PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE PALEOKRAšKI SEDIMENTI V JAMAH: PRIMERI IZ VZHODNE AVSTRALIJE IN SREDNJE EVROPE R. Armstrong L. OSBORNE1 1 Sydney School of Education & Social work, A35, The University of Sydney, NSw, 2006, Australia, e-mail: armstrong.osborne@sydney.edu.au Received/Prejeto: 01.08.2016 COBISS: 1.01 ACTA CARSOLOGICA 46/1, 19–32, POSTOJNA 2017 Abstract UDC 551.44:551.3.051(94+4) R. Armstrong L. Osborne: Palaeokarst deposits in ca�es: ��� amples from �astern Australia and Central �urope Palaeokarst deposits are most commonly found in excavations, drill holes and naturally exposed at the Earth’s surface. Some caves however intersect palaeokarst deposits. This occurs in large hypogene caves in the USA, thermal caves in Hungary and in many caves in eastern Australia. Palaeokarst deposits in caves respond to cave forming processes in the same way as hostrock: the palaeokarst deposits form cave walls. A range of palaeokarst deposits is exposed in caves including; filled tubes, walls composed of flowstone, large-scale bodies, breccia pipes, dykes, volcaniclastic palaeokarst and crystalline palaeokarst. As well as being exposed in cave walls, palaeokarst deposits can wholly or partly form speleogens (speleogens made from pal- aeokarst). Records of geological events not preserved elsewhere can occur in palaeokarst deposits in caves. These can be difficult to correlate with conventional geological histories. It is impor- tant to be able to distinguish between palaeokarst deposits, rel- ict sediments and phantom rock (in-situ weathered rock, also called ghost rock; Vergari & Quinif 1997). Relict sediments can be distinguished from palaeokarst deposits because the cave walls bound relict sediments while palaeokarst deposits form the cave walls. Palaeokarst can be distinguished from phantom rock, as palaeokarst is unconformable with the hostrock, with structures in the hostrock not continuing across the bound- ary into the palaeokarst. Hostrock structures and textures do continue across the boundary between unaltered hostrock and phantom rock. Similarly, cave sediments are unconformable or disconformable with the hostrock while phantom rock is conformable with hostrock containing hostrock structures and textures. It has been difficult to explain why palaeokarst occurs in some caves and not others. One explanation worth consider- ing is that palaeokarst deposits are not intersected by caves or sections of caves that contain large perennial streams and/or have undergone large-scale vadose fluvial development capable of escaping from the bounds of structural guidance, such as the caves in the Classical Karst. Key words: palaeokarst, caves, Australia, relict sediments, phantom rock. Izvleček UDK 551.44:551.3.051(94+4) R. Armstrong L. Osborne: Paleokraški sedimenti � jamah: primeri iz �zhodne A�stralije in Srednje ��rope Paleokraške sedimente največkrat najdemo v izkopih, vrtinah in izdankih na zemeljskem površju. Tudi kraške jame lahko seka- jo peleokraške sedimente. Najbolj znane po tem so velike hipo- gene jame v ZDA, na Madžarskem in številne jame vzhodne Avstralije. Paleokraški sedimenti tako kot prikamnina tvorijo jamske stene. V jamah najdemo različne oblike paleokraških sedimentov: zapolnjene cevi, stene iz paleokraške sige, ve- lika sedimentna telesa, cevi breče, dajke, vulkanoklastični pa- leokras in kristalinski paleokras. Poleg tega tudi jamske skalne oblike delno ali povsem nastajajo na paleokraških sedimentih. Ti v jamah nosijo tudi zapise preteklih dogajanj, ki se drugje niso ohranili, a je te zapise velikokrat težko povezati z osta- limi. Pomembno je razlikovati med paleokraškimi sedimenti, reliktnimi jamskimi sedimenti in fantomsko kamnino (in situ preperelo kamnino, v angleščini imenovano tudi ghost rock; Vergari & Quinif 1997). Reliktne jamske sedimente lahko ločimo od paleokraških, ker jamske stene predstavljajo meje prvih, medtem ko paleokraški sedimenti sestavljajo jamsko steno. Paleokraški sedimenti so praviloma v nezveznosti s prikamnino, kar velja tudi za jamske sedimente. Po drugi strani pa struktura in tekstura nespremenjene prikamnine prehajata v fantomsko kamnino. Zakaj je paleokras prisoten le v določenih jamah, je težko odgovoriti. Morda je pomenljivo, da paleokrasa praviloma ne najdemo v jamah z aktivnimi vodotoki oziroma v jamah, ki so v razvoju prešle obdobje vadoznega rečnega vrezo- vanja, kot je to značilno za jame Klasičnega krasa. Ključne besede: paleokras, jame, Avstralija, reliktni sedimenti, fantomska kamnina. ACTA CARSOLOGICA 46/1 – 201720 R. ARMSTRONG L. OSBORNE INTRODUCTION Since 1982 I have been studying palaeokarst depos- its in the caves of eastern Australia. Of the more than 300 cavernous karsts in eastern Australia 19 have caves that intersect palaeokarst deposits. These include four of the best-known and most cavernous karsts in New South wales: Jenolan, wombeyan, Bungonia and yarrangobilly. In Tasmania, caves are known to intersect palaeokarst deposits at three localities: Mole Creek, Junee-Florentine and Ida Bay. Caves in eastern Australia do not just inter- sect a single palaeokarst deposit; some intersect palae- okarst deposits of several different types and ages. During the 1980s and 1990s there were few reports of palaeokarst in caves. There was also little mention of palaeokarst in caves even in compilations dedicated to palaeokarst, such as Bosák, Ford, Głazek and Horáček (1989), and James and Choquette (1988). In 1995 Derek Ford, citing US examples, Jewel and wind Caves in South Dakota and Lechuguilla Cave and Carlsbad Caverns, in New Mexico, suggested that in- tersection of palaeokarst might be more likely in caves formed by per-ascensum processes than in caves formed by per-descensum processes. This would explain the abundance of palaeokarst deposits intersected by the Hungarian thermal caves. No one in the mid 1990s, however, would have suggested that hypogene caves oc- curred in eastern Australia and few outside Hungary and Poland would have suggested their presence in Conti- nental Europe. So there was a problem. why were palae- okarst exposures in caves apparently rare and restricted to “unusual” caves in Europe and North America but abundant in what were considered “normal” caves in eastern Australia? Had palaeokarst deposits in caves been unrecognised elsewhere or were eastern Australian caves somehow “unusual”? A visit to the Classical Karst of Slovenia in 1997 re- vealed palaeokarst in road cuttings, in the seashore and in quarries, even palaeokarst with dinosaur bones, but no palaeokarst deposits were seen exposed in the walls of caves. So it did seem that the intersection of palaeokarst by caves was rare in Continental Europe. Noticeably, in the recent review of palaeokarst by Plotnick, Kenig and Scott (2015) all of the illustrations are of palaeokarst ex- posures in quarries, so palaeokarst exposures in caves continue to be left in the dark. wHERE DO CAVES wITH PALAEOKARST OCCUR? There is good documentation of caves intersecting palae- okarst deposits in Australia, the Czech Republic, Hunga- ry, Slovakia, USA, and UK, and there are reports of caves intersecting palaeokarst deposits in Brazil, Estonia, Israel and South Africa. Some well-documented localities are given in Tab. 1, below. THE SPECIAL CASE OF EASTERN AUSTRALIA with palaeokarst deposits recognized in caves in most of the karst areas in eastern Australia where significant scientific investigation has taken place, it is worth noting the geological and geomorphic characteristics of eastern Australia. The Palaeozoic rocks of eastern Australia form the Tasman Orogen, which has been progressively accreted to the craton. In the south and the west, the Proterozoic Adelaide Orogen lies between the Tasman Orogen and the craton, forming the western half of Tasmania, where significant caves are developed in primary dolomite and magnesite. The telogenic caves of eastern Australia (Tasmanic Caves) are developed in Palaeozoic limestones of the Tas- man Orogen. Most of the caves are developed in small, impounded karsts. These caves are relatively small and shallow, with most less than 3 km long and 50 m deep. Most of these caves lack permanent streams. The Tasman Orogen is divided into two major zones, the western Lachlan-Thompson Orogen and the eastern New England Orogen now separated by the Permo-Triassic Sydney-Bowen Basin (Fig. 1). In the Lachlan-Thompson Orogen the last folding was in the Early Carboniferous while in the New England Orogen the last folding was in Permo-Triassic times. Cavernous limestones in the Lachlan-Thompson Orogen are Or- dovician to Mid-Devonian tropical island arc shelf de- posits, while cavernous limestones in the New England Orogen are Devonian olistoliths and Permian cold-water shelf deposits. The landscape of eastern Australia consists from east to west of a narrow coastal plain, the Great Escarp- ment (Ollier 1982), the Eastern Highlands Plateaux and the western Slopes and Plains. The last significant gla- ciation in mainland Australia occurred in Permo-Car- boniferous times. The Eastern Highlands Plateau was uplifted in the Late Cretaceous with localized uplifts in the Tertiary. while there has been neotectonic activity in Tasmania and Victoria, there is little evidence of neotec- tonic activity further north. with the last major uplift in the Cretaceous, cou- pled with limited and slow geomorphic development and little meteoric cave development in the Cenozoic it is not surprising that ancient caves and ancient cave de- posits could survive in eastern Australia. ACTA CARSOLOGICA 46/1 – 2017 21 PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE Fig. 1: Eastern Australia showing principal tectonic zones and 20 major cavernous karsts. (modified after Fig. 1 in Osborne (2010)). DEFINITION OF PALAEOKARST There have been many attempts to define palaeokarst. One of the best-known definitions is that of James and Choquette (1988, p. 2). They defined paleokarst as: “Ancient karst, which is commonly buried by younger sediments or sedimentary rocks and thus includes both relict paleokarst (present landscapes formed in the past) and buried paleokarst (karst landscapes buried by sedi- ments).” Osborne (2000) noted that this definition did not assist in distinguishing between relict features and what he considered to be “truly palaeokarst features formed during an ancient period of karstification and remain- ing as inherited or relict features in the landscape or rock mass” (Osborne 2000, p.114). tab. 1: Some caves and karsts with palaeokarst deposits. LOCATION SOURCE Nullarbor Plain, Australia H. Shannon, pers. communication Borenore, NSW, Australia Osborne 1984 Bungonia, NSW, Australia Osborne 1984, 1993a Jenolan, NSW, Australia Osborne 1984, 1991, 1993b Timor, NSW, Australia Osborne 1986 Wombeyan, NSW, Australia Osborne 1984, 1993c Cathedral Cave, Wellington, NSW, Australia Osborne 2007c Ida Bay, Tasmania, Australia Osborne & Cooper 2001 My Cave, Mole Creek, Tasmania Osborne 2007d Jewel and Wind Caves, SD, USA Bakalowicz et al. 1987 Grand Canyon, USA Wenrich & Sutphin 1994 Lechuguilla Cave and Carlsbad Caverns, NM, USA D. C. Ford 1995 Cayman Islands Jones & Smith 1988 Peak District, UK T. D. Ford 1989 Wet Sink Cave, Forest of Dean, UK D. Lowe, observed by author 1979 Bihor Mountains, Romania Ghergari et al. 1997 Moravia, Czech Republic J. Otava, observed by author 2013 Okno Cave, Demänovská Valley, Slovakia Osborne 2007b Budapest, Hungary Korpás 1998 Siebenhengste Region, Switzerland Häuselmann 2002 Israel A. Frumkin, pers. communication. Estonia Tõnu Pani, pers. communication. James and Choquette’s definition raises another im- portant issue. Their definition is about landscapes, but most of the features described in this paper are sedimen- tary rocks. So are the cavities these deposits fill palae- okarst, but not the deposits themselves? ACTA CARSOLOGICA 46/1 – 201722 Jennings (1982) considered that the term “palae- okarst” was “problematic”. He proposed that paleokarst should only refer to karst features that are older than the last regional orogeny. If this criterion were applied in Eastern Australia, palaeokarst could only refer to fea- tures or sediments that were older than Permian, early Carboniferous or Mid Devonian, depending in their lo- cation. Most of the palaeokarst deposits that have been identified in Eastern Australia and described in this pa- per were deposited directly after last regional orogeny and are unconformable with the host rock and their bed- ding is generally aligned to the present horizontal. These would not be palaeokarst sensu Jennings (1982). Bosák, Ford, Głazek and Horáček (1989) gave the definition that “Paleokarst refers to karst developed largely or entirely during past geological periods”. This definition relies on the ability to date karst features, both landforms and deposits, but as Osborne (2005) showed, dating karst landforms and deposits is difficult, unreliable and may produce uncertain and controversial results. Ford and williams (1989, p. 507) distinguished be- tween “relict karsts”, “paleokarsts or buried karsts” and “exhumed karsts”. This distinction was based on the de- gree of coupling between the features and the present hy- drogeological system with paleokarsts and buried karsts “completely decoupled from the present hydrogeological system”. A similar definition is found on p. 3 of Ford & williams (2007). wright and Smart (1994) defined paleokarst with the statement that: “ Paleokarst refers to karstic (disso- lution-related features) formed in the past, related to an earlier hydrological system or landsurface.” By this defi- nition many of the features described here are not palae- okarst as they are depositional (sediments, speleothems crystal masses etc.) not dissolutional features. James & Choquette (1988) were clear that palae- okarst is composed of old karst landscapes, while with wright and Smart the focus is on dissolutional features. Bosák, Ford, Głazek and Horáček (1989) just mention karst, but if we look at the chapter by Bosák, Horáček and Panoš (1989) on Paleokarst of Czechoslovakia in Bosák, Ford, Głazek and Horáček (1989) we see that their view of paleokarst is very broad and geological, in- volving large-scale geological history, paleogeography, sediments, fossils, ore minerals, caves and a time-scale extending over 500 million years. Recognising that paleokarst was difficult to define and that the term was used to describe a great range of features, Osborne (2004) provided the following inclu- sive Encyclopaedia definition: “Paleokarst is evidence for karst processes acting in the past.” Noting that: “This definition intentionally includes both karst landforms formed in the past and deposits that fill them” (Osborne 2004, p. 559). A working definition for palaeokarst deposits in caves is that in caves palaeokarst deposits react to cave forming processes in the same way as the hostrock does. Thus cave walls are continuous across palaeokarst exposures. Speleo- gens (hostrock solutional features in caves) can be wholly or partially composed of palaeokarst. Sediments in some Australian caves are hundreds of millions of years old but are not considered palaeokarst because they are not lith- ified, do not behave like hostrock and are enclosed by a cave wall rather than being part of a cave wall. PALAEOKARST DEPOSITS IN CAVES FILLED TUBES AND PHREATIC CAVES IN SPELEOTHEM The most convincing examples of palaeokarst deposits exposed in caves are sections through tubes in cave walls filled with lithified sediments. The exposure in Okno Cave, Slovakia (Fig. 2A) is an outstanding example of this type of exposure. The intersected cavity is 2.7 m high and 1.5 m at it’s widest and is filled by graded-bedded mud- stone, siltstone and sandstone (Osborne 2007b). Dif- ferential weathering of the fill has resulted in the partial exhumation of the tube in the upper part of the exposure, with the lower part of the fill protruding into the younger intersecting passage. Phreatic caves formed in speleothem are also convinc- ing examples of palaeokarst deposits exposed in caves. Fig. 2B shows part of the cave wall in Main Cave at Timor, NSw, Australia. To the left we see a complex smooth wall in massive limestone hostrock, while from the middle to the right we see the cave wall is composed of palaeoflowstone overlying sediment filling the pa- laeokarstic cavity. At the point of the red arrow we can see that the wall surface continues unchanged across the boundary between the hostrock and the palaeokarst flowstone. Timor Main Cave is located high in the land- scape and has phreatic speleogens, some developed in the palaeoflowstone. Osborne (1986) concluded that the phreatic re-excavation of a cave largely filled with palae- oflowstone was a result of the watertable rising after the adjacent valley filled with Eocene basalt. R. ARMSTRONG L. OSBORNE ACTA CARSOLOGICA 46/1 – 2017 23 LARGE-SCALE BODIES Large-scale bodies of palaeokarst fill can have signifi- cant solutional caves developed within them, without exposing the hostrock. Osborne (1991) described a mass of caymanite palaeokarst exposed continuous- ly for more than 70 metres along strike in the lower southern section of Jenolan Caves, NSw, Australia. Fragmentary exposures of the deposit suggest that it is approximately 400 metres long and 10 metres wide. Fig. 3A shows part of the type section through the deposit where a continuous cross-section, 5.5 metres thick is exposed. This section also exposes the upper contact with the enclosing steeply dipping massive limestone hostrock. The rock in this deposit is a grad- ed-bedded carbonate with crinoid ossicles in some of the coarser layers (Fig. 3B) making it a caymanite (pal- aeokarst cavity filled by predominantly marine sedi- ments in coastal settings; Jones 1992). Some parts of the contemporary cave are formed entirely within the mass of palaeokarst (Fig. 3C). BRECCIA PIPES The formation of palaeokarst breccia pipes has often been related to solution collapse processes involving the removal of gypsum, e.g. Friedman (1997). Breccia pipes can also be seen forming today in situations not involving sulfate removal. An active non-sulfate brec- cia pipe occurs in Mladeceske Cave in the Czech Re- public (Fig. 4A). The Palaeozoic limestones in eastern Australia are not interbedded with gypsum, but brec- cia pipes are developed in them. An ancient breccia pipe, probably of Late Devonian age has been inter- sected by Cathedral Cave at wellington Caves, NSw, Australia (Fig. 4B). The large clasts in the pipe are Fig. 2: A= Lithified palaeokarst deposit in Okno Cave, Slovakia; white rule for scale is 1 m. Note how the brown palaeokarst de- posit fills a hollow in the mas- sive limestone cave wall, after Osborne 2007b. b= Wall of main Cave, timor NSW, Australia. Red arrow points to boundary between the hostrock (left) and the flowstone palaeokarst (right). Green arrow points to microbat for scale. Fig. 3: A= type Section through caymanite deposit at jenolan Caves, NSW, Australia. b= Thin section of the caymanite showing graded bedding. C= Cave passage completely developed in caymanite. Note trace of horizontal bedding in left hand wall. bedding in the enclos- ing Silurian hostrock is steep to vertical. PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE ACTA CARSOLOGICA 46/1 – 201724 mostly massive limestone while the walls of the pipe are composed of intensely folded thinly bedded limestone. The red matrix between the large clasts is composed of calcite, kaolinite and quartz. DyKE-LIKE FEATURES It is not uncommon for caves in eastern Australia to intersect igneous dykes, but reports of dykes in caves elsewhere are uncommon. Dyke-like features in caves include conventional igneous dykes, altered igneous dykes, clastic dykes, caymanite dykes, pyroclastic dykes and Neptunian dykes. Figs. 5A and B show a clastic dyke intersected by Grill Cave at Bungonia, NSw, while Figs. 5C and D show a caymanite dyke intersected by Hogan Cave at Bungonia, NSw. Most dyke-like features exposed in caves are highly weathered so it is very difficult in the field to distinguish between a weathered dolerite dyke and a clastic dyke with ferruginous cement. Cavers fre- quently miss-identify dykes often confusing them with shale beds and vice-versa. In some situations, dolerite dykes can be completely altered to calcite (see Osborne 2003) and in the field could be miss-interpreted as a car- bonate palaeokarst deposit. Intrusive, clastic and caymanite dykes found in caves are not always vertical but can be dipping, following dipping conjugate joints (Fig. 5A). As well as filling with clastics; grikes and cave passages guided by joints can fill with lava, fossilizing them (making them palaeokarst features). As Os- borne (2005) discussed, it is very dif- ficult to distinguish between lava filled joint-guided cavities (palaeokarst) and intrusive dykes (not palaeokarst) in caves. Fig. 5: A= dipping clastic dyke in Grill Cave, bungonia Caves, NSW, Australia. b= Thin section of clastic dyke in “A,” note matrix and lack of carbonate cement. Sand grains are approx. 0.5 mm in diameter. C= dyke- like joint-filling of carbonate palaeokarst (dark material adjacent to author’s helmet), hogan Cave, bungonia Caves, NSW, Aus- tralia (Photo: j. Sydney). d= Thin section of joint filling of carbonate palaeokarst in “C”, red arrows point to crinoid fragments ap- prox. 1 mm in size. R. ARMSTRONG L. OSBORNE Fig. 4: A= Contemporary non-sulfate breccia pipe in mladeceske Cave, Czech Republic. base of breccia pipe is approx. 2 m wide. b= Palaeokarst breccia pipe, Wellington Caves, NSW, Australia; white post is 0.9 m high. ACTA CARSOLOGICA 46/1 – 2017 25 VOLCANICLASTIC PALAEOKARST Volcaniclastic palaeokarst is produced when tephra either from ash falls or mudflows enters caves or karst depressions, becomes lithified and is later intersected by the development of a new system of cavities. In eastern Australia volcaniclastic palaeokarst deposits in caves have been identified at two localities, wellington Caves and wombeyan Caves. In Cathedral Cave at wellington Caves (Osborne 2007c) pyroclastics with liesegang band- ing are found filling tubes in massive limestone (Fig. 6A). In thin section pesudomorphs of calcite after augite stand out against a dark grey glassy groundmass (Fig. 6B). At wombeyan Caves, marmorized Silurian lime- stone is unconformably overlain by volcaniclastics of the Bindook Porphyry Complex, a Devonian mega-Plinian assemblage (Osborne 1993c). Palaeokarst deposits found at wombeyan include intersected volcaniclastic-filled tubes (Fig. 6C) and pyroclastic dykes exposed both in the caves and at the surface (Fig. 6D). Osborne (2007a) suggested that pyroclastic palae- okarst deposits should be much more widespread than currently reported given the abundance of stratovolca- noes in both modern and ancient island arcs where much of the world’s limestone was and is being deposited. CRySTALLINE PALAEOKARST There are two main types of crystalline palaeokarst, crys- talline breccias and crystal vughs. Crystalline Breccias Crystalline breccias are composed of angular fragments, of mostly hostrock, separated by a crystalline matrix. Based largely on drill hole data, Loucks (2007), interpret- ed cave breccias, which he considered to be “paleocave facies” as the product of caves collapsing due to burial and he developed a scheme for classifying them. while Loucks envisaged these breccias as being formed at the end of a cave’s history, it is not rare to find “paleocave” breccias as the hostrock of quite intact caves. Loucks’ first category, crackle breccias, have minor displacement be- tween clasts that appear to fit together. Fig. 7A shows a body of crackle breccia exposed in the cave wall of Creek Fig. 6: A= Volcaniclastic palae- okarst deposit with liesegang banding filling a tube in the wall of Cathedral Cave, Wellington Caves, NSW, Australia. b= Thin section of palaeokarst in “A” showing calcite pseudomorph af- ter augite in glassy groundmass. Pseudomorph is 1 mm across. C= Filled tube in upper wall of bullio Cave, Wombeyan Caves, NSW, Australia. Upper dark layered de- posit is laminated volcaniclastics. Lower brown, layered deposit is younger sediment unconformably abutting the palaeokarst. tube is approx. 1 m wide at top of brown deposit. d= Surface exposure of pyroclastic dyke with lens cap (55 mm) standing out above dis- solving marble bedrock. PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE ACTA CARSOLOGICA 46/1 – 201726 Cave an ephemeral stream cave at wombeyan Caves, NSw. The clasts are composed of marble while the ma- trix is crystalline calcite. A smaller scale crystal breccia with more angular clasts is shown in Fig. 7B. In this case although they are widely separated by matrix, some clasts do fit together making it a “mosaic” breccia. Fig. 7C, an image looking up into the ceiling of Katie’s Bower, a so- lution chamber in Chifley Cave at Jenolan Caves, NSw, shows large hostrock blocks, 1 m + floating in a crystalline matrix. Traces of bedding in the blocks suggest that some have rotated. Some of the blocks are sub-rounded indi- cating solution during emplacement of the matrix. This is an example of a “matrix-supported chaotic breccia” of Loucks (2007). An alternative and older interpretation of crystal breccias in karst was given by Sass-Gustkiewicz (1974) who considered them to have a solutional origin. Unless there is clear evidence that the breccia fills a karst void, it is very difficult to be sure that these breccias are palaeokarst features rather than the products of non- karst tectonic and hydrothermal processes. Palaeokarst Crystal Vughs Cave passages can intersect crystal-lined vughs that are not related to the processes that formed the contempo- rary caves. Vughs can range in size from decimetres to tens of metres. Care must be taken not to confuse inter- sected palaeokarst crystal vughs with the remnants of more recent crystal linings that are filling indentations in the contemporary cave wall. young sediments can be de- posited in the open spaces of ancient vughs, so care must also be taken not to confuse unconsolidated ancient and modern fillings in vughs. The vugh in Fig. 7D is one of many vughs exposed over a distance of at least 50 metres in the walls and ceiling of the Imperial Cave at Jenolan Caves, NSw (Osborne 1984). Fig. 7D shows a number of important features of the crystal vughs. At the base (i) we see geopetal sediment aligned to the present horizontal, indicating that there has been no tectonism since deposi- tion. In the centre (ii) there is a bedrock relic coated with large white crystals, which are also visible at the top of the vugh (iii). SPELEOGENS IN PALAEOKARST Since palaeokarst deposits in caves behave like hostrock, it should not be surprising to find speleogens developed wholly or partly in palaeokarst. These are formed as a result of later periods of speleogenesis and include pen- dants (Fig. 8A), juts (projections from cave walls), cers (projections from cave floors), and bridges (Fig. 8B). DEMARCATION OF PALAEOKARST Distinguishing palaeokarst from other features in caves is not simple and can be controversial. Large prominent objects can prove to be the most difficult to interpret. An excellent example is the large caymanite deposit shown in Fig. 3. Before this deposit was recognised, small deposits of laminated sediment had been noted elsewhere in Je- nolan Caves and dismissed as shale beds in the Silurian limestone sequence. In 1984 it became clear that these deposits were unconformable with the Silurian hostrock. The large mass of gently dipping laminated limestone in Fig. 3A was first recognised as a palaeokarst deposit in July 1986. Its notable feature is the unconformable boundary with the hostrock, best seen in Fig. 9A. It was difficult to imagine that such a large mass of limestone could be a palaeokarst deposit so alternative explanations Fig. 7: A= Crackle breccia with party aligned clasts, Creek Cave, Wombeyan Caves, NSW, lens cap is 55 mm. b= Crystal brec- cia, basin Cave, Wombeyan Caves, NSW, black pen in centre field is 150 mm long. C= matrix- supported chaotic crystal breccia with rotated blocks in the ceiling of Chifley Cave, jenolan Caves, NSW. blocks in the breccia are 1 m+ across. d = Crystalline palaeokarst vugh with geopetal sediment, Imperial Cave, jenolan Caves, NSW, lens cap is 55 mm. R. ARMSTRONG L. OSBORNE ACTA CARSOLOGICA 46/1 – 2017 27 that it was a fault block or a tight fold in the bedrock were explored and rejected principally because it has a sutured unconformable boundary with the hostrock and showed no evidence of either folding or faulting. As the deposit was strongly lithified and behaved like hostrock it had to be palaeokarst. In the late 1980s it was thought that old sediments in caves should be high in the landscape and younger sediments should be lower in the landscape as the cave worked its way down through the limestone. This deposit was clearly very old, but it was in the low- est part of the cave system, close to the watertable. Time constraints on this deposit place it between the last fold- ing of the limestone (Early Carboniferous) and the lime- stone’s burial under the Sydney Basin (Late Carbonifer- ous to Early Permian). If this deposit had been composed of bedded terrigenous mud with marine fossils it would have not been controversial, as there are Permian marine mudstones and glacial marine sediments 17 km east of Jenolan and the caves were probably once buried under these strata. The nearest Carboniferous limestones are lo- cated 212 km northeast of Jenolan, so the occurrence of marine carbonate palaeokarst at Jenolan is controversial and an enigma. FEATURES ON THE BOUNDARy OF PALAEOKARST The bridge in Gable Cave, Cliefden Caves, NSw shown in Fig. 8B is composed of very poorly lithified clay and does not behave like cave wall material. However during the most recent phase of flooding and sediment removal it has proved sufficiently lithified to form a respectable phreatic speleogen. Fig. 9B shows poorly lithified red sediment in Swansong Cave at Cliefden Caves, NSw, the cave directly above where the bridge in palaeokarst in Fig. 8B is located. In Fig. 9B, directly above the caver’s head, we see a mass of red sediment behaving as hostrock forming the cave wall and the cave ceiling inside the old- er limestone hostrock ceiling. In both cases the materi- als involved are poorly lithified and probably relatively young, but as they respond to speleogenetic processes in the same way as hostrock, both deposits are considered to be palaeokarst. PALAEOKARST DEPOSITS VS RELICT SEDIMENT Some Australian caves contain sediments that are hun- dreds of millions of years old (Osborne et al. 2006). These sediments are not considered palaeokarst because they are not behaving like hostrock, but partly fill a cave en- closed by hostrock walls (Fig. 9C). Despite their great age these sediments are not lithified which illustrates an issue raised by Osborne (2005), the “lithification trap”. Geolo- gists often assume that lithified sediments must be older than unlithified ones, but in caves where the main proc- ess of lithification is cementation, permeable sediments like sands and gravels can become lithified much more rapidly than impermeable materials such as muds. PALAEOKARST VS PHANTOM ROCK Distinguishing palaeokarst deposits from phantom rock is a confusing issue as phantom rock can itself be palaeokarst (Quinif et al. 2006). The distinction between palaeokarst deposits, partly lithified cave sediments and phantom rock is becoming an impor- tant issue as more cases of phantom rock are reported (Dubois et al. 2014) and cave formation by the removal of weathered dolomite, dedolomite and magnesite are being investigated. Palaeokarst deposits and cave sediments should be unconformable with the hostrock and should have a different texture and composition to the surrounding Fig. 8: A= Pendant com- posed of crystalline palae- okarst, Imperial Cave, jeno- lan Caves, NSW, Australia. b= bridge composed of poorly lithified clay, Gable Cave, Cliefden Caves, NSW, Australia. PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE ACTA CARSOLOGICA 46/1 – 201728 hostrock. Traces of structures in the hostrock such as veins, bedding, fractures and alignment of clasts and fos- sils should not continue across the hostrock -palaeokarst unconformity. In contrast, phantom rock and related altered/ weathered hostrock should have the same large-scale texture and contain the same fossils as the hostrock. In phantom rock the alignment of traces of structures in the hostrock such as veins, bedding, fractures and the alignment of clasts and fossils should continue across the hostrock-phantom rock boundary. This can be seen in Fig. 9D, where sparry veins in the marble hostrock ex- tend into the adjacent muddy orange phantom rock, lo- cally described as ochre. Fig 9: A = Upper unconformable boundary of palaeokarst deposit shown in Fig. 3A (Photo: A. tyc). Red line indicates trace of bed- ding in enclosing massive Siluri- an limestone, green line indicates bedding in palaeokarst deposit. Curved unconformable boundary is the ceiling of the filled ancient cave. b = Poorly lithified sedi- ment, red behaving as cave wall and ceiling so considered to be palaeokarst, Swansong Cave, Cli- efden Caves, NSW. C = deposit of unlithified clay (yellow) dated as Carboniferous by Osborne et al. (2006), temple of baal, jenolan Caves, NSW. Note overlying pink strata and clearly defined cave wall behind. black sleeve on yel- low handle at lower left of image is 100 mm long. d = boundary between whitish hostrock and or- ange muddy phantom rock. Note sparry veins, indicated by red ar- rows extending from the hostrock into the phantom rock. E= 6.3 m deep x 10 m wide archaeologi- cal pit inside Fa-hein Cave, Sri Lanka, in phantomized gneiss mistaken for cave sediment. PHANTOM ROCK VS SEDIMENTS Just as phantom rock can be confused with palaeokarst it can also be confused with unconsolidated sediment. Osborne et al. (2013) reported that archaeologists in Sri Lanka had excavated a pit, 6.3 metres deep in phan- tomized gneiss thinking it was cave sediment, apparently confusing traces of weathered foliation with bedding (Fig. 9E). wHy DO CAVES IN SOME PLACES INTERSECT PALAEOKARST DEPOSITS? T.D. Ford (1976), when discussing the Permian and Car- boniferous limestones of the United Kingdom noted that: “any limestone of appreciable age may have gone through R. ARMSTRONG L. OSBORNE ACTA CARSOLOGICA 46/1 – 2017 29 more than one cycle of karstification”, introducing the possibility that in “older” rocks multiple karstification and the intersection of palaeokarst deposits is more likely to occur. Osborne (1984) found and described palaeokarst deposits in caves at five localities in eastern mainland Australia. These localities were originally selected for investigation because of their close proximity to major unconformities. In 1995, D.C. Ford suggested that per-ascensum caves are more likely to intersect palaeokarst than per- descensum caves. Osborne (2000, 2002, 2005 & 2013) suggested caves that are: 1. Located close to major unconformities; 2. Developed in Palaeozoic or older rocks; 3. In small, impounded, karsts with limited poten- tial pathways for cave development, where new caves are more likely to intersect older filled caves; 4. Have strong structural guidance, such as per-as- censum caves; 5. Occur in regions with low erosion rates and where the last tectonism occurred in the distant past; … are more likely to intersect palaeokarst. Exceptions to these criteria show which are the key determinants. Henry Shannon (pers. comm.) has iden- tified palaeokarst intersected by caves in the Nullarbor Plain of southern central Australia. The Nullarbor Plain is underlain by Miocene limestone and as the world’s largest karst area with an oceanic coastline, cannot be considered to be old, small or impounded. So criteria 2 and 3 are not essential, while Okno Cave in Slovakia is developed in Triassic Limestone, is in a large karst area and is in an area of high erosion with recent tectonism. So criteria 2, 3, and 5 are not essential This leaves criterion 4, structural guidance and per-ascensum development as key criteria, so D.C. Ford (1995) was right after all. The exceptions to the rules such as Nullarbor and Okno suggest that the wrong question is being asked. So perhaps it is more useful to ask: – “which caves do not intersect palaeokarst?” One possible answer worth further investigation is that: − Palaeokarst deposits and speleogens made from palaeokarst do not occur in caves or sections of caves that contain large perennial streams and/or have under- gone large-scale vadose fluvial development capable of escaping from the bounds of structural guidance. CONCLUSIONS Intersection of palaeokarst deposits by caves is common in eastern Australia and occurs in Estonia, Hungary, Isra- el, Romania, Switzerland, the UK, and in hypogene caves in the USA. These areas do not have a lot in common in terms of hostrock geology, geomorphology, hydrology or climate. Palaeokarst deposits in caves can be difficult to identify and distinguish from sediments, structural fea- tures and phantom rock. Palaeokarst deposits in caves can record events that are not represented elsewhere in the geologic record. These events can be difficult to reconcile with traditional views about regional geologi- cal history. This may result in controversy. Speleogens formed from palaeokarst are uncontroversial because if palaeokarst deposits can behave like hostrock and form cave walls, they can also form pendants and juts etc. why some caves intersect palaeokarst deposits and others do not remains a problem. One possible answer is that ex- posures of palaeokarst deposits occur in caves that have never had a perennial stream in them or in caves in which there has been minimal fluvial modification. ACKNOwLEDGEMENTS The author would like to thank the many people from different countries who have assisted in the search for and documentation of palaeokarst deposits in caves over the past thirty-four years: In Australia, Tony Allan, David Branagan, David Colchester, Ian Cooper, Ross Pogson, Henry Shannon, Joe Sydney, John wylie, Horst Zwing- mann, many cave managers, guides and national park staff. In Austria, the late Robert Seemann, In the Czech PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE ACTA CARSOLOGICA 46/1 – 201730 REFERENCES Bakalowicz, M. J., Ford, D. C., Miller, T. E., Palmer, A. N. & M.V. Palmer, 1987: Thermal genesis of dis- solution caves in the Black Hills, South Dako- ta.- Geological Society of America Bulletin, 99, 729–738. DOI: https://doi.org/10.1130/0016-7606- (1987)99<729:tgodci>2.0.co;2 Bosák, P., Ford, D.C., Głazek, J., & Horácek, I., Eds. 1989: Paleokarst. A Systematic and Regional Review. El- sevier and Academia, pp. 726, Amsterdam and Praha. Bosák P., Horáček, I. & V. Panoš.,1989 Paleokarst of Czechoslovakia.- In Bosák, P. et al. (eds.) Paleokarst. A Systematic and Regional Review. Elsevier and Ac- ademia, pp. 107–135, Amsterdam and Praha. Dubois, C., Quinif, y., Baele, J-M., Barriquand, L., Bini, A., Bruxelles, L., Dandurand, G., Havron, C., Kauf- mann, O., Lans, B., Maire, R., J. Martin, J., Rodet, J., Rowberry, M.D. Tognini, P. & A.Vergari, 2014: The process of ghost-rock karstification and its role in the formation of cave systems.- Earth-Science Re- views, 131,116–148. DOI: https://doi.org/10.1016/j. earscirev.2014.01.006 Ford, D.C., 1995: Paleokarst as a target for modern karstification.- Carbonates and Evaporites, 10, 2, 138–147. DOI: https://doi.org/10.1007/bf03175399 Ford, D.C. & P.w. williams, 1989. Karst Geomorphology and Hydrology.- Unwin Hyman, pp. 601, London. Ford, D.C. & P.w. williams, 2007: Karst Hydro- geology and Geomorphology. John wiley & Sons, 578 pp, Chichester. DOI: https://doi. org/10.1002/9781118684986.ch5 Ford, T.D., 1976: The geology of caves.- In Ford, T.D. & C.H.D. Cullingford (eds.) The Science of Speleology. Academic Press, pp. 11–60, London. Ford, T.D., 1989: Paleokarsts of Britain.- In Bosák, P. et al. (eds.) Paleokarst. A Systematic and Regional Review. Elsevier and Academia, pp. 51–70, Amster- dam and Praha. Friedman, G.M., 1997: Dissolution-collapse breccias and paleokarst resulting from dissolution of evaporite rocks, especially sulfates.- Carbonates and Evapo- rites, 12, 53–63. DOI: https://doi.org/10.1007/ bf03175802 Ghergari, L., Tamas, T., Damm, P. & F. Forray, 1997: Hydrothermal paleokarst in Pestera din Valea Rea (Bihor Mountains, Romania).- Theoretical and Ap- plied Karstology, 10,115–125. Häuselmann, P., 2002: Cave genesis and its relationship to surface processes: Investigations in the Siebenhengste region (bE, Switzerland).- PhD thesis. University of Fribourg, pp. 168. James, N.P & P.w. Choquette (eds.), 1988: Paleokarst.- Springer-Verlag, pp. 416, New york. DOI: https:// doi.org/10.1126/science.240.4855.1071 Jennings, J.N., 1982. Principles and problems in recon- structing karst history.- Helictite, 20, 2, 37–52. Jones, B., 1992: Caymanite, a cavity-filling deposit in the Oligocene-Miocene Bluff Formation of the Cayman Islands.- Canadian Journal of Earth Science, 29, 720–735. DOI: https://doi.org/10.1139/e92-062 Jones, B. & D.S. Smith, 1988: Open and filled features on the Cayman Islands: implications for the recogni- tion of paleokarst.- Canadian Journal of Earth Sci- ences, 25, 1277–1291. DOI: https://doi.org/10.1139/ e88-123 Korpás, L. 1998: Paleokarst studies in Hungary.- Occa- sional Papers of the Geological Institute of Hunga- ry, 195, 1–139. Loucks, R.G., 2007: A review of coalesced, collapsed- paleocave systems and associated suprastratal de- formation.- Acta Carsologica, 36, 121–132. DOI: https://doi.org/10.3986/ac.v36i1.214 Ollier, C.D., 1982: The Great Escarpment of eastern Aus- tralia: tectonic and geomorphic significance.- Jour- nal of the Geological Society of Australia, 29,13–22. DOI: https://doi.org/10.1080/00167618208729190 Republic, Pavel Bosák, Vaclav Cilek, Jaroslav Kadlec, Jiri Otava and Karel Zak. In Estonia, Tõnu Pani, In France, Philippe Audra, In Great Britain, Trevor Ford, John Gunn and David Lowe. In Hungary, Katalin Bolner- Takács, the late László Korpás, Szabolcs Leél-Õssy and János Móga. In Israel, Amos Frumkin, In Poland, the late Jerzy Głazek, Michał Gradziński, Andrzej Tyc and Jan Ur- ban, In Romania, Tudor Tamas. In Slovakia, Pavel Bella, Ľudovít Gaál, Peter Gažík, Jozef Hlaváč, Josef Psotka and Pavel Staník. In Slovenia, Andrej Kranjc, Andrej Mihevc, Bojan Otoničar, Tadej Slabe, France šušteršic and Nadja Zupan-Hajna, and in Switzerland, Philipp Häuselmann. Penney Osborne is thanked once again for her attention to detail in reading the proofs. The paper has been im- proved by suggestions from two anonymous referees. R. ARMSTRONG L. OSBORNE ACTA CARSOLOGICA 46/1 – 2017 31 Osborne, R.A.L., 1984: Multiple karstification in the Lachlan Fold Belt in New South wales: Reconnais- sance Evidence.- Journal and Proceedings of the Royal Society of New South wales, 107,15–34. Osborne, R.A.L., 1986: Cave and landscape chronology at Timor Caves, New South wales.- Journal and Proceedings of the Royal Society of New South wales, 119, 1/2, 55–76. Osborne, R.A.L., 1991: Palaeokarst deposits at Jenolan Caves, N.S.w.- Journal and Proceedings of the Roy- al Society of New South wales, 123, 3/4, 59–73. Osborne, R.A.L., 1993a: A new history of cave de- velopment at Bungonia, N.S.w.- Australian Geographer, 24,1, 62–74. DOI: https://doi. org/10.1080/00049189308703078 Osborne, R.A.L., 1993b: Geological Note: Cave forma- tion by exhumation of Palaeozoic palaeokarst de- posits at Jenolan Caves, New South wales.- Austra- lian Journal of Earth Sciences, 40, 591–593. DOI: https://doi.org/10.1080/08120099308728106 Osborne, R.A.L., 1993c: The history of karstification at wombeyan Caves, New South wales, Australia, as revealed by palaeokarst deposits.- Cave Science, 20, 1, 1–8. Osborne, R.A.L, 2000: Paleokarst and its significance for speleogenesis.- In Klimchouk, A.B. et al. (eds.) Spe- leogenesis, Evolution of Karst Aquifers. National Spe- leological Society, pp. 113–123, Huntsville. Osborne, R.A.L., 2002: Paleokarst: Cessation and re- birth? In Gabrovšek, F. (ed.) Evolution of karst: from prekarst to cessation. Založba ZRC, pp. 97–114, Lju- bljana. Osborne, R.A.L., 2003: Partitions, compartments and portals: Cave development in internally impounded karst masses.- Speleogenesis and Evolution of Karst Aquifers: The virtual Scientific Journal, 1,4, 1–12. DOI: https://doi.org/10.5038/1827-806x.34.1.6 Osborne, R.A.L., 2004: Paleokarst.- In Gunn, J. (ed.) Encyclopedia of Cave and Karst Sciencet. Fitzroy Dearborn, pp. 559–561, New york. DOI: https://doi. org/10.4324/9780203483855 Osborne, R.A.L., 2005: Dating ancient caves and related palaeokarst.- Acta Carsologica 34,1, 51–72. Osborne, R.A.L, 2007a: The world’s oldest caves: – how did they survive and what can they tell us?- Acta Carsologica, 36,1,133–142. DOI: https://doi. org/10.3986/ac.v36i1.215 Osborne, R.A.L., 2007b: Intensely lithified palaeokarst deposits in Okno Cave, Demänovská Valley, (Slova- kia).- Geologica Carpathica, 58, 6, 565–578. Osborne, R.A.L., 2007c: Cathedral Cave, welling- ton Caves, New South wales, Australia. A multi- phase, non-fluvial cave.- Earth Surface Processes and Landforms, 32, 2075–2103. DOI: https://doi. org/10.1002/esp.1507 Osborne, R.A.L., 2007d: Geodiversity of mole Creek Caves, tasmania.- Report to the Australian Depart- ment of Environment and water Resources, June 2007, pp. 64. Osborne, R.A.L., 2010: Rethinking eastern Australian caves.- In: Bishop, P. (eds.) Australian Landscapes. Geological Society of London Special Publica- tion 346, 289–308. DOI: https://doi.org/10.1144/ sp346.15 Osborne, R.A.L., 2013: Preservation and burial of an- cient karst.- In: Shroder, J.F. (ed) treatise on Geo- morphology. Volume 6, Academic Press, pp. 95–103, San Diego. DOI: https://doi.org/10.1016/b978-0- 12-374739-6.00132-9 Osborne, R.A.L & I. B. Cooper, 2001: Sulfide-bearing palaeokarst deposits at Lune River Quarry, Ida Bay, Tasmania.- Australian Journal of Earth Sciences, 48, 401–416. DOI: https://doi.org/10.1046/j.1440- 0952.2001.00868.x Osborne, R.A.L., weliange, w.S., Jayasingha, P., Dan- deniya, A.S, Algiriya A.K.P.P. & R. E. Pogson, 2013: Caves and karst-like features in Proterozoic gneiss and Cambrian granite, southern and central Sri Lanka: An Introduction.- Acta Carsologica, 42,1, 25–48. DOI: https://doi.org/10.3986/ac.v42i1.630 Osborne, R.A.L., Zwingmann, H., Pogson, R.E & D.M. Colchester, 2006: Carboniferous clay deposits from Jenolan Caves, New South wales, Australia.- Aus- tralian Journal of Earth Sciences, 53, 3, 377–405. DOI: https://doi.org/10.1080/08120090500507362 Plotnick, R.E., Kenig, F & A.C. Scott, 2015: Using the voids to fill the gaps: caves, time, and stratigra- phy.- In: Smith, D. G. et al. (eds.) Strata and time: Probing the Gaps in Our Understanding.- Geologi- cal Society, 233–250, London. DOI: https://doi. org/10.1144/sp404.5 Quinif, y., Meon, H., & J. yans, 2006: Nature and dating of karstic filling in the Hainaut Province (Belgium). Karstic, geodynamic and paleogeographic implica- tions.- Geodinamica Acta, 19. 2, 73–85. DOI: ht- tps://doi.org/10.3166/ga.19.73-85 Sass-Gustkiewicz, M. 1974: Collapse breccias in the ore- bearing dolomite of the Olkusz mine (Crakow- Sile- sian ore district).- Roczink Polskiego Towarzystwa Geologicznego, 44, 217–226. PALAEOKARST DEPOSITS IN CAVES: ExAMPLES FROM EASTERN AUSTRALIA AND CENTRAL EUROPE ACTA CARSOLOGICA 46/1 – 201732 Vergari, A & y. Quinif, 1997: Les paleokarsts du Hain- aut (Belgique).- Geodinamica Acta, 10, 4, 175–1 87. DOI: https://doi.org/10.1080/09853111.1997.11105 301 wenrich, K.J. & H.B. Sutphin, 1994: Grand Canyon caves, breccia pipes and mineral deposits.- Geology Today, May–June 1994, 97–104. DOI: https://doi. org/10.1111/j.1365-2451.1994.tb00880.x wright, V.P & P.L. Smart, 1994: Chapter 13 Paleokarst (Dissolution Diagenesis): Its occurrence and hy- drocarbon exploring significance.- In: wolf, K.H. & G.V. Chilingarian (eds.) developments in Sedimen- tology.- Elsevier, 51, 477–517, Amsterdam. DOI: ht- tps://doi.org/10.1016/s0070-4571(08)70447-3 R. ARMSTRONG L. OSBORNE