Journal Series 4 Number 10

Note that this is a re-print of the original publication, based on a scanned copy. During the process of converting the original paper copy to this electronic version, the original formatting, page layout and page numbers have been lost. All diagrams and surveys have been scanned from the original and are consequently of poor quality.



The Electrical Measurement of Cave Temperatures by BM Ellis

A Simple Cave Theodolite by BM Ellis

Afton Red Rift Cave, a Bibliography by EJ Watkins

Review: Men and Mining on the Quantocks


Published by the Shepton Mallet Caving Club

The Mineries, Wells Road, Priddy, Wells, Somerset, BA5 3AU



This is to be the last Journal in the present series and in the present format. The coming of metrication forces upon us yet another change of size to comply with the trend, in accordance with other Caving Club Journals.

Another change to take place, concerning the Journal, is that it will be published in the Spring and Autumn of each year. This is particularly to facilitate the ease of takeover for the incoming Editor, so that the first Journal of the year should be published before club AGM, and not immediately after.

I must apologise for the shortage of material in this issue. It was to have included material from the clubs' recent Iceland Expedition, however, it has been decided to produce an Occasional Paper concerning the latter, sometime in early Spring 1971.

The contents on this occasion concern two articles on specialised equipment used on the expedition, an appendix to an article in the previous Journal, and lastly a book review of "Mining in the Quantocks"; four short articles which you may be able to fit into your festive seasons' reading.

Finally may I take this opportunity to wish you all, "The Season's Greetings".


The Electrical Measurement of Cave Temperatures


A method of measuring temperatures is described which is more precise, more accurate and more convenient to use than the conventional mercury-in-glass thermometer. It depends on measuring the change in resistance with temperature of a thermistor by means of a simple electronic circuit.

There are occasions when it is desired to make a series of temperature readings in a cave in connection with a scientific project. For example, theories on hydrological connections between the various streams in St. Cuthbert's Swallet were made many years ago based on the later temperatures of these streams. These theories were later confirmed by dye tests. Similarly, during this club's expedition to Iceland it was planned to carry out temperature measurements along the main passage of Raufarholshellir in an attempt to explain the occurrence of the ice formations that are present throughout the year. In the past such measurements have usually been made with conventional mercury-in-glass thermometers, their accuracy sometimes being established by reference to National Physical Laboratory certified thermometers.

The use of mercury-in-glass thermometers in caves has several serious disadvantages. They are very fragile and not only is it inconvenient to have to change thermometers frequently but there is also the expense involved. The cheapest suitable thermometer, graduated only in 1°C divisions, costs about 12/-, while one graduated in 0.01°C divisions costs at least twice as much. Secondly the maximum precision with which such a thermometer can be read is 0.05°C, and this only under, ideal conditions – which are not found underground. A further disadvantage, and probably the most serious of all, is that a mercury-in-glass thermometer has to be read by a person standing in very close proximity to the thermometer itself. This means that small temperature variations cannot be measured as the ambient temperature will itself be altered to a greater or lesser extent by the presence of the experimenter reading the thermometer. It also means that it is virtually impossible to take readings in situations where it is impossible for the experimenter to climb, for example ten feet above the floor in the centre of a large passage.

It was thought that most of these disadvantages could be overcome by using some form of electrical temperature measurement. Resistors whose value changes with temperature, known as thermistors, are readily available and amongst other uses, some are designed specifically for the measurement of temperature. Commercial electrical thermometers are available but apart from their cost (prices start at £20) they usually cover too wide a temperature range and therefore their precision is usually worse than 0.1°C. Our own instrument was designed and constructed, therefore, to cover the temperature range and precision required.

The electrical circuit, shown in Figure 1, is conventional and consists of a Wheatstone Bridge composed of the thermistor, R1, and three other resistors, R2, R3 and R4. The output from the bridge is amplified by a transistor in each side, Tr1 and Tr2, and the amplified signal arising from the imbalance of the bridge is fed directly to a suitable meter. If necessary a series resistor, R5, can be added to the meter lead to adjust the value of the meter full scale deflection to that which is required. To minimise temperature variations affecting the bridge network high stability resistors were used for R2 to R5; variations in the circuit can be adjusted by means of the potentiometer, R6, placed across the transistor emitter leads. Power is derived from a battery and the voltage is stabilised for the bridge circuit by means of a Zener diode, Z1. Battery consumption is decreased by an on/off switch, S1. For compactness and long, reliable life a 7 volt Mallory battery, type TR 135, was used as the power source and the bridge circuit was stabilised at 4.7 volts. Two additional features were added to the circuit. The thermistor can be replaced in the bridge circuit by a pre-set potentiometer, R7, by means of a push switch, and the circuit can be "zeroed" against this constant value irrespective of the temperature of the thermistor itself. The other additional feature is that the range of the instrument was increased by the addition of resistor R8, which can be switched by the range switch S3, to form part of either the R2 or R3 arm of the bridge.

The values chosen for the components will depend on the thermistor used and on the temperature range over which it is intended to make measurements. For the club's expedition to Iceland we were interested in temperatures around the freezing point of water and the following values were used:

R2 270 K; R3 270 K; R4 180 K; R5 3.3 K; R6 5.0 K; R7 220 K; R8 56 K; R9 10 K; R10 2.2K; R11 1.2 K ohms. These were used with an ITT G15 thermistor as R1, and Texas Instruments BC 184's for Tr1 and Tr2.

Anyone with a knowledge of electronics will be able to change these values to suit their needs, and perhaps improve the circuit; for others they will be found suitable as a starting point from which to experiment by trial and error. The temperature ranges obtained with this circuit were: low range, -5°C, to +3.2°C; high range, +0.8°C to +9°C. The instrument was calibrated on both ranges against a mercury-in-glass thermometer, the calibration curves being shown in Figure 2. The scale of the meter was divided into 100 equal divisions each of which was sufficiently large to be divided easily by eye into quarters. This gives readings of 0.02°C but it is possible that this very simple circuit will not give this accuracy.

In use it was found necessary to check and re-set the zero reading after every three or four readings, by no means a difficult or time consuming operation, but a more stable circuit would be an obvious improvement. The major drawback was a question of engineering. Little trouble had been taken in waterproofing the instrument and after two traverses under conditions of heavy drip the suspension of the meter became sticky and unreliable. Moisture entered the switches also and made their operation unreliable. Those problems are easilyovercome but will involve additional expense. Apart from these points the instrument was found easy and convenient to use. The thermistor had a fifteen feet lead and was attached to the end of a long aluminium pole in order that it could be kept at least ten feet in front of the people making the readings. The sensitivity of the instrument meant that very small variations in temperature caused relatively large fluctuations on the meter and this resulted in some wasted time waiting for stabilisation until the significance of the fluctuations was fully appreciated.

In conclusion it can be stated that as a means of making temperature measurements in a cave, this instrument is more convenient and almost certainly more accurate than mercury-in-glass thermometers. It should be made waterproof for trouble free operation under caving conditions. Although certainly not essential, the perfectionist could improve the electronic circuit to give more stable operation.

Figure 1 – Circuit Diagram

Figure 2 – Instrument Calibration

BM Ellis

September 1970


A Simple Cave Theodolite


An instrument is described for surveying under conditions of severe local magnetic anomalies; this can be made either as a modification to the Survey Unit or as a separate instrument. The accuracy to be expected when using this instrument is discussed.

The instrument currently used by members of the club, and many other surveyors on Mendip, for the production of cave surveys has come to be referred to as a "Survey Unit". It was developed principally by club members and has been described in the caving literature, originally in a much earlier issue of this Journal (1) and much more recently in the Transactions of the Cave Research Group (2). This instrument has been found to be very convenient to use and it is thought by many to be the most suitable currently available for the production of accurate surveys in the minimum of time. However, as it is basically a compass it is of very limited accuracy if used in locations of large magnetic anomalies,  natural or man-made. From a study of the geological literature it was surmised that there was likely to be large magnetic anomalies in the vicinity of lava tunnels although this point seems to have been ignored, very conveniently, by everyone who has surveyed lava tunnels in the past. Because of this possibility it was decided to construct a suitable instrument for the preparation of non-magnetic surveys in order that these could be made by club members when in Iceland if it was found to be desirable. The only other method for measuring horizontal angles that is at least vaguely feasible when surveying in caves is some form of theodolite. There are various disadvantages in the use of this type of instrument, in particular that any error in the measurement of an angle will cause an error that becomes larger as the traverse increases in length. (When using a magnetic compass such an error will remain constant in magnitude.) This means that to obtain any reasonable degree of accuracy in the survey it is necessary to measure horizontal angles with a high degree of both precision and accuracy when using a theodolite. Despite these drawbacks, the construction of a simple theodolite was considered to be the only practical alternative to the use of a magnetic compass.


The instrument is shown in Figure 3. A right angled bracket was constructed from 1 inch x 1/8 inch brass plate. In the angle was fixed a block of brass approximately 1 inch x 1 inch x ½ inch and through this and the base of the bracket was drilled a hole to form a bearing for the vertical pivot about which the instrument rotates. Near the top of the vertical arm of the bracket was drilled a hole to take the Abney Level bearing bush as used on the Survey Unit. At the outer end of the horizontal arm was fixed a transparent plastic plate that had a line scribed on the under face; this was the rotating reference mark that was read against a fixed horizontal scale. Finally, "T"-shaped spirit levels were fastened to the horizontal arm, and a horizontal hole was drilled and tapped to meet the vertical pivot bearing so that a clamping screw could be fitted.

The horizontal scale was made from a six inch diameter protractor which was fastened with small brass screws to a disc of Bakelite out to the same diameter. This provided rigidity and strength to the scale and also protection against abrasion of the graduations. It has been found advantageous in the past to use a levelling table on the top of the ball and socket head to facilitate fine adjustment and it was an easy matter to drill the horizontal scale so that it could be fastened to the underside of the top plate of the table. Details are not given as this has since been modified to improve the versatility of the instrument. It was then only necessary to remove the Abney Level, complete with fixing plate and shaft, from the Survey Unit and fasten it through the bush at the top of the arm to complete the instrument.

An alternative method of construction, starting with a Survey Unit, is simply to add a transparent plastic pointer to the base of the Unit itself and to use this with the horizontal scale. In this way it is possible to make either a magnetic or theodolite traverse with the same instrument. It was a cave theodolite of this form that was eventually taken to Iceland. The disadvantages are the necessary bulk and weight of the unwanted compass etc., and the fact that as the pointer has to stick out beyond the side of the Unit it is very vulnerable to breakage. Several spare pointers were taken to Iceland but none were actually needed. However, by the end of the surveying programme the pointer had become scratched and this made reading of the angle more difficult.

Figure 3 – Cave Theodolite Construction

Three horizontal scales were made in order that one could be kept permanently on the top of each of the three tripods that are required for a theodolite traverse. Once the tripods were set up it was only necessary to transfer the theodolite and target lamps from one tripod to another.


The cave theodolite was used extensively by the party in Iceland and as a result it was thought that two modifications would be advantageous. The first is only a minor one but it is to change the protractors used for the horizontal scales for ones that have the degrees numbered in the more conventional clockwise direction. If the fact that they were calibrated in the opposite direction had been remembered earlier it would have saved quite a bit of head scratching when the first results were being worked out!

There were operations when surveying when it would have been an advantage to have been able to rotate the scale horizontally on top of the tripod. The scales have since been modified to enable this to be done and the present arrangement is shown in Figure 4. A bearing with a clamping screw has been fixed below the scale and there is a further pivot above the scale about which the instrument can rotate. This has been turned from a single piece of brass to ensure that the bearing and pivot are in line. It is important to make sure that the scale is clamped before any readings are taken and some form of self-locking clamp in place of the clamping screw would be a further improvement. With this facility it is possible to set up a required angle on a given line to check readings, etc. It would also be possible to choose readings that make the calculations easier by not passing through the origin when turning from one line to the other but the temptation to set one reading to zero must be resisted as attempting such a practice would almost certainly give rise to additional errors.

Figure 4 – Modified Cave Theodolite Construction


The protractors used were graduated to half a degree and it was very easy to make readings to the nearest quarter. In fact it was possible to estimate to the nearest eighth but tests showed that the cave theodolite was not sufficiently precise in itself to warrant such precision in the readings. The theodolite was set up on a tripod and two targets were set up on further tripods at a distance of approximately sixty feet. A series of readings was then taken on each of the targets in turn, both with the compass and the theodolite; the results are given in Table 1. The loss in precision with the cave theodolite is probably due largely to the relative crudeness of the system when compared with the compass. The first investigation carried out in Iceland showed that around Raufarholshellir, at least, there were very large local magnetic anomalies, in fact differences between forward and backward bearings along survey legs of up to thirty degrees were obtained. It was obvious, therefore, that theodolite surveying would have to be employed. To determine the expected accuracy four closed traverses were surveyed using the same techniques as would be used in the cave.

Table 1 – Comparison of a Series of Measurements made by the Cave Theodolite and the Survey Unit on an Inclined Angle

These ranged in length up to nearly 1000 feet and were located partly in the cave and partly on the surface. The closure errors of these traverses are given in Table 2 which also quotes the closure errors of magnetic traverses that were carried out at the same time.

Table 2 – Traverse Misclosures

It appears that there was a gross error made in the readings for traverse B, but if this traverse is ignored it will be seen that the average horizontal misclosure, by theodolite, was 1.4% and this was considerably better than those obtained magnetically in this locality. The vertical misclosure, which is common to both the magnetic and theodolite traverses, averages 0.1% and this is of the same order as the vertical misclosures obtained with the Survey Unit when it has been used in Mendip caves. The horizontal misclosure expected for a magnetic traverse, both from experience and the theoretical figures given by Warburton (3), is of the order of 0.5% when using the Survey Unit under the conditions of these traverses, so it will be seen that there is a considerable loss in accuracy. In fact, again using Warburton's figures, the cave theodolite gives an expected error similar to that expected from a good CRG grade 4 survey (i.e. a grade 4 where some form of clinometer has been used) or a poor grade 5, despite the extra care taken by the use of tripods, etc. It can be said, therefore, that the accuracy is poor in return for the effort that has had to be made but in some circumstances it is better than that which can be obtained by more conventional methods. Its use is recommended only when local magnetic anomalies make the use of a magnetic survey too inaccurate.


  1. Ellis, BM. A Mounting for Cave Survey Instruments. SMCC Journal, 3, (10), pp3-8 (November 1965)
  2. Ellis, BM. The Survey Unit - Equipment used on Mendip, England. CRG Transactions, 12, (3), pp139-147 (July 1970)
  3. Warburton, D. The Accuracy of a Cave Survey. WCC Journal, 2, (89), pp166-181 (April 1963)

BM Ellis

October 1970


Afton Red Rift Cave – A Bibliography

Appendix to the previous article in SMCC Journal 4 (9), (June 1970).


  • Bretz, JH, 1942, Vadose and Phreatic Features of Limestone Caverns. Geology, Vol. No. 50  No. 6, Part 2.
  • Brunsdon, D, 1963, The Denudation Chronology of the River Dart, Transactions of the Institute of British Geographers.
  • Butcher, AL, 1950, Cave Survey. CRG Pub. No. 3.
  • Balchin, WGV 1966, The Denudation Chronology of South-West England. The Royal Geological Society, Commemorative Volume.
  • Dineley, DL, 1961, The Devonian System in South Devonshire. Field Studies, Vol. 1. No. 3,
  • Embleton, C & King, CAN, Glacial and Periglacial Geomorphology. Pub. Arnold.
  • Green, JFN, 1936, The Terraces of Southernmost England. Quarterly Journal of The Geological Society, London.
  • Green, JFN, 1949, The History of the River Dart. Proceedings of the Geological Association.
  • Homes, A, 1965, Principles of Physical Geology. Pub. Neilson.
  • Jukes-Brown, AJ, 1907, The Age and Origins of the Platforms Around Torquay. Quarterly Journal of the Geological Society, London.
  • Jukes-Brown, AJ, 1912, The Making of Torbay. Trans. of the Devon Association.
  • Ollier, CD, Weathering. Pub. Olivier & Bord.
  • Orme, AR, 1960, The Raised Beaches and Standlines of South Devon. Field Studies.
  • Sutcliff, A, Joint Mitnor Cave, Buckfastleigh. Reprint from the Transactions of the Torquay Natural History Society. Vol. 15. Part 1.
  • Tratman, EK, 1968, The Caves of Northwest Clare, Ireland. Pub. David & Charles.
  • Usher, 1933, The Country Around Torquay. HMSO.
  • Warburton, D, Theoretical Treatise on Surveying. WCC Jnl. 8.
  • Warwick, GT, 1958, The Characteristics and Development of Limestone Regions in the British Isles, with special reference to England and Wales. Deuxieme Congress International de Speleologie Bari - Lecce - Salerno. 5-12 October. 1958.
  • Zeuner, FE, 1959, The Pleistocene Period – London.

Reference has also been made to the following publications:

  • Regional Geology of South-West England. Third Edition. HMSO. 1969.
  • Devon Speleological Society, Records, 1947 onwards.

EJ Watkins

January 1970


Review: Men and Mining on the Quantocks

By JR Hamilton & JF Lawrence, pub. Town & Country Press Ltd. (Bracknell, 1970) @ 15/-

If you have ever travelled along the A39 road from Bridgwater to Minehead, about two miles beyond the village of Nether Stowey on the north side of the road you may have noticed the remains of a tall square stone building standing alone in a field. Together, with a similar building nearby, these two engine houses are the last remains of the Buckingham Copper Mines. This group of mines were worked sporadically from 1720 to 1760, actively from 1786 to 1801, were re-opened in 1817 with a pump to overcome ground water, but finally closed in 1821.

Written by two local historians this 78 page, well produced book with photographs and maps brings to us for the first time a detailed account of the history and development of these mines, much of the material being taken from the Stowe MSc – being a complete collection of the Marquis of Buckingham's papers in a Californian museum. Apart from the Buckingham Mines this book also covers about sixteen other known and possible mining sites on or adjoining the Quantock Hills, these being mainly in search for copper but some for lead, iron and malachite. The section on mining at Broomfield for copper and its association with Andrew Crosse is from material collected by the late Jack Waddon.

Quite apart from the details of the mines and mining the book commences with a section on the economic conditions of the area at that date. During the history of the mines, miners from Derbyshire, North Somerset and Cornwall were employed and details are given of mines in an agricultural area, wages, Poor Relief, demonstrations, hunger march, agricultural revolution, medical costs and the cost of living. This background picture is very important in a full appreciation of the historical development of the mines.

The book lacks a geological map which would help the reader as it shows clearly the faulting in the area of the Buckingham Mines, where is also to be found a very small pocket of Devonian limestone. It was presumably in this that in 1795 cavities and large caverns were discovered in the course of mining under the Beech Grove, a little above Dodington House – one of which we are told was 28 yards long, 4-12 yards high and wide and the top 14 yards below the surface. The absence of a geological map can, perhaps, be justified by the fact that in all probability the miners had no such aid.

There is an interesting reference to tracing a mineral vein on the surface by 'dialling', which is presumably a similar process to dowsing or divining. The last recorded mention of the Buckingham Mines, apart from one recollection by the authors, was in 1826, but it is known that the mines were included as of the "richest copper" in Lot 1 of Particulars of an Auction Sale of the Estate of Duke of Buckingham & Chandos by Philips on 26th September 1833 at Bridgwater.

A very interesting, useful and worthwhile work.



 Journal Series 04 Number 10