road to Priddy Green. In its prime, it flowed as far as the field opposite the Queen Victoria Inn.  Since the new Belfry was built, water flowing from the Drinking Pond does not reappear on the surface.

The 1860 map showed five streams leaving the area around the present Mineries Pool.  The first was water from Fair Lady Well.  The second was drainage from a part of the Chewton Minery which joined water from Fair Lady Well, as described above.  The third flowed into the Priddy Minery, and sank in the position of the Maypole Sink, with overflow to the South.  The fourth also flowed into the Priddy Minery, to the east of the third stream, to the Washing Works, where it was joined by water captured by leets on the slopes of Stock Hill.  The fifth, Dr. Somers Course, led from an area near Fair Lady Well to the Wheel Pit near Maypole Overflow comer.  The present studies have failed to find the exact location of Dr. Somer's course; but from the comer, two courses were shown which can still be identified; west to Plantation Swallet, and east to St. Cuthbert's depression.

The history of the five streams following the construction of the Mineries Pool, and the later extension of the Chewton Minery tramway from the pool to the Priddy Minery smelter, is not clear.  By 1953, water from Fair Lady Well had been diverted into a new artificial course from the spring to close to Plantation Swallet, where it rejoined its earlier course. This is the present Ladywell Stream, as shown correctly on the 1963 1:2500 O.S. map.  Also by 1953, the water which had formerly made up the other four streams emerged from a single channel underneath the tramway.  This was Plantation Stream, which flowed in an artificial course (via a swampy pool about half-way) to Maypole Overflow comer, where it turned west into Plantation Swallet.  This stream was also shown correctly on the 1963 1:2500 O.S. map, but the 1961 6":1 m map was incorrect.  The sequence of changes between 1860 and 1963 is uncertain, and the new surveying has not removed the uncertainties.

The (approx.) century between 1860 and 1953 has given the most difficulties, when attempting to reconcile maps.  In the 1883 and 1903 maps, the depicted stream geography was different from that shown in the 1860 and 1963 maps.  Essentially the same single stream was shown in all three available maps. (Mineries Pool appeared first in the 1883 map, the tramway in the 1903 maps; locations of buddles in the depression, and small ponds associated with them, were different in all the maps).  The single stream shown in the 1883 and 1903 maps was also reproduced in the 6":1 m map of 1961.  In the 1:2500 map of 1903, Fair Lady Well was shown as a small stand-alone pool.  About 25 m to the south-east, a stream rose, close to the present Mineries Pool Outflow.  This stream followed a looping course to the south-east of the "1860" Priddy Pot stream, to close to Plantation Swallet (which was not shown in any form) from which location it continued, following the old Priddy Pot stream course (which was drawn much less sinuously than in the 1860 or the 1963 1:2500 maps, which may indicate a lack of care in draughts mans hip).  Just how this stream can be reconciled with the 1860 map or with the 1963 map is anyone's guess!  To take the 1903 map at its face value, if the Mineries Pool outflow was indeed mixed with the flow from Fair Lady Well and taken to Priddy as "potable water", the resulting water quality would have been dreadful. And if this was indeed the situation, the Pool Outflow stream was already leaving the valley.  So why was it necessary to divert this water into Plantation Swallet to drain the valley? Perhaps more information will come to light to clarify this muddle.

There is still ample evidence of a large number of minor (and some major) drainage channels in the area downstream of the Mineries Pool.  Evidence presented in connection with the Ennor/Hodgkinson legal dispute showed that at the time, large quantities of water were used by buddIes at the Priddy minery to concentrate ore-bearing material, that the process resulted in serious contamination of the Axe, and that flows of water entering swallets in the depression and Plantation swallet had been measured.  It seems likely that the outflow from the Mineries Pool could be diverted by means of sluice gates either to buddIes in the depression or to a route to Plantation Swallet.  After the conclusion of the legal dispute, which forbade the disposal of contaminated water into swallets, the water needed to prepare the ore was continually re-circulated.  For a number of years, valley floor material was smelted directly, with no pre-treatment, and the volume of slag generated by this process was substantial.  When the Chewton Minery tramway to the Mineries Pool was extended to the Priddy Minery, it covered the site of a group of buddles in the depression, and it also covered drainage routes from Stock Hill to the Priddy Minery.  The tramway bank incorporated a "siphon" for the outflow from the Mineries Pool.

Around 1900, the entire Plantation stream was diverted into the artificial course shown in the 1963 1:2500 map.  This work could not have been carried out without first diverting the Ladywell Stream from the course shown in the 1860 map, but this diversion could have been carried out at any time between 1860 and about 1900.  After the St. Cuthbert's Leadworks closed in 1908, annual maintenance of the Ladywell Stream was continued by villagers until about 1960.

Between 1985 to 1991, the flow of Plantation Stream switched completely to a "new" route from the swampy half-way pool into the depression.  Here it merged with the former St. Cuthbert's Stream, and flowed on into the cave.

During the time when Plantation Stream was sinking in or near Plantation Swallet, St. Cuthbert's Stream (about one third of the size of Plantation Stream) was an entirely separate stream sinking near the Entrance Shaft.  There was no indication of this stream on any of the O.S. maps which have been referred to, although earlier versions of this stream were shown on the 1860 map.  The stream could be followed for about 170 m to the north-east, where it emerged from a marsh.  In the St. Cuthbert's Report, Dave Irwin expressed the opinion that this water came from the Mineries Pool (Irwin, 1991, p.63-64).  Recent studies have produced new data.  Between September 1995 and January 1996, a quite unprecedented surge of sulphate ions passed through the Mineries Pool.  At the swallet, the sulphate was shown to have been diluted by mixing with water from the former St. Cuthbert's inlet stream, which had normal sulphate levels.  By the Method of Mixtures it was possible to use sulphate levels and discharge measurements, before and after mixing, to calculate the sulphate levels in the hidden inlet stream.  Sampling was continued to February 1997, and the sulphate surge was completely absent from the inlet.  This is proof that the water did not pass through the Mineries Pool.

The 1860 map showed two leets (drainage channels) on the south-western slopes of Stock Hill, where the underlying rock is at first impervious Old Red Sandstone, then Lower Limestone Shales.  One leet took water to the lower Reservoir Pool on the Chewton Minery.  Flow from the second leet was divided between the two leadworks.  The area drained by the two leets is substantial, and without artificial interception, the water would have drained as "interflow" to the Priddy Minery. Earlier flow routes are now buried under the tramway from the Pool to the Priddy Minery smelter, and under heaps of slag.  But in wet weather, water can be seen welling up from beneath slag heaps, and flowing into the very marsh from which the former St. Cuthbert's Stream emerged. The chemistry of the former inlet (measured directly from 1966 to 1973 and by 6 samples taken in the recent studies, and from recent data calculated by stream ratios) is similar to that of another spring nearby which drains Old Red Sandstone; Fair Lady Well.  It is my conclusion that the buried flow-lines from Stock Hill to St. Cuthbert's Stream still operate, and that drainage from this part of Stock Hill is indeed the major source of the St. Cuthbert's Stream inlet.


The reliability of two maps of the area was checked (they were both originals, not photo-copies). The maps in question were the 1961 edition of the 6":1 m O.S map, and the 1963 edition of the 1:2500 O.S. map. First the north-south and the east-west grids were checked. Next the diagonals were measured to check for angular distortion.  Then the checks were repeated on several parts of the maps to check for planar distortion.  Both maps passed the tests.  The co-ordinates of 11 objects (such as wall corners or junctions) were measured, and 2 point-to-point distances were measured.  Measurements were estimated to the nearest 0.1mm equivalent to approx. 2 m on the first map, and to 1/4 m on the second.  Lengths had therefore been measured to 4 m and 0.5 m respectively. From the results, the differences in east and west co-ordinates were tabulated, and the plan differences were calculated.  The mean difference was 9.3 m, with a standard deviation for 11 values of 6.40. The difference for one important corner, the one near Fair Lady Well, was 25.5 m.  Without the distortion caused by this exceptional value, the mean of the remaining 10 differences was 7.7 m (S.D. 6.04).  The difference in the bearing of the wall between the two corners was 3¢.  A wall here is known to have been moved, but it was obviously very important to know which of the sets of figures was correct.

Two distances were measured on three maps; the 1963 1:25000 map and the 1961 6":1 m O.S. map, as already mentioned, and a copy of the 1903 1:2500 O.S. map.  The lengths in the present (1996) survey were also calculated for the same two distances.  The legs were chosen so as to include both wall corners which had been chosen to calibrate the compass.  The two distances were:

1.                  From the lower wall corner to the wall junction behind the Belfry, and

2.                  From the higher wall corner to the middle of Fair Lady Well.

The results (in metres) are shown below:






Lower corner to wall junction

Upper corner to Ladywell









The figures suggest that the 1963 1:2500 map was, in this case, much more reliable and accurate than the other O.S. maps.  They also confirm that the two wall corners originally chosen to calibrate the compass were satisfactory in this particular case, but there had been a high risk of making a serious error.  Field results have once again demonstrated the crucial need for taking very great care when calibrating compasses for cave surveying.  The statistics quoted here quantify the reliability of measurements which have been made from large-scale O.S. maps.  If greater precision is required, the use of satellite location technology will be necessary.


Irwin, D.J. 1991 St. Cuthbert's Swallet. BEC, pp82.

Irwin, D.J. and Stenner, RD. 1977 Magnetic influences and cave surveying.  BEC Cav. Rep. 21 (Cave Notes 1975-1977) 35-6.

Irwin, D.J. and Stenner, RD. 1975 Accuracy and closure of traverses in cave surveying.  BCRA Trans. 2(4), 151-165.

Irwin, D.J., Stenner, RD and Tilly, G.D.  1968 The discovery and exploration of St. Cuthbert's Swallet. BEC Cav. Rep. 13(A), 36pp.

Stenner, RD. 1968 Water tracing in St. Cuthbert's Swallet, Priddy, Somerset.  Trans. CRG 10(2), 49-60.

Tilly, G.D. 1967 Manuscript draft of his contribution to Irwin, Stenner and Tilly, 1968.