Collins New Naturalist Library

Substantial caves are not by any means confined to limestone. Other notable caves are formed by solution of marble, dolomite and non-carbonate rocks such as gypsum (e.g. Optimisticheskaja in the Ukraine, 183 km long); rock salt (in arid regions); and even quartzitic sandstone (e.g. Sima Aonda in Venezuela, 362 m deep) – while lava tubes are formed by the crusting-over and subsequent draining of molten lava flows (e.g. Kazumura Cave in Hawaii, over 20 km long). The latter include the youngest of all caves, formed on the slopes of Kilauea Volcano in Hawaii within the last five years. Other significant caves may be formed in a range of different rocks and within accumulations of large talus boulders (e.g. Lost Creek in Colorado, USA), by gravity sliding (gull caves), tectonic movements, wind-, or wave-blasting and other mechanisms. Most bizarre of all is Kitum Cave on Kenya’s Mount Elgon, which has been literally eaten for 200 m horizontally into a bed of volcanic ash by generations of salt-hungry elephants.

In caving parlance, a ‘dry cave’ is one which a human visitor can explore without getting wet. While such caves may provide an easy route into an underground world of great beauty, budding cave biologists may be greatly disappointed not to find every surface festooned with strange, eyeless arthropods, armed with sweeping antennae, stalking around on matchstick legs. As stated earlier in this chapter, dry passages in tropical caves may contain huge populations of bats or birds and a wealth of guano-associated cavernicoles, but in Britain and Ireland, cave-roosting bats and the species which depend on their presence have become increasingly scarce in recent years. Nevertheless a surprisingly large number of our macrocaverns do still harbour small populations of bats and a few associated ‘battelite’ species, which will be considered in Chapters 4 and 5.

Fig. 2.7 Mexican Free-tailed Bats, Tadarida brasiliensis, stream out of the entrance of Carlsbad Caverns at dusk. (Chris Howes)

The main biological interest of our dry caves is that they may contain the only accessible populations of cavernicoles primarily adapted to other, relatively inaccessible habitats. Thus, drip- or seepage-fed pools, gour pools and wet flowstone may harbour aquatic mesocavernicoles; and rock piles, flowstone pockets and ‘deep cave’ or ‘stagnant air’ environments may give us a rare glimpse of the terrestrial mesocavernicolous fauna.

Vadose macrocaverns (wet caves)

Cave streams may contain water derived from permanently-flowing surface streams, swallowed through an open sinkhole, or percolation water which has collected gradually from a large number of mesocavernous inputs. Those fed entirely by percolation water or seasonal run-off tend to hold more biological interest, as the populations they contain (being genetically isolated from surface habitats) may have evolved specialized features.

Surface streams arrive in the cave complete with a biota of their own, most of which will survive quite well in darkness for some considerable time, though few species may succeed in reproducing. Apart from the lack of growing green plants, and a tendency to experience less seasonal temperature change, such streams do not differ significantly as a habitat from streams above ground. The terrestrial habitat in stream passages is usually draughty and moist. Its fauna may include the adult stages of aquatic insect larvae swept in with the sinking stream, and various predators, such as spiders, which feed on them, plus a range of detritivorous cavernicoles.

Where stream passages enter the phreas (cavers’ ‘sumps’, marking the end of exploration to all but divers) it is not unusual to find accumulations of organic detritus deposited by floods. Given a suitable microclimate, such places may be well populated by specialized cavernicoles, or ‘troglobites’. Indeed, in many food-poor cave systems (such as those in mountainous areas of the Pyrenees and Cantabrians) this may be the only habitat in which cavernicoles occur in any numbers.

Water-filled macrocaverns (phreatic caves)

The chemistry, food supply and fauna of phreatic waters is determined to a large extent by their mode of entry into the cave. Underground waters are classed as autogenic (originating as through-soil percolation via diffuse seeps and subcutaneous flow) or allogenic (sinking streams). The former tend to be lower in organic content, less chemically aggressive by the time it reaches the phreas and may carry a lower sediment load. Phreatic waters may have a very different chemistry to the vadose waters which feed them, because they are unable to de-gas the carbon dioxide produced by microbial oxidation of organic materials, or to replenish oxygen used up in such decomposition.

The rate of flow in large-diameter phreatic tubes is generally greater than in phreatic mesocaverns, but is still often sluggish enough to accommodate small, slow-swimming cavernicoles (such as Niphargus fontanus), which avoid fast-moving vadose streamways.

In recent years cave divers have penetrated great distances into freshwater phreatic macrocaverns, and to considerable depths, but to date no detailed studies have been made of the biota of this remote and fascinating environment.

Water-filled mesocaverns

Most of our knowledge about the structure of mesocaverns comes from looking at exposed joints or bedding planes in limestone quarries, or occasionally caves, and from the work of karst hydrologists. We know that the very earliest stages of cave development occur under phreatic conditions, eventually creating inter-connected systems of conduits which may be in any plane, from horizontal to vertical. Phreatic mesocaverns may take the form of wiggly networks of small-diameter tubes (anastomoses), or thin, but laterally-extensive cracks, or narrow shafts – and are probably as common in other cavernous rocks, such as chalk or gypsum, as in limestone. Evidence that such spaces are inhabited comes from the animals found in well-water over the centuries. It is ironic that the earliest records of our cave fauna should be from a habitat about which little more is known today than a century-and-a-half ago, when Philip Henry Gosse wrote:

“recently, investigations in various parts of the world have revealed the curious circumstance of somewhat extensive series of animals inhabiting gloomy caves and deep wells, and perfectly deprived even of the vestiges of eyes … even in this country we possess at least four species of minute shrimps [all of which] have been obtained from pumps and wells in the southern counties of England, at a depth of thirty or forty feet from the surface of the earth.”

There is no way at present of collecting information directly about how aquatic cavernicoles use mesocavernous bedrock cracks, but it seems likely that great local variation exists within this habitat in terms of oxygen concentrations, pH and food supply. Such factors are likely to influence the distributions of the fauna, and will be discussed in Chapter 4. Fortunately there are other, more accessible types of water-filled mesocavernous habitat which are easier to study. They include the deeper interstices of stream-bed cobbles (phreatic nappes), and a peculiar sub-soil phreas of mesocavernous dimensions which occurs on the surface of impervious silt or clay deposits in mountainous areas of Europe (the hypotelminorheic medium). In both these habitats, the food supply comprises dissolved or finely particulate organic material, and the waters tend to be rather low in oxygen; and both contain faunas very similar to those of limestone mesocaverns.

Amphibious mesocaverns

As no detailed investigation of the biology of air-filled mesocavernous habitats has yet been attempted, we are forced to infer what we can about the conditions within them from studies of limestone caves and other similar habitats.

As soon as mesocaverns develop an airspace, they become available for colonization by terrestrial cavernicoles. However, cracks and anastomoses are extremely flood-prone, often filling up with water each time there is heavy rainfall at the surface. Vertical cracks probably flush more violently, but remain water-filled for shorter periods than horizontal cracks, and this may result in some differences in their faunas. Less immersion-tolerant organisms may tend to inhabit the wider, better-drained vertical cracks and humid terrestrial cave habitats, while the more aquatic organisms may prefer horizontal cracks or cave pools. It is likely that particulate organic material accumulates at specific points within the cracks (perhaps at the upstream ends of permanently flooded sections), so that some patches of habitat will be better supplied with food than others. Some areas may be too anoxic to support any life other than anaerobic micro-organisms, while some patches may harbour relatively large concentrations of detritivorous invertebrates. As previously discussed, there may be an almost complete overlap in distribution between the ‘terrestrial’ and ‘aquatic’ components of the fauna of such habitats.

The French biospeleologists Juberthie, Delay and Bouillon consider that mesocavernous spaces in fractured rock immediately below the soil constitute a habitat which is separate from caves, which they have termed the ‘Superficial Underground Compartment’ (SUC). Their claim rests on differences between the fauna found here and that found in deep caves. They consider the primary cause of such differences to be the greater temperature variation experienced in the ‘SUC’ compared with the ‘Deep Underground Compartment’ (DUC) represented by deep fissures and caves. While this may be so in SUC habitats beneath shallow soils of regions which experience a strongly seasonal temperate climate, I would doubt that the microclimate in deeply-buried SUC habitats or those of tropical karst differs a great deal from that of the ‘DUC’ – and there is evidence that cave faunas migrate up into the SUC periodically in order to exploit the resources they contain. For the purposes of this classification, I propose therefore to treat the ‘SUC’ as part and parcel of other intermittently-flooding mesocavernous spaces, whether they be immediately below the soil, within cave passages, or connecting one with the other.

There would seem to be little doubt that the SUC within calcareous rocks is by far the most extensive and important of all cave habitats in terms of the numbers and diversity of its biota. Since their ‘discovery’ of the ‘SUC’ (an environment previously well-known to karst hydrologists as the ‘subcutaneous zone’), Juberthie and Delay have gone on to show that this habitat and its biota not only occurs in limestone and other cavernous rocks, but also in ‘non-cavernous’ shales, granites, schist, gneiss, sandstones, etc. My first reaction on reading the paper announcing this discovery was to attack the bottom end of my garden with a pick and shovel. There, to my delight and amazement, I found tiny-eyed cave spiders (Porrhomma egeria) frolicking among the fractured chunks of Pennant Sandstone just one metre beneath the wreckage of the flower bed. As far as I know there has been no systematic investigation to date of the fauna of ‘SUC’ habitats in Britain and Ireland – an extraordinary gap in our knowledge which surely must be remedied before long.

A better-known mesocavernous habitat is contained in talus, or scree, whose surface can frequently become covered with vegetation and soil, turning it into a fair imitation of Juberthie and Delay’s SUC. When not sealed by soil, the upper levels of talus are unsuitable as a habitat for cavernicoles, being too cold in winter, too hot in summer and too dry for much of the time. However, if the scree is deep enough, the lower levels must surely provide exactly the conditions favoured by cavernicoles, though I know of no work on this deep-talus habitat in Britain.

I know of only two accessible ‘DUC’ mesocavernous habitats within caves. One is in the spaces within rock piles (underground talus), the other is in speleothem pockets. Rock piles may, or may not provide a suitable habitat for mesocavernicoles. If the pile is in an old, dry ‘fossil’ passage, as most rock piles tend to be, it is unlikely to contain enough food to support life (unless the cave contains bats, or other vertebrates, in which case the rocks may be over-run by guano-beasts). On the other hand, if the pile is sufficiently extensive, and is traversed by percolation water carrying organic material, it is likely to harbour a rich fauna of mesocavernicoles – although the depth within it at which a searching biologist can expect to ‘strike bugs’ will increase with the increasing dryness or breeziness of the surrounding cave atmosphere, precisely as would be the case with above-ground talus. Juberthie (1983) gives an interesting example of a schist-boulder pile in the great Salle de la Verna chamber in France’s Pierre Saint Martin cave. It is inhabited by a typical SUC fauna of Aphaenops beetles which appear to be quite oblivious of the fact that their schist scree habitat lies the best part of 1000 m underground.

Speleothem pockets are essentially just spaces of mesocavernous dimensions like all the others described in this section, but where they occur in the ‘deep cave’ or ‘stagnant air’ zone of caves (see the microclimate section, later in this chapter), they will often prove to be the very best places to look for mesocavernicoles. Speleothems occur where percolation water, rich in dissolved lime, intersects a cave passage. Where such deposits are laid down over mud, pockets often form between the two; and if deposition is still in progress, trickling water maintains just the microclimate conditions favoured by cavernicoles, while also supplying a source of food. In short, they perfectly reflect conditions in the mesocaverns. The late A. Vandel, in his famous book Biospeologie la biologie des animaux cavernicoles (published in 1964), described such a habitat in the Grotte de Sainte-Catherine, at Balaguères, in the Ariège region of France:

“One side of the chamber is formed by a stalagmitic wall which is covered by a thin layer of water which flows from an opening in the roof … The constant flow brings in organic material from the exterior which nourishes the Collembola and nematoceran Diptera on which Aphaenops (a blind cave beetle) feeds … The stalagmitic covering is separated from the wall by a space of a few millimetres which contains the products of dissolution of the rock: clotted red and black clay, and black magnesium deposits.

When one enters this chamber three or four Aphaenops can usually be seen running on the wall in search of food … If they are caught, others appear. Closer examination soon explains this phenomenon. The calcareous wall is formed of stalagmitic columns arranged parallel to each other. Between these columns small holes have been hollowed and it is into these that Aphaenops disappears … It appears that the space between the rock and the stalagmitic wall constitutes the biotope in which Aphaenops lives when not in search of food, and in which they reproduce.”

Cave pools

Gours, or rimstone pools are formed by seep-fed, lime-rich waters trickling down an incline in an open cave and depositing a sequence of curved, retaining dams. Such pools presumably provide conditions akin to those in phreatic mesocaverns, and often harbour a similar fauna, which is augmented in the cave by animals washed out from seepage cracks during heavy rain. Other drip-fed pools may also contain aquatic mesocavernicoles, providing they receive a food supply.

The concave meniscus of pool surfaces may act as a deadly trap for soil and mesocavernous springtails (see Chapter 5), providing a happy hunting-ground for other specialized mites and springtails, whose feet are equipped to grip.

Cave sediments

All kinds of sediments find their way into caves. Many arrive complete with their own specialist biotas: bacteria, fungi, nematodes, earthworms and so on, and these may be exploited by cavernicoles as a source of food. Cave sediments may also serve as a habitat for the eggs, larvae, or pupae of cavernicoles.

Moonmilk

Moonmilk is a term applied to white, wet, cheese-like or dry, sticky, powdery formless masses found in limestone caves. This weird stuff may contain a cocktail of carbonate minerals – including calcite, aragonite, monohydrocalcite, magnesite, hydromagnesite, nesquehonite and huntite – some of which are alledged to be associated with particular bacteria isolated from moonmilk. A blue-green alga Synechococcus elongatus found growing in moonmilk in complete darkness must have been using alternative metabolic pathways to the normal photosynthetic ones it employs in the light, and other algae (Gleocapsa magna) may also be present. I know of no cavernicolous animals associated with moonmilk, so this is not a habitat which I propose to discuss further in this book.

Submarine and intertidal cave habitats

The submarine Green Holes and intertidal Brown Holes of Doolin, in County Clare, are ordinary limestone caves, formed in the usual way, which became inundated by rising sea levels at the end of the last glacial period – about 12,000 BP No doubt they have had a long history of intermittent sub-aerial cave-development during cold glacial periods of low sea-stance, alternating with periods of immersion in sea-water during interglacial warm spells, such as the planet currently enjoys. They provide a significant habitat for marine life – Mermaid’s Hole has been explored for over a kilometre and the Hell complex totals several hundred metres. Other significant submarine caves occur in the Brixham area of Devon and at Durness in the far north of Scotland. Otter Hole, a resurgence cave which opens on to the banks of the river Wye near Chepstow, has a tidally-flushed entrance series which may contain fresh or saline water, depending on the height of the tide and the volume of the cave river. The faunas of such caves will be discussed in Chapter 5. In other countries, an interesting fauna has been found in coastal brackish groundwaters, but I know of no such British fauna. It may exist, but no studies have been made.

Submarine caves, such as the Green Holes, should not be confused with ‘sea caves’, such as the famous Fingal’s Cave in the Hebrides, which are spray-filled holes blasted out of cliffs by wave action. The latter may harbour a few ubiquitous, fast-moving cliff-dwellers such as sea slaters (Ligia oceanica) and silverfish (Petrobius maritimus), which can dive into cracks to escape the force of the waves, as well as a few of the more hardy marine organisms found in the more extensive drowned limestone caves referred to previously. In short, the fauna of sea caves is unremarkable.

Slutch caves

Slutch is the onomatopoeic term given by Wainwright to the peat bogs of Kinder Scout, Bleaklow and Black Hill in Derbyshire. Apparently, water flowing between the base of the peat and the underlying gritstone has carved out a number of caves of explorable dimensions – one of which has been followed for 50 m underground by cavers Steve Fowler and Tony Moult. It seems that the base of the peat is riddled with such caves and with smaller air- or waterfilled passages of mesocavernous dimensions. No doubt this will prove to be a widespread habitat in upland peat deposits throughout Britain and Ireland, perhaps with a characteristic fauna all its own, which for the moment appears to have escaped investigation.

Mines, tunnels, cellars and tombs

As artificial caves, such spaces may be inhabited or visited by any cavernicoles which have both the motive and the opportunity to do so. The motive will be a suitable medium/microclimate and food supply (see following sections). The opportunity will fall to any cavernicole whose habitat intersects the artificial cave.

Food supply

Chlorophyll and sunlight are the elements of life on earth and the source of that unique green glow which identifies our living planet when seen from space. Photosynthesis lies at the base of the food chains in which all life is meshed: fish, fowl and fungus; tree, turtle and tiger – or almost all.

In 1977, an American ship on an oceanographic survey south of the Galapagos Islands, located some active volcanic vents on the sea floor at a depth of 3 kilometres. An instrument pod was sent down, armed with video cameras to record the scene. As the probe moved towards the vents, the watching scientists were amazed to see their monitor screens fill with a writhing mass of enormous worms, 10 centimetres thick and up to three metres long. Close by were beds of 30 cm-long clams. Among them swam shoals of fish, and white crabs scuttled across the black basalt rocks. At these depths, far beyond the reach of sunlight, life is generally thin on the ground. A few starfish, crinoids and crustaceans subsist on the steady drizzle of detritus, and are in turn eaten by predatory fish. To find such an abundant local concentration of large organisms clearly pointed to an unusually rich food source. The mystery was soon solved. The volcanic vents, it seems, were spouting superheated, sulphurladen water. As this cooled, clouds of black sulphides formed and were immediately consumed by great concentrations of bacteria. The worms and the clams were feeding on these bacteria and they in turn supported the scavenging fish and crabs. The bacteria concerned are chemo-autotrophs – that is, they can harness the chemical energy in the volcanic sulphides to power their own vital processes. What is more, this whole process and the mini-ecosystem which revolves around the ‘volcanic bacteria’ is quite independent of sunlight.

Sulphur bacteria, and relatives which derive energy by reducing ferric iron compounds, are common in caves – in sediment banks as remote from solar rays as are the depths of the deepest ocean trench. The muds where they live are home to nematode worms, known bacteria-feeders, which are hunted by tiny, scurrying beetles. How many visitors, catching sight of a cave beetle, appreciate that they may be watching one of the rarest of all living phenomena – a predator sustained, at least in part, by a food chain independent of sunlight.

The energy source for cave-based chemosynthesis generally originates outside the cave, as Carboniferous Limestone itself contains very little in the way of iron and sulphur minerals. These compounds, and the bacteria which exploit them have usually been washed in as part of the cave’s sediment load. But there is at least one energy source which may originate within the fabric of the cave itself. Most limestones contain detectable amounts of organic matter, largely in the form of hydrocarbons. Such material would be of no use to animals directly, but if there are bacteria present which are capable of using it as an energy source, it could be continually liberated into the cave ecosystem at the interface between the cave and the rock, imperceptibly, as the cave is dissolved out by flowing water. Since the organic matter in the limestone will have been derived from organisms present in the seas when the rock was being laid down, its energy content must originally have come from the sun. It is in fact fossil solar energy.

Interesting though they may be as a scientific curiosity, chemo-autotrophic bacteria contribute only slightly to the food base of most cave communities. By far the biggest source of energy is still the sun’s rays, but at second-hand, in the form of introduced detritus from surface communities; for no green plants can survive in the absolute dark of the cave.

The absence of green plants in caves led early biospeleologists to the conclusion that cavernicoles are starved animals. In 1886, Packard insisted that the shortage of food available to cave animals is the reason for their small size. While it is self-evident that large animals with a high metabolic rate can have no place in an entirely heterotrophic, food-poor ecosystem, in reality many cave species are actually far larger than their surface relatives. Most recent studies have shown that cave-evolved animals (troglobites) have unusually low metabolic and growth rates and that they save energy in every way possible, by streamlining their movements and by adopting highly efficient foraging and reproductive strategies. These are obvious specializations to cope with a low food supply, and seem to be a fundamental characteristic of cavernicolous evolution at temperate latitudes like ours. Some cave species are remarkable in their metabolic efficiency and consequent ability to tolerate starvation. Gadeau de Kerville (1926) reported that a specimen of the Slovenian cave salamander, or Olm (Proteus), had been kept in captivity for fourteen and a half years, and for the last eight of these had received no food. He did not report whether it eventually died of starvation or just plain old age.