Annales Societatis Geologorum Poloniae (2010), vol. 80: 303-314. SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE: ROLE OF TRAPPING AND BINDING OF DETRITAL PARTICLES IN FORMATION OF CAVE DEPOSITS Michał GRADZIŃSKI1, Maria Jolanta CHMIEL2, Anna LEWANDOWSKA1 & Beata MICHALSKA-KASPERKIEWICZ3 11nstitute of Geological Sciences, Jagiellonian University, Oleandry 2a, 30-063 Krakow, Poland; e-mail: michal.gradzinski@uj. edu.pl, anna. lewandowska@uj. edu.pl Department of Microbiology, University of Agriculture in Krakow, Aleja Mickiewicza 24/28, 30-059 Krakow, Poland; e-mail: mjchmiel@poczta. onet.pl KTJSpeleoklub Bielsko Biała, 1-go Maja 45, 43-300 Bielsko-Biała, Poland; e-mail: beata_michals@poczta.onet.pl Gradziński, M., Chmiel, M. J., Lewandowska, A. & Michalska-Kasperkiewicz, B., 2010. Siliciclastic micro- stromatolites in a sandstone cave: role of trapping and binding of detrital particles in formation of cave deposits. Annales Societatis Geologorum Poloniae, 80: 303-314. Abstract: The article deals with finely laminated microstromatolites composed of detrital siliciclastic particles, mainly quartz, feldspars and clay minerals, lintng the walls of W Sopotni Wielkiej Cave (Polt sh Outer Carpa- thians). Newly precipitated mineral phases do not contribute to their growth. The microstromatolites cover vertical and overhanging walls of the cave. They form subhorizontal ripples and tongue-shaped stepped microterracettes. The stromatolites are constructed by bacteria and Actinomycetes. Seven morphotypes of micro-organisms have been distinguished. Trapping and binding of detrital particles result in the microstromatolite growth. The growth is influenced by the relatively close distance to the soil cover which provides detrital mineral particles and by the presence of gravitationally widened fissures which guide the water transporting mineral particles down to the cave. The particles are transported only during wet periods. The episodic supply of the particles results in visible lamination of microstromatolites. The microterracettes form in zones of increased water-flow. The lack of auto- chthonous components most probably reflects too low saturation of water to precipitate any minerals. Key words: bacteria, Actinomycetes, biofilm, speleothems, Outer Carpathians. Manuscript received 18 September 2010, accepted 18 November 2010 INTRODUCTION Constructive role of micro-organisms in speleothem formation is a focus of growing interest (see review by Northup & Lavoie, 2001; Jones, 2001, 2010; Taborosi, 2006 and references quoted therein). A great body of litera- ture exists on the importance of bacteria and fungi for pre- cipitation of carbonate speleothems, especially for the growth of moonmilk, cave pisoids and some subaqueous formation (e.g., Gradziński et al., 1997b; Gradziński, 2000; Melim et al., 2001; Jones, 2009a; Curry et al., 2009). The plausible direct or indirect influence of these micro-organi- sms on the crystaltization of needle-fibre calcite in cavern environment is commonly discussed (Blyth & Frisia, 2008; Richter et al., 2008). The preciphation of speleothem-for- ming manganese and iron oxi des, opal and phosphates is also mediated by biological activity (e.g., Gradziński et al., 1995; Manolache & Onac, 2000; Aubrecht et al., 2008; Jones, 2009b). Conversely, little attention has been drawn to trapping and bindmg of detrital grains of various mineral composi- tion as a conttructive mechanism in speleothem growth. Such mechanism has been ment ioned on the occasion of studying the microbially driven precipitation of diverse speleothems in cavern environment. However, it has only been regarded as a subordinate contribution to speleothem growth. Jones and Motyka (1987) showed that algae trap and bind carbonate micrite, which eventually leads to for- mation of the well visible lamination. Cyanobacteria, repre- senting Gleocapsa sp., are capable of trapping air-born quartz grains and insect fragments, which collectively con- tribute to the formation of crayfish-like profile stalagmites located in entrance parts of Austratian caves (Cox et al., 1989a, b; James et al., 1994). Jones (1995) noted some ex- amples of trapping and binding of detrital grains to the sub- strate by organic filaments in the biofilm covering walls in twitight zones of caves. Cunningham et al. (1995, fig. 7) illustrated binding of corrosion residuum by filamentous miro-orgamsms in Lechugilla Cave (New Mexko). Cana- veras et al. (2001) mentioned trapping and binding of 304 M. GRADZIŃSKI ET AL. Fig. 1. Location of W Sopotni Wielkiej Cave micritic grains resulting from substrate breakdown in the fa- mous Altamira cave. Gradziński and Holubek (2005) found detrital silicate and dolomite grains incorporated into sub- aqueous cottonball speleothem in Zlomiskä Cave (Slo- vakia); the cottonballs presumably originated under biologi- cal mediation. Baskar et al. (2006) suggested that detrital grains organically trapped and bound are incorporated into carbonate speleothems in Borra Caves (India). Also, biolo- gicaly mediated opal speleothems from Venezuetan caves contain some grains trapped and bound by micro-organisms (Aubrecht et al., 2008). Detrital grains contribute greatly to format ion of some speleothems in sandstone caves in the PoKsh Outer Carpathians (Urban et al., 2007). In some cases they compose deposits whose relief imitates the mor- phology taken by speleothems. Some of these forms are ce- mented with opal. Clastic formations built of sand- and silt-sized grains are present in many caves (Hill & Forti, 1997, p. 219-221 and references therein). Some of them are akin to carbonate speleothems in their shape and dimension although they are only cemented by carbonate minerals (e.g., Baker, 1942). In fact, many of them represent erosional remnants after accu- mut ation of cave clastics (e.g., Gradziński & Radomski, 1960) and they do not share their origin with typical speleothems. Vermiculations represent another type of cave formation composed of siliciclastic material (Hill & Forti, 1997, p. 221-223). They are found on cave walls or ceilings as irregular spots or bifurcating stripes. Their array gives the overall impression of a group of warms or a tiger or leopard skin. In the maj ority of papers no suggestions have been made on the possible role of micro-organisms in the origin of the above clastic cave formations, excluding special type of vermiculation in sulfidic caves. Although Bini et al. (1978) considered influence of bacteria on formation of vermiculation, ultimately they ruled out this influence. This paper documents the presence of microstromato- lites composed of detrital siliciclastic particles lining the walls of a sandstone cave. The significance of micro-organ- isms to trapping and binding detrital particles, and therefore to forming the microstromatolites in question, is discussed. A hypothesis is put forward on a similar constructive role of micro-organisms in formation of clastic-rich laminae within carbonate speleothems. GEOLOGICAL AND SPELEOLOGICAL SETTING Siliciclastic microstromatolites were found in W So- potni Wielkiej Cave (Jaskinia w Sopotni Wielkiej) and re- ported during a local symposium (Gradziński et al., 2001). The cave is located in the Beskid Żywiecki Mts. From the geotogkal point of view this region betongs to the Outer Carpathians, which are built predominantly of Cretaceous- Palaeogene flysch (Fig. 1). The cave is formed in the Eocene thick-bedded sandstones of the Magura beds belonging to the Raca Unit of the Magura Nappe (Golonka & Wójcik, 1978). The sandstone is fine grained. An X-ray diffractogram of the sandstone reveals the presence of quartz, alkali feldspars and two groups of phyllosilicates, with a major order layer separation of 14.5 Ä and 10.0 Ä (see Figs 4A, 5A). The 14.5 Ä minerals are either chlorite or smectite, although fur- ther analyses are necessary for exact identification. The 10.0 Ä peak may indicate the presence of illite or micas. Observations in optical microscope revealed the presence of muscovite and biotite micas (see Fig. 5A); a confirmation of the illite presence requires < 2 pm fraction separation. The cave main entrance is located at the altitude of 840 m (Klassek, 1997). The cave is composed of a series of passages up to 2 m wide and 5 m high, with total 101 m in length (Fig. 2). The passages display rectangular or triangu- lar cross-sections. Their floors are litt ered with sandstone debris. The cave originated due to widening of fractures un- der the act ion of downslope creep of sandstone blocks to- wards the neighbouring valley. Hence, it represents a crev- ice type cave (sensu Palmer, 2007, p. 7). The cave is pro- tected as a ‘nature monument’. The microstromatotites cover the walls of the Komora Trójkątna (Triangular Chamber) located ca. 12 m from the cave entrance. They were noted by Mikuszewski (1973) and Klassek (1997) but recognized as carbonate flowstones. The chamber is completely dark; its walls are covered by drops of condensation water. The internal temperature is 5.5 °C (Klassek, 1997). The thickness of the roof rock over the Tri- angul ar Chamber is es timated at 6-8 m. The area over the cave is forested. MA TE RIAL AND METH ODS Morphology and internal structures Microstromatolites were documented in the Triangular Chamber (Fig. 2). Sampling was limńed to the minimum due to the cave protection. Samples were coltected from SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE 305 both sides of the chamber, preferably from hidden places. A thin sect ion was made from one sample sunken in a box filled with ARALDITE. Internal structures of the microstro- matotites were observed under petrographic and scanning electron microscope (SEM) JEOL 5410, coupled with a mi- croprobe (EDS) Voyager 3100 (Noran product). The sam- ples were mounted on SEM holders with silver glue and coated with C or Au. Samples were lyophitized prior to coating to prevent collapse of the organic structure. Mineral composition Samples were analysed by the powder X-ray diffracto- metry (XRD) using a vertical XPert APD Philips goniome- ter (PW 1830). Infrared absorption spectra (IR-FT) were obtained at ambient temperature and with 2.0 cm-1 resolu- tion using a BIO-RAD Fourier Transform Spectrometer (FTS 135). Microbiology Samples were aseptically collected to sterile glass flasks, transported to laborat ory and suspended in physio- logical salt sotution, shaken and inocutated in liquid and agar media. They were incubated at 20 °C and 35 °C from 1 to 21 days. The growth of micro-organisms was systemati- cally monitored. The following microbiological media were used for isotation: Beaf Extract - Nutrient Broth - Merck, Trypticase Soy Broth (Soybean-Casein Digest Medium) - BioMerieux, Nutrient Agar - Merck, TSA (Trypcase Soy Agar) - BioMerieux, Soil Extract Agar (Attas & Parks, 1997), Iron Bacteria Isolation Medium (Atlas & Parks, 1997) and Actinomycetes Isotation Agar (Att as & Parks, 1997). After incubation the clean cultures of bacteria were isol ated on agar media (Pepper & Gerba, 2004). Morphol- ogy, Gram stain and biochemical proprieties of bacteria were analyzed to identify micro-organisms. Species identi- fication was based on Bergey’s Manual of Determinative Bacteriology and Bergey’s Manual of Systematic Bacteriol- ogy (Holt, 1989, 1994). Since there are no standard bio t chemical tests for the maj ority of isolated genera, the bio- chemical tests were individually selected according to diag- nostic manuals. RE SULTS Relief The microstromatolites display a soft, pasty consistency and contain a substantial amount of water. They cover the vertical and overhanging walls of the Triangular Chamber (Fig. 3A, B, D). The dip of rocky walls covered with the mi- cro stromatotites ranges from 90° (vertical) to 110° (over- hanging). The microstromatolites occur also on vertically oriented convex bends of a cave wall, which overhangs with an angle between 93° and 120° (Fig. 3A, B). Recently, they coat around 0.5 m2 of the southern wall of the chamber (A in Fig. 2) and around 1 m2 of its opposite wall (B in Fig. 2). However, it is very plausible that they formerly occupied a much bigger area and have been destroyed by visitors. On the vertical walls, they form more or less horizontal ripples with relief (distance between the wall and the crest), up to 306 M. GRADZIŃSKI ET AL. Fig. 3. Siliciclastic microstromatolites and their morphological counterparts; if no otherwise stated photographs are taken in W Sopotni Wielkiej Cave: A - Convex corner of cave wall covered with microstromatolite, microterrces are developed along the corner (M), flat wall to the left is covered with ripples (R); B - Microteracettes with microrim fringing the micropool; C - Miroteracettes composed of traver- tine, Sivä brada, (Spis, Slovakia); D - ripples covering cave walls, bifurcation of ripples is visible; E - Microterracettes and ripples com- posed of moonmilk deposits, Szczelina Chochołowska cave (Western Tatra Mts., Poland) 2 mm (Fig. 3D). The length of particutar ripples exceeds 35 cm. The neighbouring ripples biturcate. With the wall dip changmg towards more overhangmg, the ripples be t come more sinuous and form tongue-shaped stepped micro- terracettes (Fig. 3A, B, D). The microterracettes overhang and their upper parts dip outward at a maximum angle of 45°. In some cases, a microrim is developed along their crest with a micropool formed behind it. Microterracettes are part icul arly well developed along a vert ically oriented convex bends of the cave wall (Fig. 3B). The distance be- tween the microterracette crest and rock wall reaches 1.5 cm. The vertical distance between neighbouring ripples ranges between 2 and 5 mm, whereas between microteracet- tes it is slightly higher and maximally reaches 8 mm. Microbiology Microbiological analysis reveals the occurrence of sev- eral taxa which are listed in Table 1. It seems reasonable to accept that Arthrobacter, Bacilli and Actinomycetes belong to indigenous microflora of the anatysed micro stromato- lites. A scarce presence of Micrococcus and Staphyloccus in the studied samples implies that they should be considered allochthonous. A preterred environment of their growth supports the above view (Table 1). Mineral composition Quartz is the maj or constituent of the micro stromato- lites; the alkali feldspars and clay minerals have a smaller SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE 307 Table 1 List of determined micro-organisms with their major characteristics (after Holt, 1989, 1994) Bacteria Occurrence Morphology Properties Gram positive irregular, nonsporing Aerobic, chemoorganotrophic; grows on simple media, widely Arthrobacter sp. sample 1, sample 2 rods, in old cultures cocci distributed, principally in soils; psychrothropic arthrobacters have been reported to predominate in subterranean cave silts Amycolata autotrophica sample 1, sample 2 Gram positive, Actinomycetes, branched Aerobic, chemoorganotrophic - facultatively autotrophic; vegetative hyphae isolated from diverse habitats Bacillus alcalophilus sample 1 Endospore forming, gram positive rods Aerobic, chemoorganotrophic; alcali tolerant, isolated from various material in media at pH 10 Bacillus megaterium sample 1, sample 2 Endospore forming, gram positive rods, Aerobic, chemoorganotrophic; found in wide range of cell diameter over 1 pm habitats; grows in low temperatures Bacillus mycoides sample 1 Endospore forming, gram positive rods, Aerobic but can grow anaerobically, chemoorganotrophic; cell diameter over 1 pm, forms filaments forms rhizoid colonies; widely distributed Micrococcus varians sample 2 Gram positive cocci, nonosporing Aerobic; micrococci are common on mammalian skin, soil, air Facultatively anaerobic; staphylococci are mainly associated Staphylococcus xylosus sample 1 Gram positive cocci, nonsporing with skin, but often isolated from food products, dust and water Gram positive, Actinomycetes, Aerobic; chemoorganotrophic; widely distributed and Streptomyces sp. sample 2 extensively branched vegetative hyphae abundant in soil contribution. The clay mineral content, although insignifi- cant, is slightly higher in the microstromatolites than in the host rock. Sheet sihcates show diffraction peaks at 14.5 Ä and 10.0 Ä as well as a minor peak centered at 12.0 Ä (Fig. 4A). The 14.5 Ä and 10.0 Ä peaks correspond probably to minerals present in the host rocks (smectite or chlorite and micas), whereas the 12.0 Ä peak re flects most likely the presence of mixed-layer clay minerals. This issue requires further investigation. Additional information as to the composition of the microstromatolite sample is brought by IR absorption spec- trum (Fig. 4B). A weak absorption band at 1405 cm-1 may point to the presence of only very small amounts of carbon- ates. The weak intensity of this absorption bend and the ab- sence of appropriate diffraction peaks, that is 3.03 Ä (Fig. 4A) indicate almost negligible amount of carbonates. The microstromatolites have similar composition to the sandsto- ne which hosts the cave (Fig. 5A). The XRD patterns are given for compari son of the mineralogy of both rocks (Fig. 4A). Internal structures The surface of the microstromatolites is uneven. It dis- plays elevated clumps surrounded by depressions. The clumps show lower porosity than the depressions. Also the microrim crest is characterized by a lower porosity than the micropool fringed by it (Fig. 5B). The microstromatolite is finely laminated, with the lamina thickness ca. 50-200 pm visible due to differentia- tion in mineral composition, grain size, and probably chan- ges in microporosity (Fig. 5C). The laminae are convex out- ward, irregutar. Some of them have confused boundaries. They are composed of particles of clay and fine silt fraction. Sporadic quartz and muscovite grains reach up to 150 pm across (Fig. 5C). Outsized quartz and muscovite grains are concentrated in thicker laminae. Fig. 4. A - XRD pattern of host rock and microstromatolite; Q - quartz, Al - albite, Kf - K feldspar, C14 - clay minerals with main reflection 14.5 Ä; C12 - clay minerals with main reflection 12.5 Ä, C10 - 10.0 Ä micas or illite, B - IR absorption spectrum of the microstromatolite 308 M. GRADZIŃSKI ET AL. Fig. 5. A - Host sandstone: quartz and alkali feldspar grains accompanied by biotite and muscovite flakes present in the host sand- stones, Q - quartz, Kf - potassium feldspar, M - muscovite, B - biotite, Cl - clays, thin section, crossed polars; B - Irregular clumps form- ing a microstromatolite surface, more compact arch structure to the right side is crest of microrim (arrow) seen from above, granular, more porous material fill the micropool, SEM photo; C - Finely laminated crust of the microstromatolite is composed mainly of clay minerals (Cl), with larger muscovite flakes (M) and variable size quartz (Q) grains, laminae rich in coarser-grained quartz are indicated with big ar- rows, thin section, parallel polars; D - Clumps devoid of micro-organisms from deeper part of microstromatolite, SEM photo Observation under SEM reveals the internal structures of the microstromatolites. They are composed of mineral grains, micro-organi sms and their extracellular polymeric secretions (EPS). Clay minerals compose clumpy aggregates, up to 200 pm across (Figs 5B, D, 6A). The aggregates also comprise some admixtures of bigger mineral grains. Mica grains display angular shape and tabular habit (Fig. 6B). They contain Al, Table 2 Characteristics of distinguished morphotypes Morphotype Shape Dimensions Branching Morphology Cell arrangement Comments Figure 1 coccoid to short 0 - 0.8 pm no smooth chains or three-dimensional 6C, 7A-D rods length < 1.5 pm colonies covered with EPS 2 rods 0 - 0.8 pm no smooth straight or zig-zag chains 7E length - 2-3 pm 3 spindle-shaped 0 - 1 pm no granulated chains 7F rods length - 4-5 pm 4 coccoid 0 - 0.5-0.8 pm no smooth single or in linked pairs co-occur with no 6 8D 5 irregular ovoid 0 - 6 pm no irregular but single or in linked pairs co-occur with no 6 8C smooth 0 - 1.1-1.3 pm co-occur with no 4, 5, 7, 6 filamentous length > 20 pm yes spinose filaments slightly curved forms dense mat of 8A intertwinned filaments 7 filamentous 0 < 0.5 pm no smooth some filaments are twisted co-occur with no 4, 6 8A-D length > 15 pm SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE 309 Fig. 6. A - Irregular clumps composed of clay minerals, filaments of EPS are visible (arrow), subsurafce part of stromatolite; B - Mus- covite flakes incorporated into micro stromatolite; C - quartz grain covered with partly coltapsed microbial filament, coccoid bacteria (morphotype 1) are visible to the left, D - mineral grain entombed by micro-organisms, biofilm surface seen from above; all photographs under SEM, in A-C EDS chemical composition is indicated Si and K, which, along with the results of the XRD analysis, suggests muscovite (Fig. 6B). Quartz grains are more roun- ded and devoid of cleavage planes with Si as their maj or EDS de tect able compo nent. Min eral grains co-oc cur with, and are entombed by, micro-organisms which in many cases are tightly clung to them (Figs 6C, D, 7). Seven morphotypes of micro-organisms have been dis- tinguished. Their characteristics are listed in Table 2. All morphotypes are built exclusively of organic matt er. Nei- ther EDS analyses nor observat ion under SEM reveal any traces of cell mineralization. Micro-organisms show three- dimensional morphologies, only some of them are collap- sed. The micro-organi sms and their EPS occur on the sur- face of microstromatolites and in their shall ow subsurface where they form an active biofilm. Deeper on, their amount radically decreases (Fig. 5D). Affinity of the particular morphotypes is hard to be de- termined even in the light of the results of microbiological analyses. The morphotypes 1 to 3 most probably represent bacteria (Fig. 7A-C). Chains of coccoid cells, such as those of the morphotype 1, are typical two-dimensional colonies of coccoid bacteria developed due to cell divisions (Fig. 7A-C). Similarly, chains of rods characteristic for the morphotypes 2 and 3 are the effect of cell division (Fig. 7D, E). The morphotype 2 may be assigned to Arthrobacter ge- nus, taking into account the list of determined micro-organ- isms and their shape and size (Table 1). Moreover, it forms a zig-zag chain of cells typkal of this genus (cf. Carlile, 1979). The filamentous morphotypes 6 and 7 resemble Actinomycetes. The morphotype 6 is part icul arly akin to ‘hyphae morphotype 5’ described and iltustrated by Jones (2009b, fig. 7A-C) from the Grand Cayman speleothems. This morphotype forms ext ens ive three-dimens ional, po- rous network (Fig. 8A-D). The coccoid morphotype 4 and ovoid morphotype 5 may represent both bacteria and spores of Actinomycetes (Fig. 8C, D). However, their close spatial rel at ionship with the morphotype 6 strongly suggests the latter possibility. Although the EPS form irregularly twisted filaments or a dense layer which covers micro-organisms, none of the morphotypes is associated with a copious amount of EPS. DISCUSSION Trapping and bindmg of detrital partitles cause the growth of the studied microstromatolites. Two mechanisms are evoked to entrap detrital grains into a stromatolite - ad- hesion by sticky EPS and baffling by complicated three-di- mensional microbial community (Riding, 1991). In the dis- cussed case the latter mechanism seems to be decisive, since the SEM exammation of the studted samples has not re t M. GRADZIŃSKI ET AL. Fig. 7. A-D - Morphotype 1 representing bacteria, elongated chains developed due to cell divisions; E - Chain of rod-shaped cells (morphotype 2) probably representing Arthrobacter sp., F - Morphotype 3 formed elongated chains of spindle-shaped cells with granu- lated wall; all photographs under SEM vealed copious amounts of EPS. Highly branched cells of Actinomycetes Amycolata and Streptomyces species act as a dense three-dimensional network capable of baffling the de- trital particles. Nevertheless, Arthrobacter and Bacillus cells can excrete some slimy substances, hence the former mechanism may also work, but only to a limited extent. Newly precipitated mineral phases seem not to contrib- ute to the microstromatolite growth. Bearing in mind their growth mechanism, the microstromatolites in question rep- resent agglutinated stromatolites sensu Ridtiig (1991). In this aspect they bear a strong resemblance to some modern stromatolites in marine (Schwarz et al., 1975) and lacustrine settings (Squyres et al., 1991) as well as to several fossil ex- amples (Martin et al., 1993; Braga & Martin, 2000 and ref- erences therein). Martin et al. (1993) coined the term silici- clastic stromatolite which adequately reflects the composi- tion and mode of growth of the studied examples. The size of entrapped siliciclastic grains difterentiates cave micro- stromatolites from the hitherto described marine and lacus- trine ones. Non-spelean forms are composed mainly of sand grains with some admixtures of coarser material (Martin et al., 1993); however, some silt-rich siliciclastic stromatolites are also known (Bertrand-Sarfati, 1994). Jones and Kahle (1985) introduced the term microstromatolites in order to des cribe microbolites, formed of fine carbonate part icles and displaying fine lamination, recognized in cave sedi- ments from the Cayman Isl ands. The term microstroma- tolite seems to be appropriate to the studted depostis. The mode of growth distinguishes the described forms from any other hitherto known microbial cave depostis betiig cont SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE 311 Fig. 8. A - Three-dimensional porous network composed mainly of morphotype 6 (Actinomycetes); B - Detailed view of the network, spinose-walled morphotype 6 dominate, smooth-walled morphotype 7 occur subordinately (arrows), C - ovoid morphotype 5 (arrows) as- sociated with spinose-walled filamentous morphotype 6; D - minute coccoid bodies of morphotype 4 (spores of Actinomycetes; small ar- rows) associated with filamentous morphotypes 6 and 7 (big arrow), in the centre platy mineral grain entrapped within organic filaments (EDS chemical composition is indicated); all photographs under SEM structed predominantly by minerals precipitated from solu- tion. Although the discussed depostis mimic the retief of some speleothems (see below and Fig. 3), they cannot be in- cluded into this genetically de tined group, because they contain extremely small amount of, if any, secondary min- eral phases, that is minerals preciptiated within a cave (cf. Hill & Forti, 1997, p. 45). The microstromatolites bear mor- phological resemblance to speleothems described from other sandstone caves in the Polish Outer Carpathians (Ur- ban et al., 2007). Many of the latter forms are composed of detrital grains cemented with opal, which is not a case in the discussed microstromatolites. The growth of the discussed microstromatolites is influ- enced by several factors. The influx of detrital particles is possible due to relatively close distance to the Earth surface and the weathering zone (see Urban et al., 2007). It is addi- tionally facilitated by the presence of gravitationally wid- ened fissures which guide the water transporting mineral particles down to the cave. The fine-grained nature of microstromatolites more probably depends on the preferen- tial removal of such grains from soils. The particles must be transported only during the wet periods, that is during thaw or after heavy rains, since during the vistis to the cave its walls were covered only by drops of condensat ion wat er and were devoid of seeping wat er. The episodic supply of material results in a visible lamination of microstromatolite (Fig. 5C). The microbial biofilm covering the vertical or even overhanging cave walls trap and bind the transported detrital particles (Figs 6C, D, 8D). The capacity of bacteria to stabitize sand and finer grains has been experimentally confirmed (Meadows et al., 1994; Westall & Rince, 1994; Dade et al., 1996). The particles are stabitized on the sur- face of microstromatolite and then covered with a newly de- veloped biofilm (Fig. 6D). Simultaneously, an older part of the biofilm disintegrates due to senescence, which is proba- bly facilitated by covering mineral particles. The laminae which comprise outsized quartz grains mark the especially wet episodes when relatively coarse-grained material could have been remobilized from soils and washed down into the cave (Fig. 5C). The microterracettes form in zones of more intense water flow (Fig. 3A-C). It is confirmed by their pre- ferential formation along vertically oriented convex bends of the overhanging cave wall, thus the part where water flow is concentrated due to the surface tension. The downward in- clined shape of particular microterracettes most probably re- sults from plastic deformation and creeping of soft microstro- matolite under the action of gravity (Fig. 3C). The intlux of seepage water is also important for the micro-organism assemblage forming the microstromatoli- tes, which most probably depends on the input of organic matter from the surface (cf. Groth & Saiz-Jimenez, 1999). On the one hand, the limited energy of seeping water sorts 312 M. GRADZIŃSKI ET AL. mineral grains and controls the fine-grained composition of microstromatolites. On the other hand, it allows delicate mi- crobial biofilm to exist on cave walls; the higher energy of flow would cause destruction and scrubbing off of the mi- crobial biofilm. The lack of preciptiation of minerals is most probably connected with the chemistry of the seeping water. Although direct data are lacking, we can hypothesize that the water in W Sopotni Wielkiej Cave is simüar to the water in other non-karst caves in the flysch rocks of the Outer Carpathians. Zawierucha et al. (2005) reported that the vadose water in those caves is only slightly more mineralized than the rain water. The mean concentration of the Ca ion is only 9.5 mg/l, that is definitely lower than in karst caves. For exam- ple, the Ca content in water from setected Slovak karst caves ranges between 44.1 and 132.1 mg/l (Motyka et al., 2005) whereas in the vadose zone water of karst caves on the Kraków-Wieluń Upland it is from 52 to 137 mg/l (Róż- kowski, 2006). Moreover, one can suppose that the water which quickly percolates down after heavy rains, as it is in the discussed case, has lower concentration of ions due to its lower restdence time (see discussion in Musgrove & Ban- ner, 2004). Thus, the water in the studied cave most proba- bly does not achieve the appropriate saturation to precipitate carbonate minerals. The micro-organisms detected within the microstroma- tolites have been reported from other caves and have been supposed to influence - actively or passively - precipitation of minerals. Living bacteria belonging to Arthrobater were detected within growing moonmilk deposits (Gradziński et al., 1997b) and cave pearls (Gradziński, 2000). Phosphati- zed microbial cells assigned also to this genus are reported from the speleothems of Grand Cayman Island (Jones, 2009b). Mineralized, commonly calcified, Actinomycetes, including various species of Streptomyces, are known from cave deposits of Spain (Canaveras et al., 2001), Italy (Groth et al., 2001), USA (Melim et al., 2008) and the Cayman Is- lands (Jones, 2009a, b). This leads to the suggestion that the mineralization of micro-organisms in the cavern environ- ment is strongly dependent upon chemtstry of the feedmg water. The described origin of the studted siliciclastic micro- stromatolites may shed some light on the formation of other cave deposits. The unconsolidated accumulations of fine- grained clastics on cave walls, indudmg vermicutations common in many caves, can be also formed by the trapping and binding of detrital particles by micro-organisms. Sev- eral forms of speleothems recognized in some caves of the Pohsh Outer Carpathians may also share their origin with the discussed microstromatolites. They are also composed of siliciclastic material, display lamination and contain some laminae rich in coarser quartz sands (Urban et al., 2007). The detcribed stromatolites have a low potential for preservation. The only one possibility for their preservation is a quick cementation with calcium carbonates or opal act- ing as cement or covering a stromatolite as a younger flow- stone (cf. Urban et al., 2007). Thus, clastic-rich layers oc- curring within flowstones and other speleothems (see Dzia- dzio et al., 1993; Gradziński et al., 1997a; Turgeon & Lund- berg, 2001) may be fossil counterparts of micro stromato- lites formed during the break of crystallization and intense clastic supply. Since the micro-organisms entrapped detrital grains on the vertical and overhanging walls in the studied case, it seems plausible that they also play a role in stabiliza- tion of such grains depostied on a flowstone surface and later cemented with calcite or aragonite. Non-mineralized micro-orgamsm cells have been decomposed and have not been preserved. Hence, the layers in question lack any traces of micro-organisms. The closest modern counterparts of the described forms, both in morphology and origin, are sand ripples de- velopi ng on steep sandstone crags described by Pent ecost (1999) from Kent. They are composed of sand grains stabi- lized by algae and mosses and display relief strikingly simi- lar to that of the described microstromatolites. The most im- portant difference between the relief of sand ripples and the cave microstromatolites is a slightly greater lateral distance between neighbouring ripples (5.6-8.0 mm for the sand rip- ples). Other surface counterparts are litter dams occurring on slopes. They owe their origin due to accretion of small organic and mineral particles and their stabilization by mosses (Eddy et al., 1999). The spatial arrangements and relief of the litter dams are different, mainly because of their formation on the inclined, not vertical slopes. Interestingly, a relief analogous to the described micro- stromatolites characterizes surfaces of several actively growing continental deposits. The surfaces of some cave flowstones and dripstones are crenulated (Hill & Forti, 1997, pp. 71, 105) or rippled (Ford, 1988). Simitar ripples are formed on the growi ng icicles (e.g., Ogawa & Furu- kawa, 2002). The cave draperies also display serrated edges composed of the stepped overhanging terracettes compara- ble to the terracettes formed by the described microstroma- tolites. Identical terracettes are composed of moonmilk de- posi ts (Gradziński & Radomski, 1957). On the steeply in- clined or overhanging walls their upper surfaces are down- ward inclined simi l arly to the terracettes of the described microstromatolites (Fig. 3E). Such orientation probably re- sulted from some instabili ty of moonmilk havi ng a pasty consistency and from its tendency to creeping down, which is in common with the discussed microstromatolites. Also travertines form simi l ar regul arly spaced terracettes. The lateral distance between the neighbouring travertine terrace- ttes becomes shorter on steeper slopes (Hammer et al., 2010). Hence, the travertine terrracettes developed on vertical and steep slopes may serve as a counterpart of the described rip- ples and terracettes formed by the microstromatolites. Nonetheless, in contrast with the microstromatolites and moonmilk examples, the rim of travertine terracettes is al- most always perfectly horizontal (Fig. 3C; Hammer et al., 2010). It most probably results from the robust consistency of travertine deposits. Surprisingly, although all the above mentioned calcareous deposits originated mainly by the pre- cipitation of crystals, not by the accumulation of detrital particles, they share their shape and geometric pattern with the discussed microstromatolites whose growth is governed by different factors. It adds a new dimension to the discus- sion on the factors in tluencmg the shape and spatial ar t rangements of terracettes in recently growi ng travertines and speleothems. SILICICLASTIC MICROSTROMATOLITES IN A SANDSTONE CAVE 313 CONCLUSIONS 1. The walls of W Sopotni Wielkiej Cave are covered with siliciclastic microstromatolites constructed by bacteria and Actinomycetes, which trap and bind mineral part icles washed into the cave from overiymg soil during wet ept sodes. 2. Newly precipitated mineral phases seem not to con- tribute to the microstromatolite growth. 3. The microstromatolites form ripples and microterra- cettes on vertical and overhanging cave walls. Their shape depends upon the relief of a cave wall. The microterracettes are located where water flow is more intense, mainly along vertically oriented convex bends of a cave wall. 4. It seems possible that trapping and bindi ng mecha- nism influences the origin of clastic-rich layers within cave flowstones. Clastic grains were stabilized by micro-organ- isms and later were cemented with calcite or aragonite. The micro-organisms were subsequently decomposed and have been not preserved till now. 5. Sand ripples developing on almost vertical sand- stone crags are a close genetic and morphological analogue of the cave siliciclastic microstromatolites. 6. Ripples (crenulations) on the speleothem and icicle surfaces, as well as stepped microterracettes in speleothems and travertines share the same retief with the de tcribed microstromatolites in spite of their different origin. Acknowledgements MG was supported at the beginning of the study by the Foun- dation for Polish Science in the frame of Grant for Researchers to Professor Józef Kaźmierczak. Renata Jach and Ryszard Gradziń- ski assisted in the field, Jaga Faber, Zuza Banach and Aleksandra Woźnicka operated the SEM, Renata Jach also prepared the fig- ures; the authors gratefully acknowledge this help. The manuscript benefited from the reviews of Tadeusz Peryt and Jan Urban. REFERENCES Atlas, R. M. & Parks, L. C., 1997. Handbook of Microbiological Media. CRC Press, Boca Raton, 1706 pp. Aubrecht, R., Brewer-Carias, Ch., Smida, B., Audy, M. & Kova- cik, L., 2008. Anatomy of biologically mediated opal speleo- thems in the World’s largest sandstone cave: Cueva Charles Brewer, Chimanta Plateau, Venezuela. Sedimentary Geology, 203: 181-1 95. Baker, G., 1942. Sand stalagmites. Journal of Geology, 50: 662- 667. Baskar, S., Baskar, R. & Kaushik, A., 2006. Evidences for micro- bial involvement in the genetis of speleothem carbonates, Borra Caves, Visakhapatnam, India. Current Science, 92: 350-355. Bertrand-Sarfati, J., 1994. Siliciclastic-carbonate stromatolite domes in the early Carboniferous of the Ajjers Basin (Eastern Sahara, Algeria). In: Bertrand-Sarfati, J. & Monty, C. (eds), Phanerozoic Stromatolites II, Kluwer, Dordrecht, pp. 395- 419. Bini, A., Cavalli Gori, M. & Gori, S., 1978. A critical review of hypotheses on the origin of vermiculations. International Journal of Speleology, 10: 11-33. Blyth, A. J. & Frisia, S., 2008. Mo lecul ar evi dence for bact erial mediation of calcite formation in cold high-altitude caves. Geomicrobiology Journal, 25: 101-111. Braga, J. C. & Martin, J. M., 2000. Subaqueous siliciclastic stro- matolites: A case history from Late Miocene beach deposits in the Sorbas Basin of SE Spain. In: Riding, R. E. & Awramik, S. M. (eds), Microbial Sediments. Springer, Berlin, pp. 226-232. Canaveras, J. C., Sanchez-Moral, S., Soler, V. & Saiz-Jimenez, C., 2001. Microorganisms and microbially induced fabrics in cave walls. Geomicrobiology Journal, 18: 223-240. Carlile, M. J., 1979. Bacterial, fungal and slime mould colonies. In: Larwood, G. & Rosen, B. R. (eds), Biology and Systemat- ics of Colonial Organisms. Academic Press, London, pp. 3-27. Cox, G., James, J. M., Leggett, K. E. A. & Osborne, R. A. L., 1989a. Cyanobacterially deposited speleothems: subaerial stromatolites. Geomicrobiology Journal, 7: 245-252. Cox, G., James, J. M., Leggett, K. E. A. & Osborne, R. A. L., 1989b. Stromatolitic crayfish-like statagmites. Proceedings of the Bristol Speleological Society, 18: 339-358. Cunningham, K. I., Northup, D. E., Pollastro, R. M., Wright, W. G. & LaRock, E. J., 1995. Bacteria, fungi and biokarst in Lechugilla Cave, Carlsbad Caverns National Park, New Mex- ico. Environmental Geology, 25: 2-8. Curry, M. D., Boston, P. J., Spilde, M. N., Baichtal, J. F. & Camp- bell, A. R., 2009. Cottonballs, a unique subaqeous moonmilk and abundant subaerial moonmilk in Cataract Cave, Tongass National Forest, Alaska. International Journal of Speleology, 38: 111-128. Dade, W. B., Self, R. L., Pellerin, N. B., Moffet, A., Jumars, P. A, Nowell, A. R. M., 1996. The effects of bacteria on the flow behavior of clay-seawater suspensions. Journal of Sedimen- tary Research, 66: 39-42. Dziadzio, P., Różniak, R. & Szulc, J., 1993. Origin of the Pleisto- cene calcite flowstones of two caves (Jaskinia Psia and Jaski- nia Naciekowa) in the West Tatra Mts. (In Pofish, Engtish summary). Przegląd Geologiczny, 41: 767-775. Eddy, J., Humphreys, G. S, Hart, D. M., Mitchell, P. B. & Fan- ning, P. C., 1999. Vegetation arcs and litter dams: similarities and differences. Catena, 37: 57-73. Ford, T. D., 1988. Rippled speleothems. Cave Science, 15: 86-87. Golonka, J. & Wójcik, A., 1978. Objaśnienia do szczegółowej mapy geologicznej Polski. Arkusz Jeleśnia 1:50 000. (In Polish). Wydawnictwa Geologiczne, Warszawa, 40 pp. Gradziński, M., 2000. Pizoidy jaskiniowe: uwarunkowania gene- tyczne i środowiskowe ich powstania. (In Pohsh). Unpub t lished PhD thesis, Institute of Geological Sciences, Jagiello- nian University, Kraków, 90 ms. pp. Gradziński, M., Banaś, M. & Uchman, A., 1995. Biogenic origin of manganese flowstones from Jaskinia Czarna Cave, Tatra Mts., Western Carpathians. Annales Societatis Geologorum Poloniae, 65: 19-27. Gradziński, M., Chmiel, M. J. & Michalska, B., 2001. Powstanie i rozwój detrytycznych polew naciekowych w Jaskini w So- potni Wielkiej (Beskid Żywiecki). (In Polish). In: Gradziński, M., Mleczek, T., Pukowski, J., Szelerewicz, M. & Urban, J. (eds), Materiały 35. Sympozjum Speleologicznego, Bartkowa, 26-28.10.2001. Sekcja Speleologiczna PTP, Kraków, p. 20. Gradziński, M. & Holubek, P., 2005. Cottonballs - peculiar speleothems from Zlomska Cave (Low Tatra Mountains, Slovakia) - pilot results. Slovensky Kras, 43: 187-192. Gradziński, M., Rospondek, M. & Szulc, J., 1997a. Paleoenvi- ronmental controls and microfacies variabifity of the flow- stone cover from Zvoniva Cave in the Slovakian Karst. Slo- vak Geological Magazine, 3: 299-313. Gradziński, M., Szulc, J. & Smyk, B., 1997b. Microbial agents of moonmilk calcification. In: Jeannin, P.-Y. (ed.), Proceedings 314 M. GRADZIŃSKI ET AL. of the 12th International Congress of Speleology, vol. 1. Inter- national Union of Speleology, Basel, pp. 275-278. Gradziński, R. & Radomski, A., 1957. Cavern depostis of “rock milk” in the Szczelina Chochołowska Cave. (In Polish, Eng- lish summary). Rocznik Polskiego Towarzystwa Geologicz- nego, 26: 63-90. Gradziński, R. & Radomski, A., 1960. Cementatory structures in the Miętusia Cave. (In Potish, Engtish summary). Rocznik Polskiego Towarzystwa Geologicznego, 30: 121-125. Groth, I. & Saiz-Jimenez, 1999. Actinomycetes in hypogean envi- ronments. Geomicrobiology Journal, 16: 1-8. Groth, I., Schumann, P., Laiz, L., Sanchez-Moral, S, Cańaveras, J. C. & Saiz-Jimenez, C., 2001. Geomicrobio logical study of the Grotta del Cervi, Porto Badisco, Italy. Geomicrobiology Journal, 18: 241-258. Hammer, 0., Dysthe, D. C. & Jamtveit, B., 2010. Travertine ter- racing: patterns and mechanisms. In: Pedley, H. M. & Rogerson, M. (eds), Tufas and Speleothems: Unravelling the Microbial and Physical Controls. Geological Society Special Publications, 336: 345-355. Hill, C. & Forti, P., 1997. Cave Minerals of the World. National Speleological Society, Huntsville: 1-463. Holt, J. G., ed., 1989. Bergey’s Manual of Systematic Bacteriol- ogy. Vol. 1-4. Williams & Wilkins, Baltimore, 2648 pp. Holt, J. G., ed., 1994. Bergey’s Manual of Determinative Bacteri- ology. Williams & Wilkins, Baltimore, 787 pp. James, J. M., Patsalides, E. & Cox, G., 1994. Amino acid composi- tion of stromatolitic statagmites. Geomicrobiology Journal, 12: 183-194. Jones, B., 1995. Processes associated with microbial biofilms in the twi light zone of caves: examples from the Cayman Is- lands. Journal of Sedimentary Research, A65: 552-560. Jones, B., 2001. Microbial activtiy in caves - A geotogkal per- spective. GeomicrobiologyJournal, 18: 1-13. Jones, B., 2009a. Cave pearls - the integrated product of abiotic and biotic processes. Journal of Sedimentary Research, 79: 689-710. Jones, B., 2009b. Phosphatic precipitates associated with actino- mycetes in speleothems from Grand Cayman, Briti sh West Indies. Sedimentary Geology, 219: 302-317. Jones, B., 2010. Microbes in caves: agents of calcite corrosion and precipitation. In: Pedley, H. M. & Rogerson, M. (eds), Tufas and Speleothems: Unravelling the Microbial and Physical Controls. Geological Society Special Publications, 336: 7- 30. Jones, B. & Kahle, C. F., 1985. Lichen and algae: agents of biodiagenesis in karst breccia from Grand Cayman Istand. Bulletin of Canadian Petroleum Geology, 33: 446-461. Jones, B. & Motyka, A., 1987. Biogenic structures and micrite in statactites from Grand Cayman Istand, British West Indies. Canadian Journal of Earth Sciences, 24: 1402-1411. Klassek, G., 1997. Jaskinia w Sopotni Wielkiej. In: Pulina, M. (ed.), Jaskinie Polskich Karpat fliszowych. (In Polish). Polskie Towarzystwo Przyjaciół Nauk o Ziemi, Warszawa: 215-217. Manolache, E. & Onac, B. P., 2001. Geomicrobiology of black sediments in Väntului Cave (Romania): preliminary results. Cave and Karst Science, 27: 109-112. Martin, J. M., Braga, J. C. & Riding, R., 1993. Siliciclastic stroma- tolites and thrombolites, Late Miocene, S.E. Spain. Journal of Sedimentary Petrology, 63: 131-139. Meadows, A., Meadows, P. A., Muir Wood, D. & Murray, J. M. H., 1994. Microbiological effect on slope stability: an experi- mental analysis. Sedimentology, 41: 423-435. Melim, L. A., Shinglman, K. M., Boston, P. J., Northup, D. E., Spilde, M. N. & Queen, J. M., 2001. Evidence for microbial involvement in pool finger precipitation, Hidden Cave, New Mexico. GeomicrobiologyJournal, 18: 311-329. Melim, L. A., Northup, D. E., Spilde, M. N., Jones, B., Boston, P. J. & Bixby, R. J., 2008. Reticutated filaments in cave pool speleothems: microbe or mineral? Journal of Cave and Karst Studies, 70: 135-141. Mikuszewski, J., 1973. Nowa jaskinia w Beskidzie Żywieckim. (In Polish). Wierchy, 42: 262-265. Motyka, J., Gradziński, M., Bella, P. & Holubek, P., 2005. Chem- istry of waters from selected caves in Slovakia - a reconnais- sance study. Environmental Geology, 48: 682-692. Musgrove, M. & Banner, J. L., 2004. Controls on the spatial and temporal varaibility of the vadose dripwater geochemtstry: Edwards Aquifer, Central Texas. Geochimica et Cosmochi- mica Acta, 68: 1007-1020. Northup, D. E. & Lavoie, K. H., 2001. Geomicrobiology of caves: a review. Geomicrobiology Journal, 18: 199-222. Ogawa, N. & Furukawa, O., 2002. Surface instabili ty of icicles. Physical Review E, 66: 041202-1. Palmer, A. N., 2007. Cave Geology. Cave Books, Dayton, 454 pp. Pentecost, A., 1999. A note on sand ripples developmg in sand- stone rock seepages of the Weald, UK. Earth Surface Pro- cesses andLandforms, 24: 1257-1259. Pepper, I. L. & Gerba, Ch. P., 2004. Environmental Microbiology. A Lab o ra tory Man ual. Elsevier, Amsterdam, 209 pp. Richter, D. K., Immenhauser, A. & Neuser, R. D., 2008. Electron backscatter diffraction documents randomly oriented c-axes in moonmilk calcite fibres: evidence for biologically induced precipitation - short contribution. Sedimentology, 55: 487- 497. Riding, R., 1991. Classification of microbial carbonates. In: Rid- ing, R. (ed.), Calcareous Algae and Stromatolites. Springer, Berlin: 21-51. Różkowski, J., 2006. Groundwaters of carbonate formations in the southern part of Jura Krakowsko-Częstochowska and problem with their protection. (In Polish, English summary). Wydawnictwa Uniwersytetu Śląskiego, Katowice, 264 pp. Schwarz, H.-E., Einsele, G. & Herm, D., 1975. Quartz-sandy, grazing-contoured stromatolites from coastal embayments of Mauritania, West Africa. Sedimentology, 22: 539-561. Squyres, S. W., Andersen, D. W., Nedell, S. S. & Wharton, R. A. Jr., 1991. Lake Hoare. Antarctica: Sedtinentation through a thick perennial ice cover. Sedimentology, 38: 363-379. Taborosi, D., 2006. Biologically influenced carbonate speleo- thems. In: Harmon, R. S. & Wicks, C. M. (eds), Perspectives on Karst Geomorphology, Hydrology and Geochemistry. Geological Society of America, Special Paper, 404: 307-317. Turgeon, S. & Lundberg, J., 2001. Chronology of discontinuities and petrology of speleothems as paleoclimatic indicators of the Klamath Mountains, Southwestern Oregon, USA. Car- bonates and Evaporites, 16: 153-167. Urban, J., Margielewski, W., Schejbal-Chwastek, M. & Szura, Cz., 2007. Speleothems in some caves of the Beskidy Mts., Poland. Nature Conservation, 63: 109-117. Westall, F. & Rince, Y., 1994. Biofilms, microbial mats and mi- crobe-particle interactions: electron microscope observations from diatomaceous sediments. Sedimentology, 41: 147-162. Zawierucha, L., Czop, M. & Motyka, J., 2005. Modification of precipitation chemistry in unsaturated and saturated zones of the Beskid Mały and Śląski flysch rocks. (In Polish, English summary). Współczesne Problemy Hydrogeologii, 12: 737- 742.