Environmental Geology (1995) 25:2-8

9 Springer-Verlag 1995

K. I. Cunningham 9 D. E. Northup 9 R. M. Pollastro 9 W. G. Wright 9 E. J. LaRock

Bacteria, fungi and biokarst in Lechuguilla Cave, Carlsbad Caverns National Park, New Mexico

Received: 18 March 1994 / Accepted: 21 March 1994

Abstract Lechuguilla Cave is a deep, extensive, gypsumand sulfur-bearing hypogenic cave in Carlsbad Caverns National Park, New Mexico, most of which ( > 90%) lies more than 300 m beneath the entrance. Located in the arid Guadalupe Mountains, Lechuguilla's remarkable state of preservation is partially due to the locally continuous Yates Formation siltstone that has effectively diverted most vadose water away from the cave. Allocthonous organic input to the cave is therefore very limited, but bacterial and fungal colonization is relatively extensive: (1) Aspergillus sp. fungi and unidentified bacteria are associated with iron-, manganese-, and sulfur-rich encrustations on calcitic folia near the suspected water table 466 m below the entrance; (2) 92 species of fungi in 19 genera have been identified throughout the cave in oligotrophic (nutrient-poor) "soils" and pools; (3) cave-air condensate contains unidentified microbes; (4) indigenous chemoheterotrophic Seliberius and Caulobacter bacteria are known from remote pool sites; and (5) at least four genera of heterotrophic bacteria with population densities near 5 x 105 colony-forming units (CFU) per gram are present in ceiling-bound deposits of supposedly abiogenic condensation-corrosion residues. Various lines of evidence suggest that autotrophic bacteria are present in the ceiling-bound residues and could act as primary producers in a unique subterranean microbial

food chain. The suspected autotrophic bacteria are probably chemolithoautotrophic (CLA), utilizing trace iron, manganese, or sulfur in the limestone and dolomitic bedrock to mechanically (and possibly biochemically) erode the substrate to produce residual floor deposits. Because other major sources of organic matter have not been detected, we suggest that these CLA bacteria are providing requisite organic matter to the known heterotrophic bacteria and fungi in the residues. The cavewide bacterial and fungal distribution, the large volumes of corrosion residues, and the presence of ancient bacterial filaments in unusual calcite speleothems (biothems) attest to the apparent longevity of microbial occupation in this cave. Key words Lechuguilla Cave- Biokarst. Chemolithotrophy 9Corrosion residues - Biothems

Introduction

Lechuguilla Cave is developed in the reef and back-reef facies of Guadalupian-aged carbonates on the northwestern rim of the petroliferous Delaware basin in southeastern New Mexico. The cave was first entered in 1986 and currently contains over 100 km of surveyed passage, the majority of which is developed at a depth of 300 m below the K. I. Cunningham ([g~) entrance. It is the deepest single-entrance cave in the UnitU.S. Geological Survey, Branch of Sedimentary Processes, MS 939, ed States with 475 m of known total vertical relief. Late Box 25046, DFC, Denver, Colombia, 80225-0046,USA Permian to early Tertiary phases of normal carbonic-acid D. E. Northup solution along structural and depositional trends in the University of New Mexico Centennial Scienceand Engineering carbonate rocks provided a framework for late-phase Library, Albuquerque, New Mexico, 87131-1466,USA speleogenesis (mid-Tertiary to early Pleistocene) apparR. M. Pollastro ently controlled by the episodic delivery of H2 S-bearing U.S. Geological Survey, Branch of Petroleum Geology, MS 940, fluids into the reef complex via suspected bathyphreatic Box 25406, DFC, Denver, Colombia, 80225-0046, USA flow paths (Bachman 1980; Davis 1980; DuChene and W. G. Wright U.S. Geological Survey, Water Resources Division, P.O. Box 2027, McLean 1989; Hill 1987; Hiss 1980). Solution of the Grand Junction, Colombia, 81502, USA carbonate rocks by hydrosulfuric acid explains the dramatic passage enlargement, extensive vertical solution, the E. J. LaRock widely distributed deposits of isotopically light, massive 4148 E. 19th St., Denver, Colombia, 80220, USA

3

and speleothemic gypsum and elemental sulfur, and lowpH-dependent minerals such as 10-A halloysite, dickite, alunite, and natroalunite (Hill 1987; Pahner 1991a). The cave also contains numerous unusual (rare) speleothems containing organic filaments (Davis and others, 1990) that may be biokarstic in origin. Biokarst refers to a littleunderstood array of biological and microbiological processes that result in "landforms produced largely by the direct biological erosion and/or deposition of calcium carbonate" (Viles 1984). These processes may act on surface or subterranean carbonate rocks to create depositional, erosional, and mixed depositional/erosional forms. Caves are currently known to contain few biokarst phenomena, which include predominantly microbe-assisted depositional forms such as calcitic moonmilk (Bernasconi 1960, 1980; Mason-Williams 1959), manganese- and iron-oxide deposits (Dyson and James 1981; Hill 1982; Laverty and Crabtree 1978; Moore 1981; Peck 1986; White and Ellisher 1958), and particular speleothems (Jones and Motyka 1987; Latham 1981; Went 1969; White 1981). Erosional forms in caves are rare; the only documented examples are algae-produced karren-like features (phytokarst) on speleothems near cave entrances (Bull and Laverty 1982; Jennings 1981; Waltham and Brook 1980). In most cases of reported biokarst (including the fixation of iron and manganese compounds), the microorganisms are heterotrophic (secondary producers) and are ultimately dependent upon organic matter delivered to the cave from the surface (Caumartin !963; Peck t986). The transport of organic matter into a cave is the single most important factor in the maintenance of indigenous microbial populations and the possible development of subterranean biokarst features. This factor is mostly absent from Lechuguilla Cave.

Jagnow 1989), a convincing explanation for the cave's remarkable state of preservation. The 27-m-deep pit entrance is located approximately 30 m above a dry stream bed and provides the only known, current access route into the cave for animal life, Rhadine beetles, Ceuthophitus crickets, diplurans, and Cicurina spp. spiders have been observed in the passages immediately below the entrance and have trogloxenic or troglophilic habits. Gut endosymbionts and passive or parasitic microbes may find transport into the front regions of the cave through these invertebrates. Modern use of the cave by bats has not been observed (ancient, minor guano piles exist at the base of the entrance pit), but rare bat skeletons have been found in deeper areas of the cave (Jablonsky 1991). Barometrically controlled reversals in wind direction occur at the entrance, and airborne surface microorganisms could be transported into the cave during inhalation phases. However, outside atmospheric conditions appear to have little effect on internal

Introduction of organic compounds into Lechuguilla Cave Surface-derived organic compounds and microorganisms are introduced into Lechuguilla by groundwater infiltration (vadose water), invertebrate and vertebrate animal life, and barometric pressure- and temperature-driven air exchange at the entrance. Because of its location in the arid northern Chihuahuan desert, groundwater infiltrates into the cave periodically and is organic poor (Lambert and Harvey 1987). Recent infiltration is generally confined to intermittent flow paths beneath surface valleys and can be located in Lechuguilla where confined, wet, calcitic formation areas usually feed small pools. The majority of the cave has been in the vadose zone for at least 800,000 yr, based on subaerial speleothem uranium series and electronspin-resonance dates from nearby Carlsbad Cavern (Hill 1987). Vadose solution features and allochthonous clays are rare. Massive secondary gypsum (up to 10 m thick), multiton deposits of elemental sulfur, and fragile gypsum speleothems are preserved throughout the vertical extent of the cave and suggest effective vadose water diversion by the siltstones of the overlying Yates Formation (Hill 1987;

Fig. 1 Photograph of calcitic folia located in the Far East area of Lechuguilla Cave, Carlsbad Cavern National Park, NM. The folia occur in a steeply dipping fissure passage that intersects the water table 454 m below the entrance. The graceful calcite deposits extend upward nearly 50 m before gradually disappearing. Bacteria and fungi inhabit some basal surfaces where encrustations have high concentrations of iron, manganese, and sulfur compounds. Photo courtesy of Larry David McLaughlin, 9 1990

cave air circulation patterns (microclimates), which are known to attenuate incoming air masses within 1.25 km of the entrance (Cunningham and LaRock 1991). Organic and hydrologic input to this deep, extensive, and pristine cave system is now very limited and should be capable of sustaining only limited spelean microbiological populations.

Biofilaments on calcitic folia Microorganisms in Lechuguilla were first observed on samples of nonsterile folia (Figs. 1-4) collected for mineralogic and petrographic analyses. Folia are relatively uncommon calcite speleothems formed by water-level fluctuations at the cave's two deepest points, approximately 466 m below the entrance (Fig. 1). The ventral surfaces of the folia are generally concave, with occasional windows where the limestone country rock hosts calcitic encrustations containing iron, manganese, and sulfur compounds [as determined from x-ray powder diffraction (XRD) and energy dispersive x-ray (EDX) analyses] of unknown oxidation state. Scanning electron microscopy (SEM) photomicrographs of the encrusting material show randomly spaced groupings of complex networks of 0.5- to 1.0-gin-thick filaments (Fig. 2), consistent with bacterial diameters (Buchanan 1974). EDX spectra from scans of very dense filament areas are flat and suggest carbon-based compounds because the spectrometer cannot detect elements with atomic numbers less than 11 (Na). Phosphorus is the only major element routinely detected in the filament networks in concentrations as high as 8~o oxide weight. Other areas of the folia samples have similar bacterial mats and 100- to 150-gm-

Fig. 3 Scanning electron microscope photomicrograph of Aspergillus sp. conidiophores on folia sample obtained from Lechuguilla Cave, Carlsbad Caverns National Park, NM. The 1.7- to 2.6-gm-diameter phialides are borne on 5- to 10-gin-long, uniseriate metulae that are distinctive for the AspergiIlus genus (Raper and Fennel 1965). Note smaller bacterial filaments on iron-, manganese-, and sulfur-enriched calcitic substrate

Fig. 4 Scanning electron microscope photomicrograph of suspected calcitized, organic (fungal?) filaments on a mineral-enerusted folia sample obtained from Lechuguilla Cave, Carlsbad Caverns National Park, NM. The 80-gm filament diameter suggests fungi similar to modern Aspergillus (Fig. 2) growing on the same sample 20 mm to the right of this scene. Note smaller fossilized fungi on etched and corroded calcitic substrate

Fig. 2 Scanning electron microscope photomicrograph of suspected chemolithoautotrophic bacteria on mineralized calcitic folia from Lechuguilla Cave, Carlsbad Cavern National Park, NM. 10-gm scale bar at lower left of scene. Bacterial attachment to the substrate can be observed at the left and bottom of the scene

long fungal conidiophores (Fig. 3) that are tentatively identified as Aspergillus (D. Natvig and C. Dahm, personal communication). The spatial distribution of the microbial colonies and attendant fungi is highly correlated with areas of higher concentrations of iron, manganese, and sulfur, Noncolonized areas are mainly calcite with minor magnesium, aluminum, and silica. The correlation between

colony location and higher mean concentrations of iron, manganese, and sulfur suggests that the 0.5-gm-diameter filaments may be bacteria capable of utilizing these compounds. Fungi are generally associated only with the bacterial networks and appear to be dependent on the bacteria. Calcite-encrusted segments of similar organic mats are adjacent to actively reproducing colonies (Fig. 4) and may indicate long-term colony occupation of the site (dependent on rates of calcite precipitation). Repeated sampling of the folia using aseptic techniques produced the same results: bacteria have successfully colonized inorganic substrates at the deepest in-cave point in the United States and appear to provide requisite organic matter to resident fungi.

Cave condensation-corrosion deposits as a biokarst phenomenon

The development of cave condensation-corrosion residues has been previously attributed to inorganic, corrosive gas-weathering of limestone bedrock and speleothems via carbon dioxide (and/or sulfur dioxide) charging of ambient water vapor (Hill 1987; Palmer and Palmer 1989). The deposits in Lechuguilla occur principally in the higher levels of the cave, although deposits have been identified in lower levels where rising, warm, moist air is routed. Because of Lechuguilla's sheer size, the deposits are orders of magnitude greater in extent and volume than any other occurrences known from the Guadalupe Mountains, New Mexico (Hill 1987), the Black Hills, South Dakota (Palmer and Palmer 1989), the Nullarbor Plain caves, Australia (Lowry and Jennings 1974), or the Kugitangtau caves of Microorganisms in pool waters and atmospheric southeast Turkmenistan (A. Klimchouk, personal commucondensate nication). In Lechuguilla, the residues are well developed, chestnut brown to steel gray-black in color, infrequently The extent, type, and distribution of the microbes was yellow and red, 0.25 1.5 cm thick, and occur in an arboinitially determined by sampling pool waters and atmos- rescent habit as continuous filamentous mats or tufts clingpheric condensate. Pool samples were treated in the field ing to etched and corroded wall surfaces or as minor floor with 0.1 M phosphate-buffered glutaraldehyde (2~o final deposits. Infrequent patches of fungal colonization 1-10 concentration) and prepared for SEM and epifluorescent mm in diameter have been observed on deposits throughmicroscopy. Six distinct types of bacteria were noted in out the cave. Ceiling-bound residues in deep cave areas both virgin (never touched by humans) and nonvirgin pools often overlie thick (0.1-1.5 m) floor deposits of porous, including Caulobacter sp. and Seliberia sp., which are class- white- to buff-colored residue composed of acid-insoluble ified as chemoheterotrophic bacteria (Buchanan 1974) components from the country rock. Samples of chestnutand are probably indigenous to the cave (K. LaVoie, per- brown and gray-black ceiling-bound residue were asepsonal communication). Microbial population densities are tically collected (from a site in the Far East area of LechIow, but similar, from pool to pool and suggest that most uguilla) and placed in 10-ml autoclaved Teflon or glass species are oligotrophic and that the pools are not primary vials. Sample splits were prepared for XRD and SEM/EDX sites in the cave's trophic system. Systematic resampling and for inoculating streak plates fixed with standard nutrispecifically for fungi in cave regoliths, pool waters, wall ent agar. XRD analyses of both wet and dried chestnutcrusts, calcitic formations, and corrosion residues resulted brown samples show phases ranging from amorphous (mainin the identification of 92 species in 19 genera throughout ly in wet samples) to well-ordered compounds in dry samthe known extent of the cave (Northup and others 1993). ples. The mineralogy varies among samples and is comSeveral of the fungi, such as Mucor and Rhizopus, are posed of both parent rock and corrosion products. Minerals common surficial molds and are possible human contami- from the parent rock include calcite, dolomite, ankerite, nants. Cave air-condensate water was sampled by sus- euhedral quartz, and minor illite. These same constituents pending sterile mylar sheets in the airflow path and col- are found in varying proportions in floor deposits from lecting the condensation dripwater and by pumping ambi- many areas of the cave. Common ceiling-bound minerals ent air through a coiled sterile plastic hose. Both con- are (in order of abundance) a hydrated manganese oxide densate samples contained varieties of coccoid, rod-shaped, preliminarily identified as todorokite [(Mn, Mg, Ba, Ca, K, and sheathed bacteria (D. Updegraff and N. Robbins, per- Na)aMnsOlz 93H20], well-ordered aluminum hydroxide sonal communication), many displaying opaque bodies of (mainly gibbsite), poorly ordered iron oxides and hydroxsuspected ferric hydroxides, sulfur, or polyphosphates. The ides (mainly hematite and goethite), and minor amounts microbe-containing condensate water implicates the cave's of 10-A halloysite [A12Si2O5(OH)4.2H 2O]. The todorokite large atmospheric circulation cells in the dispersal of mic- occurs in a tabular habit and is a common constituent of robes between different cave levels and areas. The humid the < 2-gm fraction of the samples. Combined heterotrophic air masses are augmented with trace amounts (< 1.0 ppm) bacterial and fungal population densities were counted of SOz, COS, and CS 2 (LaRock and Cunningham 1990) from standard nutrient agar streak plates and estimated at and may be an energy source for some of the cave's micro- 4.0 x 105-4.5 x 105 CFU/g for the chestnut-brown mateorganisms. These air masses, which contain organic mate- rial and 1.5 x 105-2.0 x 105 CFU/g for the gray-black rial, condense at cooler, higher cave levels where massive material (D. Updegraff, personal communication). Inocubedrock condensation-corrosion deposits have been lation of the same media with condensation-corrosion noted (Cunningham and LaRock 1991). samples from Wind Cave, South Dakota, and Spider Cave,

Fig. 5 Scanning electron microscope photomicrograph of suspected chemolithoautotrophic bacteria, LechuguillaCave, Carlsbad Caverns National Park, NM. Numerous bacterial filamentscan be seen within and on the todorokite substrate, particularly near the center of the scene where the bacteria give the mineralogy a feathered or ragged appearance

Fig. 6 Scanning electron microscope photomicrograph of suspected chemolithoautotrophic bacteria, LechuguillaCave, Carlsbad Caverns National Park, NM. Two distinct populations of bacterial can be discriminated based on sheath diameter. The smaller-diameter filaments are possibly engaged in the production of todorokite, a mineral associated with deep-sea vent encrustations (Jannasch and Wirsen 1981)

New Mexico, were negative. To date, nine genera of fungi

(Penicillium, 13 species; Aspergillus, three species; Cylindrocladium, Rhizopus, M ycelia s~erilia, Mucor, P aecileomyces, Fusarium, and Epicoccum) and six bacteria (Bacillus, Actinomycetes, Arthrobacter, Chryseomonas luteola, Rhodococcus, and Staphylococcus) have been identified from the samples. SEM examination of nine residue samples from the Far East revealed dense, three-dimensional filament networks composed of two populations based on filament diameter. Very fine filaments, 0.09-0.13 gm in diameter, are generally horizontally oriented and enmeshed (or penetrating) in the todorokite crystals (Fig. 5). These filaments are significantly smaller than the pseudomycelium (0.5-0.6 gm) of Parabacterium spelei (Caumartin 1963). Larger diameter (0.5-1.0 gm) filaments from the bulk of the network and appear to provide a vertical framework (Fig. 6). Corrosion residues from three sites in Lechuguilla were aseptically resampled and laboratory cultured with a nutrient mixture optimized for chemolithoautotrophic (CLA) bacteria (Baross and others 1982). Cultures from all three sites produced viable colonies in 10 days; five distinct types of suspected CLAs have been noted, and all emit an odor preliminarily identified as sulfur dioxide (T. Crocker and D. Carr, personal communication). SEM examination of the corrosion residue-bedrock interface shows greatly increased porosity beneath the colonized areas with occasional binding of rock fragments away from the surface (Fig. 7), a process that is common in numerous other SEM scenes we examined. We interpret this as mechanical and/or chemical disaggregation of the country rock and a form of erosional biokarst. We suspect that biochemical and biomechanical alteration of carbonate, aluminosilicates, silica, sulfides, and other minerals comprising the parent bedrock during

Fig. 7 Scanning electron microscope photomicrograph of Capitan Limestone-corrosion residue interface showing suspected chemolithoautotrophic bacteria, The Far East, Lechuguilla Cave, Carlsbad Caverns National Park, NM. The corrosion process can be observed in the center of the scene where the substrate has been broken down and enmeshed by the organic filaments this biokarstic erosional process formed the todorokite and halloysite. Additionally, during this process, most residual material from the biocorroded ceiling bedrock probably dropped to the floor, resulting in the thicker, gravitational deposits. In the absence of being able to detect other sources of organic matter, the CLA bacteria are probably providing the requisite organic matter to heterotrophic bacteria and fungi, similar to the trophic system described from deep-sea hydrothermal vent sites (Grassle 1985, 1986; Jannasch and Wirsen 1979; Karl and others 1980; Lonsdale

1977). No other deposits of similar organic richness are known in the cave at this time.

Biothems Beyond the earlier notions that a "sparse, but intrepid community of microorganisms" inhabits the cave (Palmer 1991b), it would appear that Lechuguilla is a veritable microbiological forest compared to other desert caves. The ecosystem may have been equally extensive in the past: numerous unusual calcite speleothems in the cave contain organic filaments such as the pool finger biothems (Davis and others 1990). The organic composition of the filaments in the pool finger was confirmed by our group (using fluorescence microscopy) and supports the idea that these filaments acted as nucleation sites for crystal growth. Nonactive webulite specimens found at an ancient pool site in the Southwestern Branch of the cave (Fig. 8) show several distinct stages of development from juvenile (millimeter- to submillimeter-diameter calcitized filaments or hyphae arranged in weblike groups) to mature specimens in which the nascent webs have been integrated into long, tubular forms. Other occurrences in the cave are similar, although not as well developed. Given the current abundance of microbial material in the cave today, it is reasonable to assume that bacterial colonization was also active in the past. We interpret all of these ancient speleothems as paleobiokarst depositional forms.

Ecological significance The literature regarding chemoautotrophic producers in caves is scant (Caumartin 1963; Christiansen 1970; G o u n o t 1967) and suggests an insignificant role for these bacteria in the cave ecosystems that were studied. The chemotroph Perabacterium spelei has been suggested as a possible primary producer in some European caves (Peck 1986), but the dependence of a secondary heterotroph has not been demonstrated. In Lechuguilla, it appears that at least five types of chemolithoautotrophic bacteria are supporting large populations of chemoheterotrophic bacteria and many varieties of fungi. This interdependency is best characterized at the corrosion-residue sites and probably accounts for most of the bacterial biomass in the cave because the suspected CLA bacteria have a virtually unlimited nutrient source. The historical role of this ecosystem as a late diagenetic process is important as evidenced by the thick deposits of corroded residues deep within the cave and the presence of bacterial psuedomycelia in ancient calcite speleothems. The former process constitutes a new type of erosional biokarst in caves and may provide unique research opportunities in the field of geomicrobiology, particularly with regard to the study of bacteria-substrate attachment.

Fig. 8 Color photograph of inactive webulite, a biologically assisted speleothem (biothem) in the Southwestern Branch of Lechuguilla Cave, Carlsbad Caverns National Park, NM. Scene is approximately 10 cm across. The delicate, weblike calcite filaments are eventually assimilated into the advanced pool finger form shown at the top of the photo. Intermediate forms can be observed in the lower left-hand part of the scene. This type of speleothem is a form of depositional biokarst. Photo courtesy of Norman Thompson, 9 1989

Acknowledgments We thank R. Kerbo, D. Pate, and J. Richards of the National Park Service for their support, and appreciate the direct contributions of J. McLean, D. Davis, A. and M. Palmer, R. Olson, D. Natvig, K. LaVoie, A. Klimchouk, D. Updegraff, T. Crocker, D. Carr, and C. Dahm. Our special gratitude to the explorationists of Lechuguilla Cave: their hard work has given the world a unique gift. E. Robbins and R. Dieke, USGS- Reston, provided bacteriologic/ hydrolologic data and assistance. This work was supported in part by the U.S. Geological Survey, Onshore Oil and Gas Program, D. Gautier and T. Ahlbrandt, Program Chiefs. The manuscript benefited from the careful reviews of John McLean and John Watterson.

References Bachman GO (1980) Regional geology and Cenozoic history of the Pecos region, southeastern New Mexico. Washington, DC: US Geological Survey Open-File Report 80-1099

Baross JA, Dahm CN, Ward AK, Lilley MD, and Sedell JR (1982) Initial microbiological response in lakes to the Mr. St. Helens eruption. Nature 296:49-52 Bernasconi R (1960) L'6volution physico-chimique du Mondmilch. Varenna, Spain: Symposium Internationale de Speleologia, pp 328 Bernasconi R (1980) Mondmilch (moonmilk)--two questions of terminology. In: Beck BF (Ed), Proceedings of the 8th International Congress of Speleology Americus, GA, pp 113-116 Buchanan RE (1974) Bergey's manual of determinative bacteriology, 8th ed. Baltimore, MD: Williams & Wilkins Bull PA and Laverty M (1982) Observations on phytokarst. Z Geomorphol. 26:437-457 Caumartin V (1963) Review of the microbiology of underground environments. Natl Speleol Soe Bull 25:1-14 Christiansen K (1970) Survival of collembola on clay substrates with and without food added. Ann Speleol 25:849-852 Cunningham KI and LaRock EJ (1991) Recognition of microclimate zones through radon mapping, LechuguillaCave, Carlsbad Caverns National Park, New Mexico. Health Phys 61:493-500 Davis DG (1980) Cave development in the Guadalupe Mountains-a critical review of recent hypotheses. Natl Speleol Soc Bull 42: 4248 Davis DG, Palmer MV, and Palmer AN (1990) Extraordinary subaqueous speleothems in Lechuguilla Cave, New Mexico. Natl Speleol Soc Bull 52:70-86 DuChene HR and McLean JS (1989) The role of hydrogen sulfide in the evolution of caves in the Guadalupe Mountains of southeastern New Mexico. In: Harris PM and Grover GA (Eds), Subsurface and outcrop examination of the Capitan shelf margin, northern Delaware basin. San Antonio, TX: SEPM Core Workshop No. 13, SEPM. pp 475-481 Dyson HJ and James JM (1981) The incidence of iron bacteria in an Australian cave. In: Beck BF (Ed), Proceedings of the 8th International Congress of Speleology. Americus, GA, pp 79-81 Gounot AM (1967) La microflore des limons argileux souterrains: sone activite productrice dan la biocenose cavernicole. Ann Speleol 22:23-146 Grassle JF (1985) Hydrothermal vent animals--distribution and biology. Science 229:713-717 Grassle JF (1986) The ecology of deep-sea hydrothermal vent communities. Adv Mar Bio123 : 301 Hill CA (1982) Origin of black deposits in caves. Natl Speleol Soc Bull 44:15-19 Hill CA (1987) Geology of Carlsbad Cavern and other caves in the Guadalupe Mountains, New Mexico and Texas. N M Bur Mines Miner Resour Bull 117:150 pp Hiss WL (1980) Movement of groundwater in Permian Guadalupian aquifer systems, southeastern New Mexico and western Texas. In: Dickerson PW and Hoffer JM (Eds), New Mexico Geological Society 31st Annual Field Conference Guidebook. Albuquerque, NM, pp 289-294 Jablonsky P (1991) Distribution and identification of bat skeletons in Lechuguilla Cave. Voidspeak 1: 3-4 Jagnow DH (1989) The geology of Lechuguilla Cave, New Mexico. In: Harris PM and Grover GA (Eds), Subsurface and outcrop examination of the Capitan shelf margin , northern Delaware basin. San Antonio, TX: SEPM Core Workshop No. 13. pp. 478-481 Jannasch HW and Wirsen CO (1979) Chemosynthetic primary production at East Pacific seafloor spreading centers. BioScience 29 : 592- 598

Jannasch HW and Wirsen CO (1981) Morphological survey of microbial mats near deep-sea thermal vents, Appl Environ Microbiol 41 : 528-538 Jennings JN (1981) Morphoclimatic control a tale of piss and wind or a case of the baby out with the bathwater? In: Beck BF (Ed), Proceedings of the 8th International Congress of Speleology, Americus, GA. pp 367-368 Jones B and Motyka A (1987) Biogenic structures and micrite in stalactites from Grand Cayman Island, British West Indies. Can J Earth Sci 24:1402-1411 Karl DM, Wirsen CO, and Jannasch HW (1980) Deep-sea primary production at the Galapagos hydrothermal vents. Science 207: 1345-1347 Lambert SJ and Harvey DM (1987) Stable-isotope geochemistry of groundwaters in the Delaware basin of southeastern New Mexico. Sandia National Laboratories Report Series SAND87-0138 LaRock EJ and CunninghamKI (1990)The October 1989 Lechuguilla Microclimate study field trip. Rocky Mount Caving 7:27 33 Latham AG (1981) Muck spreading on speleothems. In: Beck BF (Ed), Proceedings of the 8th InternationalCongress of Speleology, Americus, GA. pp 356-357 Laverty M and Crabtree S (1978) Rancieite and mirabilite--some preliminary results in cave mineralogy. Trans Br Caving Res Assoc 5:135-142 Lonsdale P (1977) Clustering of suspension-feeding macrobenthos near abyssal hydrothermal vents at oceanic spreading centers. Deep Sea Res 24:857-863 Lowry DC and Jennings JN (1974) The Nullarbor karst, Australia. Z Geomorphol 18:35 81 Mason-Williams MA (1959) The formation and deposition of moonmilk. Trans Cave Res Group GB 5:133-138 Moore GW (1981) Manganese deposition in limestone caves. In: Beck BF (Ed), Proceedings of the 8th International Congress of Speleology, Americus, GA. pp 642-644 Northup DE, Carr DL, Crocker MT, Cunningham KI, Hawkins LK, Leonard P, and Welbourn WC (1993) Biological investigations in Lechuguilla Cave, Carlsbad Caverns National Park, New Mexico. Nat SpeIeol Soc Bull (in press) Palmer AN (1991a) Framework geology mapping in Lechuguilla Cave. VoidSpeak 1 :2 Palmer AN (1991b) Geology and origin of Lechuguilla Cave. In: Taylor M (Ed), Lechuguilla--Jewel of the underground. Basel, Switzerland: SpeleoProjects, 144 pp Palmer AN and Palmer MV (1989) Geologic history of the Black Hills caves, South Dakota. Natl Speleol Soc Bull 51:72-99 Peck SB (1986) Bacterial deposition of iron and manganese oxides in North American caves. Natl Speleol Soc Bull 48 : 26-30 Raper KB and Fennel DI (1965) The genus Aspergillus. Baltimore, MD: Williams and Wilkens Viles HA (1984) Biokarst--review and prospect. Prog Phys Geogr 8: 523-542 Waltham AC and Brook DB (1980) Cave development in the Melinau limestone of the Gunung Mulu National Park. Geogr J 146:258 266 Went FW (1969) Fungi associated with stalactite growth. Science 166:385-386 White WB (1981) Reflectance spectra and color in speleothems. Natl Speleol Soc Bull 43 :20-26 White WB and Ellisher AM (1958)The occurrence of unusual minerals in Pennsylvania caves. Netherworld News 6:77-80