Earth-Science Reviews 60 (2003) 175 – 194 www.elsevier.com/locate/earscirev

The occurrence and significance of biogenic opal in the regolith Jonathan Clarke * Department of Geology, CRC LEME, Australian National University, ACT 0200, Canberra, Australia Received 23 November 2001; accepted 28 February 2002

Abstract Biogenic opal produced by vascular plants, diatoms, and siliceous sponges have been found in soils and terrestrial sediments of all continents except Antarctica since the middle of the 19th century. The opal particles range in size from fine silt to fine sand. Almost all soils contain detectable opal up to levels of 2 – 3%, and a significant number contain values in excess of 5%. Even higher values have been found from soils and sediments of all continents in a wide range of soil types. The most important factor is poor soil drainage and seasonal to permanent water logging. This encourages the proliferation of silica producing organisms. Such conditions have been found in the soils and aquatic sediments of the monsoonal tropics, tropical rain forests, temperate forests, tropical savanna, tropical islands, semi-arid grasslands and savanna, and temperate woodland and grassland. The presence of a volcanic substrate also appears favourable in some cases, but is not necessary. Biogenic opal preferentially collects in the A horizon of soils and, to a lesser extent, in the B horizon. This preferential distribution facilitates identification of palaeosols in stacked soil sequences. Biogenic opal is also a component of windblown dust, even in arid environments. Biogenic opal is significant to regolith processes in a number of ways. Firstly, as in the case in marine environments, it is likely to be important in silica cycling and storage because of its greater lability compared to quartz. Secondly, dissolution and reprecipitation of opal A as opal CT or micro-quartz may play a role in cementation and silicification of regolith to form silica hardpans and silcrete. Thirdly, the organisms that form biogenic opal can have considerable palaeoenvironmental significance and be valuable in reconstructing regolith evolution. Finally, some forms of biogenic silica, in particular sponge spicules, can present a health hazard. Their high abundance in some soils and sediments needs to be considered when assessing the health implications of airborne dust. D 2002 Elsevier Science B.V. All rights reserved. Keywords: regolith; opal; phytolith; soil science; micropalaeontology

1. Introduction There is an increasing awareness of the role of biota in the regolith generally. Examples include the precipitation of iron and manganese (Skinner and Fitzpatrick,

* Fax: +61-2-6125-5544. E-mail address: [email protected] (J. Clarke).

1992) and interaction between bacteria and minerals (McIntosh and Groat, 1997). Part of this role is the deposition of biogenic silica in the form of opal. Examples include phytoliths, diatoms, and siliceous sponge spicules. Brewer et al. (1993) described phytoliths in Australian soils as ‘‘almost ubiquitous’’. In contrast, the same authors describe diatoms as ‘‘have been recorded’’ and sponge spicules as ‘‘rare’’ and ‘‘almost certainly inherited’’. Wilding et al. (1989)

0012-8252/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 0 1 2 - 8 2 5 2 ( 0 2 ) 0 0 0 9 2 - 2

176

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

regarded most soils as containing up to 3% biogenic opal and cited examples where up to 20% was present. These values mean that opal comprises a significant part of the silica component of many soils and can commonly exceed the abundances of potassium, calcium, magnesium, sodium and phosphate. However, biogenic silica is normally regarded as a minor component and the potential role of organisms in the large-scale deposition of silica in the regolith is largely ignored. Despite this, there is considerable evidence that opal from phytoliths, diatoms, sponge spicules, and other organisms are often abundant in the regolith in many localities. Widespread and common accumulations >2% by grain abundance are known from every continent except Antarctica in environments as diverse as coastal and inland swamps, forests, grasslands, and flood plains. They are present in the tropical, temperate, and semi arid regions. Assessing the actual abundance of biogenic silica in regolith materials is difficult. Most studies concentrate on one component, such as phytoliths, and do not discuss the presence of other forms. Some authors (e.g., Brewer, 1955) give actual percentages (Table 1),

others list proportions (Table 2) (e.g., Schwandes and Collins, 1994) in opal grains per gram, still others (e.g., Hart, 1992) as percentage of a particular size fraction, such as silt, or (Clark et al., 1992a) as cm2 cm3. None of the works, to date, appears to have used the various chemical techniques used in the study of biogenic opal in marine sediments (see Muller and Schneider, 1993 and references therein). This paper reviews occurrences of biogenic opal in the regolith and their potential importance to regolith processes. Far from being of minor importance, I argue that biogenic opal is a key constituent of the regolith.

2. Status of terrestrial biogenic opal research For more than 150 years, biogenic silica has been known to be a component of the regolith. The earliest mention of biogenic opal in soils was in 1841 by the German microscopist Ehrenherg (Piperno, 1988). A few years later, Gregory (1855) reported spicules and diatoms as well as phytoliths in soils. Early studies on spicules were listed by Smithson (1959), while Oehler

Table 1 Abundances of opaline material in selected soils (representative only) % Opal

Forms

Setting

Locality

Source

19 3.2

Spicules Phytoliths, spicules, diatoms Phytoliths, diatoms

Flooded forest Coastal swamps

Central Amazonia Dalmore, Victoria, Australia Mt. Gellibrand, Victoria, Australia Duntroon, ACT, Australia Deep Creek, NSW, Australia Doughboy Creek, NSW, Australia Reunion Island

Chauvel et al. (1996) Baker (1959b)

2.1 24

Forested hills

43

Spicules, diatoms, phytoliths Spicules, diatoms, phytoliths Spicules, diatoms, phytoliths Phytoliths

0.9

Phytoliths

Bamboo forested slope Forested plain

4.5

Phytoliths

Swamp

20 100 30 – 60 48

Phytoliths Phytoliths Phytoliths Phytoliths, spicules, diatoms Phytoliths, spicules, diatoms

Grassland Basalt flow Volcanic slopes Seasonal wetlands Seasonal wetlands

11 2

100

Alluvial plain Alluvial plain Alluvial plain

Amazonia Northern Sydney, Australia Oregon, USA East Africa Japan Magela Creek, NT, Australia Magela Creek, NT, Australia

Baker (1959b) Brewer (1955) Brewer (1955) Brewer (1955) Meunier et al. (1999) Kondo and Iwasa (1981) Hart (1992) Hart (1992) Hart (1992) Hart (1992) Clark et al. (1992b) Clark et al. (1992b)

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

177

Table 2 Other selected abundances of opaline material in soils (non-compatible units) Amount

Units

Forms

Setting

Locality

Source

0.5 – 6

Phytoliths

Runge (1999)

D. R. Congo

Runge (1999)

Central African Republic

Runge (1999)

Central African Republic

Runge (1999)

Spicules

Tropical forest soils Tropical forest soils Tropical forest soils Tropical forest soils Histsol

D. R. Congo

500,000

% silt fraction % sand fraction % silt fraction % sand fraction Grains/gram

Florida

4600

Grains/gram

Spicules

Entisol

Florida

48,000

Grains/gram

Spicules

Spodsol

Florida

9500

Grains/gram

Spicules

Mollisol

Florida

23,000

Grains/gram

Spicules

Ultisol

Florida

f 10,000

Grains cm2 cm3

Phytoliths, spicules, diatoms

Black soils

Magela Creek, Northern Territory

Schwandes and Collins (1994) Schwandes and Collins (1994) Schwandes and Collins (1994) Schwandes and Collins (1994) Schwandes and Collins (1994) Clark et al. (1992a)

>0.1 – 2 14 1.65

Phytoliths Phytoliths Phytoliths

(1979) provided a review of the significance, origin, and biogeochemical cycling of silica in the terrestrial environment. Much has been documented on the taxonomy of biogenic opal in soils, and some detailed palaeoenvironmental and palaeoecological studies have been carried, most on local scales (e.g., Gasse, 1987; Volkmer-Ribeiro, 1992). However, much of this work has been fragmentary and focused on specific applications, such as palaeoenvironmental and archaeological analysis. There has been little integration of data from soil science, palaeoenvironmental, palaeontological, sedimentological, and regolith studies.

3. Sources of biogenic silica in soils 3.1. Plant opal Many groups of plants produce silt-sized opal grains known as phytoliths (Fig. 1A). These form structural elements and provide a defense against herbivores (Lewin and Reimann, 1969). They are extremely abundant in grasses, which can contain up to 10% silica by weight (Lovering, 1959) and horsetails (Equisetum) can contain up to 16% dry weight of biogenic opal (Lewin and Reimann, 1969). However, phytoliths also occur in a wide range of other plant

groups (see Piperno, 1988) and it would be a mistake to assume that they are abundant only in grasses and horsetails. Phytoliths are found in most soils, including those from temperate forests of North America (Wilding and Drees, 1971), the savanna landscapes of Kenya (Runge, 1999) and tropical forests of Amazonia (Kondo and Iwasa, 1981) and Africa (Runge, 1999). They locally reach abundances of 43% in the B-horizon of some soils on Reunion Island (Meunier et al., 1999). They offer considerable potential for palaeoenvironmental reconstruction, owing to different morphologies derived from different plant communities (Runge, 1999; Wilding and Drees, 1971). Phytoliths are the most common form of biogenic silica in most terrestrial environments. A much rarer occurrence is tabashir, massive bodies of opal found in bamboo (Jones et al., 1966). Phytoliths may be confused with sponge spicules, especially in older literature (see review in Hart and Humphreys, 1997). Both have a rod- or needle-shaped morphologies, however sponge spicules typically have a central canal, absent in phytoliths. Some spicules are highly complex and irregular in shape, especially those of the lithistid sponges, common in Eocene marine sediments in southern Australia, these, along with gemmule bodies and microscleres lack a central canal, and may be confused with phytoliths.

178

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Fig. 1. Representative examples of opal secreting organisms showing the immense range in morphology. (A) Phytoliths (after Baker, 1959a). (B) Marine and freshwater diatoms (after Braiser, 1980). (C) Marine and freshwater sponge spicules (after de Laubenfels, 1955). (D) Silicoflagellates, (E) chrysomonads, (F) radiolaria, and (G) ebridians (all after Braiser, 1980). (H) Helizoans (after Moore, 1964).

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Fig. 1 (continued).

179

180

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Fig. 1 (continued).

3.2. Diatoms

3.3. Sponges

Diatoms (Fig. 1B) are found in lakes, rivers, and soils. They also occur subaerially on plant and rock surfaces in moist environments (Patrick, 1977). Diatoms comprise the bulk of the 24% of siliceous remains reported by Brewer (1955) from the A2 horizons of soils in Canberra, Australia. Comparatively pure accumulations of diatoms (diatomites) are known only from lake and swamp basins, however such sediments may form part of the regolith, especially where they occur in stable continental environments. Modern Australian examples include the Holocene diatomites of the Swan coast plain of Western Australia (Gibson, 1976) and the Neogene diatomites of Victoria (Kenley, 1976) which are intimately associated with lakes and sediments formed by Quaternary basaltic volcanism and associated drainage diversion. Miocene Bonnie Doon Diatomite from New South Wales is not, however, associated with volcanism (Taylor et al., 1990). All these deposits contain varying amounts of clastic material and sponge spicules in addition to the diatom frustrules. Secondary silica mobilisation is common in the Victorian deposits, as they have been exposed to percolating groundwater, forming opaline bands. Diatoms are known from all continents. The expansion and contraction of large lakes in low relief continental environments, such as the modern lake Chad, can result in deposition of diatomaceous sediments over large areas and then expose them to pedogenic processes (Thiry, 1999).

Siliceous sponges (Fig. 1C) are normally only a minor component of marine and terrestrial ecosystems. Their presence in sediments and soils (Smithson, 1959) are similarly normally minor. Under conditions not fully understood they can, however, proliferate and even dominate. Late Eocene spicular marine and marginal marine sediments, locally with up to 100% sponge spicules (Clarke, 1994a), form the land surface over large areas of southern Australia. The reasons for this proliferation during a narrow time interval are not fully known, but probably related to a unique confluence of runoff, turbidity, and nutrient conditions (Gammon et al., 2000). Freshwater sponges produce much smaller spicules than marine sponges, typically silt rather than sandsized. Thus, they are often missed by those looking only at the larger fraction. They are known not only from lakes (Harrison, 1988) and rivers (Chauvel et al., 1996) but also bogs (Volkmer-Ribeiro, 1992) and waterlogged soils (Schwandes and Collins, 1994). They have been reported as composing more than 20% of the silt fraction of the soil in parts of Amazonia (Chauvel et al., 1996) and in numbers of more than one million spicules to the gram (Schwandes and Collins, 1994). They have been found to date in the soils of all continents except Antarctica. Even though the presence of such spicules has been taken to indicate high levels of dissolved silica (see Turner, 1985), this need not be the case. Spicule-rich

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

sediments were reported from rivers in Amazonia by Chauvel et al. (1996) that have extremely low dissolved silica levels of 2.1 ppb. In comparison, Aston (1983) gives an average global silicon value for river water of 13.1 ppm, while Wollast and Mackenzie (1983) say 5.42 ppm. 3.4. Other organisms (Figs. 1 D –G) Many groups of regolith bacteria are known to dissolve silicates (see review of Silverman, 1979) through enzyme and organic acid secretion. Some, such as Proteus mirabilis, also store the silica within their cells and in slime layers (Tesic and Todorovic, 1958; Lauwers and Heinen, 1974) as monomeric silica. The fate of such silica in the regolith is not known but it may provide both a source of silica for uptake by higher plants and, potentially, as a means of silica cementation should the silica in the slime layers and dead bacterial cells between regolith grains undergo polymerisation. The abundance of bacterial remains in hydrothermal and marine chert deposits suggests that bacteria do play a role in silica deposition (Ferris, 1997). Shaw et al. (1990) reported that desiccation of formerly floating colonies of the filamentous cyanobacteria Chloriflexus provided a template for silica deposition in a silica-depositing alkaline saline pan in Botswana. Widespread silicification of bacterial cells has been reported (Ferris, 1997) and strongly supports a bacterial role in the deposition of silica in the regolith. Radiolaria are significant siliceous organisms in the pelagic realm (Braiser, 1980). Not found in terrestrial aquatic environments, they are likewise rare in the coastal sediments likely to be incorporated into the terrestrial regolith. However, the Australian regolith includes epicontinental sediments of Cretaceous age, and these can include radiolarian-rich deposits. An example is the radiolarian-bearing Darwin Member of the Bathurst Island Formation of the Northern Territory (Nott, 1994) which may be the source of silica in siliceous bands in the weathering profile. Helizoans are Protozoa with a siliceous test similar to radiolaria but restricted to freshwater environments. They are rarely preserved because their fragility normally results in the rapid disintegration of the test after death (Braiser, 1980). Heliozoans however have occasionally been reported from Quaternary sedi-

181

ments and are thought to indicate swampy to lacustrine environments (Piperno, 1988; Moore, 1964). Another group of Protozoa are the testate amobae known as rhizopods (Charman, 2000). While some rhizopods form tests (xenosomic tests) of cemented grains from the local environment, others have very weak siliceous tests. These are known as idiosomic tests and are formed by the parent amobae during reproduction. Rhizopods are found in lakes and peatlands and in recent research has demonstrated their potential for palaeoecological studies. Their larger scale significance to silica cycling is unclear, as they are only weakly mineralised. Silicoflagellates are also important silica depositing organisms in pelagic marine environments. They are not, however known from terrestrial environments, nor are they, unlike diatoms and sponges, known at present to be locally common enough in coastal sediments to become significant components of the regolith as a result of sea level changes (Braiser, 1980). Whether they, as radiolaria do, occur in significant numbers in marine epicontinental deposits such as the Cretaceous of the Australian regolith, is not known. Chrysomonads are mostly non-marine Chrysophyte algae related to silicoflagellates (Braiser, 1980). They appear to be of minor importance in siliceous nonmarine sediments. Ebridians are a small and comparatively minor group of siliceous marine plankton allied to dinoflagellates (Braiser, 1980) and are not presently known to form significant accumulations on their own. They are common in some diatomite deposits (Bohaty and Harwood, 2000). Some Ebridians produce external siliceous scales or spines while all produce siliceous resting cysts termed statosphores (Smol, 1987).

4. Spatial distribution in the regolith 4.1. Opal in the landscape No detailed study has been carried on the spatial distribution of biogenic opal in soils. However, following Hart’s (1992) analysis of the data of Stace et al. (1968), detectable opal is common in many Australian soils profiles. Stace et al. (1968) compiled 147 representative soil profiles from 43 soil groups. Micromorphological data was provided for 85 of

182

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Table 3 Abundance of biogenic opal in Australian soil profiles. After Stace et al. (1968) Abundance

Number of profiles

Percentage all profiles

Soil groups

Fraction of profiles in each soil group

Frequent (>5%)

12

14.1

Common (2 – 5%)

14

16.5

Occasional (0.5 – 2%)

24

27.1

Rare ( < 0.5% but easily seen under microscope)

10

12.9

Very rare (hard to find under microscope)

21

24.7

Grey, brown and red clays Solodized solonetz and sodic soils Soloths Krasnozems Red podzolic Yellow podzolic Humic gleys Grey, brown and red clays Prairie soils Soloths Red-brown earths Non-calcic brown soils Chocolate soils Yellow earths Yellow podzolic Gleyed podzolic Podzols Humic gleys Siliceous sands Earthy sands Desert loams Grey, brown and red clays Chernozerms Prairie soils Solodized solonetz and sodic soils Solonized brown soils Red-brown earths Chocolate soils Calcareous red earths Red earths Terra rossa soils Brown podzolic Lateritic podzolic Humic gleys Grey, brown and red clays Solonized brown soils Brown earths Calcareous red earths Red earths Euchrozems Yellow podzolic Lateritic podzolic Humic gley Grey, brown and red calcareous soils Desert loams Grey, brown and red clays Black earth Solodized solonetz and sodic soils Solonized brown soils Calcareous red earths Red earths Xanthrozems

1/13 2/4 4/5 1/2 1/1 1/3 3/4 2/13 1/4 2/5 2/5 1/1 1/2 1/1 1/3 1/1 1/1 1/4 1/2 1/1 1/5 4/13 1/2 2/3 1/4 1/6 3/5 1/2 2/6 1/4 1/1 1/1 2/4 1/5 1/13 1/6 1/1 1/6 1/4 1/1 1/3 2/4 1/5 2/2 4/5 5/13 1/1 1/2 2/6 1/6 2/4 1/1

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

183

Table 3 (continued) Abundance

Absent

Number of profiles

5

Percentage all profiles

4.7

Undescribed

profiles. Opal was detected in 81 of these (Table 3) representing over 95% of soils studied. Just over 30% (26 examples) contained more than 2% opal, and just over 56% (47 examples) contained more than 0.5%. Abundant (>2%) opal was found in soils from western Australia, Victoria, New South Wales, Queensland, South Australia, and the Australian Capital Territory. All but three of these soils (88.5%) were characterised by slow or impeded, drainage. In contrast, poor drainage was a feature of only 41.6% profiles containing occasional opal grains. Profiles with rare and very rare grains of biogenic opal were poorly drained in only 30% and 33% of cases, respectively. These results indicate that there is a strong correlation between the occurrence of high levels of biogenic opal in soils with poor drainage. Stace et al. (1968) performed no soil petrography on alluvial or organic-rich soils. Data on occurrences of exceptionally high levels of biogenic opal indicate that they often occur in such soils also. Examples include those of the flood plains (Chauvel et al., 1996) and bogs (Volkmer-Ribeiro, 1992) of Brazil, and the Magela Creek flood plain of the Northern Territory of Australia (Clark et al., 1992a,b) (Fig. 2). As post-organicrich soils, such as histosols are poorly drained, the correlation with poor drainage and water logging,

Soil groups

Fraction of profiles in each soil group

Krasnozems Alpine humus soils Siliceous sands Solonized brown soils Calcareous red earths Solonchalks Alluvial soils Lithosols Red-brown hardpan soils Redzinas Wiesenboden Calcareous sands Grey-brown podzolic Humus podzols Peaty podzols Neutral to alkaline peats Acid peats

1/2 1/1 1/1 2/6 2/4

noted by Brewer (1955), is thus probably much stronger than these data indicate. 4.2. Opal in soil profiles There have been many studies noting the presence of biogenic opal in soils, and the following references provide an outline only. Biogenic opal is normally preferentially concentrated in the A-horizon of individual soil profiles (Oehler, 1979). It is also present less commonly in the B-horizon, and sometimes both (Simons et al., 2000). Accumulation in the B-horizon is typically due to downward movement resulting from bioturbation and percolating soil water (Hart and Humphreys, 1997; Boettinger, 1994; Piperno, 1988). In some cases the biogenic opal may be concentrated entirely in the B-horizon (see Meunier et al., 1999; Schwandes and Collins, 1994), but this is exceptional. Bobrova and Bobrov (1997) and Bobrov and Bobrova (2001) reported the concentration of phytoliths in illuvial and eluvial horizons. Because phytoliths are commonly perceived to be concentrated in the A horizon of soils, increases in their abundance within a profile have sometimes been used as indicators of palaeosols in a succession (see Dormaar and Lutwick, 1969). Care should be taken in such interpretations, because of the downward move-

184

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Fig. 2. An opal factory (1): wetlands of Kakadu National Park, Australia contain up to 100% opal. Open water billabongs (A) and seasonal grass and sedge wetlands (B) have the highest opal productivity. Paperbark wetlands (C) are less productive. Biogenic opal is visible as pale streaks in the black soils of the seasonal wetlands (D). All photographs courtesy of R. Wasson.

ment in such circumstances noted above by Simons et al. 2000) and should only be carried out with good soil stratigraphic control.

5. Composition and chemistry Biogenic silica is formed mainly as amorphous opaline silica (opal A), but plant opal is also known to contain opal CT (Wilding and Drees, 1974). Diagenesis of the opal in soils results in further production of opal CT and eventually its stablisation as quartz (Wilding et al., 1989). Studies of the subsequent precipitation of dissolved silica from biogenic opal are rare, but amorphous opal, opal CT and quartz are all possible secondary phases, with crystalline forms becoming more likely with time.

Plant opal can contain significant amounts of Al, Fe, Ti, Mn, P, Cu, N, and C (Wilding et al., 1989). The Al is known to play a key role in the surface chemistry of plant opal, influencing dissolution and interaction with organic acids (Bartoli, 1985). Much or all of the Al is chemisorbed on the surface, rendering the plant opal as reactive in the soil as iron and aluminium hydroxides and allophanes. Nitrogen, phosphorous, and carbon may be the result of inclusion of lignin and cellulose during formation of the phytoliths. Trace element composition of other forms of biogenic opal, such as spicules and diatoms, is not known. Biogenic opal also shows high delta O18 values. Diatoms typically show approximate values of + 29 to + 32 relative to SMOW, and phytoliths values of + 36 to + 39 (Wilding et al., 1989). This strongly suggests that the isotopic value of regolith opal, at

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

least in its particulate or dispersed form, can be used to demonstrate biological origin. Webb and Longstaffe (1997) argued that progressive depletion reflected increasing values of evapotranspiration. Bombin and Muehlenbachs (1980) earlier noted that the oxygen isotope values varied according to both temperature and humidity. This makes the use of these isotopes as direct palaeoclimatic indicators problematic, although they certainly appear to have potential as evapotranspiration indicators. It is not known whether these signatures persist through diagenesis of the opal. Webb and Longstaffe also reported extreme depletion in deuterium ( 125), the reasons for which were not known at the date of publication. This contrasts with the study of Fredlund (1993), who found deuterium values consist with that of the waters that supported plant growth, allowing for the usual levels of biological fractionation. Carbon isotope ratios of phytoliths were reported by Fredlund as reflecting the C3/C4 ratio within the source vegetation. As is the case with oxygen isotopes, it is not yet known whether these hydrogen and carbon survive diagenesis. As noted by Oehler (1979), non-crystalline forms of silica are much more soluble than crystalline forms. Wilding et al. (1989) said that amorphous silica was more soluble than quartz by a factor of 10 or more. Biogenic opal showed a great range in solubility. Generally, those remains containing less organic carbon were more soluble than those that still contained significant organic matter. In addition, some types of biogenic opal are more soluble than others. Wilding and Drees (1974) found that forest opal was 10 – 15 times more soluble than grass opal, owing to its greater surface area. In sediments, biogenic opal is more labile than quartz. Kosters and Bailey (1983) identified sponge spicules as among the most chemically mobile silica sources in sediments from the Mississippi Delta. Once silica has been dissolved from biogenic opal grains, it can then be re-precipitated lower in the soil profile. Alexandre et al. (1997) described incipient cementation of soil particles in the lower part of a soil profile by opal remobilised from phytoliths and other organisms. Dissolution and cementation of silica in soils can occur quite rapidly under favourable conditions, as shown by Breese (1960) in studies of aeolian dust.

185

6. Biogeochemical cycling of silica in the regolith Organisms play a vital role in both the dissolution (Silverman, 1979) and deposition (Oehler, 1979) of silica in the regolith. Ambivalent results were reported in early studies cited by Jones and Handreck (1967) on the effect of growing plants in silicon-free substrates. Some studies showed no ill effect whereas others did. One of the main functions of biogenic opal in plants is in improving pest resistance, thus under field conditions improved silicon availability to plants is desirable. A more recent work cited in Epstein (1999) shows that silicon-deprived plants are structurally weaker, have various abnormalities, and are more susceptible to biotic and abiotic stress. Results from experimental studies in silicon-free media replicate studies of plant pathologies found in silicon-poor soils, especially lateritic and bauxitic soils of tropical regions. In Epstein’s words, ‘‘the evidence is overwhelming that silicon should be included among the elements having a major bearing on plant life (Epstein, 1999). Because of its greater chemical mobility than crystalline silica phases, biogenic opal plays a major role in the cycling of silica in soils (Alexandre et al., 1997), just as it does in aquatic environments (Konhauser et al., 1992). As pointed out by Heinen and Oehler (1979), the cycling of silica through the biosphere has evolved through time. Silica bacteria may have been present since the Precambrian. Marine sponges are known since just before the end of the Neoproterozoic (Braiser et al., 1997) and have formed strandline accumulations since the Carboniferous (Carlson, 1994). Sponges have been reported from freshwater environments of at least Carboniferous age (De Laubenfels, 1955) and probably form a significant source of silica in coals throughout geological history (Davis et al., 1984). Equisetum, the single modern representative of the Equisetales, a dominant component of land vegetation from the Devonian through to Permian, contains abundant silica (Lewin and Reimann, 1969). If these values of silica content reflect a characteristic of the taxon as a whole, evolution of these plants would have significantly increased the rates of silica cycling in the regolith. Rates of silica cycling would have increased still further with the evolution of grasses and their ecological dominance from the Oligocene onwards (Braiser, 1980). Although diatoms may have appeared in the

186

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Jurassic, they are common in terrestrial environments only from the Miocene (Braiser, 1980). Thus, the pattern rate of silica cycling and the potential for accumulations of biogenic silica in the regolith has increased markedly with time. Major increases would have occurred in the Mid to late Palaeozoic and near to the Paleogene – Neogene boundary. Alexandre et al. (1997) reported that 92% of the biogenic silica in the soil is recycled by plants and is the main source of this nutrient. Only the remaining 8% accumulates in the soil. Konhauser et al. (1992), in a study of the seasonally flooded forests of the Amazon basin reported that diatoms played a key role in silica exchange between the dissolved and precipitated state, with the precipitated silica being stored as gel coatings on wood. The significance of sponges in silica cycling is not known, but the abundance of siliceous spicules in some soils in the flooded forests of the Amazon (19%) indicates that they, along with diatoms, are likely to be significant in at least some water logged environments.

7. Application to understanding and study of regolith processes 7.1. Setting of exceptionally opal-rich environments Although small values of opal are fairly ubiquitous in many soils and sediments significant quantities (e.g., >5%) occur less commonly. Apart from the unusual stranded Eocene sediments of southern Australia, examples are non-marine. Examples reviewed in this paper include the Magela Creek flood plain, (Northern Territory, Australia; Clark et al., 1992a,b), Amazonia (Chauvel et al., 1996), Bungendore (NSW) and Duntroon (ACT; both Brewer, 1955) (Fig. 3). The Okavango Delta has abundant dissolution and precipitation of silica (Shaw et al., 1990; Shaw and Nash, 1998), and high levels of opal productivity. Peat formed from grasses and sedges contain up to 30% phytoliths, along with minor diatoms and sponge spicules (McCarthy et al., 1989). These opal-depositing environments differ in many respects. With respect to vegetation, Amazonia consists of flooded tropical rainforest, Bungendore and Duntroon formerly temperate eucalypt woodland,

Magela Creek monsoon grassland, woodland, and wetland, and the Okavango Delta seasonally to permanently flooded arid wetland. With respect to water chemistry, Magela Creek and Amazonia waters are acidic, the Okavango region strongly alkaline, and the NSW and ACT examples near neutral. The Okavango waters are also comparatively saline whereas the others are dilute. The common feature in all of these environments is seasonal to permanent flooding. Under such conditions the diatoms and sponge bloom in sufficient quantities so as to make a significant component of the sediment. The sponges and diatoms may be epiphytic (Chauvel et al., 1996; Konhauser et al., 1992; Clark et al., 1992b), or planktonic (VolkmerRibeiro, 1992). Furthermore, in many wetlands the plant taxa include those with high phytolith production, such as grasses and sedges (Clark et al., 1992a,b). This association of poor drainage and high levels of opal reinforces the conclusions drawn above from a review of soil micromorphology in Stace et al. (1968) discussed above. Under ideal conditions, these environments can produce sediments as rich in biogenic silica as those of the marine realm. Magela Creek forms part of the East Alligator River drainage system in the Northern Territory of Australia. The black alluvial soils, typically between 30 cm and 2 m thick, can average up to 35% biogenic opal over lateral distances of several km and up to 48% in a single soil core. The mineral fraction of individual beds within the black soils can be composed entirely of biogenic opal (Clark et al., 1992a,b). Other unusually opal-rich environments are found on volcanic soils in Africa, Japan (see references in Hart, 1992) and Reunion (Meunier et al., 1999). The high levels of opal in these soils may reflect unusually rapid release and uptake of silica into plants as a result of the fast weathering of soils derived from the weathering of volcanic debris and rocks, especially those low in silica. Although high rainfall encourages water logging and thus potentially high productivity of opal, it is not necessary for it. Water logged soils and aquatic environments occur in arid and semi arid environments, provided there is sufficient water. More critical is salinity, but provided it is not excessive, these environments are potentially highly productive, even in dry

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

187

Fig. 3. An opal factory (2): soils of the Canberra region contain significant amounts of opal. (A) Podzolic soils in gently undulating landscapes north of Duntroon, Canberra, Australia contain up to 22% biogenic opal. (B) Alluvial soils at Barrack Flat, Queanbeyan, contain up to 7% opal (Angela Harrison, unpublished data). (C) Soils at Deep Creek, east of Bungendore, ACT, Australia reportedly contain 11% biogenic opal (D). Some 2% biogenic opal has been reported from Doughboy Creek, east of Bungendore.

environments. Examples discussed in this paper include Lake Chad and the Okavango Delta. 7.2. Dust Most dust contains a small proportion of opal, this being almost entirely of biogenic origin. Australian windblown dusts (Baker, 1960) contain phytoliths, sponge spicules, and diatoms. Drees et al. (1993) reported that the main type of biogenic opal in dusts from Niger is sponge spicules. These dusts have been deflated from the Sahara or Sahel regions of Africa, not the most obvious habitat for freshwater sponges. Jones and Beavers (1963) and Wilding and Drees (1968) used the presence of sponge spicules in soils from ridge tops to indicate contributions from windblown materials. Given the ubiquity of spicules and

spicule-like structures in many soils, this criterion may prove to be of doubtful value. However, recognising windblown components to soils is important to the regolith geology of many areas (Greene, in press), such as eastern Australia, where dryland soil salinity is believed to be the result of accession of aeolian material and potential tools in its recognition should not be ignored. In addition, Wilding and Drees (1974) suggested the clay-sized quartz particles in some soil profiles that have been attributed to aeolian accession may, in fact, be due to the recrystallisation of phytoliths. 7.3. Palaeoenvironmental reconstruction Opal-forming organisms are highly sensitive to environmental variations and, properly interpreted,

188

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

can be a useful guide in environmental reconstruction. Both phytoliths and diatoms have been widely used in palaeoenvironmental interpretation. Sponge spicules have been used less because of difficulties in taxonomic identification, but are good indicators of permanently waterlogged conditions and, where taxa are recognisable, of marine influence. Beavers and Stephen (1958) showed that phytoliths in the soils of Illinois varied according to vegetation type. Such patterns can be used to reconstruct ancient vegetation patterns. One such example is the report of Barboni et al. (1999), which provides an example of phytoliths in the environmental reconstruction of archaeologically significant regions in Ethiopia. Because phytoliths are commonly perceived to be concentrated in the A horizon of soils (Oehler, 1979), increases in their abundance within a profile have sometimes been used as indicators of palaeosols in a succession (Dormaar and Lutwick, 1969). As noted above, care should be taken in such interpretations as exceptions are known. Piperno (1988) gave an extensive review of the use of phytoliths in palaeoenvironmental reconstruction and archaeology, with the collection edited by Meunier and Colin (2001) providing a summary of the state of the art. The use of diatoms to reconstruct lacustrine sedimentary environments is well known (Smol, 1987). However, diatoms can be used even in arid and semiarid regions, as illustrated by the study of Gasse (1987) from sub-Saharan Africa. In this example, the diatoms live in a wide range of environments, including former mega-Lake Chad, dilute swamps, and small hypersaline lakes, fed by groundwater discharge or ephemeral streams. Harrison (1988) reviews the use of sponge spicules from a Canadian lake. Sponge gemmules are also valuable in reconstructing environments and have been found in diverse environments such as bogs of the Puget Lowland, Washington (Turner, 1985) and the flood plain forests of Amazonia (Junk, 1984). Piperno (1988) provides one of the few examples of using a wide range of siliceous organisms in environmental reconstruction. She was able to extract a diverse assemblage of phytoliths, diatoms, sponge spicules, and helizoans from a series of terrestrial sediment cores from Panama. Piperno was able to identify moist tropical forest, marine swamp, freshwater swamp and cropland vegetation and therefore

reconstruct an environmental and floral history of the last 11,300 years. The record closely matched that obtained from palynology and indicates the potential utility of siliceous remains to document palaeoenvironments in sediments that might not preserve organic microfossils. Similarly, Clark et al. (1992a,b) used palynology, diatom taxonomy, and distribution of phytoliths and sponge spicules to determine the Holocene evolution of the Magela Creek floodplain in the Northern Territory of Australia. These authors were able to demonstrate how the environment evolved from a mangrove swamp to seasonal freshwater wetland. Opal production was a clear indication of wetland development. However, very high levels of opal productivity (>8% for bulk soil composites) were associated with grass and sedge wetlands, rather than wooded wetlands or mangroves. Sponges were the main source of opal in the mangrove environment. Despite these studies, the use of biogenic, opal, apart from phytoliths, as a tool in regolith geology is in its infancy. Most studies using siliceous remains have concentrated on the Quaternary, rather than the much longer time necessary in many regolith studies as illustrated by Ollier and Pain (1996). Some work has been done on siliceous remains in older sediments, such as that of Folk (1964) on Cainozoic phytoliths. Marine sponge spicules have been important in understanding Eocene environments of southern Australia (Clarke, 1994a,b; Gammon et al., 2000). Phytolith geochemistry is another possible but poorly explored avenue for palaeoclimatic research. Webb and Longstaffe (1997) indicated that increasing enrichment in O18 reflected increasing degrees of evapotranspiration. More needs to be known about the range of O18 values of different taxa before this can be routinely applied. 7.4. Silcretes siliceous hardpans, and hardsetting soils Silcretes are silica-rich duricrusts found in many parts of the globe (Thiry, 1999). They are enigmatic features in that, unlike other duricrusts such as calcretes, bauxites, and ferricretes, there are few known modern analogues. Silcrete appears to form by a range of different processes in a diverse range of landscape contexts and eras (Thiry, 1999; Alley, 1996; Firman, 1994; Ollier, 1991), matched only by the diversity of opinions as to its formative conditions. An important

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

series of observations, however is that it commonly appears to have formed on former valley slopes and floors, often, although not always, in quartz-rich colluvium or alluvium. Importantly, to distinguish silcretes from siliceous hardpans (see below), there is also silicification of the host material, as well as silica cementation. Once formed, such silcretes typically form inverted relief (Ollier and Pain, 1996). Such silcretes need not have extended across the entire landscape, but rather have formed in selected localities by the lateral migration of silica-rich groundwater. The greater lability of biogenic opal compared to other forms of silica means that it is a prime source of silica for precipitation in the regolith. Several authors have suggested a connection between silcretes and biogenic opal. Gunn and Galloway (1978) raised the possibility that biogenic silica leached from marine organisms was an important component in some Australian silcretes developed on Cretaceous marine shales. Oehler (1979) similarly suggested that biogenic opal may play a role in silcrete formation, postulating that dissolution of the siliceous remains of terrestrial organisms may have contributed to the formation of silcretes. Thiry (1999) noted silica deposition by diatoms in the regolith during pluvial highstands of Lake Chad (see also Gasse, 1987). Few examples of recognisable biogenic silica remains have been noted from silcretes, however. One example is that of Clarke (1994b), which noted that almost all the silcretes present in the Kambalda and Norseman regions of WA were developed on Eocene spicular marine sediments, such as the Princess Royal Spongolite and Pallinup Siltstone. The formation of silcrete has largely obliterated the spicules and they are preserved only as ghosts. These marine spicules are quite large, typically 1 –5 mm in length and 100 Am in diameter. It is likely that the smaller freshwater spicules (100 – 300 Am in length and 3– 10 Am across) originally present in a silcreted sediment would be completely destroyed or rendered unrecognisable, as would the even more delicate diatoms and phytoliths. Therefore, the remains of siliceous microfossils may have been originally much more common in what are now silcreted surficial sediments. A possible additional link between biogenic opal and silcretes is the high levels of titanium found in many silcretes (Thiry, 1999) and the observation than phytoliths are com-

189

monly also enriched in this element (Wilding et al., 1989). Much more work must be done to elucidate this relationship, as detrital Ti from insoluble residules or adsorbed Ti on clays and iron oxides are alternatives to a biological origin. Clearly, the accumulation of biogenic silica in the regolith alone is not the key to silcrete formation. As the literature reviewed in this paper has shown, such accumulations are not uncommon. Modern analogies for silcrete are, however, difficult to find. What remains unanswered is what conditions facilitate the large-scale cementation and replacement by silica in the regolith. If biogenic opal indeed plays a role in this, it is through the stablisation of that opal as quartz. It is possible however that accumulation of biogenic silica, followed by major shifts in physical and chemical hydrology associated with the Cainozoic climate changes, may have been a factor. Siliceous hardpans (Wright, 1983) are also common in many parts of inland Australia and also from the Paris Basin (Thiry, 1999). They consist of silicacemented alluvium and colluvium. Other cementing agents, including iron oxides, carbonate, and clay are also present (Bettenay and Churchwood, 1974). Typically, they contain less silica and more iron and carbonate than silcretes. Hardpans may form a continuum with silcretes, or represent an early stage in silcrete development. Unlike silcretes, they are often less indurated, lacking in silicification, and appear more likely related to contemporary or near-contemporary landscape processes (Milnes et al., 1991) in arid and semi-arid environments. Also, hardpans form through siliceous cementation, rather than whole-scale silicification. Despite this, their genesis and relationship to silcrete is still obscure. Nor can they always be related to overlying soil profiles (Wright, 1983). However, the presence of common mobile silica from biogenic opal would clearly be a favourable precursor to their formation. As for silcretes, the fact that biogenic opal is more common that hardpans suggests that the key question is not the presence of opal in the soil, but what factors encourage its stablisation in the regolith as a quartz cement. Hardsetting soils (Greene, in press) may be an early precursor to hardpans. Hardsetting occurs through cyclic wetting and drying of the soils. Slaking of soils aggregates under conditions of rapid wetting, and/or dispersion of the clay fraction alone. However, in some

190

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

cases, cementation of the subsoil can also occur as a result release of soluble silica during wetting and reprecipitation as the soil dries (Chartres et al., 1990). The presence of potentially labile silica in the form of biogenic opal may facilitate this process. Another possible link is the relationship sometimes noted between some silcretes and volcanic rocks, especially basalts (see Ollier, 1991). In the past this was normally attributed to silica released during hydrothermal alteration or rapid weathering. While a hydrothermal origin is ruled out by modern workers (Taylor and Eggleton, 2001), the exceptionally high levels of biogenic opal found in soils developed on some volcanic rocks (Meunier et al., 1999, and references in Hart, 1992) suggest that it is the organisms that are responsible for this association. 7.5. Health The size, geometry and composition of biogenic silica bodies in the regolith is a potential health issue. The large marine spicules of the Eocene of WA can cause contact dermatitis among geologists logging percussion and air core holes in the palaeovalley fills of the area. The irritation caused by the spicules is increased by the hypersaline waters which often saturates these deposits. The numbers of open pits through the Eocene cover is also increasing, although no effects among mine waters are as yet reported. Sponge spicules in soils of southwestern Western Australia, probably also Eocene in age, have been linked to hoof disease in horses (Carroll, 1932). Stone et al. (1970) mentioned similar contact dermatitis amongst agricultural workers in areas of freshwater spicule-rich soil of New York, as did miners of freshwater spiculites in Brazil (Volkmer-Ribeiro, 1992). Although respirable dust (must with a median aerodynamic diameter of 10 Am and less) is normally regarded as an urban problem, it is also a major health issue in rural areas (Clausnitzer and Singer, 1996). The lower size limit of freshwater sponge spicules ( < 5 Am across and some less than 3 Am) places them within the definition of hazardous mineral fibre (Skinner et al., 1988). Some of the cellular damage caused by fibrous materials is from fibre penetration. Many sponge spicules have sharper terminations than non-biogenic mineral fibres, which greatly increases their penetration potential. Other cellular damage is caused by

mineral reactivity, however the biological reactivity of opal is unknown. The presence of spicules in windblown dust raises the possibility that they may present a respirable silica hazard in dust-affected areas. Silicified root hair cells (Piperno, 1988) also have a high aspect ratio, are sharp, and potentially fall into the hazardous size range. Most are probably too large to pose a cancer risk, although respiratory irritation has been reportedly caused by them (Parry et al., 1984).

8. Conclusions Biogenic opal in soils was first recognised in the 1840s. The potential importance of silica-secreting plants in the formation of silcretes was suggested in the 1950s. Despite periodic reviews in the 1970s and 1980s, and considerable research of aspects of biogenic silica in soils and terrestrial sediments since then, regolith science is not much closer to understanding the larger scale significance of biogenic opal than we were 50 years ago. What is known is that juvenile biogenic opal is present in most soils in at least trace amounts and common (up to 2 – 3% in the A horizons many soils on all continents except Antarctica. Phytoliths, sponges, and diatoms largely produce the opal. Furthermore, in a significant number of cases, it can be present in large amounts comprising more than 10% of the mineral fraction, and in some examples 100%. There is a strong correlation between poorly drained or at least seasonally water logged conditions, and abundant biogenic opal. In some cases, volcanic substrates are also conducive to abundant soil opal. Biogenic opal may play an important role in the cycling of silica in soils and aquatic sediments, in the genesis of the siliceous cements of hardpans, and silcretes, and may be significant in environmental health. A number of approaches for future research appear promising. The first is taxonomy; what opal-secreting organisms are present in the regolith, in what numbers, and how can they be recognised from their remains. The second is the ecological limits and roles represented by each organism, which requires both drainage basin and catena studies. Thirdly, we need to know the stratigraphic distribution of these remains especially through the Cainozoic. These studies will enable a better understanding of the both the regolith signifi-

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

cance of biogenic opal in a wide range of environments and also constrain the conditions under which they may become the dominant part. Fourthly, we need to identify the possible geochemical signatures of biogenic opal in mature regolith profiles. Fifthly, research is needed into the physical, chemical, and biological conditions necessary to convert labile opal into stable quartz, leading to the formation of silica cements, hardpans, and silcretes. Sixthly, the role of opal in nutrient cycling needs better documentation. Finally, the potential health hazards of biogenic opal in respirable dusts would also be a fruitful research area. Study of the importance of opal producing organisms in the regolith is clearly a multidisciplinary process. After more than 160 years, the importance of biogenic opal in the regolith is surely an idea whose time has come. Acknowledgements I would like to thank Doreen Bowdery, Robin Clark, Tony Eggleton, John Field, Dianne Hart, Megan Kilby, Ian Roach, and Bob Wasson for their help and suggestions in preparing this paper. Graham Taylor and J. Meunier made several helpful and encouraging comments in their reviews of the manuscript. The paper is published with the permission of CRC LEME. References Alexandre, A., Meunier, J.D., Colin, F., Koud, J.M., 1997. Plant impact on the biogeochemical cycle of silicon and related processes. Geochim. Cosmochim. Acta 61 (3), 677 – 682. Alley, N.F., 1996. Cainozoic stratigraphy, palaeoenvironments, and geological evolution of the Lake Eyre Basin. Palaeogeogr. Palaeoclimatol. Palaeoecol. 144, 239 – 263. Aston, S.R., 1983. Natural water and atmospheric chemistry of silicon. In: Aston, S.R. (Ed.), Silica Geochemistry and Biochemistry. Academic Press, New York, pp. 39 – 76. Baker, G., 1959a. Opal phytoliths in some Victorian soils and ‘red rain’ residues. Aust. J. Bot. 7, 64 – 87. Baker, G., 1959b. A contrast in the opal phytolith assemblies of two Victorian soils. Aust. J. Bot. 7, 88 – 96. Baker, G., 1960. Phytoliths in some Australian dusts. Proc. R. Soc. Vic. 72, 21 – 40. Barboni, D., Bonnefille, R., Alexandre, A., Meunier, J.D., 1999. Phytoliths as paleoenvironmental indicators, west side middle Awash valley, Ethiopia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 152, 87 – 100.

191

Bartoli, F., 1985. Crystallography and surface properties of biogenic opal. J. Soil Sci. 36, 335 – 350. Beavers, A.H., Stephen, I., 1958. Some features of the distribution of plant opal in Illinois soils. Soil Sci. 86, 1 – 5. Bettenay, E., Churchwood, H.M., 1974. Morphology and stratigraphic relationships of the Wiluna hardpan in arid Western Australia. J. Geol. Soc. Aust. 21 (1), 73 – 80. Bobrov, A.A., Bobrova, E.K., 2001. Phytolith assemblages in podzol and podzolic soils of the south-west coast of the Ohotsk Sea. In: Meunier, J.D., Colin, F. (Eds.), Phytoliths; Applications in Earth Sciences and Human History. A.A. Balkema, Rotterdam, pp. 365 – 370. Bobrova, E., Bobrov, A., 1997. Phytoliths in soils: species composition, distribution along a soil profile, and value as environmental indicators. In: Pinilla, A., Juan-Tresserras, J., Machado, M.A. (Eds.), First European Meeting on Phytolith Research. Centro de Ciencias Medioambientales del Consejo Superior de Investigaciones Cientificas Serrano, 115B. 2800, Madrid, pp. 5 – 13. Boettinger, J.L., 1994. Biogenic opal as an indicator of mixing in an alfisol/vertsol landscape. Proceedings of the 9th International Working Meeting on Soil Micromorphology, 17 – 26. Bohaty, S.M., Harwood, D.M., 2000. Ebridian and silicoflagellate bisotratigraphy from Eocene McMurdo erratics and the Southern Ocean. In: Stilwell, J.D., Feldmann, R.M. (Eds.), Paleobiology and Paleoenvironments of Eocene rocks, McMurdo Sound, East Antarctica. Antarct. Res. Ser., vol. 76. American Geophysical Union, Washington, pp. 99 – 160. Bombin, M., Muehlenbachs, K., 1980. Potential of 18O/16O ratios in opaline plant silica as a continental paleoclimatic tool. American Quaternary Association Abstracts and Program, Sixth Biennial Meeting at Orono, Maine, 43 – 44. Braiser, M.D., 1980. Microfossils George Allen and Sons, London. Braiser, M.D., Green, O., Sheilds, G., 1997. Ediacarian sponge spicule clusters from southwestern Mongolia and the origins of the Cambrian fauna. Geology 25 (4), 351 – 354. Breese, G.F., 1960. Quartz overgrowths as evidence of silica deposition. Aust. J. Sci. 23, 18 – 20. Brewer, R., 1955. Diatom skeletons and sponge spicules in the soils of New South Wales. Aust. J. Sci. 17, 177 – 179. Brewer, R., Sleeman, J.R., Foster, R.C., 1993. The fabric of Australian soils. Soils: An Australian Viewpoint. Division of Soils, CSIRO, pp. 439 – 476. Carlson, E.H., 1994. Paleoshoreline patterns in the transgressive – regressive sequences of Pennsylvanian rocks in the northern Appalachian Basin, USA. Sediment. Geol. 93, 209 – 222. Carroll, D., 1932. Mineralogy of fine sand fractions of some Australian soils. J. R. Soc. West. Aust. 18, 125 – 144. Charman, D.J., 2000. Biostratigraphic and palaeoenvironmental applications of testate amobae. Quat. Sci. Rev. 20, 1753 – 1764. Chartres, C.J., Kirby, J.M., Raupach, M., 1990. Poorly ordered silica and aluminosilicates as temporary cementing agents in setting soils. Soil Sci. Am. J. 54, 1060 – 1067. Chauvel, A., Walker, I., Lucas, Y., 1996. Sedimentation and pedogenesis in a central Amazonian black water basin. Biogeochemistry 33, 77 – 95. Clark, R.L., East, T.J., Guppy, J., Mahon, D., Mcbride, P., Wasson, R.J., 1992a. Late Quaternary stratigraphy of the Magela Plain.

192

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

In: Wasson, R.J. (Ed.), Modern Sedimentation and Evolution of the Magela Creek Plain. Res. Rep. Supervising Scientist for the Alligator Rivers Region, vol. 6. pp. 28 – 80. Clark, R.L., East, T.J., Guppy, J., Mahon, D., Mcbride, P., Wasson, R.J., 1992b. Late Quaternary evolution of the Magela Plain. In: Wasson, R.J. (Ed.), Modern Sedimentation and Evolution of the Magela Creek Plain. Res. Rep. Supervising Scientist for the Alligator Rivers Region, vol. 6. AGPS, Canberra, pp. 81 – 157. Clarke, J.D.A., 1994a. Evolution of the Lefroy and Cowan Palaeodrainages, Western Australia. Aust. J. Earth Sci. 41, 55 – 68. Clarke, J.D.A., 1994b. Geomorphology of the Kambalda region. Aust. J. Earth Sci. 41 (3), 229 – 240. Clausnitzer, H., Singer, M.J., 1996. Respirable-dust production from agricultural operations in the Sacramento Valley, California. J. Environ. Qual. 25, 877 – 884. Davis, A., Suzanne, J.R., Rimmer, S.M., Yeakel, J.D., 1984. Some genetic implications of silica and aluminosilicates in peat and coal. Int. J. Coal Geol. 3, 293 – 314. De Laubenfels, M.W., 1955. Porifera. Treatise on Invertebrate Paleontology Part E. Geological Society of America, pp. E21 – E112. Dormaar, J.F., Lutwick, L.E., 1969. Infrared spectra of humic acids and opal phytoliths as indicators of paleosols. Can. J. Sci. 49, 29 – 37. Drees, L.R., Manu, A., Wilding, L.P., 1993. Characteristics of aeolian dusts in Niger, West Africa. Geoderma 59, 213 – 223. Epstein, E., 1999. Silicon. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 641 – 664. Ferris, F.G., 1997. Formation of authigenic minerals by bacteria. In: Mcintosh, J.M., Groat, L.A. (Eds.), Biological – Mineralogical Interactions. Mineral. Assoc. Can. Short Course, vol. 25, pp. 187 – 208. Firman, J.B., 1994. Paleosols in laterite and silcrete profiles. Evidence from the south east margin of the Australian Precambrian shield. Earth Sci. Rev. 36, 149 – 179. Fredlund, G.G., 1993. Paleoenvironmental interpretation of stable carbon, hydrogen, and oxygen isotopes from opal phytoliths, Eustis Ash Pit, Nebraska. In: Pearsall, D.M., Piperno, D.R. (Eds.), Current Research in Phytolith Analysis: Applications in Archaeology and Paleoecology. MASCA Res. Pap. Sci. Archaeol., vol. 10, pp. 37 – 46. Folk, R.L., 1964. Note on the occurrence of phytoliths in some Cenozoic sedimentary rocks. J. Paleontol. 38, 773 – 775. Gammon, P.R., James, N.P., Bone, Y., Clarke, J.D.A., 2000. Sedimentology and lithostratigraphy of a Late Eocene sponge-dominated sequence, southern Western Australia. Aust. J. Earth Sci. 47 (6), 1087 – 1103. Gasse, F., 1987. Diatoms for reconstructing palaeoenvironments and palaeohydrology in tropical semi-arid zones. Hydrobiologia 154, 127 – 163. Gibson, A.A., 1976. Diatomite—Western Australia. In: Knight, C.L. (Ed.), Economic Geology of Australia and Papua new Guinea. Australas. Inst. Min. Metall. Monogr., vol. 8, pp. 111 – 112. Greene, R.S.B., 2002. Hardsetting soils. In: Lal, R. (Ed.), Encyclopedia of Soil Science. Marcel Dekker, New York, USA (in press). Gregory, W., 1855. On the presence of diatomacaeae, phytolitharia, and sponge spicules in soils which support vegetation. Proc. Bot. Soc., Edinb., 69 – 72.

Gunn, R.H., Galloway, R.W., 1978. Silcretes in south-central Queensland. In: Langford-Smith, T. (Ed.), Silcrete in Australia. Department of Geography, University of New England, Armidale, pp. 51 – 71. Harrison, F.W., 1988. Methods in Quaternary ecology #4. Freshwater sponges. Geosci. Can. 15 (1), 193 – 198. Hart, D.M., 1992. A field appraisal of the role of plant opal in the Australian environment. PhD thesis, Department of Earth Sciences, Macquarie University, Australia (unpublished). Hart, D.M., Humphreys, G.S., 1997. The mobility of phytoliths in soils: pedological considerations. In: Pinilla, A., Juan-Tresserras, J., Machado, M.A. (Eds.), First European Meeting on Phytolith Research. Centro de Ciencias Medioambientales del Con sejo Superior de Investigaciones Cientificas Serrano, 115B, 2800, Madrid, pp. 93 – 100. Heinen, W., Oehler, J.H., 1979. Evolutionary aspects of biological involvement in the cycling of silica. In: Trudinger, P.A., Swain, D.J. (Eds.), The Biogeochemical Cycling of Mineral-Forming Elements. Elsevier, Amsterdam, pp. 431 – 443. Jones, R.L., Beavers, A.H., 1963. Sponge spicules in Illinois soils. Proc. Soil Sci. Soc. Am. 27, 438 – 440. Jones, L.H.P., Handreck, K.A., 1967. Silica in soils, plants, and animals. Adv. Agron. 19, 107 – 149. Jones, L.P.H., Milne, A.A., Saunders, J.V., 1966. Tabashir: an opal of plant origin. Science 151, 464 – 466. Junk, W.J., 1984. Ecology of the varzea, floodplain of Amazonian whitewater rivers. In: Sioli, H. (Ed.), The Amazon. Monogr. Biol., vol. 56. Kluwer, Dordrecht, pp. 214 – 244. Kenley, P.R., 1976. Diatomite—Victoria. In: Knight, C.L. (Ed.), Economic Geology of Australia and Papua New Guinea. Australas. Inst. Min. Metall. Monogr., vol. 8, pp. 111 – 112. Kondo, R., Iwasa, Y., 1981. Opal phytoliths in humic yellow laterosols and yellow laterosols in the Amazon region. Res. Bull. Obihiro Univ. 12, 231 – 239. Konhauser, K.O., Mann, H., Fyfe, W.S., 1992. Prolific organic SiO2 precipitation in a solute deficient river, Rio Negro, Brazil. Geology 20, 227 – 230. Kosters, E.C., Bailey, A., 1983. Characteristics of peat deposits in the Mississippi River Delta Plain. Trans. Gulf Coast Assoc. Geol. Soc. 33, 311 – 325. Lauwers, A.M., Heinen, W., 1974. Biodegradation and utilization of silica and quartz. Arch. Microbiol. 95, 67 – 78. Lewin, J.C., Reimann, B.E.F., 1969. Silicon and plant growth. Annu. Rev. Plant Physiol. 20, 289 – 304. Lovering, T.S., 1959. Significance of accumulator plants in rock weathering. Bull. Geol. Soc. Am. 70, 781 – 800. McCarthy, T.S., McIver, J.R., Caincross, B., Ellery, W.N., Ellery, K., 1989. The inorganic chemistry of peat from the Maunachira channel swamp system, Okavango Delta, Botswana. Geochim. Cosmochim. Acta 53, 1077 – 1089. McIntosh, J.M., Groat, L.A. (Eds.), 1997. Biological – Mineralogical Interactions. Mineral. Assoc. Can. Short Course, vol. 25. Meunier, J.D., Colin, F. (Eds.), 2001. Phytoliths; Applications in Earth Sciences and Human History. A.A. Balkema, Dordrecht, 378 pp. Meunier, J.D., Colin, F., Alarcon, C., 1999. Biogenic silica storage in soils. Geology 27, 835 – 838.

J. Clarke / Earth-Science Reviews 60 (2003) 175–194 Milnes, A.R., Wright, M.J., Thiry, M., 1991. Silica accumulations in saprolites and soils of South Australia. In: Nettleton, W.D. (Ed.), Occurrence, Characteristics, and Genesis of Carbonate, Gypsum and Silica Accumulations in Soils. Soil Sci. Soc. Am. Spec. Publ., vol. 26, pp. 121 – 149. Moore, R.C., 1964. Protista 3: Protozoa (chiefly Radiolaria and Tintinnina). Treatise on Invertebrate Paleontology Part D. Geological Society of America, Lawrence, KS. Muller, P.J., Schneider, R., 1993. An automated method for the determination of opal in sediments and articulate matter. Deep-Sea Res., Part I 40, 425 – 446. Nott, J., 1994. Long-term landscape evolution in the Darwin region and its implications for the origin of landsurfaces in the north of the Northern Territory. Aust. J. Earth Sci. 41, 407 – 415. Oehler, J.H., 1979. Deposition and diagenesis of biogenic silica. In: Trudinger, P.A., Swain, D.J. (Eds.), The Biogeochemical Cycling of Mineral-Forming Elements. Elsevier, Amsterdam, pp. 467 – 479. Ollier, C.D., 1991. Aspects of silcrete formation in Australia. Z. Geomorphol. 35 (2), 151 – 163. Ollier, C., Pain, C., 1996. Regolith, Soils, and Landforms Wiley, Chichester. Parry, D.W., Hodson, M.J., Sangster, A.G., 1984. Some recent advances in the study of silicon in higher plants. Philos. Trans. R. Soc. London, Ser. B 304, 537 – 549. Patrick, R., 1977. Ecology of freshwater diatoms – diatom communities. In: Wener, D. (Ed.), The Biology of Diatoms. Bot. Monogr., vol. 13. Blackwell, London, pp. 284 – 332. Piperno, D.R., 1988. Phytolith Analysis: An Archaeological and Geological Perspective. Academic Press, San Diego. Runge, F., 1999. The opal phytolith inventory of soils in central Africa—quantities, shapes, classification, and spectra. Rev. Palaeobot. Palynol. 107, 23 – 53. Schwandes, I.P., Collins, M.E., 1994. Distribution and significance of freshwater sponge spicules in selected Florida soils. Trans. Am. Microsc. Soc. 113 (3), 242 – 257. Shaw, P.A., Nash, D.J., 1998. Dual mechanisms for the formation of fluvial silcretes in the distal reaches of the Okavango delta fan, Botswana. Earth Surf. Processes Landf. 23, 705 – 714. Shaw, P.A., Cooke, H.J., Perry, C.C., 1990. Microbialitic silcretes in highly alkaline environments: some observations from Sua Pan, Botswana. S. Afr. J. Geol. 93 (5/6), 803 – 808. Silverman, M.P., 1979. Biological and organic chemical decomposition of silicates. In: Trudinger, P.A., Swain, D.J. (Eds.), The Biogeochemical Cycling of Mineral-Forming Elements. Elsevier, Amsterdam, pp. 445 – 465. Simons, N.A., Hart, D.M., Humphreys, G.S., 2000. Phytolith depth functions in a podzol and a texture contrast soil, Sydney. In: Adams, J.A., Metherell, A.K. (Eds.), Soil 2000: New Horizons for a New Century. Aust. and NZ Second Joint Soils Conference, vol. 3. New Zealand Society of Soil Science, Canterbury, pp. 193 – 194. Skinner, H.C.W., Fitzpatrick, R.W. (Eds.), 1992. Biomineralization: processes of iron and manganese—modern and ancient environments. Catena. (Supplement 21).

193

Skinner, H.C.W., Ross, M., Frondel, C., 1988. Asbestos and Other Fibrous Materials. Oxford Univ. Press, New York. Smithson, F., 1959. Opal sponge spicules in soils. J. Soil Sci. 10 (1), 105 – 109. Smol, J.P., 1987. Methods in freshwater ecology #1. Freshwater algae. Geosci. Can. 14 (4), 206 – 215. Stace, H.C.T., Hubble, G.D., Brewer, R., Northcoate, K.H., Sleeman, J.R., Mucahy, M.J., Hallsworth, E.G., 1968. A Handbook of Australian Soils. Rellim Technical Publications, Glenside, South Australia. Stone, E.L., Dawson, J.E., Zeeuw, C.D., 1970. Dermatitis caused by sponge spicules. Soil Sci. Soc. Am. Proc. 34, 151 – 152. Taylor, G.T., Eggleton, R.A., 2001. Regolith Geology and Geomorphology. Wiley, Chichester. Taylor, G., Gasse, F., Walker, P.H., Morgan, P.J., 1990. The palaeoecological and palaeoclimatic significance of Miocene freshwater diatomite deposits from southern New South Wales, Australia. Palaeogeogr. Palaeoclimatol. Palaeoecol. 77 (2), 127 – 143. Tesic, Z.P., Todorovic, M.S., 1958. Contribution to knowledge of the specific properties of silicate bacteria. Zemljiste, 233 – 240. Thiry, M., 1999. Diversity of continental silicification features: examples from the Cenozoic deposits of the Paris Basin and neighbouring basement. Spec. Publ. Int. Assoc. Sedimentol. 27, 87 – 127. Turner, J., 1985. Sponge gemmules from lake sediments in the Puget Lowland, Washington. Quat. Res. 24, 240 – 243. Volkmer-Ribeiro, C., 1992. The freshwater sponges in some peatbog ponds in Brazil. Amazonia 12 (2), 317 – 335. Webb, E.A., Longstaffe, F.J., 1997. Oxygen isotope geochemistry of silica phytoliths. In: Pinilla, A., Juan-Tresserras, J., Machado, M.A. (Eds.), First European Meeting on Phytolith Research. Centro de Ciencias Medioambientales del Consejo Superior de Investigaciones Cientificas Serrano, 115B. 2800, Madrid, pp. 233 – 243. Wilding, L.P., Drees, L.R., 1968. Distribution and implications of sponge spicules in surficial deposits in Ohio. Ohio J. Sci. 68, 92 – 99. Wilding, L.P., Drees, L.R., 1971. Biogenic opal in Ohio soils. Soil Sci. Soc. Am. Proc. 35, 1004 – 1010. Wilding, L.P., Drees, L.R., 1974. Contributions of forest opal and associated crystalline phases to fine silt and clay fractions of soils. Clay Clay Miner. 22, 295 – 306. Wilding, L.P., Smeck, N.E., Drees, L.R., 1989. Silica in soils: quartz, crystabolite, tridymite and disordered silica polymorphs. In: Dixon, J.B., Weed, S.B. (Eds.), Minerals in Soil Environments, 2nd edn. Soil Sci. Soc. Am. Book, vol. 1. pp. 913 – 974, Madison, WI. Wollast, R., Mackenzie, F.T., 1983. The global cycle of silica. In: Aston, S.R. (Ed.), Silica Geochemistry and Biochemistry. Academic Press, New York, pp. 39 – 76. Wright, M.J., 1983. Red-brown hardpans and associated soils in Australia. Trans. R. Soc. South Aust. 107, 252 – 254.

194

J. Clarke / Earth-Science Reviews 60 (2003) 175–194

Jonathan Clarke graduated with a BSc (Hons.) from the University of Tasmania and completed his PhD degree at Flinders University in South Australia. He worked for 10 years at WMC Resources Ltd. in mineral exploration and research before joining the geology department at the Centre for Landscape, Environment, and Mineral Exploration at the Australian National University. His past research interests have included Cambrian reef, platform, and slope carbonates, cool water carbonate deposition, coal geology, modern sediments on the Australian shelf, evolution of the Great Australian Bight, and regolith geology in Western Australia. Other current research interests include the history of aridity of the Atacama Desert, Eocene carbonate, clastic, and biosiliceous sedimentation in the Eucla Basin, and terrestrial Mars analogues.