Pen Park Hole has been formed in a buried ridge of Carboniferous limestone, the north west side of the Westbury Anticline, surrounded on three sides by a layer of Rhaetic clays. The limestone, Clifton Down limestone (Kellaway and Welch, in litt.) dips at approximately 50° to the north west. The cave has been strongly influenced by a high angle reverse fault which has been formed parallel to the bedding.
The present account draws heavily on the previous work of Bristow (1963) especially in terms of field work, but attempts to give fuller answers to some ot the points that he only touched upon. This account was originally written some twenty years ago, shortly after the cave was reopened. Some of the stratigraphic nomenclature may well, therefore, be out of date.
Bristow states that the cave is old, pre-Miocene at the least but possibly as early as pre-Liassic. He bases this on the observations that the cavity pre-dates both the mineralisation found within it, and certain earth movements which have subsequently affected it. He also remarks that, unlike most Permo-Triassic cave systems, it has not been infilled with later, usually liassic, sediment. What he does not do is make any statement about the conditions under which the cave was actually formed, except for a mention of 'solutional activity'. It is the opinion of the present writer that this cave has been formed by hydrothermal processes and that this conclusion explains many of its unusual features.
According to Ford and Williams (1989), certain erosional features are diagnostic of a hydrothermal origin. These include a tree-form of effluent chimneys as the form of the entire cave; deep, rounded solution pockets and highly corroded patches where steam has condensed above hot pools. All these are present here. The tree like layout of the cave can be seen from the survey, particularly on the east-west projection. This form has been largely constrained by the guiding fault zone and gives the appearance of being two-dimensional. It can be seen from the north-south projection that the cave is confined to a quite narrow span of beds, no more than 20 m in thickness. The detailed form of most of the passage walls has been obscured by the later mineral deposits, but deep rounded pockets divided by smoothly rounded pendants can be particularly well seen in the entrance passage, especially in the second chamber and the approach to the main pitch. Highly corroded patches of rock can be seen at several points in the higher reaches of the cave. A fine example is to be found close to the belay bolts for the main pitch.
Ford and Williams also state that a better indication of such an origin would be given by a cave's mineral deposits, amongst which this cave's all-over covering of walls, floor and ceiling by a crust of dog-tooth spar calcite would undoubtedly be included. Virtually the whole cave is lined with this deposit. It is only not seen low in the Main Chamber, where the walls are obscured by later silt and clay deposits, in the higher parts of the Main Chamber, presumably above the level reached by the thermal waters during that phase of deposition, and on the north wall of the main chamber. In the recently entered Upper East passage the crust has been found to have peeled away from the southern wall and to be lying on the floor. This may also explain its absence at other points, the fallen pieces having been removed by previous visitors.
Taken together, these features indicate a long history of hydrothermal conditions, during which the differing process zones of solution and precipitation have oscillated across the cave several times in response to changing external conditions. There have been at least three different phases: the initial phase was one of solution, during which the cave largely attained its present shape. There then followed several periods of deposition.
According to Bristow, the first phase of mineralisation was the deposition of galena, partly replacing the limestone and partly as a layer on top of it. The galena is mainly found in the First Chamber and dies out to the west and lower down. All the phases of mineralisation can be seen to post date the formation of the cave, as the deposits are all within the cavity and do not pass out of it into the cave walls. The second phase of deposition was of the coarsly crystalline calcite. This takes the form of large, 2-3 cm, scalenohedral crystals in a thick layer covering most of the floor, walls and roof of the cave. The deposit is about 15 cm thick in the eastern parts of the cave, but, as with the lead, thins out, to 8-10 cm, in the western part. The calcite is by no means pure and galena may be found within it along, probably, with other components. A second phase of solutional activity followed after this. It had little effect on the mineral deposits, on which its main result appears to have been a 'softening' of the outlines of the large calcite crystals; but where this deposit was absent, or had been penetrated, a significant amount of preferential solution of the bedrock took place. The last major phase of activity in the cave, before present day conditions were reached was an episode of limonite deposition, a thin layer of this mineral being present throughout most of the cave, overlying the massive calcite layer.
The final phase in the development of the cave was reached with the onset of present day conditions. This has seen the fairly minor deposition of thin calcite films and small stalactites in the upper part of the cave and of extensive clay deposition in the lower part of the cave. This latter material appears to have been brought in via the lake at the bottom of the cave as its water level fluctuates.
It is not known what, if any, role is played by the cave in the present hydrology of the area. It is known that the level of the lake is capable of quite large variation. The highest level recorded is approximately 51.5 m AOD, on 29th January 1993. The lowest recorded level is approximately 24 m AOD, on 18th October 1957 and there are possibly apocryphal tales of the lake drying up completely. Thus there is a known range of 27.5 m and a possible range of over 34.5 m. Tratman (1963) argued that there were likely to be connections of a fair size to adjoining conduits, to explain an apparent rapid and turbulent rise in water level noted in 1959, but direct exploration, by diving, has failed to find them. If they exist they must lie under the debris that completely covers the floor of the main chamber and the lake. The small spring in the main chamber noted by Glennie in 1957 (Tratman, 1963) has not been seen during recent explorations.
Recent fieldwork on the surface has identified one, possibly two, sink points in ditches to the east of the site, on the border with Filton Golf Course (P.L. Smart, pers com.) but no signs of flowing water have been seen in the passages above the level of the Lake on this side of the cave.