Clay Minerals (2001) 36, 447–464
The nature and origin of non-marine
10 Å clay from the Late Eocene and
Early Oligocene of the Isle of Wight
(Hampshire Basin), UK
J. M. HUGGETT1,2,*, A. S. GALE2,1
AND
N. CLAUER3
1
School of Earth Sciences, University of Greenwich, Grenville Building, Central Parade, Chatham Maritime,
Chatham, Kent ME4 4AW, UK, 2 Department of Palaeontology, Natural History Museum, Cromwell Road, London
SW7 5BD, UK, and 3 Centre de Géochimie de la Surface (EOST, CNRS-ULP), 1 rue Blessig, 67084, Strasbourg Cedex,
France
(Received 29 November 1999; revised 11 April 2001)
AB ST R ACT : Variegated palaeosols, which formed from weathering of clays, silts and brackish to
freshwater limestones, are present in the Late Eocene–Early Oligocene Solent Group of the
Hampshire Basin, southern UK. The composition and origin of the clay in three segments of the
lower part of the Solent Group have been investigated by X-ray diffraction, microprobe analysis,
inductively coupled plasma-mas spectrometry, K/Ar dating, high resolution scanning electron
microscopy, analytical transmission electron microscopy and wet chemistry. The detrital clay
mineral suite is dominated by illite and smectite with minor kaolinite and chlorite. Seasonal wetting
and drying in gley soils has resulted in replacement of smectite by Fe-rich, or illite-rich illitesmectite. Illite has also formed with gypsum and calcite in ephemeral hypersaline alkaline lakes that
periodically dried out. This illite may have precipitated directly from solution. X-ray diffraction data
and probe analyses indicate that the neoformed illite is Fe-rich. The K and Fe for the illitization are
thought to be derived from weathered glauconite reworked from the underlying Bracklesham Group
and Barton Beds.
KEYWORDS: non-marine clay, illite, smectite, XRD, EPMA, ICP-MS, SEM, TEM, K-Ar dating, Isle of
Wight, UK.
The Late Eocene–Early Oligocene Solent Group of
the Isle of Wight has been the subject of diverse
palaeontological research (plant macrofossils:
Collinson, 1983; ostracods, Keen 1977, 1978;
mammals, Hooker, 1992; dinoflagellates,
Liengjarren et al., 1980) and palaeoenvironmental
analysis (Daley, 1972, 1973). These studies indicate
that deposition occurred in a brackish to freshwater
setting with occasional marine incursions. Brief
mention of the clays, which make up the bulk of the
sediment was made by Gilkes (1968).
* E-mail: [email protected]
The data presented here are part of an ongoing
multidisciplinary study of the Solent Group. Huggett
& Gale (1997) demonstrated that the green clay of
the underlying Barton Group is reworked, finely
disseminated glauconite. Olive-green clays at the
base of the Solent Group are marine or brackish and
contain fine silt-size fragments of variably weathered
glauconite (our unpublished data) but through most
of the Solent Group the green clays are non-marine
and have a blue tinge, quite unlike any of the
glaucony-bearing sediments in the earlier Eocene
sediments of the Isle of Wight. Some of these bluegreen clays weather to brown in a few weeks; others
retain their colour.
# 2001 The Mineralogical Society
448
J. M. Huggett et al.
GEOLOGICAL BACKGROUND
The Solent Group on the Isle of Wight (Fig. 1)
comprises ~250 m of clays, silty clays, silty
calcareous clays, limestones and a few thin units
of fine sand. Faunal, floral and palaeoenvironmental
evidence (Daley, 1972, 1973; Keen 1977, 1978;
Liengjarren et al., 1980; Collinson, 1983; Hooker,
1992; Armenteros et al., 1997) indicates that these
sediments were deposited in quasi-marine, brackish
and lacustrine environments in and around the
East–West proto-Solent estuary, during the Late
Eocene (Priabonian) and Early Oligocene
(Stampian). Tectonic inversion of the Isle of
Wight–Purbeck structure during the Middle and
Late Eocene (Lutetian–Bartonian) resulted in the
uplift of chalk hills in the southern part of the Isle
of Wight, which, together with the overlying
Palaeocene and Eocene sediments, were undergoing
active erosion during deposition of the Solent
Group (Gale et al., 1999).
The Headon Hill Formation (the oldest part of
the Solent Group) is 90 m thick in Whitecliff Bay,
and was described in considerable detail by Forbes
(1853) and Bristow et al. (1889). These primarily
lithological and palaeontological accounts formed
the basis of the widely read description in the
Geological Survey Memoir by White (1921). The
lithostratigraphy of the Headon Hill Formation was
revised extensively by Daley & Edwards (1974),
who identified eight members within the Formation
(Totland Bay, Colwell Bay, Cliff End, Lacey’s
Farm, Fishbourne, Osbourne, Seagrove Bay). The
succession is conspicuously cyclic; each cycle
includes: (1) laminated, shelly silts and silty clays
deposited in quasi-marine, brackish and fluviatile
environments, overlain by (2) units including
numerous palaeosols developed in freshwater silts
and silty clays. Four such cycles are present. The
numerous palaeosols are characteristically mottled
in bright green, red, yellow and brown hues and can
be described as gleys and pseudogleys (J. Skipper,
pers. comm.). These palaeosols are concentrated in
the Totland Bay, upper Colwell Bay, Lacey’s Farm,
Osbourne and Seagrove Bay Members.
The Bembridge Limestone Formation (8 m),
which succeeds the Headon Hill Formation, is
made up of discrete beds of freshwater limestones,
several of which are capped by palaeosols,
alternating with beds of brackish and freshwater
clay, deposited in palustrine and lacustrine environments (Armenteros et al., 1997).
MATERIALS AND METHODS
Samples were taken at 0.2–1 m intervals from a
10 m thick interval extending from the Colwell Bay
Member to the lower part of the Cliff End Member
(Figs 2, 3), a 5 m thick interval of the Lacy’s Farm
Member (Figs 2, 4) and a 3 m thick interval of the
Bembridge Limestone Formation (Figs 2, 5). X-ray
diffraction (XRD) analysis of the clay fraction
FIG. 1. Location map showing the Isle of Wight and the position of Whitecliff Bay.
449
Isle of Wight non-marine green clays
FIG. 2. Late Eocene and Early Oligocene stratigraphy
at Whitecliff Bay, with the stratigraphic positions of
the sampled intervals marked.
(<2 mm) was carried out on every sample using a
Philips 1820 automated X-ray diffractometer with
Cu-Ka radiation. Air-dried samples were scanned
after spraying with glycol and heating for 2 h at
400 and 5508C. Peak area weightings for semiquantification were determined using known
mixtures of standard clays. Limestones were first
leached with 30% acetic acid and the <2 mm
fraction of the resulting insoluble residue was
analysed. Inductively coupled plasma-mass spectrometry (ICP-MS) analysis was carried out on the
<0.5 mm and <2 mm fractions of five illite-rich
samples. Only from the sample from the Bembridge
Limestone Formation (44 m) and the Colwell Bay
Member sample from 96.9 m was it possible to
extract sufficient mass of the <0.2 mm fraction to
analyse by ICP-MS. Calcareous samples were
leached with 30% acetic acid prior to size
fractionation.
One sample (from 44 m) was re-analysed by
ICP-MS after removal of Fe oxides and hydroxides
using the method described by Stucki (1981). As
this showed a decrease in Fe content of 0.1%, all
further analyses were carried out without Fe oxide/
hydroxide removal. K-Ar dating was performed on
the <0.5 and 0.2 mm fractions of five illite-rich
samples. Using a procedure similar to that described
by Bonhomme et al. (1975). Fe(II) determinations
were made, using Stucki’s (1981) method on six
<0.5 and 0.2 mm fractions of the illite-rich samples
used for ICP-MS and K-Ar dating. Carbon-coated
polished blocks of selected illite-rich and illite-poor
samples were examined using back-scattered electron imaging (BSEI) and gold-coated fracture
surfaces using secondary electron imaging (SEI)
in a Jeol 3510LV scanning electron microscope
(SEM) with Oxford Instruments ISIS microanalysis
software. Analytical transmission electron microscopy (ATEM) was used to obtain quantitative
energy dispersive X-ray analyses of single illite
particles and to check for K-feldspar, prior to K-Ar
dating. The TEM was a Philips 200 kV instrument
with Oxford Instruments ISIS thin-film analytical
software. The thin-film analytical software calculates analyses as normalized totals. In order to
compare our data with published data, we have
therefore recalculated the published data to 100%.
Assistance with the description and interpretation of
the soils was provided by J. Skipper (Natural
History Museum, London).
RESULTS
Sedimentology
Sedimentological logs, together with detailed
descriptions of the sediments, our interpretation of
the sedimentary environments and XRD data from
analyses of the <2 mm clay fraction are presented in
Figs 3 5. Stages of palaeosol development are
based on Retallack (1990).
Lake sediments include green clays associated
with gley soils and biogenic limestones, exemplified by the Lacey’s Farm Member of the Headon
Hill Formation and the Bembridge Limestone
Formation. Caliche formation on limestone indicates exposure (Fig. 5), while pseudomorphs of
calcite after gypsum in clays indicate desiccation of
lake margins at times of marine influence (cf.
Armenteros et al., 1998), and possibly total
desiccation of lakes. Gastropods of the pulmonate
FIG. 3. Lithological log for part of the Colwell Bay and Cliff End Members of the Headon Hill Formation with semi-quantitative XRD data plotted as bars, a
description of the sediment, and an interpretation of the environments of deposition.
450
J. M. Huggett et al.
Isle of Wight non-marine green clays
451
FIG. 4. Lithological log for part of the Lacy’s Farm Member of the Headon Hill Formation with semi-quantitative
XRD data plotted as bars, a description of the sediment, and an interpretation of the environments of deposition.
family Lymnaeidae are abundant in the lake
sediments and may indicate that the lakes were
ephemeral because pulmonates are air-breathing
snails (Paul, 1989).
Brackish sediments are identified on the basis of
their fauna and are found in the lower part of the
Colwell Bay Member and at the base of the Cliff
End Member. These diverse sediments include
laminated silts and sands, and bioturbated shelly,
sandy clays; both facies contain typical brackish
water gastropods and bivalves. The sediments were
deposited in estuarine channels and intertidal
lagoons.
The soils are predominantly gleys, very wet or
waterlogged soils in which anaerobic conditions
prevail. Bacteria proliferate and utilize dissolved
organic matter and ferric iron to respire. The ferric
iron is microbially reduced to the soluble ferrous
state, and forms blue-green ferrous salts, most
commonly ‘green rust’, an Fe(II)-Fe(III) hydroxy
carbonate (Newman, 1987). Most of the Headon
Hill Formation gleys later underwent seasonal
wetting and drying. In those parts of the profile
which periodically dry out, allowing air to
penetrate, the ferrous iron is redeposited as ferric
iron. Because oxygen diffuses more rapidly through
roots and fractures than through water-saturated
soil, ferric oxides are commonly concentrated in
these features, resulting in colour mottling. Pseudogleys (surface water gleys) are also present, in
which gleying occurs when water accumulates over
the soil, and is therefore most intense at the top of
the soil profile, the reverse of true gleys. The most
intense oxidation (red mottling) is characteristic of
pseudo-gleys: less intense oxidation results in
brown and orange mottling, and is characteristic
of gleys which have undergone some seasonal
drying out. In gleys, some pyritization of roots has
occurred. Slickensiding, which is widespread in
both the gley and pseudogley soils, is caused by
movement of peds (a block or crumb of soil) during
wetting and drying (Fitzpatrick, 1980). Calcite
nodules and gypsum crystals formed during
periods when evaporation exceeded precipitation
452
J. M. Huggett et al.
FIG. 5. Lithological log for part of the Seagrove Member of the Headon Hill Formation and the Bembridge
Limestone Formation with semi-quantitative XRD data plotted as bars, a description of the sediment, and an
interpretation of the environments of deposition.
(Ollier & Pain, 1996). Illutriation (accumulation of
clay layers and drapes) has occurred between 96.5
and 97.4 m. The A and B horizons are most often
preserved, the C horizon less often. The O horizon
is absent in all cases. Most of the soils show
evidence of having been through more than one
stage of development, i.e gleying followed by
seasonal drying out or vice versa.
XRD
In lacustrine, mixed-salinity and permanently
water-logged gley environments, the clay assemblage comprises roughly equal proportions of illite
and smectite, with minor kaolinite and rare chlorite
(Figs 3, 4). These sediments are brown, orange, red
or blue-green and are frequently mottled. The bluegreen sediments change to orange in a few weeks.
Clay assemblages dominated by illite, or illite with
illite-smectite, are green, almost without exception
(stable during the time since sampling which is up
to 5 y) with some colour mottled, slickensided and
calcareous (either nodular calcite or bioclasts) gley
or pseudogley palaeosols (Figs 3 5). Illite-rich
clay is found in the A horizon of gleys, less
often, as at 96.8 m, in the Bt horizon, or in
laminated lake sediment at 98 m (Fig. 3). In the
lacustrine limestone and clay of the Bembridge
Limestone the clay fraction is almost pure illite
with very minor kaolinite, smectite and illitesmectite (Fig. 5). The only illite-rich clay fraction
which is not from a palaeosol occurs in the
laminated brackish sediments between 97.3 and
98.7 m (Fig. 3), and this illite is inferred to have
been reworked from the major soil (only partly
shown in Fig. 3) beneath. Whole-rock XRD
analysis of the acid-insoluble residue of claystone
samples shows that the proportion of illite in the
bulk-rock clastic fraction correlates positively with
the proportion of illite in the clay fraction
(Table 1, Fig. 6). Apart from clays, the wholerock mineralogy is predominantly quartz, with
occasional K-feldspar, ankerite, siderite, pyrite,
and in the marl from the Bembridge Limestone
limestone
marl
limestone
?
?
?
claystone
"
"
"
42.8
44.0
44.8
70.2
70.6
75.0
90.0
92.4
93.3
96.9
Sample Lithology
depth (m)
82
66
86
66
95
96
96
70
98
87
48
36
59
60
59
68
58
48
78
42
6
4
7
1
0
3
0
0
0
0
20
18
5
11
24
0
0
0
0
0
9
0
3
3
3
12
0
trace
15
0
13
24
25
15
10
8
42
23
7
45
0
10
0
10
0
0
0
0
0
0
0
0
0
0
0
6
0
0
0
0
% of <2 mm fraction Illite Kaolinite Chlorite Smectite Quartz K-feldspar Albite
which is illite
and mica
%
%
%
%
%
%
%
0
2
0
0
0
0
0
0
0
0
%
0
2
0
0
0
0
0
0
0
0
%
3
3
0
0
4
3
0
0
0
trace
%
0
0
0
0
0
0
0
29
0
0
%
0
0
0
0
0
0
0
0
0
trace
%
Ankerite Dolomite Pyrite Gypsum Anhydrite
TABLE 1. Whole-rock XRD analyses of insoluble residues, with % illite in the clay fraction shown for comparison.
Isle of Wight non-marine green clays
453
454
J. M. Huggett et al.
FIG. 6. Relationship between the percentage of illite in
the whole clastic fraction (carbonates omitted from the
semi-quantitative analysis of the XRD data) and the
percentage of illite in the <2 mm fraction.
Formation, calcite pseudomorphs after gypsum are
present.
The 001/002 intensity ratio of mica was
measured on all samples. This provides an
indication of Fe content as Fe-rich micaceous clay
has a weak 002 reflection. Values typically range
from 1.48 to 7.8 and increase with the percentage of
illite in the clay fraction (Fig. 7). The value
measured for illite-rich marl from 44 m is
exceptionally high, at 12.1. In the illite-dominated
assemblages, the illite 001 reflection is typically
broad (0.7 0.982y) for samples with >80% illite in
the <2 mm fraction), indicating the presence of
illite-rich mixed-layer clay. This mix of illite and
illite-rich illite-smectite is hereafter referred to as
illite-dominated or illitic. In mixed-clay assemblages (<80% illite in the <2 mm fraction), the illite
001 reflection is typically 0.3 0.682y. The width of
the 001 reflection shows a positive correlation with
the 001/002 ratio (Fig. 8).
FIG. 7. Relationship between the percentage of illite
and the 001/002 10 Å intensity ratio for all samples,
including some not shown in Figs 1, 2.
clear trend in chemistry with particle size. We have
assumed that all Fe is in clay, as removal of Fe
oxides and hydroxides showed a decrease in Fe
content of just 0.1%. Fe(II) determination shows
that typically 10% of the total Fe in the <0.5 mm
and <0.2 mm fractions is Fe(II). The Fe-rich illite
may therefore be more accurately described as
ferric illite. Only the <0.5 mm fraction sample from
90 m has a significantly different ferrous iron
content (23.5%).
Age determinations for the <0.5 mm and <0.2 mm
fractions of four samples are shown in Table 2. The
K/Ar ages are considerably older than the age of
Bulk chemical and isotope data
The ICP-MS data show a range for Fe2O3
(6.72 19.20%), for Al2O3 it is 12.45 19.81% and
for K2O it is 3.22 5.72% (Table 2). For the
Bembridge Limestone sample these show a
progressive increase in Fe2O3 (10.41 19.2%) and
K2O (5.03 5.44%), and a decrease in Al2O3
(16.11 12.45%), but for other samples there is no
FIG. 8. Relationship between the width of the 10 Å 001
reflection and the 001/002 10 Å intensity ratio for all
samples, including some not shown in Figs 1 and 2.
91.85
Total
: analysis not performed
57.19
16.11
10.41
0.72
0.02
2.14
0.07
0.16
5.03
–
–
–
–
Wt.% total Fe content
Std. dev.
Wt.% Fe(II)/Fe total
Std. dev.
ICP–MS data
SiO2
Al2O3
Fe2O3*
Ti2O
MnO
MgO
CaO
Na2O
K2 O
34+2.1
5.25
98
44 m
<2 mm
K/Ar age (Ma+2s)
Wt.% K2O
Sample
Size-fraction
mm
% 10 Å clay by XRD
91.30
51.1
13.07
18.5
0.1
0.01
2.11
0.9
0.15
5.36
13.5
0.1
11
0.1
–
–
–
93
70.5 m
<2 mm
–
70.5 m
<0.5 mm
–
70.5 m
<0.2 mm
96
75.3 m
<2 mm
–
75.3 m
<0.5 mm
–
75.3 m
<0.2 mm
90.58
50.00
12.45
19.20
0.13
0.01
2.09
1.17
0.09
5.44
13.8
0.1
10.9
<0.1
88.06
49.09
19.46
9.35
0.45
0.11
2.78
1.23
0.23
5.36
–
–
–
–
91.55
52.41
19.81
8.90
0.72
0.20
3.05
1.05
0.23
5.18
6.5
<0.1
10.9
0.7
–
–
–
–
–
–
–
–
–
–
–
–
–
49.83
18.22
11.85
0.34
0.03
2.77
1.28
0.15
5.72
–
–
–
–
89.39
54.31
14.86
8.89
0.67
0.03
2.24
3.47
0.16
4.76
6.1
<0.1
9.1
<0.1
–
–
–
–
–
–
–
–
–
–
–
–
–
39.7+1.9 181+12.6 162+13.8 101.7+3.6 149+4.6 155.4+4.2 85.3+4.3
5.08
3.92
3.34
5.52
4.60
4.56
5.48
–
44 m
44 m
<0.5 mm <0.2 mm
88.43
54.49
16.74
8.40
0.81
0.30
2.65
–
0.29
4.75
–
–
–
–
–
–
82
90 m
<2 mm
TABLE 2. Chemical and K/Ar data for various size-fractions of illite-rich samples.
91
96.9 m
<2 mm
–
87.70
60.91
13.77
6.72
0.65
0.25
2.03
–
0.15
3.22
4.2
<0.1
23.5
1.3
91.35
58.85
16.07
9.08
0.85
0.03
1.87
–
0.12
4.48
–
–
–
–
91.41
49.68
19.71
12.42
0.27
0.03
2.8
0.72
0.16
5.62
7.3
<0.1
7.7
0.1
92.30
53.39
18.81
10.71
0.85
0.03
2.36
0.70
0.15
5.30
–
–
–
–
–
–
–
96.9 m 96.9 m
<0.5 mm <0.2
204.5+9.5 173.9+4.5 167.9+4.5
3.00
4.73
4.99
–
90 m
<0.5 mm
Isle of Wight non-marine green clays
455
456
J. M. Huggett et al.
deposition for all illite-rich samples except that
from the Bembridge Limestone Formation (44 m).
The age for the illite-rich sample from the
Bembridge Limestone Formation, which has a
negligible clastic component (Table 1), is
34+3 Ma for the <2 mm fraction and
39.7+1.9 Ma for the <0.2 mm fraction. For the
two clastic intervals sampled, the K-Ar age
decreases with reduction in size fraction, and
increase with the % illite in the size fraction
(Table 2). The age range for these two intervals is
85.3+4.4 Ma for the <0.2 mm fraction of the
almost pure illite clay from 75.3 m, to
204.5+9.5 Ma for the <0.5 mm fraction of the
least pure illite-rich sample from 90 m.
Scanning electron microscopy
The SEM images reveal that the illite-rich clay is
extremely fine grained, consisting of aggregated
platy particles of 0.1 0.2 mm maximum diameter,
with a high proportion of face-to-edge contacts
(Fig. 9a c). Illite-poor intervals have a strong
preferred orientation of clay particles (Fig. 9d).
BSEI of polished samples indicate that in addition
to clay minerals, fine silt-grade quartz is abundant
(visual estimates range from 30 50%), trace clay
grade and fine silt-grade K-feldspar, chert, zircon
and rutile are also present. Low-Mg calcite is the
principal authigenic mineral. Bioclasts frequently
have micritic overgrowths, and the limestones are
>50% authigenic low-Mg calcite. Rare but ubiquitous authigenic siderite is present in palaeosols
regardless of their clay mineralogy, and at two
horizons in the Colwell Bay to Cliff End Member
section, siderite concretions are present. Pyrite is
present in the mixed-salinity Balinid Bed (90.8 m)
but is rare in the non-marine sediments.
Transmission electron microscopy
The TEM images confirm that the particles in the
illite-rich clays are typically <0.2 mm, even for
what was nominally the <0.5 mm fraction of the
sample from the Bembridge Limestone Formation
green clay (44 m). Normalized analyses of single
particles of illite from the <0.2 mm fraction of
samples from 44 m and 90 m (Cliff End Member)
are listed in Table 3. These data show a wide range
of Fe2O3, Al2O3 and K2O values: 4.99 23.39% for
Fe2O3, 9.4 29.79% for Al2O3 and 3.04 6.76% for
K2O. As the presence of aluminous illite was only
unequivocally identified by analytical TEM, the
relative proportions of Fe-rich and Al-rich illite
could not be determined. The Al substitution for Si
in the octahedral layer is greater in the sample from
the Cliff End Member (Table 3a) than in the sample
from the Bembridge Limestone Formation
(Table 3b). In the octahedral layer, MgO substitution is generally higher and total Fe lower in the
Cliff End Member than in the Bembridge
Limestone Formation. Some particles from the
Bembridge Limestone Formation have higher total
Fe contents than any of the published analyses
included for comparison. Most particles, both Ferich and Al-rich, have irregular outlines, though a
few laths and pseudohexagonal platelets are present
(Fig. 9e). Calculated values for the number of
cations on the basis of an illite formula with
O20(OH)4 and assuming full tetrahedral occupancy,
show close to full octahedral occupancy and
variable interlayer cation content and totals
(Table 3). Interlayer cations in the Bembridge
Limestone Formation illite are Ca, K and Na; in
the Colwell Bay member illite, Na is absent. The
proportion of interlayer K and the layer charges are
low for illite, though XRD data for these two
samples indicate a very low percentage of swelling
layers.
DISCUSSION
XRD data
The association of illite-dominated clay mineral
assemblage with slickensided, generally green and
colour-mottled gleys, and the absence of illitesmectite from any other lithology in the studied
sections suggests a genetic link. The generally
green colour of the illite-rich clay assemblages
suggests that an Fe-rich clay is present. This was
investigated indirectly by measuring the 001/002
intensity ratio of the 10 Å clay. For Fe-rich illite
the ratio is >2, for aluminous illite it is <2
(Brindley & Brown, 1980). The percentage of
illite in the clay fraction increases with the 001/
002 intensity ratio (Fig. 7), which implies that
illitization also involves Fe uptake. The 001/002
intensity ratio for the illite-rich marl from 44 m is
exceptionally high, at 12.1, which implies either a
greater Fe content than in any other sample, or an
absence of detrital aluminous illite. In the course of
ATEM analysis, very few aluminous particles were
detected in the <0.2 mm fraction of the sample from
Isle of Wight non-marine green clays
457
FIG. 9. Scanning electron micrographs: (a) Greyish-green silty clay from a pseudo-gley horizon at 70.6 m. Illite
plus minor kaolinite are the only clay minerals present. Field of view = 25 mm; (b) Green palaeosol claystone
from 44 m, comprising illite and minor calcite only. Field of view = 8 mm; (c) Clayey lacustrine limestone from
43 m. The clay minerals are illite plus minor kaolinite and smectite. Field of view = 25 mm; (d) Partly laminated,
olive green clay from 72.25 m, with a ‘detrital’ clay assemblage of illite and smectite, plus minor kaolinite. Field
of view = 25 mm; (e) Transmission electron micrograph of the <0.2 mm fraction from 44 m. Particles are a
mixture of Fe-rich and Al-rich illite. Very few equant particles of either composition are present. Field of view =
1.1 mm; (f) Back-scattered electron micrograph of claystone from 44 m showing illite-rich clay with calcite
pseudomorphs after gypsum. Field of view = 250 mm.
458
J. M. Huggett et al.
TABLE 3a. Energy dispersive X-ray analyses of single illite particles from
the <0.2 mm fraction for the sample from 90 m (Cliff End Member),
expressed as percentage oxide and as number of cations based on
(O20(OH)4).
SiO2
Al2O3
FeO (estimated)
Fe2O3
TiO2
MgO
CaO
Na2O
K2O
64.44
14.57
1.31
10.62
63.29
19.52
1.17
9.47
.
3.34
0.48
.
.
2.48
0.32
.
.
3.5
0.23
.
.
4.63
0.45
.
62.2
21.99
0.55
4.44
.
2
4.38
.
4.41
4.44
Number of cations on the basis of O20(OH)4
Si
8.00
7.76
7.83
Al
0.00
0.24
0.18
S tetrahedral
8.00
8.00
8.01
7.64
0.36
8.00
7.62
0.38
8.00
2.51
2.8
2.14
3.74
62.26
29.79
0.93
7.54
5.03
Al
Ti
Fe2+
Fe3+
Mg
S octahedral
5.24
63.58
18.2
1.04
8.43
2.58
.
.
2.46
.
.
.
0.26
0.86
0.62
3.88
0.23
0.75
0.64
4.20
0.21
0.67
0.64
3.98
0.18
0.60
0.85
4.14
0.11
0.35
0.37
3.63
0.06
0.83
0.04
0.59
0.03
0.79
0.05
0.69
0.54
0.7
Ca
K
Na
Total interlayer
0.89
0.63
0.82
0.74
1.24
tetrahedral charge
octahedral charge
interlayer charge
0
0.98
0.98
0.24
0.42
0.66
0.18
0.7
0.88
0.36
0.43
0.79
0.38
1.48
1.86
.
.
44 m compared with the sample from 90 m, which
suggests that the 44 m clay assemblage has a higher
proportion of Fe-rich illite.
In the illite-dominated assemblages, the illite 001
reflection is typically broad, suggesting a small
degree of mixed layering. The width of the 001
reflection shows a positive correlation with the
001/002 ratio (Fig. 8), which indicates that the Ferich clay has a higher proportion of smectite layers
than the more aluminous illite. The smectite layers
may be all that remains of detrital smectite which
underwent illitization to Fe-rich illite. The aluminous illite is interpreted as being detrital. This
suggests that although the Solent Group sediments
were reworked from glauconite-rich Eocene sediments (Huggett & Gale, 1997) the detrital 10 Å
clay is not glauconite partially stripped of K+ and
Fe2+/3+. The positive correlation between the
.
.
.
proportion of illite in the clay fraction and the
proportion of illite in the bulk rock clastic fraction
suggests that neoformation of illite has occurred as
well as replacement of smectite by illite (Fig. 5).
The blue-green colour of the ‘low illite’ gley
soils, when fresh, changes to orange in a few
weeks. This is believed to be due to the presence of
Fe(II)-Fe(III) hydroxy carbonate, ‘green rust’, a
mineral associated with gley soils (Newman, 1987).
However it has not been possible to verify this by
XRD.
Chemical and isotope data
The K-Ar ages are considerably older than the
age of deposition for all illite-rich samples from the
Colwell Bay/Cliff End and Lacey’s Farm Members.
459
Isle of Wight non-marine green clays
TABLE 3b. Energy dispersive X-ray analyses of single illite particles from the <0.2 mm fraction for the sample
from 44 m (Bembridge Limestone Formation), expressed as percentage oxide and as number of cations based
on (O20(OH)4).
SiO2
Al2O3
FeO (estimated)
Fe2O3
TiO2
MgO
CaO
Na2O
K2O
57.97
9.14
2.57
20.82
55.13
15.92
2.10
16.99
0.14
2.04
0.07
0.85
6.76
57.64
14.76
1.89
15.30
1.96
0.47
2.03
5.94
2.13
0.41
2.4
5.52
3.58
3.04
2.22
0.65
1.1
5.76
Number of cations on the basis of O20(OH)4
Si
7.62
7.68
Al
0.38
0.32
S tetrahedral
8.00
8.00
7.2
0.8
8.00
7.46
0.54
8.00
7.71
0.29
8.00
7.43
0.57
8.00
7.32
0.68
8.00
7.49
0.51
8.00
Al
Ti
Fe2+
Fe3+
Mg
S octahedral
1.03
0.00
0.26
2.06
0.50
3.85
1.15
1.71
2.03
2.56
2.66
2.32
0.23
1.85
0.44
3.67
1.65
0.01
0.21
1.66
0.4
3.93
0.18
1.50
0.38
3.77
0.14
1.11
0.36
3.64
0.12
1.01
0.39
4.08
0.12
0.97
0.44
4.19
0.12
0.98
0.41
3.83
Ca
K
Na
0.09
0.95
0.17
0.07
1.1
0.49
0.01
1.13
0.21
0.07
0.98
0.51
0.05
0.89
0.58
0.07
0.56
0.02
0.48
0.08
0.93
0.27
Total interlayer
1.21
1.66
1.35
1.56
1.52
0.63
0.50
1.28
tetrahedral charge
octahedral charge
interlayer charge
0.38
0.95
1.33
0.32
1.43
1.75
0.8
0.64
1.44
0.54
1.07
1.61
0.29
1.44
1.73
0.57
1.07
1.64
0.68
0.13
0.55
0.51
0.92
1.43
.
2.54
0.62
0.67
5.67
58.28
9.47
2.24
18.15
.
2.25
0.52
1.9
6.63
.
This indicates that the clay is either detrital or a
mixture of authigenic and detrital. The fact that the
K-Ar ages decrease with decrease in size-fraction
suggests that authigenic illite is concentrated in the
fine fraction. In the ‘detrital’ clay assemblages,
approximately half of the clay is smectite
(Figs. 3, 4), if this is presumed to have been
replaced by illite in the illite-rich samples, then
approximately half of the illite should have an age
equal to the latest Eocene (~40 Ma), and the rest
should be much older.
For the measured range of 85.3+4.3 Ma to
204.5+9.5 Ma, and presuming 50% of the illite
to be neoformed, the detrital clay must ultimately
be derived from Palaeozoic or Mesozoic rock.
Evidently if the clay is reworked from older Eocene
sediments, they were also reworked from older
sediments, possibly the Permo-Trias and the
.
.
60.91
15.53
1.44
11.66
.
.
60.15
21.5
1.35
10.88
.
0.54
2.21
.
.
.
59.5
23.08
1.30
10.50
.
2.42
0.15
.
.
.
59.57
19.07
11.62
11.62
.
.
Carboniferous sediments of western Britain. The
oldest age is for the <0.5 mm fraction sample from
90 m. This sample also has a higher ferrous Fe
content, and the lowest total Fe oxide and K2O
values, which suggest that there is a significant
detrital component, either illite-smectite or illite or
both.
The much younger age of 34+3 Ma for the
<2 mm fraction of the illite-rich marl from the
Bembridge Limestone Formation is close to the
astronomically calibrated date of 33.9 Ma for the
base of the Oligocene (Shackleton et al., 1999). The
older age (39.7+1.9 Ma) for the <0.2 mm fraction
suggests that a very small proportion of detrital clay
is present which is finer grained than the neoformed
clay, though this was not apparent from TEM or
SEM observations. These ages are consistent with
almost all the illite having formed at the time of
460
J. M. Huggett et al.
deposition, or shortly afterwards, and it implies that
in this sample, almost all the illite, ferric and
aluminous, is neoformed (i.e. there was negligible
detrital clay present), or most detrital clay has
recrystallized into more coarsely crystalline authigenic clay.
The Fe and K determined by ICP-MS is
presumed to be in illite, as the illite-rich samples
which were analysed include only a few percent
chlorite or smectite, and no detectable K-feldspar
(in the <2 mm fractions analysed by XRD, or in the
<0.2 mm fraction analysed by ATEM). The ICP-MS
data are consistent with the XRD interpretation that
the illite in the illite-rich samples is enriched in Fe
relative to the illite in samples which are not illiterich (Tables 2, 3). The negative correlation between
Fe and Al, for ICP-MS and ATEM data is
interpreted as indicating variable proportions of
neoformed Fe-rich clay and Al-rich detrital clay
(Fig. 10a). K2O and Fe2O3 increase and Na2O and
CaO decrease with decreasing size-fraction for the
one pure illite sample from 44 m (Table 2). This
implies that neoformed Fe-rich illite is concentrated
in the fine fraction. The lack of correlation between
chemistry and particle size for other samples may
be due to variable proportions of detrital illite,
smectite and kaolinite, both in the samples and in
the various size fractions.
Interlayer cation totals are rather low for illite,
which typically has ~1.8 interlayer cations
(Newman, 1987). In the case of the ATEM
analyses, K2O and Na2O may be low due to beam
damage (Ferrell & Carpenter, 1990). However, as
the values are similar for the ICP-MS data it is
more likely that the Fe-rich illite has smectite
interlayers, which would result in both lower K2O
and lower interlayer cation totals. The breadth of
the 10 Å XRD reflections for illite-rich <2 mm
fractions suggests the latter rather than the former,
though beam damage cannot be discounted. The
absence of Na2O from ATEM analyses of illitic
clay from 90 m suggests that the mode of origin of
the clay is not identical to that from 44 m. The
proportion of SiO2 in ICP-MS analyses decreases
with decreasing size fraction (Tables 2, 3), which
suggests the presence of clay-grade silica minerals,
concentrated in the coarser fractions.
Most illite single-particle ATEM analyses from
the Bembridge Limestone Formation and Colwell
Bay Member show a similar spread of Fe oxide and
Al2O3 values to previously published data for Ferich illite (Jung, 1954; Keller, 1958; Gabis, 1963;
Parry & Reeves, 1966; Deconinck et al., 1988,
Baker, 1997) (Fig. 10a), though some unusually
high Fe concentrations were recorded from the
Bembridge Limestone.
Fe(II)/Fe(III) measurements indicate that the
Fe-rich illite is predominantly ferric. Jung (1954),
Keller (1958), Gabis (1963), Parry & Reeves
(1966), Jeans et al. (1994) and Baker (1997) have
also identified non-marine Fe-rich illite as ferric,
though only Jeans et al. and Baker indicate the
method by which the Fe2+/3+ was determined. Jeans
et al. (1994) described neoformation of ferric illite
in arid Permo-Triassic soils, while Baker (1997)
suggested that formation may occur specifically in
well drained soils. The proportion of Fe(II) is over
twice as high for the sample from 90 m, as for all
other samples for which Fe(II)/Fe(III) measurements were made. As reduction of Fe in smectite
has been linked to K+ fixation (Stucki et al., 1984;
Chen et al., 1987), it is possible that where
significant Fe reduction has occurred, as at 90 m,
Na has been replaced by K.
Illitization
Neoformed low-T Fe-rich illite from green or
greenish-grey marls is known from lacustrine or
lagoonal environments (Jung, 1954; Keller, 1958;
Parry & Reeves, 1966; Gabis, 1963; Porrenga,
1968; Singer & Stoffers, 1980; Deconinck et al.,
1988; Hay et al., 1991) and infrequently from
palaeosols (Robinson & Wright, 1987; Inglès &
Ramos-Geurrero, 1995), though Huggett & Laenen
(1996) have since demonstrated that the example
described by Porrenga (1968) was in fact reworked
glauconite, and others of these examples invoke
alteration of igneous rock or volcanics as the source
of the green clay (Jung, 1954; Gabis, 1963).
Deconinck et al. (1988) reviewed the possible
mechanisms of illitization of smectite in soils and
concluded that the most probable is wetting and
drying in an environment with an adequate supply
of K+ and Fe, or formation in an alkaline, hypersaline lake. The presence of authigenic siderite,
low-Mg calcite and the almost total absence of
pyrite from these sediments is consistent with a
low-sulphate, alkaline pore-fluid at the time of
precipitation.
In the Hampshire Basin, illitization cannot have
occurred through burial diagenesis as maximum
burial has not exceeded 500 m, and it is unlikely
that the green 10 Å clay-dominated claystones
Isle of Wight non-marine green clays
461
FIG. 10. (a) Relationship between Fe (total Fe(II) and Fe(III)) and Al in illite-rich samples from the Colwell Bay
Member (90 m) and the Bembridge Limestone Formation (44 m), plotted together with previously published data.
The latter have been normalized so as to be comparable with the Isle of Wight data which are normalized. The
ICP-MS data are from the <0.5 mm fraction. (b) Relationship between Fe (total Fe(II) and Fe(III)) and K in illiterich samples from the Colwell Bay Member (90 m) and the Bembridge Limestone Formation (44 m). Both
ATEM and ICP-MS data are included.
could consist of finely disseminated detrital
glauconite as the absence of other clay species
from a detrital clay would be difficult to explain.
Only in the marine and brackish interval in the
lower part of the Solent Group (the Colwell Bay
Member) is fragmented and dispersed glauconite
thought to be the source of any of the green
colouration (Fig. 4).
We believe that the correlation between slickensiding and mottling in gley soils and high illite
content (Figs 3 5) is the key to the origin of the
illite. Slickensiding is caused by movement of peds
462
J. M. Huggett et al.
during wetting and drying. Colour mottling is also a
consequence of drying out of the soil profile and
reprecipitation of ferrous Fe as ferric, a process
which is not uniform and hence the mottled
appearance (Fitzpatrick, 1980). Experimental work
has shown that wetting and drying of K smectite, at
surface temperature and pressure, fixes K irreversibly, with most layers collapsing in fewer than 40
cycles (Środoń & Eberl, 1984; Eberl et al., 1986).
These authors found that smectite layers are
converted to interstratified illite-smectite by internal
reorganization of smectite layers (substitution of
lower-charge cations for higher-charge cations in
the tetrahedral and octahedral layers) and irreversible K fixation. As described above, K fixation
may also be linked to Fe(III) reduction in smectite,
but the low proportion of ferrous Fe in all samples
analysed, except that from 90 m, suggests this has
not been an important process in the Solent Group.
The wide variation in the ATEM data possibly
indicates a higher proportion of smectite interlayers
in the illite-rich samples than do the XRD data.
This suggests that the illitization is more physical
than chemical. We propose that the illite-smectite in
the gley soils of the Colwell Bay Member represent
this intermediate stage between smectite and illite.
The face-to-edge morphology of the illitic clay
aggregates suggests that the clay may be
neoformed, though in the marl from 44 m, which
apparently has the most neoformed Fe-rich illite,
the morphology of the clay is partially masked by
micrite (Fig. 9b). There is no evidence for the
release of large amounts of silica, as in the Alconserving reaction for illitization of smectite
proposed by Boles & Franks (1979). The mobile
Al mechanism for illitization of smectite (Hower et
al., 1976) which releases very little silica is more
appropriate to the Solent Group. However, given
that both these equations are for elevated temperatures, neither may be applicable. Whatever the
precise mechanism, it clearly necessitates a source
of K+ and a means of concentrating it. There is no
palaeontological evidence of marine ingressions in
the Solent group soils, which suggests that seawater
was not a major source of K+, though both lakes
and sub-aerial sediments may have been inundated
by the sea during storms. Iron-rich illite formation
is most often associated with a high input of K into
freshwater, either through erosion of K-rich igneous
bedrock (Jung, 1954; Gabis, 1963) or through
evaporative concentration, at lake margins or on
floodplains, of K+ derived through weathering of
detrital K-feldspar and mica (Inglès & RamosGuerrero, 1995). Alteration of smectite to illite
through wetting and drying may have occurred
during periods of lake regression and transgression.
Wetting and drying would clearly have been most
effective at lake margins and may have occurred
during intervals of highly seasonal climate, i.e. wet
winters and hot dry summers. Alternatively it may
have occurred on an alluvial plain which was
seasonally flooded by river water rich in K+ and
Fe2+/3+ derived through weathering of Bracklesham
Group or Barton Clay Formation glauconite.
For the Solent Group, the most likely mechanism
is evaporative concentration of K+ derived through
weathering of detrital glauconite, K-feldspar and
mica. Sufficient Fe for Fe-rich illite formation is
present as oxyhydroxides and attached to detrital
clay in normal freshwater. In these sediments
weathered glauconite provides an in situ source of
Fe, as well as K+. The association of shell debris or
carbonate soil nodules with the illite suggests that
the illitization occurred in an alkaline environment.
The association of calcite pseudomorphs after
gypsum, with illite-rich clay in the Bembridge
Limestone indicates that this clay may have formed
with evaporites. The difference in mineralogy,
chemistry and K-Ar age between this sample and
those from palaeosols in the Colwell Bay to Cliff
End Members, suggests that the mode of illitization
may not be identical in the two intervals studied.
Formation of Fe-rich green clays in the late
Eocene–Oligocene has been reported from other
localities world-wide (Jung, 1954, Gabis, 1963) and
from the Jurassic in the Jura (Deconinck et al.,
1988). These deposits formed during lowstands
(Haq et al., 1988), when non-marine sedimentation
would have been more widespread than during
highstands.
CONCLUSIONS
The association of illite-dominated intervals with
slickensided, generally green and mottled gleys or
pseudogleys, and the absence of illite-smectite from
any other lithology in the Colwell Bay Member is
interpreted as evidence of illitization resulting from
repeated wetting and drying of the clay-rich
sediment, probably during periods of highly
seasonal climate.
Alteration of smectite to illite through wetting
and drying may have occurred during periods of
lake regression and transgression, this process
Isle of Wight non-marine green clays
would clearly have been most effective at lake
margins and may have been either climatically or
eustatically controlled. Alternatively it may have
occurred on an alluvial plain which was seasonally
flooded by river water rich in K+ and Fe2+/3+
derived through weathering of Bracklesham Group
or Barton Clay Formation glauconite. Further
concentration of Fe and K would occur through
evaporation during dry seasons.
The blue-green colour of the ‘low illite’ gley
soils, when fresh, changes to orange in a few
weeks. This is believed to be due to the presence of
Fe(II)-Fe(III) hydroxy carbonate, ‘green rust’, a
mineral associated with gley soils (Newman, 1987).
However it has not been possible to verify this by
XRD.
The difference in mineralogy, chemistry and
K/Ar age between the green lake/soil clay in the
Bembridge Limestone Formation and those from
gleys and pseudogleys which have undergone
seasonal wetting and drying in the Lacy’s Farm,
Colwell Bay and Cliff End Members of the Headon
Hill Formation suggests that the mode of illitization
may not be identical in the two intervals studied.
The K/Ar ages for Bembridge Limestone Formation
illite indicate that no detrital illite is present now,
and may never have been, whereas in the Headon
Hill Formation, detrital clays, including illite are
still present.
In the Hampshire Basin, illitization cannot have
occurred through burial diagenesis as maximum
burial has not exceeded 500 m.
ACKNOWLEDGMENTS
We wish to thank David Wray and his technicians at
the University of Greenwich for carrying out the ICPMS analyses, Joe Stucki of the University of Illinois for
the Fe(II) determinations, Martin Gill of Imperial
College for carrying out the XRD analyses, Jackie
Skipper for her thoughts on the soils, Raymond
Wendling of the Centre de Géochimie de la Surface
for technical assistance in the K-Ar determinations and
Harry Shaw for reading the manuscript. The ATEM
was carried out at the Centre for Microscopy and
Microanalysis at the University of Surrey.
REFERENCES
Armenteros I., Daley B. & Garcia E. (1997) Lacustrine
and palustrine facies in the Bembridge Limestone
(late Eocene, Hampshire Basin) of the Isle of Wight,
southern England. Palaeogeography,
463
Palaeoclimatology, Palaeoecology, 128, 111 132.
Baker J.C. (1997) Green ferric clay in non-marine
sandstones of the Rewan Group, southern Bowen
Basin, Eastern Australia. Clay Miner. 32, 499 506.
Boles J.R. & Franks S.G. (1979) Clay diagenesis in the
Wilcox sandstones of southwest Texas: Implications
of smectite diagenesis on sandstone cementation. J.
Sed. Pet. 49. 55 70.
Bonhomme M.G., Thuizat R., Pinault Y., Clauer N.,
Wendling R. & Winkler R. (1975) Méthode de
datation potassium-argon. Appareillage et technique.
Note Tech. Inst. Géologie, Univ. Strasbourg, 3, 53.
Brindley G.W. & Brown G. (1980) Crystal Structures of
Clay Minerals and their X-ray Identification.
Monograph 5, Mineralogical Society, London.
Bristow H.W., Reid C. & Strachan A. (1889) The
geology of the Isle of Wight (2nd edition). Mem.
Geol. Surv. U.K.
Chen S.A., Low P.F. & Roth C.B. (1987) Relation
between potassium fixation and the oxidation state of
octahedral iron J. Soil Sci. Soc. Am. 51, 82 86.
Collinson M.E. (1983) Palaeofloristic assemblages and
palaeoecology of the Lower Oligocene Bembridge
Marls, Hamstead Ledges, Isle of Wight. Botanical J.
Linean Soc. 86, 177 255.
Daley B. (1972) Macroinvertebrate assemblages from
the Bembridge Marls (Oligocene) of the Isle of
Wight, England and their environmental significance. Palaeogeography, Palaeoclimatology,
Palaeoecology, 11, 11 32.
Daley B. (1973) The palaeoenvironment of the
Bembridge Marls (Oligocene) of the Isle of Wight.
Proc. Geol. Assoc. 84, 83 93.
Daley B. & Edwards N. (1974) Weekend field meeting:
the Upper Eocene–Lower Oligocene Beds of the Isle
of Wight. Proc. Geol. Assoc. 85, 281 291.
Daley B. & Edwards N. (1990) The Bembridge
Limestone (Late Eocene), Isle of Wight, southern
England: a stratigraphical revision. Tertiary Res. 12,
51 64.
Deconinck J.F., Strasser A. & Debrabant P. (1988)
Formation of illitic minerals at surface temperatures
in Purbeckian sediments (lower Berriasian, Swiss
and French Jura). Clay Miner. 23, 91 103.
Eberl D.D., Środoń J. & Northrop H.R. (1986)
Potassium fixation in smectite by wetting and drying.
Pp. 296 326 in: Geochemical Processes at Mineral
Surfaces (J.A. Davis & K.F. Hayes, editors). ACS
Symposium Series 323/14, American Chemical
Society, Washington, D.C.
Ferrell R.E. & Carpenter P.K. (1990) Application of the
electron microprobe and image analysis in the study
of clays. Pp. 108 132 in: Electron Optical Methods
in Clay Science (I.D.R. Mackinnon & F.A.
Mumpton, editors). CMS Workshop Lectures, 2.
Clay Minerals Society, Boulder, CO.
Fitzpatrick E.A. (1980) Soils, their Formation,
464
J. M. Huggett et al.
Classification and Distribution. Longman, London.
Forbes E. (1853) On the fluvio-marine Tertiaries of the
Isle of Wight. Q.J. Geol. Soc. London, 9, 259 270.
Gabis V. (1963) Etude minéralogique et géochimique de
la série sédimentaire oligocène du Velay. Bull. Soc.
Franç. Minéral. Cristallogr. 86, 315 354.
Gale A.S., Jeffery P.A., Huggett J.M. & Connolly P.
(1999) Eocene inversion history of the Sandown
Pericline, Isle of Wight, southern England. J. Geol.
Soc. London, 156, 327 339.
Gilkes R.J. (1968) Clay mineral provinces in the
Tertiary sediments of the Hampshire Basin. Clay
Miner. 7, 351 361.
Haq B.U., Hardenbol J. & Vail P.R. (1988) Sea level
changes: an integrated approach. Pp. 71 108 in:
Mesozoic and Cenozoic Chronostratigraphy and
Eustatic Cycles (C.K. Wilgus, H. Posamentier, J.V.
Wagoner, C.A. Ross & C.G.St.C. Kendal, editors).
SEPM Spec. Publ. 42. Society of Economic
Palaeontologists and Mineralogists, Tulsa, OK.
Hay R.L., Guldman S.G., Mathews J.C., Lander R.H.,
Duffin M.E. & Kyser T.K. (1991) Clay mineral
diagenesis in core KM-3 of Searles Lake, California.
Clays Clay Miner. 39, 84 96.
Hooker J.J. (1992) British mammalian paleocommunities across the Eocene–Oligocene transition and
their environmental implications. Pp. 494 515 in:
Eocene–Oligocene Climatic and Biotic Evolution
(D.R. Prothero & W.A. Berggren, editors). Princeton
University Press, Princeton, NJ.
Hower J., Eslinger E.V., Hower M.E. & Perry E.A.
(1976) Mechanism of burial metamorphism of
argillaceous sediment: Mineralogical and chemical
evidence. Geol. Soc. Am. Bull. 87, 725 737.
Huggett J.M. & Gale A.S. (1997) Petrology and
palaeoenvironmental significance of glaucony in
the Eocene succession at Whitecliff Bay,
Hampshire Basin, UK. J. Geol. Soc., London, 154,
897 912.
Huggett J.M. & Laenen B. (1996) Green clays from the
lower Oligocene of Aardebrug, Belgium, a reevaluation. Clay Miner. 31, 557 562.
Inglès M. & Ramos-Guerrero E. (1995) Sedimentological control on the clay mineral distribution in the
marine and non-marine Palaeogene deposits of
Mallorca (Western Mediterranean). Sed. Geol. 94,
229 243.
Jeans C.V., Mitchell J.G., Scherer M. & Fisher M.J.
(1994) Origin of the Permo-Triassic clay mica
assemblage. Clay Miner. 29. 575 590.
Jung P. (1954) Les illites du bassin oligocène de Salins
(Cantal). Bull. Soc. Franç. Minéral. Cristallogr. 77,
1231 1249.
Keen M.C. (1977) Ostracod assemblages and the
depositional environments of the Headon, Osborne
and Bembridge Beds (Upper Eocene) of the
Hampshire Basin. Palaeont. 20, 405 445.
Keen M.C. (1978) The Tertiary–Palaeogene. Pp.
385 450 in: A Stratigraphical Index of British
Ostracoda (R. Bate & E. Robinson, editors). Bell
House Place, London.
Keller W.D. (1958) Glauconitic mica in the Morrison
Formation in Colorado. Clays Clay Miner. 5,
120 128.
Liengjarren M., Costa L. & Downie C. (1980)
Dinoflagellate cysts from the Upper Eocene-Lower
Oligocene of the Isle of Wight. Palaeont., 23,
475 499.
Newman A.C.D. (1987) Chemistry of Clays and Clay
Minerals. Monograph 6. Mineralogical Society,
London.
Ollier C. & Pain C. (1996) Soils and Landforms. Wiley,
Chichester, UK.
Parry W.T. & Reeves C.C. (1966) Lacustrine glauconitic
mica from pluvial Lake Mound, Lynn and Terry
Counties, Texas. Am. Miner. 51, 229 235.
Paul C.R.C. (1989) The molluscan faunal succession in
the Hatherwood Limestone Member (Upper
Eocene), Isle of Wight, England. Tertiary Res. 10,
147 162.
Porrenga D.H. (1968) Non-marine glauconitic illite in
the lower Oligocene of Aardenburg, Belgium. Clay
Miner. 7 421 429.
Retallack G.J. (1990) Soils of the Past. Harper Collins
Academic, London.
Robinson D. & Wright V.P. (1987) Ordered illitesmectite and kaolinite-smectite: pedogenic minerals
in a lower Carboniferous paleosol sequence, South
Wales? Clay Miner. 22, 109 118.
Shackleton N.J., Crowhurst S.J., Weedon G.P. & Laskar
J. (1999) Astronomical calibration of OligoceneMiocene time. Phil. Trans. R. Soc. Lond. A, 357,
1907 1929.
Singer A. & Stoffers P. (1980) Clay mineral diagenesis
in two East African lake sediments. Clay Miner. 15,
291 307.
Środoń J. & Eberl D.D. (1984) Illite. Pp. 495 544 in:
Micas (S.W. Bailey, editor). Reviews in Mineralogy,
13. Mineralogical Society of America, Washington,
D.C.
Stucki J.W. (1981) The quantitative assay of minerals
for Fe2+ and Fe3+ using 1, 10 phenanthroline. II A
photochemical method. J. Soil Sci. Soc. Am. 45,
638 641.
Stucki J.W., Low P.F., Roth C.B. & Golden D.C. (1984)
Effects of iron oxidation state on clay swelling.
Clays Clay Miner. 32, 357 362.
White H.J.O. (1921) A short account of the geology of
the Isle of Wight. Mem. Geol. Surv. G.B.