MICROBIAL BIOFACIES IN HOT-SPRING SINTERS: A MODEL BASED ON OHAAKI POOL, NORTH ISLAND, NEW ZEALAND BRIAN JONES1, ROBIN W. RENAUT 2, AND MICHAEL R. ROSEN3 1 Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta T6G 2E3, Canada 2 Department of Geological Sciences, University of Saskatchewan, Saskatoon, Saskatchewan S7N 5E2, Canada 3 Institute of Geological and Nuclear Sciences, Wairakei Research Centre, Private Bag 2000, Taupo, New Zealand ABSTRACT: Ohaaki Pool was the main hot spring in the Broadlands– Ohaaki geothermal area before recent anthropogenic modification. The alkaline Na–HCO3–Cl water, which discharged at 958C with a flow rate of 10 l .s21, flowed down a broad discharge apron into the Waikato River. The discharge apron was inhabited by thriving microbial communities that included Calothrix, Phormidium, and numerous Synechococcus. These microbes mediated the construction of columnar stromatolites around the edge of Ohaaki Pool, oncoids in the discharge channel, and intercalated stratiform stromatolites, ‘‘Conophyton’’, and coccoid microbial mats on the distal part of the discharge apron. All the microbes were variably replaced and encrusted by amorphous silica while alive or shortly after death. Consequently, fabrics in the siliceous sinters around Ohaaki Pool are controlled by the growth patterns and composition of the microbial community. For example, the Calothrixdominated community gave rise to stratiform stromatolites that are characterized by alternating erect and prostrate laminae. Conversely, the Synechococcus-dominated communities formed mats that produced dense, white siliceous laminae. The Phormidium–Synechococcus community constructed the ‘‘Conophyton’’. Platy, skeletal, and blocky calcite crystals are found in, around, and between the siliceous stromatolites that formed around Ohaaki Pool, the proximal part of the discharge apron, and in the oncoids. Although minor calcite is found locally in the stratiform stromatolites on the discharge apron, there is no evidence that microbes played any role in calcite precipitation. INTRODUCTION Although rare on a global scale, siliceous sinters are significant from several perspectives. Economically, they commonly host epithermal gold and silver deposits. Studies on the taphonomy of silicified Precambrian microbes have commonly relied on comparisons with microbes preserved in the modern hot-spring sinters (e.g., Knoll et al. 1975; Golubic and Hofmann 1976; Bartley 1996). Further emphasis has been placed on the modern systems by the speculation that extraterrestrial microfossils may resemble thermophilic microbes found in hot-spring environments (Bock and Goode 1996). Siliceous sinters are present at many modern geothermal fields, and fossil paleosinters are known from Australia (e.g., Cunneen and Sillitoe 1989; White et al. 1989; Walter 1996; Walter et al. 1996) and Scotland (Trewin 1994, 1996). Nevertheless, apart from the recent work by Cady and Farmer (1996), there are few studies of modern sinter-forming spring systems that have attempted to produce a model with which ancient systems can be compared. Thus, the main aim of this paper is to provide a model that outlines the factors that control the formation of sinters in a modern spring setting. This study is based on Ohaaki Pool, a hot spring in the Ohaaki–Broadlands Geothermal Field on North Island, New Zealand (Fig. 1). The spring pool is surrounded by a broad sinter apron of unknown thickness that covers an area of ; 10,000 m 2. This sinter incorporates a broad array of fabrics that were controlled by the filamentous and coccoid microbes that acted as templates for amorphous silica (opal-A) precipitation. In this paper JOURNAL OF SEDIMENTARY RESEARCH, VOL. 68, NO. 3, MAY, 1998, P. 413–434 Copyright q 1998, SEPM (Society for Sedimentary Geology) 1073-130X/98/068-0413/$03.00 we describe the facies of the sinter from pool margin to the distal discharge apron and emphasize (i) the stromatolites, (ii) the microbial communities found in the stromatolites, (iii) the style of microbe preservation, and (iv) the manner in which the microbes controlled the fabrics in the sinters. The model provided by this study should have broad application to the interpretation of ancient sinter-forming systems. GEOLOGICAL AND ENVIRONMENTAL SETTING The Broadlands–Ohaaki system is located on the eastern margin of the Taupo Volcanic Zone (Lonker et al. 1990). In this area, shallow-dipping Quaternary volcanic deposits and lacustrine sediments unconformably overlie Mesozoic sediments (Grindley 1970). The hydrology of the Broadlands– Ohaaki system is controlled by an alternating sequence of nearly horizontal, permeable and less permeable units that are crossed by vertical fractures and normal faults that act as feeder channels (Browne and Ellis 1970). Ohaaki Pool is ; 45 m long and ; 15 m wide with a surface area of ; 850 m 2 (Figs. 1, 2). The base of the pool has two irregular funnelshaped depressions, up to 12 m deep, that are separated by a ridge that rises up to 1.5 m below water level (Glover et al. 1996; Hunt 1997, plate 3). In its original state (pre-1980s), Ohaaki Pool discharged water at 958C with a flow rate of 10 l .s21 (Browne 1970; Browne and Lloyd 1987). Before anthropogenic modification related to the development of the geothermal power plant, Ohaaki Pool was commonly ebullient above the centers of the depressions with surges up to 1.7 m high (Lloyd 1957; Glover et al. 1996). Before 1957, a shallow ditch was dug in the sinter (Fig. 1D) so that water could drain from the pool into the open-air baths (Lloyd 1957) of the nearby Marae (central area of a Maori settlement). Even before anthropogenic modification, flow from Ohaaki Pool was irregular (Lloyd 1957). Today, water level is maintained by pumping waste geothermal fluid into it from the power plant (Glover et al. 1996; Hunt 1997). Water no longer flows freely across the sinter apron, but is confined to the narrow excavated ditch and a broad, unconfined channel on the southwest part of the area (Fig. 1D). Fortunately, analyses of the original hot-spring water are available (Table 1). The sinters around Ohaaki Pool were precipitated from the discharged alkaline Na–HCO3–Cl dilute spring waters (Mahon and Finlayson 1972; Glover and Hedenquist 1989). Although water composition was relatively uniform, significant, short-term changes did take place. In mid-1957, for example, orange-red precipitates (‘‘metastibnite’’) containing Au, Ag, Hg, Tl, and As formed on the sinter surface around the pool (Weissberg 1969; Goldie 1985; Browne and Lloyd 1987). By 1966, however, those precipitates had been covered by newly precipitated siliceous sinter (Weissberg 1969). METHODS Samples were collected from the rim around Ohaaki Pool (Fig. 2A), the floor of the discharge channel, and the discharge apron that stretches from the pool to the Waikato River (Figs. 1, 2). In addition, 22 cores (up to 25 cm long, 2.5 cm diameter) were drilled along a transect across the discharge apron (Figs. 1D, E, 2B). Recovery varied from 0 to 100%. Large hand samples and cores were slabbed and polished. Large thin sections (7 cm x 5 cm) were made from selected samples. Following pet- 414 B. JONES ET AL. FIG. 1.—A, B, C) Location of Ohaaki Pool in the Taupo Volcanic Zone, North Island, New Zealand. D) Map of area around Ohaaki Pool showing location of siliceous sinter apron, the discharge drain, and core transect. The area labeled FS, which is covered with silicified filamentous streamer microbes (Calothrix?), may be a recent feature related to a temporary diversion of water from the excavated ditch. E) Positions of cores relative to fence around Ohaaki Pool. rographic analysis, fractured pieces were examined using scanning electron microscopy (SEM). Some samples were coated with gold and examined on a Cambridge S250 SEM with an accelerating voltage of 25 kV whereas other samples, which were uncoated or coated with a thin layer of gold, were examined on a JEOL 6301F Field-Emission SEM (FE-SEM) with an accelerating voltage of 1.5–2.5 kV. Elemental content was determined using an energy dispersive X-ray analyzer attached to each SEM. Mineralogy was confirmed by X-ray diffraction of powdered cavity mounts using a Rigaku rotating-anode diffractometer. TERMINOLOGY AND MICROBE IDENTIFICATION The term microbialite is applied to ‘‘. . . organized structures that have accreted as a result of a benthic microbial community . . . . forming the locus of mineral precipitation’’, whereas the term ‘‘stromatolite’’ is used for microbialites with fine, approximately planar internal laminae (cf. Burne and Moore 1987). A filament is formed of a trichome that is surrounded by a sheath (cf. Riding 1991). Laminae formed of (sub)erect filaments are called ‘‘erect laminae’’ whereas those formed of filaments that lie parallel to laminae boundaries are called ‘‘prostrate laminae’’ (cf. Riding 1991, fig. 3). Although microbe preservation in the Ohaaki Pool sinters is superb, many critical features required for their identification were not preserved or were disguised by thick coatings of amorphous silica (cf. Schultze-Lam et al. 1995; Jones et al. 1997a). Features evident for most of these microbes include the (i) internal and external diameters, (ii) presence or absence of sheaths, and (iii) overall morphology of the filamentous microbes in terms of their branching style. Determining filament length is problematical because of the difficulty of tracing filaments on SEM samples and the low probability of a filament remaining in the plane of a thin section. Additional information for some, but not all, microbes includes color (as seen in plane polarized light [PPL] in thin section), sheath structure, and cell dimensions in the trichome (cf. Jones et al. 1997a). Although this information allows MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 415 FIG. 2.—A) Edge of Ohaaki Pool showing raised rim formed of stromatolites (arrow) and upper surface of laminated sinter (SS) behind rim (; 0.5 m wide). B) General view down discharge apron from Ohaaki Pool. The tape measure indicates line along which most cores were taken. C) General view across gently sloping surface of discharge apron. D) Inactive terrace pools on discharge apron south of Ohaaki Pool. The pools are generally 10–50 cm wide. E) Outflow channel from Ohaaki Pool; note steam rising from water in channel. F) Oncoids on floor of outflow channel. The largest oncoids are ; 25 mm long. 416 B. JONES ET AL. TABLE 1.—Composition of water from Ohaaki Pool. All samples are from the main pool. Analysis Date G M&F B&L pre-1937 1965 pre-1987 Flow l/sec T8C pH 10 94 98 95 8.1 7.0 7.0 Li Na K 7.4 926 860 860 84 82 82 Rb 0.1 Cs Ca Mg 1.2 12.0 8.5 2.6 0.5 0.2 0.5 F Cl 5.2 1048 1040 1030 Br 3.0 I SO4 0.6 67 100 100 B SiO2 32.1 305 338 338 As NH3 HCO3 H2S 1.5 4.3 3.8 768 680 1.0 G 5 Grange (1937, p. 97, 104) original data expressed in parts per 100,000; converted to ppm for this table. M & F 5 Mahon and Finlayson (1972, Table 1). B & L 5 Browne and Lloyd (1987, Table 1), expressed in mg·kg. determination of the general affinity of some microbes, it is generally insufficient to permit accurate identification. OHAAKI POOL MICROBIALITES The main physiographic elements in the Ohaaki Pool system are the spring pool (Fig. 2A), a broad, low-gradient discharge apron that lies between the spring pool and Waikato River (Fig. 2B–D), and an outflow channel (Fig. 2E). The discharge apron is constructed by a series of very shallow, arcuate terraced pools (‘‘microterraces’’; Fig. 2D). Each element of this system is characterized by distinctive microbialites that formed in response to the temperature and hydrodynamic conditions of an area. Columnar stromatolites are found around the lip of Ohaaki Pool (Fig. 2A); stratiform stromatolites, ‘‘Conophyton’’, and coccoid mats are present on the discharge apron; and oncoids are found in the outflow channel and some terrace pools near the spring (Fig. 2F). Columnar Stromatolites Morphology and Composition.—The rim around Ohaaki Pool is formed of upward-expanding columnar stromatolites, up to 2 cm in diameter, that curve gently and radiate outward from a basal area, 3–4 cm wide, on the pool wall (Figs. 2A, 3, 4A). Although this ‘‘. . .unique fretted scroll-design sinter lip. . .’’ (Browne and Lloyd 1987) was mentioned by Grange (1937) and illustrated by Ellis (1975), Jones et al. (1997a) showed that it is formed of columnar stromatolites. Calcite crystals are found on the pool-rim walls just below and onlapping the basal parts of the stromatolites, between the stromatolites, and inside the stromatolites. The platy calcite crystals on the pool-rim walls and basal part of the stromatolites are up to 4 mm long, 3 mm wide, and 0.5 mm thick (Fig. 4B). These crystals, with their c axis parallel to the short axis of the plates, are morphologically like the calcite crystals found in the subsurface of the Ohaaki–Broadlands area (Simmons and Christenson 1994). Although scattered or small clusters of silicified microbes (, 4 mm long, 0.4 mm diameter) are present on some crystal faces, they do not form any organized structures (Fig. 4C). The stromatolites are formed of interlaminated amorphous silica sheets (up to 25 mm thick, generally , 5 mm thick) and silicified microbial mats (Fig. 4D, E). Thicker parts of the amorphous silica laminae contain irregular-shaped pores (, 20 mm long, 5 mm high) that represent gaps between vertically stacked wavy laminae. There is no evidence that microbes controlled their formation. In contrast the silicified microbial mats, , 1 mm thick, are composed of intertwined filamentous microbes (Fig. 4E). The porosity of these laminae is a function of the amount of silica encrusting the filamentous microbes. Hexagonal skeletal calcite crystals and blocky calcite crystals are present inside the columnar stromatolites. The skeletal crystals (, 50 mm long, 5 mm wide), which are associated with the silicified filaments, are commonly encrusted by thin layers of amorphous silica (Jones and Renaut 1996a). Blocky calcite crystals, up to 75 mm long, are rooted on pore walls in the amorphous silica laminae. Many crystals are coated with amorphous silica laminae. Microbe Morphology, Affinity, and Diversity.—The silicified microbial mats (Fig. 4E) are formed of interwoven, nonbranching, sheathed filaments that are at least 50 mm long with an external diameter of 1.25– 1.50 mm, an internal diameter of 0.5–0.6 mm, and walls , 0.5 mm thick (Fig. 4E). Locally, silica encrustation increased wall thickness to 5 mm. Although many silicified sheaths are hollow, others contain silicified trichomes up to 1 mm in diameter. Locally, patches of mucus (EPS: extracellular polysaccharide films), that envelop collapsed spores up to 5.5 mm in diameter, are present between and around the filaments. Although the morphological similarity of the silicified microbes indicates a monospecific community, their taxonomic affinity is unknown. Comparisons.—The columnar stromatolites around Ohaaki Pool are akin to those found in the sinter spine around the orifice of Dragon’s Mouth Geyser in the Wairakei Thermal Valley (Jones et al. 1997a). Oncoids FIG. 3.—View of columnar stromatolites (; 1 m total height) around edge of Ohaaki Pool. View taken from inside pool when it was fully drained. Photograph courtesy of Contact Energy Ltd., New Zealand. Morphology and Composition.—Numerous silica oncoids, up to 25 mm long, 20 mm wide, and 12 mm high, are found on the floor of the shallow outflow channel (; 8 m wide, , 15 mm deep) that crosses the discharge apron to the southwest (Fig. 2E). Today, the outflow water is a mixture of natural waters and waste water from the Ohaaki Power Station (Glover et al. 1996). Unfortunately, we do not know if the oncoids are a product of the modern flow regime or if they predate anthropogenic modifications, which began in 1965. The height of these discoid oncoids, which have a ‘‘rough’’ exterior (Fig. 2F), is limited by water depth because their length and width seem to increase without limit whereas their height is consistently , 15 mm (Fig. 5). The porous oncoids, formed of amorphous silica, calcite, and minor gypsum, have a nucleus, and inner and outer cortical zones (Figs. 6, 7). The irregular-shaped nuclei, up to 4 mm long and 3 mm wide, are formed of dense, low-porosity amorphous silica. Some nuclei contain well-preserved filamentous microbes whereas others are formed of finely laminated silica. Calcite crystals are found in some pores in the nuclei. MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 417 FIG. 4.—Columnar stromatolites from edge of Ohaaki Pool. A) Vertical cross section showing stromatolite columns (SC), stratiform stromatolites (SS), and platy calcite crystals (C). B) General view of platy calcite crystals that cover basal part of columnar stromatolites shown in Figure 4A. C) Scattered silicified microbes resting on surfaces of platy calcite crystals. D) Interior of columnar stromatolite showing thin layer of structureless amorphous silica overlying mat formed of filamentous microbes. E) Enlarged view of filamentous microbes that form mat shown in Figure 4D. For further illustrations of these stromatolites see Jones et al. (1997a). In many oncoids, the inner and outer cortical zones are separated by a thin band (, 0.5 mm) of thinly laminated amorphous silica. The inner cortical zone is more porous and contains more calcite and gypsum than the outer zone. In addition, the outer zone contains numerous laminated, columnar structures. The outer part of the inner zone generally has a higher silica/calcite ratio than the inner part and there are isolated columnar struc- FIG. 5.—Bivariate graphs for length versus width and length versus height of oncoids from outflow channel (Fig. 2E, F). tures formed of medium- to dark-gray (PPL) arched laminae of amorphous silica. The anhedral to subhedral calcite crystals (, 0.4 mm long) of the inner zone commonly form pseudodendritic arrays that radiate from the nucleus (Fig. 6A, B). The pseudodendrites formed by the growth of one crystal on top of another. Branches formed where two or more crystals nucleated on an underlying crystal (Fig. 6A, B). Further growth involved the addition of more crystals to each of those crystals. Although many calcite crystals are anhedral, others are euhedral with a blocky to rhombic form. Some euhedral crystals are zoned, with the zones being delineated by minute inclusions (Fig. 6C). Severe etching of many calcite crystals produced spiky calcite (Fig. 7F; cf. Jones and Pemberton 1987). The calcite arrays are not true dendrites because they are formed of crystal aggregates rather than a single crystal that branches (cf. Jones and Renaut 1995). The pseudodendritic calcite arrays are encased by colorless (PPL) layers of amorphous silica up to 0.2 mm thick (Figs. 6A–C, 7A–C). Locally, silicified filamentous microbes are present around and between the calcite crystals (Fig. 7D, E). Banding in the silica laminae around the calcite crystals (Fig. 7B), which can be traced from the calcite crystals to nearby filamentous microbes (Fig. 7D), indicates that these laminae probably developed as isopachous cements. Subhedral to euhedral gypsum crystals, , 0.4 mm long (Fig. 6D), are rooted on the silica that coats the pseudodendritic calcite. Other gypsum crystals are found inside the pseudodendritic arrays, typically in cavities inside a shell of amorphous silica (Fig. 7G, H). The silica casing, however, does not follow the contours of the gypsum crystals as it does with the calcite crystals. This, plus the angularity of the cavities (Fig. 7G), indicates that they were once filled with subhedral to euhedral calcite crystals. These relationships indicate that the gypsum crystals formed after the calcite had dissolved to leave cavities in the silica. This notion is supported by the 418 B. JONES ET AL. FIG. 6.—Thin-section photomicrographs showing textures in oncoids from outflow channel. A, B) General views of inner part of oncoid showing pseudodendritic array of calcite crystals (C) encased with amorphous silica (arrows). C) Enlarged view of zoned calcite crystals (arrow). D) Gypsum crystals (arrow) associated with calcite pseudodendrites in inner part of oncoid. E) General view of outermost part of inner zone showing pseudodendritic array of calcite crystals (C), laterally discontinuous laminae of amorphous silica (arrow), and amorphous silica (S) encasing calcite crystals. F, G) General views of outer zone showing pseudodendritic arrays of calcite (C) that are encased by amorphous silica (S). Note thin, laterally discontinuous laminae of silica between layers of calcite crystals. → FIG. 7.—SEM photomicrographs of oncoids from Ohaaki Pool. A–C) General views of calcite crystals surrounded by amorphous silica. Note banding in silica (B). D, E) Example of filamentous microbe replaced and encrusted by amorphous silica. Note banding in the silica (D) similar to that around the calcite crystals (B). F) General view of calcite that has been severely etched to produce spiky calcite. Note amorphous silica shell. G) Amorphous silica with angular cavities, formed by dissolution of calcite, that contain small gypsum crystals. H) Gypsum crystals that partly fill cavities shown in Figure 7G. I) Silicified spherical body from outer part of an oncoid. J) MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 419 Cross section through spore (?) from outer part of oncoid. K) Outer surface of subspherical body with a trilete structure similar to that found in some spores. L) Wall of spore (?) shown in Figure 7J. M) Cluster of small spherical bodies on outer surface of oncoid. N) Spherical body covered with numerous small opal-A spheres. O) Disarticulated diatom frustule from outer part of oncoid. 420 B. JONES ET AL. fact that silicified filaments are commonly encased by gypsum crystals whereas other filaments span the gaps between the gypsum crystals and the inner wall of the silica shell. The ‘‘solid’’ silica laminae that divide the inner and outer zones are formed of stacked laminae, 50–100 mm thick, that are separated from each other by laterally discontinuous pores, 20–50 mm thick. Individual laminae are structureless with no evidence of microbes. Locally, the laminae are crenulated and arched toward the exterior of the oncoid. Cavities beneath these arches are commonly contain small (7–8 mm long, 1–2 mm wide) euhedral gypsum crystals. The outer cortical zone is formed of columnar structures that are separated from each other by narrow gaps (Fig. 6E–G). Many columns are rooted on the high points of the crenulations in the underlying laminae. The core of each column is formed of gray (PPL) amorphous silica laminae that are separated from each other by blocky calcite crystals and (or) colorless amorphous silica (Fig. 6E–G). Some calcite crystals are arranged in pseudodendritic arrays. The amount of calcite decreases toward the outer edge of the oncoid. Silicified filaments are present in parts of these columns. Small (, 0.15 mm) angular detrital grains are trapped in some parts of the outer zone. Microbe Morphology, Affinity, and Diversity.—The dominant filamentous microbes, 8–10 mm in diameter, are disguised by thick (up to 20 mm) amorphous silica encrustations (Fig. 7D, E). Another microbe, up to 14 mm long and 2 mm in diameter, is found on the surface of the amorphous silica that encrusts the larger filaments. The taxonomic affinity of these microbes is unknown. On the outer surfaces of oncoids and in the outer cortical zone there are (sub)spherical silicified spores and (or) cysts. Some spores, 7–8 mm in diameter, have a smooth, featureless surface and an attachment site (Fig. 7I). Other (sub)spherical bodies, up to 60 mm long and 50 mm wide (Fig. 7J, K), have an outer shell formed of three layers (Fig. 7L). The distinct trilete ridge (Fig. 7K) found in one specimen indicates that they are probably spores that came from plants around the spring. Some spherical bodies, 7–9 mm in diameter, are covered with amorphous silica beads up to 0.3 mm (average 0.15 mm) in diameter (Fig. 7M, N). Rare diatom fragments are scattered throughout the cortex of the oncoids (Fig. 7O). In summary, the diverse microbial community is formed of filamentous microbes, diatoms, spores and algal cysts, and spores (?) derived from the surrounding vegetation. Comparisons.—Unlike the siliceous oncoids from Orakeikorako, New Zealand (Renaut et al. 1996) and El Tatio, Chile (Jones and Renaut 1997), the oncoids at Ohaaki Pool contain calcite and gypsum. The Orakeikorako oncoids are formed of concentrically arranged porous erect laminae and low-porosity prostrate laminae. The El Tatio oncoids grew through development of spicules composed of alternating erect and prostrate laminae. There is, however, no evidence that growth of the Ohaaki Pool oncoids was controlled by the attitude of the filamentous microbes. The oncoids from Ohaaki Pool are akin to the ‘‘pisoliths’’ from Laguna Pastos Grandes, Bolivia, which are formed of alternating calcite and silica laminae that contain numerous diatoms and scattered filaments (Jones and Renaut 1994). Calcite laminae in the Pastos Grandes oncoids, however, are formed of merged bundles of radiating acicular crystals, not the pseudodendritic calcite arrays found in the Ohaaki Pool oncoids. ‘‘Conophyton’’ Morphology and Composition.—These stromatolites, formed of amorphous silica, are restricted to small areas scattered across the discharge apron. They form beds, 1–12 cm thick, that are intercalated with stratiform stromatolites. They are dominated by vertical isodiametric columns, up to 3 cm high and 2–5 mm in diameter, that have (sub)rounded transverse sections (Fig. 8). Individual columns taper to a rounded or conical summit (Fig. 8A–C). Neighboring columns, typically , 5 mm apart, are connected by undulatory horizontal sheets (, 0.5 mm thick) of amorphous silica (Figs. 8A–D, 9A, B). Vertical ridges (, 0.75 mm thick) that radiate from the columns originate from the outermost laminae of the columns (Fig. 9A, B). Column interiors are formed of steeply inclined laminae that arch over the apex of the column (Fig. 8). The porous core is formed of stacked laminae that are separated by gaps that are commonly thicker than the silica laminae (Fig. 8C, E). The core is encased by an outer zone that is formed of alternating ‘‘solid’’ and porous microlaminae (Figs. 8A, C, E, 9C). The ‘‘solid’’ laminae contain numerous coccoid (single, double, and chains of four cells; terminology cf. Copeland 1936) microbes and scattered filamentous microbes, , 1 mm in diameter, that are replaced and encrusted by amorphous silica (Fig. 9D–K). Long axes of elongate coccoid microbes are typically aligned parallel to the laminae. The filamentous microbes, however, had little influence on the fabric of the laminae. Many ‘‘solid’’ sheets are broken into polygonal masses by desiccation cracks (Fig. 9E– G). The ‘‘porous’’ microlaminae that overlie the desiccated ‘‘solid’’ laminae are formed of numerous filamentous microbes (, 2 mm in diameter) that are partly silicified and collapsed or silicified but not collapsed (Fig. 9E–G). Locally, silicified mucus (?), spores, and spores attached to filaments are present (Fig. 9M, N). Minute beads of amorphous silica grew in the mucus (Fig. 9O). The fact that the collapsed filaments commonly bridge the desiccation cracks with no sign of disruption (Fig. 9G) indicates that the ‘‘solid’’ silica sheets were desiccated before colonization by the filaments. Most horizontal and vertical sheets between the columns are formed of dense amorphous silica that is essentially featureless (Fig. 9B). Where there is less silica precipitation, however, interwoven filamentous microbes are apparent (Fig. 10A, B). Locally, vertical sections through the horizontal laminae reveal numerous double-celled and four-celled coccoid microbes (Fig. 10D–F). Thus, these sheets appear to have formed by amorphous silica precipitation around filamentous and coccoid microbes. Microbe Morphology, Affinity, and Diversity.—The nonbranching isodiametric filaments (, 0.5 mm diameter), which are commonly encrusted with small (, 0.2 mm) silica beads, are similar to Phormidium as described by Nash (1938), Cassie (1989), and Cassie-Cooper (1991). Although some filaments could be the flexibacterium Chloroflexus, it is difficult to separate it microscopically from Phormidium (cf. Walter et al. 1972). The coccoid microbes are unicellular, double celled, or form chains of four cells (Figs. 9I–K, 10C–E). Unicellular forms are , 2.5 mm in diameter with silicified walls that are 200 nm thick, whereas the doublecelled forms are , 4.5 mm long and 1–2 mm wide (Fig. 10E, F). The rare chains of four cells are , 4 mm long and 1.5 mm wide (Fig. 10D). These cells commonly contain numerous (sub)spherical granules, 300–400 nm in diameter, that are penetrated by a small (20–30 nm) hole (Fig. 10F–H). These microbes are similar to Synechococcus, a common photosynthetic cyanobacterium in thermal systems (e.g., Copeland 1936; Brock 1969; Meeks and Castenholz 1971; Zhang 1986). Silicified granules inside the specimens from Ohaaki Pool may be equivalent to granules found in living Synechococcus (cf. Copeland 1936) and other cyanobacteria (cf. Stanier 1977). In size and cell morphology, the Ohaaki specimens resemble Synechococcus elongatus that Copeland (1936) described from Yellowstone National Park. Comparisons.—Following White et al. (1989, figs. 4d–e), we use the term ‘‘Conophyton’’ based on the similarity of these stromatolites to the Yellowstone Conophyton described by Walter et al. (1976). The Conophyton from Ohaaki Pool resemble Conophyton weedii in terms of their external morphology (cf. Walter et al. 1976, figs. 20–22) and internal structure (cf. Walter et al. 1976, figs. 23, 24, 27–29). At Yellowstone, the microbial community associated with modern Conophyton includes Phormidium tenue var. granuliferum with lesser numbers of Chloroflexus aurantiacus, Synechococcus, Pseudanabaena, and Isocystis (Walter et al. 1976). In the Yellowstone examples, Phormidium tenue var. granuliferum is more com- MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 421 FIG. 8.—Thin-section photomicrographs showing textures in ‘‘Conophyton’’ from Ohaaki Pool. A) Column with porous core (dark colored area) encased by dense, finely laminated outer zone (light colored area). B) ‘‘Stacked Conophyton’’ columns and horizontal mats that spread laterally from base of each column (arrows). C) Porous core of ‘‘Conophyton’’ encased by finely laminated outer zone. D) Neighboring columns joined by thin horizontal mats (arrows). E) Outer part of ‘‘Conophyton’’ column showing contrast between porous core and finely laminated outer part. F) Enlarged view of horizontal mat showing cross sections of filamentous microbes that are responsible for their formation. mon in the cones than in the adjacent mats, whereas Chloroflexus is less abundant on the cones than in the mats. Construction of the cones was attributed to Phormidium. In Yellowstone, Conophyton is found in water with T of 32–598C and a pH of 7–9 in settings that range from intermittent flow in outflow channels, to gently and irregularly sheet flow in shallow pools, to nearly still pools up to 30 cm deep (Walter et al. 1972, 1976). Stratiform Stromatolites Morphology and Composition.—The discharge apron around Ohaaki Pool is largely constructed of stratiform stromatolites (Figs. 11–14) that are formed of intercalated erect (, 5 mm) and prostrate (, 2 mm) gray laminae. The porous erect laminae (; 60–70%) are formed of pillars up to 25 mm in diameter that curve and locally branch or merge upward (Figs. 11, 12A–D). Larger, composite pillars formed by the merger of narrow pillars. Each pillar is formed of up to 10 (average 5) sheathed filaments (. 2 mm long) that are encased by amorphous silica (Fig. 11C–E). In PPL, the sheaths are commonly highlighted by their yellowish-orange color. Gaps between the vertical pillars are commonly spanned by individual filaments (, 5 mm diameter) or bundles of filaments, 10–15 mm in diameter (Fig. 12C, D) that have been replaced and encrusted by amorphous silica (Fig. 12D). Locally, filaments at least 200 mm long and 10 mm in diameter are attached to the sides of the vertical pillars and lie parallel to bedding (Fig. 12E). Thin films of silicified mucus commonly span the gaps between the filaments or cover the filament surfaces. Locally, these films form ‘‘curtain-like’’ structures between the pillars. Collectively, the fila- 422 B. JONES ET AL. FIG. 9.—SEM photomicrographs of ‘‘Conophyton’’. A) Transverse section though a ‘‘Conophyton’’ column showing its laminated interior and thin vertical sheets radiating from its exterior. B) Junction between vertical sheet and column exterior. C) Vertical cross section through outer part of ‘‘Conophyton’’ column showing alternating microlaminae of ‘‘solid’’ and porous amorphous silica. D) Porous microlaminae showing silicified filamentous microbes. E) Exterior of thin amorphous silica microlaminae showing desiccation cracks. F) Enlarged view of desiccated amorphous silica lamina that has partly engulfed a spherical, spore-like body. G) Oblique view of amorphous silica lamina with a filamentous microbe on its surface that is not broken by the desiccation crack. H) Fragmented diatom resting on surface of amorphous MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS ments and mucus strengthened the (sub)vertical bundles of filaments that dominated the microbial mats. The prostrate laminae, ; 1 mm thick, are formed of densely packed sheathed filaments (Figs. 11A, B, 13A–C). Their low porosity is due to crowding of the filaments and extensive silica encrustation (Figs. 11B, 13C). Individual filaments change attitude as they pass from the underlying erect laminae into the prostrate laminae, or from the prostrate laminae into the overlying erect laminae (Fig. 11C). This indicates that the dominant microbes in the prostrate laminae are the same as those in the erect laminae. Surfaces of the stratiform stromatolites are locally characterized by ‘‘streamers’’ that lie subparallel to each other and, presumably, the paleocurrent direction (Fig. 15). They are similar to the ‘‘streamer mats’’ in paleosinters from Queensland (Walter et al. 1996, fig. 18a, b) and Steamboat Springs (White et al. 1964, fig. 11), and modern microbial mats in Yellowstone National Park (Walter et al., 1996, fig. 18c), where they are formed mainly by Phormidium. Microbial Morphology, Affinity, and Diversity.—Large-diameter, filamentous microbes encased by pigmented sheaths dominate the erect and prostrate laminae (Figs. 12–14). Smaller epiphytic microbes colonized these encrusted filaments (Fig. 14). Transverse sections through the large-diameter filaments in the erect laminae show that they commonly encompass four zones (Fig. 12H, I). Zone 1, the filament core, is open or filled by a silicified trichome (Fig. 12H, I). Although typically 8–10 mm in diameter, it is only ; 4 mm in diameter in some specimens. Zone 2, evident only if Zone 1 is vacant, is a thin (, 0.5 mm), smooth silicified membrane that covers the inner sheath wall. In some specimens this membrane has a ‘‘bumpy’’ appearance that mimics the surfaces of the silica spheres in Zone 3. Zone 3 (2–3 mm thick) is formed of loosely to tightly packed silica spheroids (; 1 mm diameter) that have a rough exterior (Fig. 12J). In some specimens, there is an innermost ring, ; 0.5 mm thick, formed of densely packed spheres (Fig. 12I, J). Commonly, there are laterally continuous or discontinuous laminae that probably mimic the structure of the original sheath (Fig. 12J). Zone 4, of variable thickness, is formed of dense, largely structureless silica that was precipitated around the sheath (Fig. 12H, I, K). Where filaments are crowded together, the outermost zones of neighboring filaments merge to form a larger mass of structureless silica that encases numerous filaments (Fig. 12G, H). In thin section (PPL), some trichomes are outlined rarely by a thin black line. In most examples, the trichome segments are shorter than the encasing sheath. Some trichomes taper toward their rounded apical ends. These filaments are tentatively assigned to Calothrix on basis of the large trichome diameter, the apical tapering of the trichome, the laminated sheath, and the lack of branching. The yellowish-brown to orange color of the sheath is consistent with the tangerine orange-yellow sheaths of C. thermalis in modern hot springs in New Zealand (Cassie 1989). These filaments are similar to C. thermalis as described by Copeland (1936) and Cassie (1989), and Calothrix parietina var. thermalis described by Nash (1938). Small-diameter filaments, encased in silica, are commonly wrapped around the outer surfaces of the silica encrusted Calothrix filaments (Fig. 13A–D). These microbes, at least 15 mm long, unbranched, and ; 0.7 mm in diameter, lack a sheath, and have linked cells ; 1.5 mm long (Fig. 13E, F). They are similar to Phormidium as described by Cassie-Cooper (1991). Transverse sections through the Calothrix filaments commonly reveal tubular pores, up to 1 mm in diameter, embedded in the silica of Zone 4 (Fig. 12H, I, K, L). Some tubes are surrounded by a silicified sheath that 423 has merged almost imperceptibly with the surrounding silica. These pores probably represent Phormidium that once lived on the outer surfaces of the larger filaments but were subsequently interred as more silica was precipitated. Unicellular microbes (, 2.75 mm long, 2.5 mm wide) and double cells (, 5.0 mm long, 2.5 mm wide) are common (Fig. 12L). Although most double-celled forms are symmetrical about their median constriction, some are asymmetrical with one cell being smaller and less bulbous than the other. Mixed with these microbes are chains of four cells with transverse constrictions separating adjacent cells. The middle transverse constriction, however, is more pronounced than the other constrictions. As with the double-celled forms, the terminal cells in some chains are smaller and less bulbous than others. The similarity between the smaller cells in the double-celled forms and chains of four cells indicates that they are probably a natural growth feature. Individual cells are hollow or contain silica granules that are , 0.25 mm in diameter (cf. Fig. 10F–H). The Calothrix filaments with the epiphytic flora of filamentous and coccoid microbes form a complex community. This is similar to modern mats in Yellowstone National Park, where small microbes like Synechococcus survive in fast-flowing water only by attaching themselves to filamentous microbes such as Chloroflexus (Brock 1969, 1978). Evidence from Ohaaki Pool (e.g., Fig. 12H, K) shows that silica precipitation around the Calothrix filaments and subsequent colonization by epiphytic microbes was a cyclic, time-separated process. Thus, during periods of silica precipitation, the epiphytic microbes could not colonize the substrate because the rate of silica precipitation exceeded their growth rate. When silica precipitation ceased, however, the epiphytic microbes colonized the substrate. In some areas there are unsheathed filaments, at least 75 long and , 5 mm in diameter, with cells up to 15 mm long. Their original septae are now denoted by hairline gaps (Fig. 13G, H). These gaps must be part of the original filament structure because they do not cut through the amorphous silica that encrusts the filament (Fig. 13H). The prostrate laminae are formed of the Calothrix that have been encased in amorphous silica (Fig. 14). Compared to those in the erect laminae, they are more crowded and there are only scattered epiphytic microbes associated with them (Fig. 14). Comparisons.—Dark olive-green mats of Calothrix thermalis (Rivulariaceae) are present at Naike Hot Springs, Artist’s Palette (Waiotapu), and Rainbow Terrace (Orakeikorako) on North Island, New Zealand (Cassie 1989). At those localities, filaments of Calothrix thermalis, up to 3 mm long, are encased by tangerine-orange sheaths that are commonly encrusted with silica (Cassie 1989). Although the cells are typically 8–9 mm wide and 3–5 mm long in the basal region of the filament, they taper toward their distal ends. This species inhabits waters with a T of 25–288C and a pH of 6.0–6.5, and commonly contributes to the formation of white and gray sinter terraces (Cassie 1989). The alternating erect-prostrate laminae in the Calothrix mats at Ohaaki Pool resemble the laminated microbialites found at Spectacles Spring in China (Zhang 1986, figs. 6H–I). In Yellowstone National Park, Calothrix species are abundant where the water T is 20–408C (Copeland 1936; Walter 1976a; Walter et al. 1996; Cady and Farmer 1996) and play an important role in the formation of hot-spring terraces. Calothrix, for example, is the dominant element of the ‘‘Calothrix–Scytonema–Schizothrix’’ community that covers much of the Lower Geyser Basin (Copeland 1936). The orange color of cyanobacterial sheaths has been attributed to scytonemin (e.g., Golubic and Hofmann 1976; Garcia-Pichel and Castenholz 1991; Garcia-Pichel et al. 1992), a pigment found in cyanobacteria from ← silica microlaminae. I) Vertical section through a ‘‘solid’’ silica lamina in outer part of ‘‘Conophyton’’ column showing numerous coccoid microbes that are encased with amorphous silica. J, K) Enlarged views of coccoid microbes that are encased with amorphous silica. L) Outer surface of filamentous microbial mat that forms the porous laminae between the ‘‘solid’’ silica laminae. M) Small spore-like bodies encased with mucus (?) that has been replaced by amorphous silica. N) Spore-like body attached to end of filamentous microbe. O) Small spheres of amorphous silica that probably formed in mucus associated with filamentous microbial mats. 424 B. JONES ET AL. FIG. 10.—SEM photomicrographs of horizontal mats found between ‘‘Conophyton’’ columns. A) Small patch of microbial mat that has not been obscured by amorphous silica precipitates. B) Enlarged view of filamentous microbes from area shown in Figure 10A. C) General view of horizontal mat that is formed of amorphous silica that was precipitated around coccoid microbes. D) Chain of four-celled coccoid microbe embedded in amorphous silica. E) Double-celled coccoid microbes embedded in amorphous silica. F) Double-celled coccoid microbe filled with silicified granules. G, H) Enlarged views of coccoid microbes with granules. Each granule is pierced by a small hole (H). I) Cross section through a filament or a coccoid microbe showing wall that has been replaced by amorphous silica. diverse habitats and locations, but particularly near the surface of Calothrix mats. The pigment may provide a sunscreen against ultraviolet light (Garcia-Pichel and Castenholz 1991; Garcia-Pichel et al. 1992) or protection against high light intensities (Fogg et al. 1973; Whitton and Potts 1982; Pentecost 1985). COCCOID MICROBIAL MATS Morphology and Composition.—Coccoid microbial mats, which form white laminae up to 3 mm thick, rest on top of erect laminae or, in rare cases, on top of prostrate laminae in the stratiform stromatolites (Fig. 16). These mats are formed of dense amorphous silica that is devoid of organized structures or contains indistinct microlaminae that are visible as light brown hues in PPL (Fig. 16). Some of these features highlight undulatory laminae that resemble microstromatolites (Fig. 16). Microbial Morphology, Affinity, and Diversity.—Some laminae and localized areas of other laminae lack recognizable microbes. Conversely, other laminae contain dense populations of coccoid microbes and scattered filaments (Fig. 17). Locally, the microbial community comprises (sub)spherical, unicellular (, 1.5 mm long), double-celled (, 3 mm long, 1.5 mm wide), and four-celled (, 7 mm long, 2.5 mm wide) microbes (Fig. 17A–D). These microbes, which resemble the coccoid microbes in the ‘‘Conophyton’’ and on the Calothrix filaments in the erect laminae of the stratiform stromatolites (e.g., Figs. 9J, K, 10C–H, 13I), are probably Synechococcus. Associated with these microbes are elongate microbes (, 6 mm long, 1.25 mm diameter) that may be Synechococcus lividus (Fig. 17D). Filamentous microbes are present on the surfaces of some laminae. The mode of preservation of the coccoid microbes in the white laminae is critical to laminae formation and their apparent lack of structure. Walls of the coccoid microbes have been replaced with amorphous silica and appear structureless even when viewed at magnifications . 50,000x. These microbes appear to have acted as nucleation centers for silica precipitation (Fig. 17B). Thus, the white laminae formed as the silica precipitated around neighboring microbes merged to form a solid, seemingly structureless mass of silica. CALCITE AND GYPSUM IN THE OHAAKI POOL SYSTEM Calcite has long been recognized in the subsurface precipitates of the Ohaaki–Broadlands area (e.g., Browne and Ellis 1970; Mahon and Finlayson 1972; Blattner 1975; Lonker et al. 1990; Simmons 1991; Simmons and Christenson 1994), but its presence in the sinter around Ohaaki Pool has MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 425 FIG. 11.—Thin-section photomicrographs of stratiform stromatolites from discharge apron around Ohaaki Pool. A) Alternating prostrate (P) and erect (E) laminae. B) Enlarged view of prostrate lamina showing numerous Calothrix filaments that lie (sub)parallel to lamina boundary. C) Transition from prostrate to erect laminae with individual filaments showing a change in attitude (arrow). D) General view of package of alternating erect (E) and prostrate (P) laminae. E) ‘‘Calothrix’’ filaments that have been encrusted with amorphous silica. received little attention (e.g., Tulloch 1982; Nicholson and Parker 1990). Aragonite has been precipitated from local borehole water (Browne 1973), and calcite is found in the columnar stromatolites that fringe Ohaaki Pool (Jones and Renaut 1996a; Jones et al. 1997a). At Ohaaki, calcite is common in the proximal sinter deposits but decreases in abundance with distance from the spring. Platy calcite crystals are found in the peripheral rim of Ohaaki Pool and the basal regions of the columnar stromatolites (Jones et al. 1997a) whereas skeletal and blocky calcite crystals are found inside the stromatolites (Jones and Renaut 1996a). In the proximal discharge apron, blocky and branching calcite crystals are intercalated with thin amorphous silica laminae (Fig. 18). Locally (e.g., Cores 1, 1A), calcite exceeds amorphous silica in abundance (Fig. 18). Calcite is also common in the oncoids from the main discharge channel (Fig. 5). Although calcite is generally absent from the microbialites in the 426 B. JONES ET AL. FIG. 12.—SEM photomicrographs of erect laminae in stratiform stromatolites that form much of the discharge apron around Ohaaki Pool. A) Erect laminae separated by thin prostrate laminae. B) Erect laminae formed of (sub)vertical pillars that have smaller-diameter filamentous microbes attached to their sides. C) Enlarged view of vertical pillars and narrow bridges that span the gaps between the pillars. D) Enlarged view of narrow bridges that span gaps between neighboring pillars. E) Long filamentous microbe (arrow) attached to sides of vertical pillars. F) Oblique view of erect laminae showing pillars with hollow cores. G) Upper parts of pillars shown in Figure 12F. H) Transverse section through large pillar showing concentric pattern that is defined by distribution of pores and coccoid microbes that are centered around the hollow filaments. I) Transverse section through a pillar showing hollow core, laminated sheath (replaced by beaded silica), and outer zone of encrusting amorphous silica. J) Sheath that has been replaced by small beads of amorphous silica that have a ‘‘rough’’ exterior. The thin, laterally discontinuous laminae are probably inherited from the original laminated sheath. K) Transverse section through pillar showing numerous coccoid microbes embedded in amorphous silica that encrusted filamentous microbes. L) Enlarged view of coccoid microbes shown in Figure 12K. discharge apron, it is present in some cavities in the porous, erect laminae of the stratiform stromatolites. The calcite precipitated upon rapid degassing of CO2 when the fluids reached the surface. Further CO2 loss resulted from agitation in the pool, proximal terraces, and outflow channel(s). Platy calcite is restricted to the hottest sites around the pool margins because it is precipitated from fluids that have undergone boiling and a rapid rise in pH (Simmons and Christenson 1994). Skeletal calcite, a high disequilibrium form (Jones and Renaut 1996a), is found in the proximal sinters where rapid CO2 degassing takes place. Blocky calcite probably formed at somewhat lower tempera- MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 427 FIG. 13.—Epiphytic microbes associated with pillars that form erect laminae in stratiform stromatolites. A) Small-diameter filamentous microbes on surface of pillar. B) Various filamentous microbes coating pillar surfaces. Note variation in diameter of these microbes. C) Small-diameter filament attached to surface of filament shown in Figure 13B (arrow in Figure 13B). D) Small-diameter filamentous microbes on outer surface of pillar. E) General view of Phormidium? that are found between pillars. F) Enlarged view of Phormidium?, showing constriction that indicates position of septa. Note small silica beads on filament surface. G) Filamentous microbe encased with silica. Transverse gaps denote positions of septa. H) Enlarged view of filament shown in Figure 13G. The gap that denotes the position of the septa does not cut the encrusting silica. I) Coccoid microbes encased with amorphous silica. tures around the pool margin and on the discharge apron. The decrease in calcite with distance from the pool may reflect the progressive downslope depletion of Ca 21 in the outflow waters. Minor amounts of gypsum are present in the oncoids. Although gypsum could result from evaporative concentration of low-volume pore waters, some gypsum is closely associated with calcite etching and dissolution (Fig. 7G), implying a possible genetic link. Oxidation of metallic (R) sulfides could account for local calcite dissolution and subsequent gypsum precipitation by reactions of the type 4RS2 1 15O2 1 8H2O 1 16CaCO3 5 2R2O3 1 16Ca 21 1 8SO4 22 1 16HCO32 Gypsum would then precipitate from the pore fluids. Although sulfides are known to have been precipitated episodically from the Ohaaki Pool discharge waters (Weissberg 1969; Goldie 1985), we did not find any in our cores. Gypsum was also found in microstromatolitic sinter at Champagne Pool, Waiotapu, where sulfide oxidation was also inferred to have taken place (Jones et al. 1997b). MORPHOSEDIMENTARY SETTING AND BIOFACIES The main morphosedimentary units in the Ohaaki Pool system are the (i) spring pool, (ii) outflow channel, and (iii) broad, low-gradient discharge apron (Figs. 1, 2, 19). Knowledge of the original ecological conditions is limited because the natural system is now inactive and historical records ignored the outflow system. The spring pool water was ejected at a temperature of 94–988C (Table 1), but the temperature and hydrodynamic conditions for the discharge zone can be inferred only from the sinter morphology and facies. Spring Pool.—The steep rim of Ohaaki Pool is dominated by columnar stromatolites (Fig. 3) that formed through the activities of filamentous microbes of uncertain affinity. Given their location, the stromatolites were probably not immersed continuously in the hot-spring waters during growth, but were bathed in steam that rose from the pool surface. Periodically their columns may have been inundated or splashed when the hot spring water rose in level and spilled across the pool rim onto the adjacent discharge apron (Jones et al. 1997a). Although sinter also lines the submerged pool floor (Hunt 1997), we could not sample it. Proximal Discharge Apron.—The discharge apron around Ohaaki Pool can be divided into proximal and distal zones by the distribution of different types of microbialites and calcite (Fig. 19). Oncoids are restricted to the proximal microterraces below sites of spillover and outflow channels where flow rates, turbulence, and CO2 degassing were significantly higher than on the main discharge apron. The narrow (, 5 m) proximal zone of 428 B. JONES ET AL. FIG. 14.—SEM photomicrographs of prostrate laminae in stratiform stromatolites that form most of discharge apron around Ohaaki Pool. A) Oblique view of prostrate filaments. B) Prostrate laminae formed of numerous closely spaced filaments. C) Almost solid prone laminae formed by extensive silica precipitation around the filaments. D) Transverse view through filamentous microbe showing sheath and encrusting silica. E) Oblique view of filamentous microbe in prostrate laminae. F) Enlarged view of sheath that has been replaced by small beads of amorphous silica. the discharge apron is characterized by thin silica laminae and a high calcite content (Fig. 18). Apart from scattered columnar stromatolites that resemble those around the edge of Ohaaki Pool, microbial activity seems to have been minimal in this high-temperature zone. Distal Discharge Apron.—The distal part of the discharge apron, which dominates the system, is characterized by shallow (, 6 cm), narrow (10– 50 cm), arcuate terraced pools (Fig. 2B, C). Their architecture is very similar to the ‘‘micro-terracettes’’ around Grand Prismatic Spring in Yellowstone National Park (e.g., Walter et al. 1996, fig. 17A) that formed through the activity of Calothrix under sheetflow conditions (Walter et al. 1996). Although spring water in the vent pool had a T of 94–988C, there is no record of temperatures across the discharge apron. Where the outflow took FIG. 15.—Silicified ‘‘streamer’’ filaments on surface of discharge apron, ; 100 m from Ohaaki Pool. place in channels, hot waters (. 508C) may have extended downstream tens of meters from the vent pool. Where the water flowed over the lip of the pool-rim dam as a broad unconfined sheet, however, the rate of temperature decline may have been much quicker given the shallower flow, the lower discharge velocity, and the higher surface area of the water exposed to atmospheric cooling. At times or in certain places, cooling was probably rapid because microbialites that formed around Calothrix, which typically live in temperatures of , 358C (Copeland 1936; Walter 1976a; Walter et al. 1996; Cady and Farmer 1996), are found 15 m from the edge of Ohaaki Pool. The distal parts of the discharge apron around Ohaaki Pool are dominated by stratiform stromatolites that have Calothrix as their dominant biotic component. The ‘‘Conophyton’’, which are dominated by Phormidium (?) and Synechococcus, are locally intercalated with these stromatolites. The limited distribution of the ‘‘Conophyton’’ indicates that they probably grew in response to minor environmental fluctuations that affected the composition of the microbial community. Temperature was probably responsible for the changing composition of the microbial community given their similar hydrodynamic settings. The Phormidium–Synechococcus association is found at higher temperatures than Calothrix at Yellowstone National Park (Walter et al. 1972; Cady and Farmer 1996). ‘‘Conophyton’’ does not indicate a particular environment because modern forms in Yellowstone National Park grow in a broad array of settings ranging from outflow channels to still, deep pools (cf. Walter et al. 1972, 1976). The coccoid microbial mats are irregularly distributed across the discharge apron. In some areas, individual mats can be traced between neighboring cores that are 10 m apart. The similarity in the biotic composition of the coccoid microbial mats and ‘‘Conophyton’’ may indicate growth on areas of the terrace where temperatures were slightly elevated above those with Calothrix. There is a strong correlation between the forms and fabrics of microbialites (biofacies) and their local environmental setting in the spring system (Fig. 19). This correlation, which is related to the ecological requirements of the organisms that constructed the microbialites, operates on large and small scales. On a large scale, stromatolite type may correlate with the MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 429 FIG. 16.—Thin-section photomicrographs of coccoid mats (‘‘white laminae’’) that are locally intercalated with stratiform stromatolites on discharge apron around Ohaaki Pool. A) White laminae sandwiched between prostrate and erect laminae in stratiform stromatolites. B, C) Enlarged views of white laminae showing ‘‘stromatolite-like’’ structures that are highlighted by darker zones. Note lack of obvious filamentous microbes. morphosedimentary unit (e.g., stratiform stromatolites on the medial to distal discharge apron). On a small scale, however, stromatolite type can be related to the various ecological niches that existed in any single morphosedimentary unit. The causal relationship in the latter case is more difficult to establish because the environmental controls responsible for the changes in the stromatolite types are not always readily apparent. For example, ‘‘Conophyton’’ and the coccoid microbial mats are highly localized in their distribution across the discharge apron. Although temperature may be the controlling factor, the reason for the change from stratiform stromatolites to ‘‘Conophyton’’ and coccoid microbial mats cannot be determined from the rocks. SILICA PRECIPITATION AND MICROBIAL SILICIFICATION Silica solubility increases with increasing temperature, pH, and pressure. Most silica precipitation at hot springs and geysers takes place when fluids that have equilibrated with respect to quartz, chalcedony, or volcanic glass at elevated temperatures in the subsurface rise rapidly to the surface where they attain high levels of supersaturation at lower temperatures and pressures. Silica then precipitates at the vent and on the discharge apron as sinter or laminated geyserite, usually as amorphous silica (Walter 1976b; Walter et al. 1996; Jones et al. 1997a). This is triggered by rapid cooling at ambient air temperature, evaporative concentration, or an abrupt change in pH of the spring waters (e.g., Weed 1889a, 1889b; Allen 1934; Krauskopf 1956; White et al. 1956; White et al. 1988; Rimstidt and Cole 1983; Fournier 1985). At high levels of supersaturation amorphous silica may precipitate from hydrothermal fluids abiotically, as shown by silica precipitation in the shallow, subsurface feeder conduits to hot springs and geysers. At the surface, precipitated silica may incorporate organic (mainly microbial) remains into siliceous sinter and geyserite. The role of the microbes in this process, however, is not always completely passive. Recent studies have shown that organic tissues in living and recently dead microbes commonly seed precipitation of sinter and geyserite by providing the initial sites for nucleation of amorphous silica (opal-A) (Ferris et al. 1986; Ferris et al. 1988; Schultze-Lam et al. 1995; Konhauser and Ferris 1996; Jones and Renaut 1996b; Jones et al. 1997a, 1997b). Microbes may act as templates for silica nucleation because their electronegative (hydroxyl and carboxyl) surfaces act as magnets for silica in solution as monosilicic or polysilicic acid. Having seeded silica precipitation, the microbial community may subsequently control the development of the sinter fabrics and facies. DISCUSSION Microbial filaments from New Zealand hot springs include trichomes containing cells that have been replaced by silica, and sheaths that have been replaced and externally encrusted by amorphous silica (Jones et al. 1997a, 1997b). The fact that these processes took place before the trichomes and sheaths had collapsed and desiccated indicates that silicification took place while the microbes were alive or soon after their death (Renaut et al. 1996; Jones et al. 1997a, 1997b). Bartley (1996) showed experimentally that filamentous cyanobacteria begin to degrade a few days after death. Thus, naturally preserved filaments that lack evidence of collapse or degradation indicate rapid silicification. This is viable because living filaments of Calothrix thermalis from Orakeikorako (15 km WNW of Ohaaki Pool) are partly encased with silica (Cassie and Cooper 1989, text-fig. 12) and Calothrix filaments at Yellowstone are commonly heavily silicified (Walter 1976a). Similarly, silicification of modern microbes in Icelandic hot springs begins while they are alive (Schultze-Lam et al. 1995). Clues to the timing of silica precipitation around the filamentous microbes on the discharge apron at Ohaaki Pool are provided by the heavily encrusted Calothrix filaments. Transverse sections through these filaments reveal concentric zones of epiphytic coccoid and filamentous microbes intercalated with zones of amorphous silica (Fig. 12H, K). This alternation shows that silica was not continuously precipitated around the Calothrix sheaths. When silica was not being precipitated, epiphytic microbes colonized the substrate and grew until the next phase of silica precipitation began. Breaks in silica precipitation may have been due to changes in water chemistry or its physical properties (e.g., temperature), or reflect changes in the patterns of outflow (e.g., local diversion). The sinter on the discharge apron around Ohaaki Pool is largely composed of stratiform stromatolites which have fabrics that are largely con- 430 B. JONES ET AL. FIG. 17.—SEM photomicrographs of coccoid mats (‘‘white laminae’’). A) Numerous coccoid microbes embedded in amorphous silica groundmass. B) Enlarged view of coccoid microbes embedded in amorphous silica. C) Chain of four-celled coccoid microbe. D) Synechococcus lividus (?) embedded in amorphous silica. E) White laminae with microlaminae highlighted by numerous ‘‘pores’’. F) Enlarged view of ‘‘porous’’ microlaminae shown in Figure 17E. These holes are molds of either filamentous or coccoid microbes. FIG. 18.—Thin-section photomicrographs of calcite crystals (C) and thin bands of amorphous silica (arrow) laminae from 1 cm to 5 cm in Core 1A (Fig. 1E). Note that some calcite crystals form pseudodendritic arrays (B). MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS 431 FIG. 19.—Schematic block diagram of the Ohaaki Pool system showing microbialitic fabrics that developed in different parts of the system. Abbreviations: O. Zone 5 outer zone; N 5 nucleus; P 5 pore. trolled by the type and attitude of the filamentous microbes around which amorphous silica was precipitated. In this respect, the alternating erect and prostrate laminae were important because they controlled the fabric of the microbialites. The alternation between prostrate and erect filaments has been attributed to the motility of phototactic microbes, which respond to light variations (Monty 1967, 1976). Monty (1976) showed that trichomes of Phormidium hendersonii glide upward by day to produce laminae, up to 900 mm thick, formed of erect bundles of filaments. At night, the filaments assume a prostrate attitude. Golubic and Focke (1978) confirmed this conclusion in their studies of modern marine stromatolites that are formed of Phormidium hendersonii. Not all erect laminae, however, result from daytime growth because some microbes (e.g., Chloroflexus) respond to low rather than high light levels (Monty 1976). At Yellowstone, filament orientation in stromatolites has been attributed to phototactic controls, with each couplet representing one day of growth (Walter et al. 1972, fig. 2B; Walter 1976a, fig. 5). These stromatolites, however, do not grow continuously, and only ; 70% of the summer days were represented. Even fewer pairs of laminae formed during the winter. Thus, a seasonal control is superimposed upon the diurnal growth pattern. Hinman and Lindstrom (1996) confirmed seasonal growth of sinter laminae at a Yellowstone hot spring. Alternating light and dark colored laminae in calcareous tufas and stromatolites have also been attributed to seasonal variations. Thus, light colored laminae were attributed to summer precipitation and darker laminae were attributed to winter growth (e.g., Irion and Müller 1968; Pentecost 1978; Pentecost and Riding 1986; Chafetz et al. 1991). The cause of the alternating erect and prostrate laminae in the stratiform stromatolites at Ohaaki Pool is difficult to ascertain because there are no independent means of assessing the time that it took for them to form. Furthermore, the thicknesses of the erect and prostrate laminae vary from area to area. Nevertheless, in some cores the patterns of laminae alternation may indicate seasonal and shorter-period changes. These features are best seen in Cores 5, 5A, and 5B, each 5 m from the other along a N–S transect that is ; 60 m from Ohaaki Pool (Figs. 1E, 20). In the 10–14 cm interval of Core 5A, there are two packages, each 1.6 cm thick, that are formed of thick erect laminae that are separated by thin (, 1 mm) prostrate laminae. At the base and top of each package are layers, ; 3.5 mm thick, formed of numerous prostrate laminae but few thin erect laminae. This variance in lamina packaging suggests that the thick packages, dominated by erect laminae, formed during the summer whereas the thinner packages, dominated by prostrate laminae, formed during the winter. The number of alternating prostrate and erect laminae in the ‘‘summer’’ packages, however, is less than the number of summer days. Thus, it is unlikely that the alternation between erect and prostrate laminae is related to diurnal changes in light levels. Transverse sections through the pillars in the erect laminae show that they developed through cycles of silica precipitation and epiphytic microbe colonization. Such cycles could not develop during a 24hour period. 432 B. JONES ET AL. In places, individual laminae and (or) packages of laminae can be traced between neighboring cores that are up to 10 m apart. Elsewhere, however, there is little similarity in the vertical sequence of laminae from neighboring cores. For example, packages of laminae in Cores 5A and 5B, 5 m apart, are easy to correlate (Fig. 20). Core 5, 5 m south of Core 5A, differs significantly in terms of its laminae packages, and correlation with Core 5A is virtually impossible (Fig. 20). Core 5 was originally in a different terrace pool than those from which Cores 5A and 5B were taken, so correlation is likely to be difficult. Outflow waters cross only parts of the apron at one time while adjacent areas remain dry (Fig. 19). During active sinter precipitation, the terraced surface of the discharge apron undergoes both vertical aggradation and outward progradation. During such phases, outflow waters may occasionally be diverted to lower areas, temporarily abandoning parts of the terrace surface. The fabrics in the stromatolites are a function of the morphology and growth patterns of the microbes, the volume of silica precipitated around them, and the hydrodynamic conditions that controlled silica delivery. The microbes, irrespective of their morphology, acted as templates for silica precipitation. Thus, the size, shape, and density of the microbes controlled the fabrics of the siliceous laminae. The densely crowded coccoid microbes in the coccoid mats, for example, produced dense, apparently structureless siliceous layers. These laminae formed through the merger of the amorphous silica cements that rapidly encrusted the microbes. In contrast, filaments in the erect laminae of the stratiform stromatolites have more open fabrics because the filaments are widely spaced. The different delivery mechanisms of silica to growth sites also controlled laminar fabrics. At sites of oscillation and splash around the pool margin, narrow columnar stromatolites grew where water preferentially precipitated silica on the tops of individual columns (cf. Jones et al. 1997a). Conversely, in areas with fast-flowing currents, horizontally aligned filaments (streamers) produced distinctive fabrics. In lower-energy sites of unchannelled outflow across the microterraces, stratiform stromatolites dominate. The results of this study show that sinters have several distinct microbial facies which can be related to specific locations in the spring system (Fig. 19). Following Walter (1976a), Cady and Farmer (1996) described sinter facies from thermal springs at Yellowstone National Park. Although every hot-spring system is unique, facies at Yellowstone and Ohaaki Pool bear many similarities. As the microbial communities change with temperature and hydrodynamic conditions, so the sinter fabrics change accordingly. For example, at both locations, Calothrix produces similar fabrics in the cooler (, 408C) distal discharge apron, whereas the Phormidium–Synechococcus association gives rise to ‘‘Conophyton’’ and other laminated fabrics in areas with higher temperatures. Unlike most Yellowstone hot springs, the stromatolitic fabrics at Ohaaki Pool include abundant calcite. This can be attributed to the higher CO2 concentration of the discharged water. Nonetheless, despite local differences, these and other hot springs are revealing common facies sequences that can be applied to ancient spring deposits (e.g., Walter et al. 1996). Most modern sinters are composed of amorphous silica (principally opalA). During diagenesis, this silica transforms to chalcedonic silica and quartz and some of the critical biological information and original microfabrics is destroyed. As yet, little is known of the effects of diagenesis on subaerial sinters. Although quartzose paleosinters preserve macrofabrics, plant remains, and some morphological details (Trewin 1994, 1996; Walter et al. 1996), the overall diagenetic effects of the transition of opal-A to quartz on the early silicified microbes still remain to be studied in natural spring → FIG. 20.—Cores OP5, OP5A, and OP5B taken 56 m downslope from fence that surrounds Ohaaki Pool. Note similarity in sequence of laminae in cores OP5 and OP5A. Core OP5B, taken 5 m north of Core OP5A, however, is characterized by a significantly different sequence of laminae that cannot be correlated with Cores OP5 and OP5A. Note thin bed of ‘‘Conophyton’’ in Core OP5B and thin coccoid mats near the tops of Cores OP5 and OP5A. MICROBIAL BIOFACIES IN NEW ZEALAND SINTERS systems. The evidence from Precambrian cherts, however, indicates than much of the original microbial detail and fabric may survive (cf. Knoll and Walter 1996). CONCLUSIONS The siliceous sinters around Ohaaki Pool are microbialites that formed by precipitation of amorphous silica around various types of microbes. Locally, calcite was precipitated in and around some of the microbialites. This study has yielded the following important conclusions: ● Columnar stromatolites, formed of opaline silica and calcite, grew around the edge of Ohaaki Pool under the influence of steam and hot splashing water. ● Oncoids formed of calcite and amorphous silica developed in the outflow channels under the influence of flowing water. ● The discharge apron is formed principally of stratiform stromatolites with localized areas of ‘‘Conophyton’’ and coccoid microbial mats. ● The fabrics in the sinter around Ohaaki Pool were controlled by the composition of the microbial community and the orientation of the filamentous microbes. ● The exceptional preservation of the filamentous and coccoid microbes can be attributed to the fact that silicification probably began while they were alive. Collectively, the evidence from Ohaaki Pool demonstrates that the microbial community exerted considerable influence over the fabrics of the sinters, primarily because they acted as templates for silica precipitation. ACKNOWLEDGMENTS This research was supported by grants from the Natural Sciences and Engineering Research Council of Canada to Jones (No. A6090) and Renaut (No. GP0000629), and the Central Research Fund, University of Alberta (to Jones). We are indebted to members of the Institute of Geological and Nuclear Sciences (IGNS), Wairakei, for their generous support, Contact Energy Ltd. for permission to examine the deposits at Ohaaki Pool, George Braybrook, who took most of the SEM photomicrographs, Duncan Graham (IGNS, Wairakei), who guided us around Ohaaki Pool, and Drs. K. Brown, H. Chafetz, and P. Wright, who critically evaluated an earlier version of this manuscript and Dr. J. Southard for his careful editing of the penultimate version of the manuscript. REFERENCES ALLEN, E.T., 1934, The agency of algae in the deposition of travertine and silica from thermal waters: American Journal of Science, v. 27, p. 373–389. BARTLEY, J.K., 1996, Actualistic taphonomy of cyanobacteria: implications for the Precambrian fossil record: PALAIOS, v. 11, p. 571–586. BLATTNER, P., 1975, Oxygen isotopic composition of fissure-grown quartz, adularia, and calcite from Broadlands Geothermal field, New Zealand: American Journal of Science, v. 275, p. 785–800. BOCK, G.R., AND GOODE, J.A., EDS., 1996, Evolution of Hydrothermal Ecosystems on Earth (and Mars?): Ciba Foundation Symposium No. 202, Chichester, U.K., Wiley. BROCK, T.D., 1969, Vertical zonation in hot spring algal mats: Phycologia, v. 8, p. 201–205. BROCK, T.D., 1978, Thermophilic Microorganisms and Life at High Temperatures: New York, Springer-Verlag, 465 p. BROWNE, P.R.L., 1970, Hydrothermal alteration as an aid in investigating geothermal fields: Geothermics, Special Issue 2, p. 564–570. BROWNE, P.R.L., 1973, Aragonite deposited from Broadlands geothermal drillhole water: New Zealand Journal of Geology and Geophysics, v. 16, p. 927–933. BROWNE, P.R.L., AND ELLIS, A., 1970, The Ohaaki–Broadlands hydrothermal area, New Zealand: mineralogy and related geochemistry: American Journal of Science, v. 269, p. 97– 131. BROWNE, P.R.L., AND LLOYD, E.F., 1987, Water dominated geothermal systems and associated mineralisation: International Volcanological Congress, 1987, Active Volcanoes and Geothermal Systems, Taupo Volcanic Zone: New Zealand Geological Survey, Record 22, p. 85–146. BURNE, R.V., AND MOORE, L.S., 1987, Microbialites—organosedimentary deposits of benthic microbial communities: PALAIOS, v. 2, p. 241–254. CADY, S.L., AND FARMER, J.D., 1996, Fossilization processes in siliceous thermal springs: trends in preservation along thermal gradients, in Bock, G.R., and Goode, J.A., eds., Evolution of Hydrothermal Ecosystems on Earth (and Mars?): Ciba Foundation Symposium No. 202, Chichester, U.K., Wiley, p. 150–173. 433 CASSIE, V., 1989, A taxonomic guide to thermally associated algae (excluding diatoms) in New Zealand: Bibliotheca Phycologica, v. 38, p. 161–255. CASSIE, V., AND COOPER, R.C., 1989, Algae of New Zealand thermal areas: Bibliotheca Phycologica, v. 78, p. 1–159. CASSIE-COOPER, V., 1991, Micro-algae of thermal areas, in Clarkson, D.D., Smale, M.C., and Ecroyd, C.E., eds., Botany of Rotorua: Rotorua, New Zealand, Forest Research Institute, p. 73–78. CHAFETZ, H.S., UTECH, N.M., AND FITZMAURICE, S.P., 1991, Differences in the d18O and d13C signatures of seasonal laminae comprising travertine stromatolites: Journal of Sedimentary Petrology, v. 61, p. 1015–1028. COPELAND, J.J., 1936, Yellowstone thermal Myxophyceae: New York Academy of Sciences, Annals, v. 36, p. 1–229. CUNNEEN, R., AND SILLITOE, R.H., 1989, Paleozoic hot spring sinter in the Drummond Basin, Queensland, Australia: Economic Geology, v. 84, p. 135–142. ELLIS, A.J., 1975, Geothermal systems and power development: American Scientist, v. 63, p. 510–521. FERRIS, F.G., BEVERIDGE, T.J., AND FYFE, W.S., 1986, Iron-silica crystallite nucleation by bacteria in geothermal sediment: Nature, v. 320, p. 609–611. FERRIS, F.G., FYFE, W.S., AND BEVERIDGE, T.J., 1988, Metallic ion binding by Bacillus subtilis: implications for the fossilization of microorganisms: Geology, v. 16, p. 149–152. FOGG, G.E., STEWART, W.D.P., FAY, P., AND WALSBY, A.E., 1973, The Blue-Green Algae: London, Academic Press, 459 p. FOURNIER, R.O., 1985, The behavior of silica in hydrothermal solutions, in Berger B.R., and Bethke, P.M., eds., Geology and Geochemistry of Epithermal Systems: Society of Economic Geologists, Reviews in Economic Geology, v. 2, p. 45–61. GARCIA-PICHEL, F., AND CASTENHOLZ, R.W., 1991, Characterization and biological implications of scytonemin, a cyanobacterial sheath pigment: Journal of Phycology, v. 27, p. 395–409. GARCIA-PICHEL, F., SHERRY, N.D., AND CASTENHOLZ, R.W., 1992, Evidence for an ultraviolet sunscreen role of the extracellular pigment scytonemin in the terrestrial cyanobacterium Chorogloeopsis sp.: Photochemistry and Photobiology, v. 56, p. 17–23. GLOVER, R.B., AND HEDENQUIST, J.W., 1989, A brief history of chemical exploration at Ohaaki– Broadlands, in Browne, P.R.L., and Nicholson, K., eds., 11th New Zealand Geothermal Workshop, Proceedings, p. 73–79. GLOVER, R.B., HUNT, T.M., AND SEVERNE, C.M., 1996, Ohaaki Ngawha; Ohaaki Pool: 18th New Zealand Geothermal Workshop, Proceedings, p. 77–84. GOLDIE, R., 1985, The sinters of the Ohaki and Champagne Pools, New Zealand: possible modern analogues of the Hemlo Gold deposit, Northern Ontario: Geoscience Canada, v. 12, p. 60–64. GOLUBIC, S., AND FOCKE, J.W., 1978, Phormidium hendersonii Howe: identity and significance of a modern stromatolite building microorganism: Journal of Sedimentary Petrology, v. 48, p. 751–764. GOLUBIC, S., AND HOFMANN, H.J., 1976, Comparison of Holocene and mid-Precambrian Entophysalidacae (Cyanophyta) in stromatolitic algal mats: cell division and degradation: Journal of Paleontology, v. 50, p. 1024–1082. GRANGE, L., 1937, The Geology of the Rotorua–Taupo Subdivision, Rotorua and Kaimanawa Divisions: New Zealand Department of Scientific and Industrial Research, Bulletin, v. 37, p. 86–105. GRINDLEY, G.W., 1970, Subsurface structure and relation to steam production in the Broadlands geothermal field., New Zealand: Geothermics, Special Issue 2, p. 248–261. HINMAN, N.W., AND LINDSTROM, R.F., 1996, Seasonal changes in silica deposition in hot spring systems: Chemical Geology, v. 132, p. 237–246. HUNT, T.M., 1997, Mitigating the impact of geothermal development: an example from New Zealand (in Japanese): Chishitsu News, No. 516, p. 43–49. IRION, G., AND MÜLLER, G., 1968, Mineralogy, petrology and chemical composition of some calcareous tufa from the Schwabische Alb, Germany, in Müller, G., and Friedman, G.M., eds., Recent Developments in Carbonate Sedimentology in Central Europe: New York, Springer-Verlag, p. 157–171. JONES, B., AND PEMBERTON, S.G., 1987, Experimental formation of spiky calcite through organically mediated dissolution: Journal of Sedimentary Petrology, v. 57, p. 687–694. JONES, B., AND RENAUT, R.W., 1994, Crystal fabrics and microbiota in large pisoliths from Laguna Pastos Grandes, Bolivia: Sedimentology, v. 41, p. 1171–1202. JONES, B., AND RENAUT, R.W., 1995, Noncrystallographic calcite dendrites from hot-spring deposits at Lake Bogoria, Kenya: Journal of Sedimentary Research, v. A65, p. 154–169. JONES, B., AND RENAUT, R.W., 1996a, Skeletal crystals of calcite and trona from hot-spring deposits in Kenya and New Zealand: Journal of Sedimentary Research, v. 66, p. 265–274. JONES, B., AND RENAUT, R.W., 1996b, Influence of thermophilic bacteria on calcite and silica precipitation in hot springs with water temperatures above 908C: evidence from Kenya and New Zealand: Canadian Journal of Earth Sciences, v. 33, p. 72–83. JONES, B., AND RENAUT, R.W., 1997, Formation of silica oncoids around geysers and hot springs at El Tatio, northern Chile: Sedimentology, v. 44, p. 287–304. JONES, B., RENAUT, R.W., AND ROSEN, M.R., 1997a, Biogenicity of silica precipitation around geysers and hot-spring vents, North Island, New Zealand: Journal of Sedimentary Research, v. 67, p. 88–104. JONES, B., RENAUT, R.W., AND ROSEN, M.R., 1997b, Vertical zonation of biota in microstromatolites associated with hot springs, North Island, New Zealand: PALAIOS, v. 12, p. 220– 236. KNOLL, A.H., AND WALTER, M.R., 1996, The limits of palaeontological knowledge: finding the gold among the dross, in Bock, G.R., and Goode, J.A., eds., Evolution of Hydrothermal Ecosystems on Earth (and Mars?): Ciba Foundation Symposium No. 202, Chichester, U.K., Wiley, p. 198–209. KNOLL, A.H., BARGHOORN, E.S., AND GOLUBIC, S., 1975, Paleopleurocapsa wopfnerii gen. et sp. 434 B. JONES ET AL. nov.: a late Precambrian alga and its modern counterparts: National Academy of Science [U.S.A.], Proceedings, v. 72, p. 2488–2492. KONHAUSER, K.O., AND FERRIS, F.G., 1996, Diversity of iron and silica precipitation by microbial mats in hydrothermal waters, Iceland: implications for Precambrian iron formations: Geology, v. 24, p. 323–326. KRAUSKOPF, K.B., 1956, Dissolution and precipitation of silica at low temperatures: Geochimica et Cosmochimica Acta, v. 10, p. 1–26. LONKER, S., FITZGERALD, J., HEDENQUIST, J., AND WALSHE, J., 1990, Mineral–fluid interactions in the Broadlands–Ohaaki Geothermal System, New Zealand: American Journal of Science, v. 290, p. 995–1068. LLOYD, E.F., 1957, Ohaki Hot Spring: New Zealand Department of Scientific and Industrial Research Geothermal Circular EFL2 (unpublished). MAHON, W., AND FINLAYSON, J., 1972, The chemistry of the Broadlands geothermal area, New Zealand: American Journal of Science, v. 272, p. 48–68. MEEKS, J., AND CASTENHOLZ, R.W., 1971, Growth and photosynthesis in an extreme thermophile, Synechococcus lividus (Cyanophyta): Archiv für Mikrobiologie, v. 78, p. 25–41. MONTY, C.L.V., 1967, Distribution and structure of recent stromatolitic algal mats, Eastern Andros Island, Bahamas: Société Géologique de Belgique, Annales, v. 96, p. 585–624. MONTY, C.L.V., 1976, The origin and development of cryptalgal fabrics, in Walter, M.R., ed., Stromatolites: Amsterdam, Elsevier, Developments in Sedimentology 20, p. 193–250. NASH, A., 1938, The cyanophyceae of the thermal regions of Yellowstone National Park, U.S.A., and of Rotorua and Whakarewarewa, New Zealand; with some ecological data [unpublished Ph.D. thesis]: University of Minnesota, 220 p. NICHOLSON, K., AND PARKER, R.J., 1990, Geothermal sinter chemistry: toward a diagnostic signature and a sinter geothermometer: 12th New Zealand Geothermal Workshop, Proceedings, p. 97–102. PENTECOST, A., 1978, Blue-green algae and freshwater carbonate deposits: Royal Society [London], Proceedings, v. 200, p. 43–61. PENTECOST, A., 1985, Association of cyanobacteria with tufa deposits: identity, enumeration, and nature of the sheath material revealed by histochemistry: Geomicrobiology Journal, v. 4, p. 285–209. PENTECOST, A., AND RIDING, R., 1986, Calcification in cyanobacteria, in Leadbeater, B.S.C., and Riding, R., eds., Biomineralization in Lower Plants and Animals: The Systematics Association, Special Volume 30, Oxford, U.K., Clarendon Press, p. 73–90. RENAUT, R.W., JONES, B., AND ROSEN, M.R., 1996, Primary silica oncoids from Orakeikorako hot springs, North Island, New Zealand: PALAIOS, v. 11, 446–458. RIDING, R., 1991, Calcified Cyanobacteria, in Riding, R., ed., Calcareous Algae and Stromatolites: Berlin, Springer-Verlag, p. 55–87. RIMSTIDT, J.D., AND COLE, R.R., 1983, Geothermal mineralization I: The mechanism of formation of the Beowawe, Nevada, siliceous sinter deposit: American Journal of Science, v. 283, p. 861–875. SCHULTZE-LAM, S., FERRIS, F.G., KONHAUSER, K.O., AND WIESE, R.G., 1995, In situ silicification of an Icelandic hot spring microbial mat: implications for microfossil formation: Canadian Journal of Earth Sciences, v. 32, p. 2021–2026. SIMMONS, S., 1991, Isotopic evidence for calcite formed from steam-heated waters at the Broadlands–Ohaaki, Wairakei and Waiotapu Geothermal Systems, New Zealand: 13th New Zealand Geothermal Workshop, Proceedings, p. 85–89. SIMMONS, S., AND CHRISTENSON, B., 1994, Origins of calcite in a boiling geothermal system: American Journal of Science, v. 294, p. 361–400. STANIER, R.Y., 1977, The position of cyanobacteria in the world of phototrophs: Carlsberg Research Communications, v. 42, p. 77–98. TREWIN, N.H., 1994, Depositional environment and preservation of biota in the Lower Devonian hot-springs at Rhynie, Aberdeenshire, Scotland: Royal Society of Edinburgh, Transactions, v. 84, p. 433–442. TREWIN, N.H., 1996, The Rhynie cherts: an early Devonian ecosystem preserved by hydrothermal activity, in Bock, G.R., and Goode, J.A., eds., Evolution of Hydrothermal Ecosystems on Earth (and Mars?): Ciba Foundation Symposium no. 202, Chichester, U.K., Wiley, p. 131–145. TULLOCH, A., 1982, Mineralogical observations on carbonate scaling in geothermal wells at Kawerau and Broadlands: Pacific Geothermal Conference, Incorporating the 4th New Zealand Geothermal Workshop, Proceedings, Part 1, p. 131–134. WALTER, M.R., 1976a, Hot-springs sediments in Yellowstone National Park, in Walter, M.R., ed., Stromatolites: Amsterdam, Elsevier, Developments in Sedimentology 20, p. 489–498. WALTER, M.R., 1976b, Geyserites of Yellowstone National Park: an example of abiogenic ‘‘stromatolites’’, in Walter, M.R. ed., Stromatolites: Amsterdam, Elsevier, Developments in Sedimentology 20, p. 87–112. WALTER, M.R., 1996, Ancient hydrothermal ecosystems on Earth: a new palaeobiological frontier, in Bock, G.R., and Goode, J.A., eds., Evolution of Hydrothermal Ecosystems on Earth (and Mars?): Ciba Foundation Symposium no. 202, Chichester, U.K., Wiley, p. 112–124. WALTER, M.R., BAULD, J., AND BROCK, J.D., 1972, Siliceous algal and bacterial stromatolites in hot spring and geyser effluents of Yellowstone National Park: Science, v. 178, p. 402–405. WALTER, M.R., BAULD, J., AND BOCK, T.D., 1976, Microbiology and morphogenesis of columnar stromatolites (Conophyton, Vacerrilla) from hot springs in Yellowstone National Park, in Walter, M.R., ed., Stromatolites: Amsterdam, Elsevier, Developments in Sedimentology 20, p. 273–310. WALTER, M.R., DESMARAIS, D., FARMER, J.D., AND HINMAN, N.W., 1996, Lithofacies and biofacies of mid-Paleozoic thermal spring deposits in the Drummond Basin, Queensland, Australia: PALAIOS, v. 11, p. 497–518. WEED, W.H., 1889a, Formation of travertine and siliceous sinter by the vegetation of hot springs, U.S. Geological Survey, 9th Annual Report, p. 613–676. WEED, W.H., 1889b, On the formation of siliceous sinter by the vegetation of thermal springs: Science, v. 37, p. 351–359. WEISSBERG, B., 1969, Gold–silver ore-grade precipitates from New Zealand thermal waters: Economic Geology, v. 64, p. 95–108. WHITE, D.E., BRANNOCK, W.W., AND MURATA, K.J., 1956, Silica in hot-spring waters: Geochimica et Cosmochimica Acta, v. 10, p. 27–59. WHITE, D.E., HUTCHINSON, R.A., AND KEITH, T.E.C., 1988, The geology and remarkable thermal activity of Norris Geyser Basin, Yellowstone National Park, Wyoming: U.S. Geological Survey, Professional Paper 1456, 84 p. WHITE, D.E., THOMPSON, G.A., AND SANDBERG, C.H., 1964, Rocks, structure, and geologic history of Steamboat Springs thermal area, Washoe County Nevada: U.S. Geological Survey, Professional Paper 458-B, 63 p. WHITE, N.C., WOOD, D.G., AND LEE, M.C., 1989, Epithermal sinters of Paleozoic age in north Queensland, Australia: Geology, v. 17, p. 718–722. WHITTON, B.A., AND POTTS, M., 1982, Cyanobacteria of the marine littoral, in Carr, N.G., and Whitton, B.A., eds., The Biology of Cyanobacteria: London, Blackwell, p. 512–542. ZHANG, Y., 1986, Thermophilic microorganisms in the hot springs of Tengchong Geothermal area, West Yunnan, China: Geothermics, v. 15, p. 347–358. Received 29 April 1997; accepted 20 December 1997.