Middleton G.V. (Ed.) Encyclopedia of Sediments and Sedimentary Rocks
Подождите немного. Документ загружается.
IMPREGNATION
375
tacky acetate sheet applied to the sediment surface is an
alternate approach that avoids the subsequent shrinkage
problem. However, the acetate peel can only bond to sur-
face grains and may not provide a complete, representative
sample.
A more common approach to imbedding involves filling the
entire sample with a cement to preserve the sedimentary
structures. In practice, a large number of possible embedding
materials exist, and availability varies in different areas. In
general, embedding media can be characterized as either water-
soluble or insoluble. Soluble materials are useful when the
sample is wet, such as recent lacustrine or marine sediments.
The two most common soluble materials include diluted
ordinary glue and polyethylene glycol (PEG). The sample is
immersed in the liquid embedding material and allowed to
soak for several days or longer to allow the embedding fluid to
fully infiltrate the sample. Generally, the PEG used for
embedding is a high molecular weight (> 800) variety and is
a waxy solid at room temperature. The PEG flakes must be
melted by the application of heat during the soaking process.
Although the water-soluble media are nontoxic and require no
specialized equipment to use, they often result in some
shrinkage, are generally too soft for thin sectioning, require
non-aqueous grinding fluids, and they have poor optical
properties.
Water insoluble media are generally either epoxy or acrylic
resins that require the sample to be dehydrated prior to
embedding to obtain full polymerization of the resin com-
pounds. While air or oven drying may be suitable for sandy or
gravelly samples, finer textured samples will crack and shrink
during drying, and the dry permeability of the sample will
often be too low to allow full infiltration of the embedding
material, especially into the center of larger samples.
Two common methods for dehydrating samples to minimize
shrinkage are to sublimate the sample water under vacuum
(freeze drying) or by using repeated liquid-liquid replacements.
Freeze drying effectively d.ehydrates samples of all textures,
although many commercial units are limited to small samples
(less than a few centimeters). Where suitable equipment is
available, freeze-drying produces minimal disturbance of
the sample if the sample is first frozen with liquid nitrogen
(Bouma, 1969). Pre-freezing is particularly important for clay-
rich samples, as the rapid freezing in nitrogen prevents the
formation of large ice crystals in the sample that result in voids
in the embedded sample. However, some ice crystal formation
tends to occur in very fine clay units, particularly if the
sediment sample .is more than
1
cm thick. Depending on the
rate of sample drying and the sediment composition, some
minor shrinkage may also occur and cracks may form at
sedimentary contacts. Moreover, the freeze-dried samples are
extremely fragile and coarse units may require support to
prevent collapse during handling.
The second approach to dehydrating samples involves
replacing the pore water with a water and resin miseible
solution. Most epoxy resins are compatible with acetone and
various alcohols. By replacing the pore water with such a fluid,
there is no shrinkage or disturbance to the sedimentary
structures. This method is suitable for a variety of sediments
including those from clay-rich lacustrine, marine, and glacio-
genic environments. The wet samples are submerged in the
replacement fluid that is left to fully infiltrate into the sample,
often for one or more days. The fluid is then drawn off and
replaced with fresh solution until the specific gravity of the
supernatant indicates that the water content is below
1
percent
(or less). Liquid-liquid replacement dehydration of samples
results in the least disturbance of samples, as there is no
shrinkage or danger of ice crystal casts in the final embedded
sample. However, the process is time consuming and uses a
large amount of the exchange fluid and resin. In some cases,
application of a light vacuum during the resin infiltration can
avoid extra resin waste.
Selection of the embedding resin varies considerably among
users,
ranging from slow curing acrylics (months) to rapid
curing epoxies (hours). Some epoxy resins that cure rapidly are
unsuitable because they are strongly exothermic and can
disturb the sample. Depending on circumstances, a variety of
resins can be used (Tippkotter and Ritz, 1996). Low viscosity
resins used for embedding histologieal and material samples
are readily available and suitable for fine-grained lacustrine
and marine sediments. Less specialized, higher viscosity resins
are suitable for coarser samples. Most require curing in an
oven to fully polymerize. The advantages of impregnating
samples with resin include standard handling procedures for
thin section preparation, excellent optical characteristics, and
permanent preservation of the sample. Compared to water-
soluble media, impregnating samples with resin is often time
consuming and relatively expensive. Moreover, more specia-
lized equipment is required, including facilities to handle what
are typically toxic chemicals.
Partial curing of an impregnated block, especially with
epoxy resins, usually indicates the presence of water in the
sample or poor resin infiltration. Increased infiltration
times using resins with long working times can avoid this
problem, and light vacuum may also accelerate infiltration. In
some situations, samples with incomplete infiltration can still
be cut and the exposed face treated with resin to complete the
stabilization prior to mounting (Carr and Lee, 1998).
Impregnation of sedimentary samples has proven to be a
valuable technique for geologists working in a wide range of
environments where sample preservation and microscopic
analysis has been limited by poorly consolidated materials. A
number of methods are available for impregnation, depending
on sample texture and moisture content. Samples that are
dehydrated and embedded with epoxy resins are suitable for
most applications, although simpler methods are available
as well.
Scott F. Lamoureux
Bibliography
Bouma, A.H., 1969. Methods for the Study of Sedimentary Structures.
John Wiley & Sons.
Carr, S.J., and Lee, J.A., 1998. Thin-section production of diamicts:
problems and solutions. Journal of Sedimentary Research, 68:
217-220.
Murphy, C.P., 1986. Thin Section Preparation of Soils and Sediments.
Berkhansted; A B Academic.
Tippkotter, R., and Ritz, K., 1996. Evaluation of polyester, epoxy and
acrylic resins for suitability in preparation of soil thin-sections for
insitu biological studies. Geoderma, 69: 31-57.
Cross-references
Bedding and Internal Structures
Fabric, Porosity, and Permeability
Relief Peels
376 IRON-MANGANESE NODULES
IRON-MANGANESE NODULES
Blaek
to
dark brown spheroidal
to
diseoida! coneretions of iron
and manganese oxyhydroxides. generally
a tew em in
diameter.
that cover extensive areas ot'the oeean floor
in ail
water depths
as well
as
the bottoms of some temperate latitude lakes (Figures
15
and
16).
The
ferromanganese oxyhydroxides
are
intimately
mixed with mineral
and
rock fragments
and
fossil debris,
and
occur
as
concentric layers arranged around
a
central nucleus
or
core consisting
of
rock fragments (generally volcanic), fossils
(e.g.. fish teeth, whale
ear
bones,
etc.) or
broken nodules.
Surfaces
are
either smooth
or
granular. Nodules
are
highly
porous
and
have
dry
bulk densities ranging from l.2gcm
to
1.6gcm~"
depending
on the
chemical composition.
Distribution and abundance
Nodules
are
most abundant
on the
sediment surface
and
occur
sporadically wilhin
the
seafloor sediment. Sampling
and
bottom photography
has
shown that average surface concen-
trations
of
lOkgm
~.
reaching
a
maximum
of
about
35kgm~^,
are
typical
of
vast areas
of the
Indian
and
Pacilic
seafloors between
4
km
and
5 km water depth.
The
highest
concentrations
are
found where
the
sediment accumulates very
slowly,
so
that
the
nodules
are not
covered
and
buried. Hence,
dense pavements
of
nodules
are
eharacteristic ol"
the
central
ocean basins most remote from continental sediment supply.
Similar
Mn and Fe
deposits
are
found
on
northern European
shelves,
the
Bailie
Sea and
some tjords. Ferromanganese
oxyhydroxides also oeeur
as
crusts
and
coatings
on
rock
and
mineral surfaees, especially
on
seamounts
and
similar topo-
graphic elevations throughout
the
oeean basins.
Ferromanganese nodules
are
locally abundant
on the
floors
of temperate zone lakes, such
as
those
of
Scandinavia
and
eastern North America (Cronan
and
Thomas, 1970). They
occur
as
spherical
and
discoidal nodules
and as
coatings
on
rock surfaces.
Growth rate
The rate
of
accretion
of
ferromanganese nodules
and
crusts
has been determined
by
dating rock nuclei
by the
K/Ar method
(Barnes
and
Dymond, 1967),
and by
measuring
the
decay
of
natural uranium series (""""Th.
""Pa) or
cosmogenie
(' Be)
nuclides
in the
oxyhydroxide layers (Ku.
in
Glasby. 1977).
The
median rate obtained with these different methods
is
around
5 mm/Ma, with
a
range
of <1
mm/Ma
to >
50 mm/Ma.
In
contrast,
the
rate
of
accumulation
of
associated sediment
is
of order I mm/Ka
to
5mm/Ka. This discordance,
and ihe
puzzling fact that most deep-sea nodules
arc
found
at
the sediment surface, implies that they must
be
maintained
at
the
sediment-water interface
for
very long time periods.
Occasional disturbance
of the
seafloor.
by
currents, slumping
or
the
activities
of
benthic fauna,
is
probably sufficient
to
remo\e sediment from
the
tops
of the
nodules
and to
turn
them over from time
to
time (Piper
and
Fowler, 1980).
Chemical composition
The chemical composition
of
ferromanganese nodules
and
crusts reflects their constituent mixtures
of
mineral phases,
comprising oxyhydroxides, aluminosilicates. carbonates
and
phosphates. They
are
significantly enriched
in Mn. Fc and
other transition metals compared with crustal rocks. Mn/Fe
ratios
are
generally greater than unity
and Cu and Ni
coneentrations
are
highest
in the
Pacific Ocean, whereas
Mn/Fe ratios
are
less than unity
and Zn and Pb
concentrations
are highest
in the
Atlantic Ocean (Table 12).
On an
ocean-wide
cm
Figure 15 Photographs
of
Pacific seafloor ferromanganese nodules. The upper
p^iir
of
nodules hnve smooth surfaces <ind were collected from
ihe northern cenlral region where sedjmenlalion rates
are
very
low. The
lower pair
of
nodules have granular, botryoidal surfaces and were
collecled from
ihe
northern equatori.il region where sedimentation rates are higher.
IRON-MANGANFSE NODULB 377
Figure 16 Scal'loor pholograph from the Southern Ocean {81 '
41
S;
1
70' 4O'E; water depth 5160 m) ^howinj; dense pavement ol nodules 2 cm
to ticm in diameter with a li^hl sediment coating. Surface concentration is approximately 25kgm ". Photograph t.iken by S.E. Calvert
using equipment owned by the US Navy,
Table 12 Average chemical composition uf uceanit terromanganese deposits (wt. per cent). From
( ronan (1%0)
Eilenu'iit
Pacific Ocean
Atlantic
Crustal abundance
Mn
Fe
Co
Ni
Cu
Zn
Ph
19.8
12.0
0.33
0.63
0.39
0.07
15.8
20.8
0.32
0.33
0.17
0.08
0,13
0.1
5.6
t).()02
0.007
0,005
0.007
0,001
scale, minor element concentrations appear lo be related to
the Mti or the Fe contents; Ba, Cu. Mo. Ni. and Zn vary
proportionally wilh the Mn conteni, whereas Co. Pb, Ti. and
Ce concentrations correlate with the Fe content. These
relationships reflect the substitution of the minor metals for
Mn or Fe in the tnajor oxyhydroxide mineral phases. Buserite
is a (ibrous tnanjzanate in which a number of divalent cations,
including Mn" ' . Cu, Ni. Zn, etc., substitute for Mn"^' in an
hexagonal array of edge-shared Mn(IV)O(, octahedra thereby
stabilizing the structure (Giovanoli and Arrhenius, in Halbaeb
el
III..
1988); vevniulite is a highly disordered manganate witb
[he approxitnate IbrtnuUi MnO2 which is intimately assoeialed
with ferrihydrite (FeOOH). the principal Fe phase (Burns and
Burns, in Glasby. 1977). MnO^ and F'eOOH host minor tnctals
Ihat can exist in their higher oxidalion states iti the oeean
(Co''.
Pb"^'. Ti"*'. Ce"*"*), but lack sites of the appropriate size
for the substitution of divalent eations. Tbus. buserite is only
Ibiind in nodules with high Mn. Cu. and Ni concentrations,
whereas vernadite oeeurs in nodules with high Fe and Co
eontents. Nodules from shallow water environments eonsist of
a mineral phase similar to buserite. whieh contains very low
minor element contents and is unstable when dried in air
(Calveri and Priee. in Glasby, 1977).
Regional variations
The chemical cotnposilion o^ ferromanganese nodules varies
markedly within a given ocean basin. This is best illustrated in
the Paeitic {Figure 17). where nodules that form in the
marginal areas and in east west zones north and south of
the equator, have the highest Mn/Fe ratios. The abyssal Mn-
rich nodules also have the highest Cu and Ni contents and they
eontain stable buserite as the prineipal Mn mineral phase
(Crotian. 1976; Piper and Williamson. 1977; Calvert. 1978).
The marginal Mn-rich nodules are impoverished in minor
eletnents. and consequently they contain metastable buserite.
Nodules and crusts that form in areas of the seafloor with very
low sedimentation rates and on sediment-starved elevations
abo\e Ihe abyssal seafloor (seamounts and banks) have the
lowest Mn/Fe ratios, they lack buserite and have the highest
378
IRON MANGANESE NODULES
Figure 17 Ocean-wide variation in (Al the Mn/Fe ratio and IB) the Ni
concentration in seal'loor nodules from the Pacific Ocean. From
ICronan,
1980).
Co eontents. This compositional variability reliccts the two
processes by which ferromungiinese deposits are fortned.
Origin
Hydrogciietic nodules and crusts fortn by the very slow
precipitation of seawater Mn and Fe. supplied to ocean waters
by rivers and sealloor hydrothcrmal sources, as very line-
grained or colloidal oxyhydroxides onto sohd substrates
(Halbach el<il.. 1981). The nodules and crusts formed in this
tnanner generally have smooth surfaces, and vernadite is the
principal Mn phase. They have the lowest Mn/Fe ratios and
Cu and Ni contents, and the highest Co. Pb. Ti. and Ce
contents. Diugeiwtic nodules form by the tiiore rapid
precipitation of Mn and divalent metals supplied to the
seatloor in association with settling planktonic organic matter
and recycled from the underlying sediment. Sueh nodules
generally have granular surfaces and buserite is the principal
Mn phase. They have much higher Mn/Fe ratios, and the
highest Cu and Ni contents (Calvert and Priee. 1977). These
two types of precipitates can occur in the satne nodule; in this
case a distinctly discoidal nodule from the equatorial region of
the Pacific Ocean has a smooth, vernadite-bearing upper
surface with a low Mn/Fe ratio and high Co content, and a
granular, buserite-bearing lower surface with a much higher
Mn/Fe ratio and high Cu and Ni contents growing uithiti the
surlicial sediment (Raab, 1972). Distinct cotnpositional regions
in the Pacific therefore identify environments where metals are
either supplied mainly from bottom waters or additionally
from the bottom sediment (Halbach cl iiL. 1988). Marginal
nodules with high Mn/Fc ratios have low tninor metal contents
because diagenetic reactions in the subsurface anoxie scditnent.
fueled by high burial rates oi organic matter, trap them as
sulfides or cause them to be adsorbed on panicle surfaces,
thereby preventing their recycling to the sediment surface
where they would otherwise be incorporated into surlicial
oxyhydroxides (Calvert and Price, in Glasby 1977). Lacustrine
nodules form by very similar processes and have compositions
that reflect the relative importance of metal precipitation from
lake waters and the bottom sediments (Harriss and Troup.
1970).
Summary
Ferromanganese nodules and crusts represent a comtnon type
of authigenic deposition in all oxygenated aquatic environ-
ments in all water depths. They grow very slowly by the
accretion of very fine-grained and colloidal oxyhydroxides
onto solid substrates, and often form dense pavements over
large areas of the seafloor. They can be greatly enriched in
transition metals, which are derived from norrnal ocean waters
and from sediment pore waters. Their tiiineralogies and bulk
compositions vary from ocean to ocean, within an ocean basin
and on the scale of a single nodule. There are two end-member
nodule and crust types, with all gradations between these
forms.
Hydrogenetic nodules and crusts are formed from
bottotn seawater. whereas diagenetic nodules also have a metal
source from the associated sediment via diagenetie metal
reeyeling. These processes are also responsible for the
formation of similar deposits on shallow areas oi the seafloor
and in lakes.
Stephen E. Calvert
Bibliography
ofBarnes. S.S. ;ind Dymond. J.R.. 1967, Rates o\' itccumulatiori
tcrromangiincsc luidttlcs. Nalure. 213: I21K 1219.
Btinis. R.G.. and Burns. V.M.. 1977. Mineralogy. In Glasby. G.P.
(cd.).
Marine SUtn^diicse Deposits.
EISCVILM'.
pp. 185-248.
Calvert. S.E.. 1978. Geochemistry ot oceatiic teiromangatiese deposits.
Philosophical Transcations of the Royal Society of Londoti. 290:
43-73.
CalvLTt. S.E.. and Price. N.B.. 1977a. GcocliL-mical variation irt
lerromanganese tiodtilcs and associated sediments tVotu the Pacific
Oceati, MitriiieClienii.slrv.S: 43-74.
IKONSTONES AND IRON FORMATIONS
379
Ciilvcrt. S.E,. atid Price. N.B.. I977h. Shallow water, contitiental
margin deposits: disirihulioti and geochemistry. In Glasby. G.P.
(ed.).
MaiiiH-
Man^{im-\v
DeptisiiK.
Amsterdam: Elsevier. pp. 45
86.
Cronan.
D.S,. 1^76, Manjianese tiodtilcs atid other fcrromatigancsc
oxide deposits. In Riley. J.P.. and Chester. R. (eds.). Cliemitul
OceiiiKigrtipliy.
Sati Diego. California: Academic, pp. 217 263.
Cronan.
D.S..
198(1.
Vii<lcn\(ticr Mincnils. Londoti: Academic Press.
362 p.
Cronan.
D.S,. and TliomLis. R.L,. l')7(}, Eerromanganese concretions
in Lake Ontario. Canadiati Jourtiat of Earlh Sciences. 1346-1349.
Giovanoli.
R.. and Arrhenius. G,. I9KS. Structural chemistry oftnarine
manganese and iron minerals and synthetic model compounds. In
Halbach.
P.. (Viedricii, G,. and von Staekelberg. li. (eds,}. The
Munfiauese Ncilule Bell tif ihc Ritiju- Ocean. Stuttgart. Enke,
pp.
2(1
37,
(ilasby. G.P.. (ed,). 1977, Muriiie
Maiiiiaiiesc
Dcpcsil.s.
Elsevier,
Halha'ch.
P,. Enedrich. G,. and von Stackelherg. U.. 1988. TheSUiii-
t-aiiese
Nodule
Belidj
the Faciiic
Ocean.
Stiittuart. Enke. 254 p.
M.ilbacli. P.. Sclierluig. C. Hehlsch. V,, and Marchig. V,. 1981,
Geochemical ant! mineralogical control oj dilterent genetic types oT
deep-sea nodules from the Pacific Oceati: MineruliaDeposiui. 16:
59-84.
Harriss. R,C,. and Troup. A.G,. 1970, Chetnistry and origin of
tVeshuater ferrotnanganese concretions
:
Liiiiiwtagv ciml Oceiiati-
grtiphv . 15: 702 712.
Ku.
T,L,. 1976. Rates of accretion. In Glasby. G.P. (cd.). Mayiue
\ftiii,ii(tiiesi'
Dcpiisii.^.
New York: Elsevier. pp. 249 267.
Piper. D.Z.. iitid Fowler. B. 1980.. New constraint on the iiiaintcnimce
oi"
Mn nodules at the sediment surface
:
.\anirc. 286
:
880-883.
Piper. D.Z.. and Williamson. M.E.. 1977. Regional variations in the
elemental and mineralogical composition of terromanganese
nodules from the pelagic environment of the Pacific Ocean, Mar-
ine Geology. 2i: 285 303.
Raab.
W,. 1972. Physical and chemical features of Pacific deep-sea
manganese
nod tiles
and their implications to the genesis of nodules.
In Hoin. D.R,
(cd.l.
Feyiiimtnigauc.sv Deposits un
ihe
OvcLin
Fttuir.
Washington. DC: National Science I-oiindatioii. pp. 31 49.
Cross-references
Hydro.xidcs and Oxyhydio,\ide Minerals
Oecanic Sediments
IRONSTONES AND IRON FORMATIONS
Iron-rich sedimentary rocks cotitain >15 percetit metallic iron
by weight (James. 1966) utid form a class of chemical
sedimetits comparable to evaporiles and phosphorites. Most
workers recognize two maiti categories. Iron
Formations
are
generally cherty. thinly laminated, and Precambrian in age.
ironstone.s
are generally less siliceous, more aluminous, not
hmiinated,
smaller, and Phanerozoic in age. This distinction
highlights titne-relaled changes whieh constitute one of the
most interesting aspects of iron-rich sedimentary rocks and has
impoitant itnplications for the evolution of Earth's atmo-
sphere and hydrosphere. Both types have long been sources of
iron lor hutnan use. Ironstones wctx' mined in Rotnan limes
and were a key resource in the industrial t'evokttion (Young.
1993). Mineralogical investigations of ironstone began in the
tnid-lSOfls. Studies of iron formations were begun in the late
lSOOs by geologists from the fledgling U.S. Geological Survey
in the Lake Superior region. This work resulted in classic
tiionographs focused oti specific "iron ranges" as well as Van
Hise and Leith (1911). Fresh studies triggered by the demand
for iron during World War [l yielded a new round of U.S.G.S.
publications and the seminal article by James (1954). Large
iron fortiiations and associated deposits of iron ore were still
being discovered. The subsequent discovery and study of large
iron (brmations in Western Australia have been particularly
influential in shaping current percepiions of iron formations,
even though they are exceptional in tnany ways (Trendall,
2002).
The vast majority of iron ihat is ever mined will
be
extracted
from Preeambrian iron formations or derivative deposits.
Large iron formations are eurrently tiiined all over the world.
In
200(1.
Australia produced over 160 tnillion metric tons of
iron ore worth in excess of USS2..'' billion, and China atid
Brazil produced even more. The largest ore deposits occur
where silica was leached and iron was oxidized in iron
formations during the Precatiibrian (Morris. 1987). Ores
produced by post-Precambrian weathering generally have
lower iron contents and more impurities. Iron can also be
extracted from unaltered iron formations by crushing ihem
and separatitig the magtietite.
Iton fortnations have also yielded eeonotnic quatitities ui'
other materials. Iron formations in Archean greenstone belts
host world-class gold deposits, and manganese is extracted in
large quantities from iron formations in Brazil. India, and
especially South Africa. Blue asbestos (erocidolite) was
extracted from iron formations during the mid-[900s. but
production dropped off precipitously as the highly eareino-
genie nature of blue asbestos becatne clear.
iron formations
The main iron minerals fall into four groups which James
(1954) used to define four "facies" of iron fonnation
:
oxide,
silieate, carbonate, and sulfide. In low-gtade iron formations,
the dorninant minerals are magnetite and hematite in the oxide
faeies; greenalite and tiiitinesotaite in the silicate facies: siderite
and ankerite in the carbonate facies: atid pyrite in the sulftde
facies. Where detrita! textures are not obscured by tneta-
morphism. iron formations can be sub-divided into handed
versus granular varieties. Banded iron formations or BIFs.
by far the more abundant of the two. were originally chemical
muds.
Most BIFs exhibit a quasi- to highly rhythtiiie
alternation of layers rich in iron and ehert on a scale of
millimeters to centimeters (Figure IS(A)). Granular iron
fortnations or GIFs originated as well-sorted chetnical sands
(Figure 18(B)) in which the clasts were largely derived via
intrabasinal erosion and redepositioti of pre-existing BIF.
GIFs lack the
even,
continuous banding of BIFs and generally
belong to the oxide or silicate mineral facies. whereas BIFs
show a much broader spectrum of iron minerals, including
a greater abundance of reduced facies (James. 1954; Simonson.
1985). Layers of pure GIF thicker than a few meters are rare,
whereas BIF can continue uninterritpted by GIF for tens to
hundreds of meters stratigr;tphical!y. Iron formations with a
mixture of BIF and GIF are actually more abundant than pure
GIFs,
and they show bedding that is tnore irregular than pure
BIF,
but less massive than pure GIF. In tnixcd iron
formations. GIF takes the form of discontinuous lenses that
probably represent bedforms generated by storm waves and
currents (Simonson. 1985). Chert is not used to classify iron
formalions because it is a ubiquitous eomponeni. but its
abundance helps distitiguish them from ironstoties. Another
380 IRONSTONES AND IRON FORMATIONS
Figure 18 Photomicrographs of iron-rich sedimentary rotks in plane polarized lif>ht. (A) Banded iron lorrtiation from the early Paleoproterozoic
Dales Gorge Memlier of the Brocknian Iron Forrtiation, Hamersley basin of Western Ausiralid. Clear part i5 chert pod with opaque hematite
impurities, riark parl is dominantly magnetite; note the contrast in thickness and sharpness of laminae as ihey exit the chert pod into the
surrounding B!F. Field of view is --5 mm wide. IB) CIranuliir iron formation from ihe Paleoproterozoic Sokoman Iron Formation, Labrador trough.
Peloid-like granules consist of chert and hematite with lesser amounts of magnetite; the interslices are filled with slightly recryslallized
chalcedonic cement. Field of view is -'2.'imm wide. (Cl Oolitic ironstone from the late Ordovician to early Silurian Don Braulio FormtUiun,
Sierra de VJllictJm of the Argentine Pre-Cordillera. The ooids consist of silicate, show nice pseudouniaxial extinction crosses between crossed
polarizers, and are locally truncated along dark styiolitic seams; the composition of the fine-grainetJ matrix is undetermined, but it is probably a
mixture of fine quartzose silt and chert. Field of view is -
5
mm wide.
IRONSTONFS AND IRON FORMATIONS
381
scheme subdivides iron formations into Superior-lype versus
Algoma-type, as described below. Given their great age. most
iron [brmations are tec'tonicaily deformed or rnetatnorphosed
to some degree, yet iron formations retain their distinctive
chemical cotnposition. even at high grades. The metamorph-
istii of iron formations is a complex topic in its own right
(Trendall and Morris, 1983, pp. 417-469).
Intrinsic characteristics
Despite their great antiquity, many iron formations have well-
preserved sedimentary features. Like ail arenites. GIFs consist
of a cotiibination of: (I) satid-size clasts (long referred to as
"granules. " for exatnple. Van Hise and Leith. 191 I): (2) fmer
grained material (matrix), which is relatively rare; and {3)
cetnents (interstitial, void-filling authigenie minerals)
(Dimroth. 1976). In addition, diagenetic phases replace these
primary constituents in part or whole in all iron formations.
The granules, which are analogous to carbonate peloids and
intraelasts. are mostly fine to coarse sand size (Metigel. 1973)
and range from well-rounded to angular (Figure I8(B)). Ooids
with concentrically laminated cortices are abundant locally
(Diniroth. 1976: Simonson. 1987). but much rarer than gra-
nules.
GIF's had depositional porosities of 40 percent to
50 percent, much o\' which was occluded by siliceous cement
early in diagenesis (Figure I8(B); Mengel, 1973; Simonson.
1987).
Other GIFs are heavily compacted and show tight
frameworks and distorted clasts. Many GIFs have intergra-
nular cavities, cracks, and vugs that are largely filled with
chalcedonic and drusy quartz cetnents prior to compaction.
Similar cracks and vugs arc widely developed in the strotna-
loiitic cherts found in some GIFs. These cracks and vugs
probably (triginated via post-depositional shrinkage due to the
dewatering of gelatinous precursors, and sotne contain internal
seditnent that is distributed geopetally (Simonson. 1987).
Depositional stritetures are often obscured by diagenetically
redistribitted minerals, but dunc-scaie cross-stratification is
widespread in (ilFs (Sitiionson. 19S5). The few paleocurrents
which have been measured show complex polymodal patterns
with hints of herringbone typical of shallow marine sands
(Ojakangas. 1983). Flat pebble conglomerates are a minor but
widespread component of GIFs with the pebbles generally
derived from more siliceous layers. Stromatolites that vary in
width from a centimeter to over a meter and range in
morphology frotn columnar to domal are another tninor but
distincti\e sedimentary structure found in GIFs but not BIFs.
Although they are usually interpreted as products of seditnent
trapping and/or precipitation by microbial mats, some
stromatolites in GIFs have characteristics like those of
siliceous sinters deposited in and around hot springs. Thanks
to the early silica cementation, iron formations contain some
o\'
the best-preserved early Precambrian biotas in the world
(Ttctidall and Morris. 1983, pp. 373-400).
In contrast to GIFs. BIFs generally lack any obvious
detrital textures. Fven though they are diagenetically reorga-
nized, tnany Bil-"S are still retnarkably iine-grained and
unitbrtn (Figure I8(A)). As noted above. BIFs show more
diversity oi' iron minerals than GIFs, including substantial
thicknesses ol'all four of James' facies. Most of these minerals
are thought to be derived from phases precipitated from
the original basin waters, but some BIFs eontain aluminous
silicate phases, most notably stilpnomelane. due to the
presence of volcaniclastic contaminants. Exquisite volcanic
shards replaced by stilpnomelane occur in some iron
formations (LaBerge, 1966a.b).
.As the name itnplies. most BIFs have well-developed thin
latninaiion lo thin bedding (Figure 18(A)). This is no surprise,
given the lack oi' burrowcrs in the Precambrian. but the layers
in BIFs (particularly those rich in iron oxides) can be quite
striking, and in sotne cases they can be correlaled for hundreds
of kilometers (Trendall and Blockley. 1970; Ewers and Morris,
1981).
The bedding in BIFs can also be highly cyclic via
alternations of iron-rieh and iron-poor precipitates or alter-
nations of BIF and fine shaly or volcaniclastic strata (Trendall
and Blockley. 1970; Ewers and Morris. 1981; Beukes. 1984).
Trendall (1972) attempted to relate these cycles to orbital
parameters, but they have yet to be tested for the periodicities
typical of Milankovitch forcing. Sedimentary structures other
than banding are rare in BIFs, except for chert pods
(F'igure I8(A)). whieh are best interpreted as localized pockets
of early cementation. Chert-poor BIF surrounding the pods
offers textbook examples of differential compaction following
localized early cementation (Dimroth. 1976; Beukes, 1984;
Simonson. 1987) which indicate chert-poor BIE-'s lost up to 90
percent or more of their original thickness during cotnpaction:
this is typical of muds in general. The tninerals shielded within
chert pods also indicate the origitial sediment had a range of
compositions similar to the ones shown by present-day BIFs.
Stratigraphic context and secular variation
Although they are associated with a wide variety of lithologies.
the nature of the rocks associated with iron formations
changes progressively through time. Gross (1965. 1983) used
this and other criteria lo subdivide iron formations into
Algoma-type and Superior-type. His original definitions have
evolved through time, but the modifier Superior-type is still
used for uniquely large iron formations of the late Arehean
and Paleoproterozoie that are associated with non-volcanic
sedimentary rocks. The modifier Algoma-typc is used for iron
formations that are generally smaller and occur in Archean
greenstone belts where they are associated with volcanic rocks
(Fralick and Barrett. 1995). Virtually all of the Algoma-type
iron formations are BIFs, and they run the full gamut from
oxide to sulfide facies mineralogies. In contrast, the thicker,
more laterally extensive Superior-type iron formations contain
GIFs as well as BIFs; the BIFs again show a full range of iron
mineralogies, but the GIFs aredotninated by oxide and silicate
facies rocks. Some of the Superior-type iron formations are
associated with shallow*water deposits, but many of them as
well as most of the Algoma-type iron formations are
associated with deeper water deposits.
Iron formations range from early Archean to Paleoproter-
ozoic in age and are found among the oldest sedimentary
strata on earth in the Isua supracrustal belt in western
Greenland. Iron accumulation clitnaxed during an interval of
<80(l million years in the Late Archean to Paleoptoicro/oic
and ended rather abruptly on or before 1.8Ga (Gole and
Klein. 1981; Trendal! and Morris. 1983. pp. 471 490).
Algoma-type iron fortnations were forming in North America
as the oldest Superior-type iron-formations formed in Western
Australia and South Africa, indicating a gradual transition
between the two.
A noteworthy episode of global sedimentary iron accumula-
tion also took place in the Neoproterozoic. at which time iron-
rich rocks often referred to as iron formations were deposited
382
IRONSTONES AND IRON FORMATIONS
on several continents. These late Precambrian iron fortnations
tend to be richer in iron and have a simple iron mineralogy
dominated by hematite (Beukes and Klein, 1992). Many of
these iron formations are also closely associated with glacio-
genic sediments (Young. 1988), so they may differ in origin.
Origins
and
implicafions
for
earth history
The slow rise and more rapid collapse of iron formation
deposition early in Earth's history must be linked to co-
e\olution of the atmosphere, hydrosphere, biosphere, and
lithosphere. but controversy continues over the details. Many
hypotheses championed by early workers have been
abandoned. For example, suggestions that Iron formations
originated via the replacetnent of carbonates, represent
evaporites unique to the early Precambrian. or were deposited
in lacustrine basins are no longer tenable. Sedimentological
studies have made it clear that iron formations are primary
precipitates deposited mostly in open marine settings
(Ojakangas. 1983; Beukes, 1984; Simonson, 1985). The genetic
model curtently ptesented in tnost textbooks is likewise at odds
with otir current knowledge of iron fortnations. Cloud (1973)
first suggested thai dissolved ferrous iron was ubiquitous in the
early ocean and iron fortnations formed wherever photo-
synthesizing microbes produced free oxygen, causing iron
oxides to precipitate. It was an elegant hypothesis at the time.
but we now know the iron minerals in iron formations are
much more than jusf oxides and the iron concentrations in
carbonate sediments contemporaneous with iron formations
are not high enough to fit this scenario (Veizer cl uL. 1990,
1992;
Simonson and Hassier. 1996)
It is now widely believed that hydrothermal sources in the
deep ocean supplied the iron to make iron formations. BIF
deposition has been linked to hydrothermal activity with
reasonably high confidence via strafigraphic context and facies
relationships for some Algoma-type iron formations, and all
types of iron formations show various geochemical signatttres
of hydrothcrmal sources (Klein and Beukes, 1992). It is also
widely believed that iron formations precipitated along
chetnoclines in a stratified ocean (Beukes and Klein. 1992;
Sitnonson and Hassier, 1996), which would e.xpiain why iron
formations occur in many different teefonic settings and in
association with many difierent rock types (Gross, 1983;
Fralick and Barrett, 1995). Isley (1995) outlined a plausible
scenario whereby iron-rich plumes released from deep-sea
hydrothermal systems could travel to distant depocenters. The
main point of contention is the mechanism(s) responsible for
precipitating such great quantities of iron undiluted by other
types of clastic or chemical sediment. Recent isotopic work
indicates microbes played a role in the precipitation and/or
diagenetic reorganization of the iron (Beard eial.. 1999).
Consensus has also been reached on the cause of the shift
from Algoma- to Superior-type iron formations through time.
The appearance of large iron-formations probably reflects the
first developtnent of extensive shelf environmetits in the Late
Archean (Eriksson. 1995). This is consistent with the
simultaneous appearance of laterally extensive exatnples of
other stable-shelf deposits such the first large carbonate
platforms (Grotzinger, 1994). Age differences among the
largest iron formations on different continents could reflect
the fact that the "cratonization" of shields was a highly
diachronous process (Eriksson and Donaldson. 1986). The
smaller size of Algoma-type iron formations is generally
believed to signal smaller depositional environtnents, but Gole
and Klein (1981) point out thai they are more highly deformed
and some "may have been quite extensive prior to deformation
and disruption." The growth in the size of iron formations
through time cottld also involve changes in the bathymetry
and/or chemistry of the world ocean that permitted the
dispersal of iron over longer distances. However, this would
require an increase in the tnobility of dissolved
iron,
which
runs counter to the prevailing view that the concentration of
free oxygen in the atmosphere was increasing at that time
(Eriksson, 1995). Even within the Superior-type category of
iron lbrmations. secular changes are evident. The oldest
Superior-type iron fortnations are almost pure BIF. whereas
GIF's reached their acme around 2.0Ga. This suggests the
locus of iron formation deposition shifted into progressively
shallower environments, which may signal a progressive
shallowing of a global chemocline through time.
The abrupt end to the deposition of iron formations in the
Paleoproterozoic is presumably related more to atmospheric
and hydrospheric than lithospheric evolution. Many lines of
evidence signal a dramatic rise in atmospheric oxygen around
2.2Ga to !.9Ga (Hollatid. 1994). which at sotne point
ventilated the world ocean sufftciently well to prevent the
storage and long distance transport of large quantities of
dissolved
iron.
However, the disappearance of iron formations
around 1.8Ga may have actually been a result of the deep
oceans becoming sulfidic rather than oxic (Anbar and Knoll,
2002). In either case a brief reappearance by iron formations
late in the Precambrian tnay signal a return to these earlier
conditions. The Neoproterozoic glaciations were so extretne
that Hoffman elal. (1998) have argued they gave rise to a
"snowball earth." Perhaps the ice capped the world ocean in
such a way that it could become highly stratified for the lirst
time in over an eon, thereby reactivating some of the
mechanisms at work in the Paleoproterozoic. Like the older
iron formations, the source of the iron for the Neoproterozoic
iron formations appears to have been hydrothermal (Young.
1988).
Iron formalions also differ frotn younger iron-rich
sediments in their high silica content. Various lines of evidence
including the cement-filled cracks and vugs in GIFs and the
apparent syneresis of the original sediment indicate the silica in
iron formation is a primary component. Given the absence of
silica-secreting organisms, higher silica concentrations should
have been present in Precambrian seawater to co-precipitatc
with the
iron,
but again the precise tnechanism for this is
unclear. On average. GIFs are probably tnorc siliceous than
BIFs.
and oolitic and stromatolitic layers probably have the
lowest iron contents of
all.
This is consistent with a deep-ocean
source of
iron,
but it could also reflect a greater abundance of
silica cement in GIFs.
Ironstones
The tenn ironstones encompasses a diverse group of rocks
ranging in age from Protcrozoic to Pliocene (Van Houton.
2000) which consist of materials ranging in size from sand to
mud.
Ironstones are one of several types of iron-rich
sedimentary rocks formed during the Phanerozoic. The latter
includes exhalativc sediments formed directly by hydrothermal
activity and "green sands." which are clastic seditncnts rich in
glauconitic silicates. Such deposits have traditionally been
differentiated from ironstones and that practice is followed
IRONSTONES AND IRON FORMATIONS
383
here,
even though glauconitic sediments and ironstones have
much in common (Van Houton and Purucker, 1984; Odin,
1988). Given their young geologic age. ironstones are much
less likely to be deformed or metamorphosed than iron
fortnations and comtnonly contain fossil debris. This has
permitted detailed analyzes of the depositional setting of
iron-
stoties
(e.g..
Hunter, 1970; Van Houton and Bhattacharyya.
1982). bitt consensus on the specific mechanisms and timing of
iron precipitation has retnained elusive.
Intrinsic characteristics
Ironstones cotitain representatives of all four of James" (1954)
tnineral facies. but differ from iron formations in some ways.
Most ironstones consist of oxides and/or silicates, although
siderite is also common in some (Odin. 1988). This is similar to
the assemblages found in GIFs, but more oxidized than iron
formations as a whole, given the widespread occurrence of
reduced facies in BIFs. The oxides and silicates found in
ironstones versus iron formations also differ systematically.
Hetnatite is found in both, but gocthitc is only comnmn in
ironstones whereas magnetite is only common in iron
fortnations. Likewise, chamosite and berthierine arc the chief
iron-rich silicates in ironstones but rare in iron formations. In
addition to iron-rich phases, ironstones typically contain
quartz sand, calcite, dolomite, and authigenic phosphorite
and chert. All of these phases are rare in iron formations with
the exception of chert. These tnineralogical differences arc
retlecJed in lower iron and silica and higher alutnina contents
in ironstone versus iron fonnation. which is partly, but not
entirely, a rellection of the age disparity between the two (sec
belov\).
Ooids arc a distinctive feature of many ironstones
(Figure 18(C)). but they are rarely its main constituent. Like
those in iron formations, the ooids in ironstones consist of
a nucleus plus a cortex with thin concentric laminations; no
radial textures have been reported from ooids in cither iron
formations or ironstones. Ironstone ooids consist of iron
silicates and/or iron oxides: these two minerals occur in
alternating laminae in some ooids. The nuici of the ooids are
diverse in composition, rangitig from calcareous fossil debris
to detrital quartz sand grains to fragments of oolitie cortices.
Replacetnent textures are commonly observed in the ooids.
involving both iron-rich and iron-poor materials (including
fossils) as both replacer and rcplacee. Despite the attention
lavished on the ooids, other constituents probably form the
bulk of ironstones. One common constituent is interstitial
calcite or dolomite ccmcnl. but the most abundant arc
probably fitiely crystalline iron oxides, iron-rich silicates, and/
or siderite. Much of this finer tnaterial is authigenic rather than
detrital in origin (Odin. 1988). Many oolitic ironstones are
cross-bedded, a sure sign they originated as well-sorted sands,
but other ironstones are highly bioturbated. and show evidence
of hardground-like induration in the form of local sediment
reworked as intraelasts (Odin. 19S8).
Strafigraphic context
and
secular variation
Most oolitic ironstones take the form of thin interbeds amidst
successions of clastic and carbonate strata deposited in shallow
marine environments. These are typically sediments deposited
close to paleoeoastlines in water on the order of a few tens of
meters deep. In general., ironstones are Ibrmed closer to shore
and in shallower water than time-equivalent glaueonitic
deposits
(e.g..
Hunter. 1970). but the abundant remains of
nonnal marine organisms and the evidence of repeated
reworking indicate ironstones accumulated on the open
sealloor rather than in restricted settings. Nevertheless, most
workers believe ironstones did not accumulate in the highest
energy environments, which is consistent with their common
association with clastic mudroeks. Ironstones also tend to
occur at stratigraphic breaks which coincide with trans-
gressions (Van Houton and Bhattacharyya, 1982). Essentially,
ironstones are now believed to constitute condensed sequences
formed largely during times of ma.ximum transgression and
reduced clastic inllux. This is also true oti a grander scale: the
two tnain pulses of ironstone accutnulation coincide with
Fischer's two '"greenhouse" phases of sea-level highstand frotn
the Ordovician to the Devonian and from the Jurassic to the
Paleogene (Van Houton and Arthur, 1989).
Ironstones also vary in mineralogical composition as a
function of age. Goethite and berthierine are both relatively
comtnon in the Mcsozoic to Tertiary ironstones, whereas
hematite and chamosite dotninate the Paleozoic ironstones.
This contrast probably reflects a higher degree of diagenetic
equilibration by the older ironstones.
Origins
and
implicafions
for
earfh history
The fundamental problem in the origin of ironstones is
reconciling the evidence for deposition under open-marine,
oxidizing conditions with the fact thai dissolved iron is only
mobile under reducing conditions (Odin. 1988). This leaves
three options: (I) ironstones were originally deposited as
something entirely different and replaced by iron-rich phases
after burial: (2) conditions fluctuated between oxidizing and
reducing during deposition; or (3) high concentrations of iron
somehow managed lo accumulate under oxidizing conditions.
Sotne of the earliest in\cstigators took the presence of ooids in
ironstones to mean that they were diagcneticalK replaced
carbonates. Fven though carbonate fossils are locally replaced
by iron-rich minerals, this interpretation has fallen out of favor
due to the evidence of synseditnentary reworking of various
types of ironstone and the apparent lack of any units of
carbonate oolite that are half-converted to oolitic ironstone.
Faced with a lack of sedimentological evidence for bottom
water anoxia during deposition, most workers now envision
ironstones as forming under a fully oxidized water column.
This is clearly happening today in the case of glauconitic
sediments (Van Houton and Purucker. 1984). and thin oolitic
coatings of chamosite have been reported from pellets on the
modern seafloor at one site (Rohrlich etal... 1969). The latter
are well below wave base, which suggests the coatings may be
able to grow in .situ without benclit of regular physical
movement. Some researchers have even proposed the ooids
originated entirely in siiu. but the widespread breakage and
regrowth of the ooids argue against this. Other workers have
suggested they are pedogenic ooids that were transported then
redcposited on the seafloor, but the highly concentrated nature
of ironstones and their stratigraphic context makes this seem
highly unlikely also. The current thinking is that ironstones
originally accumulated mainly as ferric materials, then were
modified considerably via early diagenetic authigencsis of the
iron-rich phases very close to the sediment water interf^ice
(Odin,
1988).
384
IRONSTONES AND IRON FORMATIONS
In summary, ironstones can probably originate in different
ways,
but mosf formed more or less where they were ultimately
buried via a process of slow iron precipitation during times
of highly reduced seditnentation rates. Proximity to a river
mouth is also thought to be advantageous because the source
of the iron is generally attributed to deep weathering on
adjacent landmasses. However, this is somewhat at odds with
the temporal correlation of ironstone and black shale
deposition (Van Houton and Arthur. 1989). Whatever the
source of the iron and the mechanism of deposition, both the
iron-rich and iron-poor phases in ironstones were clearly
affected by a host of early diagenetic transformations; many
took place so close to the sediment/walcr interface that new
phases were detritally reworked. In conclusion, the separation
of iron formations and ironstones into two separate categories
seems to be quite justifiable, but a closer comparison between
these two might be very instructive in terms of revealing what
processes were taking place on or near the early Precambrian
seafloor.
Bruce M. Simonson
Bibliography
Aiihar. A.D., iiiid Knoll. A.H..
21J02.
Proterozoic ocean chemistry
and
evolution:
A
hioirtoryanic hridge?
Science.
297
:
11.^7 -1192.
Beard.
B.L..
Johnson.
CM.. Cox. L., Sun. H..
Nealsoti.
K.H.. and
Agtiitar,
C. 1999.
Iron isotope biosienatures.
.Science.
285:
1889
1892.
Beukes.
N.J..
1984. Sedimentology
of the
Kuruman atid Griquatown
Iron-formations. Transvaal Stipcrgroap. Griquakind West. SoiHh
Africa.
Precimthrktn
Reseurch.
24:
47-84.
Beukes,
N.J.. and
Klein.
C. 1992.
Models
for
iron-formalion
deposition.
In
Schopf,
J.W.. ;ind
Klein.
C
(eds.).
The Prolerozoie
Biiisphere.
.-)
Mitllitlisciplinarv
Situly.
Cambridge University Press.
pp.
147-151.
Cloud.
P.. 1973.
Paleoecological signiticiincc
of the
banded
iron-
formalion.
Economic Geoliifry.
68: 1135 1143.
Dimroth,
E.,
1976. Aspects
of
the
sedimentary petrology
of
eherly iroti
formation.
In
Wolf.
K.H.
(ed.), Hantlhook
cj
Siratu-Botind itnd
Slniliffirm
Ore
Depo.siis,
7.
Elsevier.
pp. 203 254.
Eriksson.
K.A.. 1995.
Crustal giowtli. surface processes
and
alnio-
spheric evolution
of the
eiirly Earth.
In
Coward,
M.P.. and
Ries.
A.C.
(eds.), Etirly
Precunihriun
Processes.
Geological .Society
of London. Special Puhlictttion,
95. pp.
11
-25.
Eriksson.
K.A..
atid Dotialdson.
J.A. 1986.
Basin;tl aitd shelf
sedimentation
in
reiaiion
to the
Archaean-Prolerozoic boundary.
Preccimhniin
Reseairh.
33:
103-121.
Ewers.
W.E.,
atid Morris.
R.C.. 1981.
Studies
of the
Dales Gorge
Member
of the
Brockman Iron Fortnation. Western Aiistralia.
EeonomicGeology.
76:
1929-1953.
Fralick.
P.W.. and
Barrett.
T.J., 1995.
Depositional controls
on
iron fortnalion associations
in
Canada.
In
Plint.
A.G.
(ed.),
Seclimeiiiiiry
Em ies
Analysis. Internatiomil Association
of
Sedimentologisi, Special Publication,
22. pp. 137 156.
Gole.
M.J.. and
Klein.
C. 1981.
Banded iron-formations through
much
ol'
Pieeambrian time, .louvnal of
Geology.
89
:
!69 183.
Gross.
G.A..
1965.
Geology of Iron Depositsoj
Ciinuda.ihlunie
I.
General
GeologyundEnilualionaflron
Deposits.
Ottawa: Geological Survey
of Canada. Economic Geology. Report
22.
Gross.
G.A., 1983.
Tectonic systems
and the
deposition
of
iron-
formation.
Preitnnhyiein
Reseiireh.
20:
171-187.
Grotzinger.
J.P.. 1994.
Trends
in
Precambrian carbonate sediments
and their implications
lor
undetstanding evolution.
In
Bcngsion.
S.
(ed.).
Early Li/eon
Eitrlh.
Nobel
Symposiiii)i,S4.
Columbia Ihiiversity
Press,
pp.
245-25S.
Hoffman.
P.F..
Kaufman.
A.J..
Ilalverson.
CJ.P..
and
Schrag.
D.P..
1998.
A
Neoprolero/oic snowball earth.
.Seieiiee.
281
:
1.342-1.346.
Holland.
H.D.. 1994.
Early Proierozoic atmospheric change.
In
Bengston.
S.
(ed.). Early Life
on
Eurlh.
Nobel Symposium
No.
S4.
Columbia University Press,
pp. 237 244.
Hunter.
R.E.. 1970.
Facies
of
iron sedimentation
in ihe
Clinton
Group.
In
Fisher.
G.W..
Pettijolin.
F.J..
Reed.
J.C. Jr.. and
Weaver,
K.N.
(eds.).
Siiulies
of
Appalaehian
Geology:
Cenlnilanil
Southern.
Wiley,
pp. iOl 124.
Isley,
A.E.. 1995.
Hvdrothermal plumes
and
tlie delivery
of
iron
lo
banded iron fortnalion. Jourmil of
Geology.
103:
169-185.
James.
H.L., 1954.
Sedimentary facies
of
iron-formalion. Economic
Geology.
49: 235 293.
James.
H.L.. 1966.
Chemistry
of the
Iron-Rich Sedimentary Rocks.
L^nileil
Slates Geological Survey Professional Paper
440-W.
l.aBerge.
G.L..
I96(ia. Altered pyroclastic rocks
in
iron-formation
in
the MamerslL-y Range, Western Australia.
Economic
Geology.
61 :
147-161.
LaBerge.
G.L..
1966b. Pyroclaslic rocks
in
South African
iron-
formations.
£•((»((»»«•
Ctvi/(',t,'r. 61
: 572 581.
Klein.
C. and
Beukes.
N.J.. 1992.
Proterozoie iron fortnations.
In Condic.
K.C.
(ed.).
Prolerozoie
Crusia! Evolution. Elsevier.
pp.
383 418.
Mengel,
J.T., 1973.
Physical sedimentation
in
Precatnbrian cherty
iron fomiaiions
of the
Lake Superior type.
In
Amstutz,
G.C..
and Bernard.
A.J.
(eds.). Ores
in
Sedintfiits. Springer-Verlag,
pp.
179 193.
Morris.
R.C.. 1987.
Iron ores derived
by
enrichment
of
banded
iroti-
formation.
In
Hein.
J.R.
(ed.).
Siliceous Sedimentary Rock-hosled
Ores and
Pelroleiini.
Van
Nostrand Reinhold
Co.. pp.
231
267.
Odin.
G.S.
(ed.).
19H8.
Green Marine Clays. Developments
in
Seditnentology
45.
Elsevier.
Ojakangas.
R W..
1983.
Tidal deposits
in
the early Protero/oic basiti
of
the Lake Superior region tlie Palms and Pokegama I-ormaiions:
evidence
for
subtidal shelf deposition
of
Superior-type banded
iron-formation.
In
Medaris.
L.G. Jr.
(ed.). Early Proteroroic
Geology
of
the Great Eakes
Region.
Geological Socieiy
of
Atnerica
Memoir.
60. pp. 49 66.
Rohrlicli.
V,.
Price.
N.B . and
Calveri.
S.H., 1969.
Chaniosiie
in
tlie
Recen!
scdimenls
of
Loch Etive. Scotland.
Journal
of Sedimentary
Petrology.
39: 624 631.
Simonson.
B.M..
1985. Sedimentoiogieai constraints
on ihe
origins
of
Precambrian iron-i'ormiitions. Geological Society
of
Amereriai
Bulletin.
96:
244-252.
Sitnonson.
B.M., 1987.
Early silica cementation
and
subsequent
diagenesis
in
arenites from four early Proterozoie iron forma-
tions
ol"
Nortli America. Jotirnal
of
Sedimentary
Petrology.
57:
494
511.
Simonson.
B.M.. and
Hassier.
S.W..
1996. Was the depositioti oflarge
Precambrian iron fortnations linked
to
major marine trans-
gression.'
Journal oj
Geology.
104: 665 676.
Trendall.
A.F.. 1972.
Revolution
in
earth history. Journal
t)f
the
Geological Society
ojAuslralia. 19
:
287-31!.
Trendall.
A.F..
2002.
The
significance
of
iron-form at ion
in the
Precambrian stratigraphic record.
In
Altermann.
W.. and
Corcoran.
P.L.
(eds.). Precanthrian Sedimentary Environments:
A Modern
Approach
to
Ancient
Deposilional
.Systems.
International
Association
of
Sedimeniolosiists. Special Piiblicatiim.
33. pp.
33
66.
Trendall.
A.F.. and
Blockley.
J.G.. 1970. The
lion Formations
of
the Precambrian Hamersley Group. Western Australia. Perth:
Geological Survey
oji{['.\lern
.Australia
Bulletin.
119.
Trendall.
A.F.. and
Morris.
R.C.
(eds.).
1983.
Iron-Formaiion. Facts
and
Prohlems.
Elsevier.
Van Misc.
C.R.. and
Leith,
C.K..
1911.
Geology
of
the Lake Superior
Region.
United Slates Geological Survey Monograph
52.
Van Houton.
F\B.,
2000. Ooidal ironstones
and
phosphorites—a
comparison from
a
stratigraplier's view.
In
Glenn.
C.R..
Prevol-
Lucas.
L . and
Lucas. J. (eds.).
Marine
Authigenesis:
FrontGlohalto
Mierobial. Society
of
Economic Paleontologists
and
Mineralogists.
Special Publication. 44.
pp. 79 106.
Van Houton.
F.B.. and
Arthur.
M.A.. 1989.
Temporal patterns
among Phanerozoic oolitic ironslones
and
oceanic anoxia.
In
Young.
T.P.. and
Taylor. W.E.G. (eds.).
Phanerozoic
Ironslones.
Geoloiiical Society
of
London, Special Publication,
46, pp.