Clays and Clay Minerals, Vol. 44, No. 3, 408-416, 1996. [630629]

Clays and Clay Minerals, Vol. 44, No. 3, 408-416, 1996.
ORIGIN OF CRETACEOUS AND OLIGOCENE
KAOLINITES FROM THE IWAIZUMI CLAY DEPOSIT,
IWATE, NORTHEASTERN JAPAN
C. MIZOTA 1 AND E J. LONGSTAFFE 2
Faculty of Agriculture, Iwate University, Ueda 3-18-8, Morioka 020, Japan
2 Department of Earth Sciences, University of Western Ontario, London, Ontario N6A 5B7 Canada
Abstract–Hydrogen- (3D = -106 to -97%0 and oxygen- (3180 = +14.0 to +16.6%0 isotope com-
positions of kaolinite from late Cretaceous and Oligocene deposits at Iwaizumi, northeastern Japan, in-
dicate that these clays formed by weathering of volcanic parent rocks, rather than during hydrothermal
(>100 ~ alteration. The Iwaizumi kaolinites also are depleted of D and 180 relative to kaolinite formed
during modern, tropical weathering, suggesting that the kaolinite developed under cool or cool-temperate
conditions. The oxygen-isotope compositions of the kaolinite increase slightly upward through the de-
posits, perhaps implying a modest increase in temperature from late Cretaceous to Oligocene time. The
8D and 8~sO results for kaolinite from the Oligocene deposits closely follow the kaolinite weathering
line. However, a small but systematic deviation from this line for the Cretaceous kaolinites is most simply
explained by post-formational, hydrogen-isotope exchange between these clays and downward percolating
meteoric water.
Key Words–Climate, Cretaceous, Hydrogen-isotopes, Iwate, Japan, Kaolinite, Oligocene, Oxygen-iso-
topes.
INTRODUCTION
The Iwaizumi clay deposit, located in northeastern
Honshn, is one of several highly productive, late Cre-
taceous or Oligocene fire-clay and flint-clay deposits
in Japan. These deposits are characterized by well-
crystallized kaolinite, and are very pure and highly
refractory. On the basis of mineralogical and paleo-
botanical data, Iijima (1972) and Tanai et al. (1978)
suggested that this kaolinite formed by lateritic weath-
ering and intensive desilication of parent volcanic
rocks under warm-temperate to subtropical, humid
conditions. This conclusion is consistent with the com-
mon view that extensive weathering, leading to ka-
olinite formation and enrichment in gibbsite and he-
matite, is indicative of such climates (Biscaye 1965;
Oilier 1969). Bird and Chivas (1988) have challenged
the universality of such an assumption, using oxygen
isotopes to show that Permian kaolinitic weathering in
eastern Australia occurred at relatively low tempera-
tures.
A hydrothermal origin has also been proposed for
the Iwaizumi clay deposits, based on the microscopic
presence of dickite pseudomorphs after feldspar (Hu
and Zhang 1988). Dickite is a common hydrothermal
alteration product of volcanic rocks. However, it is dif-
ficult to demonstrate unequivocally the presence of
dickite in the Iwaizumi clay deposits from the X-ray
diffraction (XRD) and differential thermal analysis
patterns of Hu and Zhang (1988). Furthermore, dickite
formation under sedimentary conditions has been re-
ported, for example, the Ashfield shale in Australia (Davey et al. 1975) and Permo-Triassic sandstones in
Spain (Ruiz Cruz and Moreno Real 1993).
In this paper, we use the hydrogen- and oxygen-
isotope compositions of kaolinite from a stratigraphic
section (Figure 1) through the Iwaizurni mine to de-
termine the origin of these clays. Results for a single
sample reported by Marumo et al. (1979, 1982) plot
near the line that describes kaolinite formation during
weathering (Savin and Epstein 1970). Our data con-
firm this earlier observation and provide further insight
into the evolution of these kaolinites.
The hydrogen- and oxygen-isotope compositions of
kaolinite are determined primarily by temperature and
the isotopic composition of water present during crys-
tallization (Savin and Epstein 1970; Lawrence and
Taylor 1971, 1972). Because the ~D and ~lsO values
of meteoric water vary systematically with latitude and
altitude (Craig 1961; Dansgaard 1964; Yurtsever and
Gat 1981), the stable isotope composition of kaolinite
formed during weathering is normally considered to
reflect the location, and hence mean temperature, of
the landscape surface (Lawrence and Taylor 1971,
1972). For example, lower kaolinite ~-values are char-
acteristic of cooler regions, typically located at higher
latitudes and/or altitudes. For oxygen, post-deposition-
al isotopic exchange between kaolinite and water is
virtually non-existent (O'Neil and Kharaka 1976).
Thus, this climatic signature is almost always pre-
served. However, hydrogen-isotope exchange can oc-
cur under some conditions. These results should be
interpreted more cautiously (O'Neil and Kharaka
1976; Bird and Chivas 1988; Longstaffe and Ayalon
1990).
Copyright 9 1996, The Clay Minerals Society 408

Vol. 44, No. 3, 1996 Stable isotope geochemistry of kaolinite, Iwaizumi clay deposit, Japan 409
Nameiri Formatic
(Oligocene)
Komatsu Formatic
(Oligocene)
Yokomichi Formatiq
(Late Cretaceousl 9 9 No. 2, 34 & 35 fire clays
No. 1 fire clay
Flint clay
White hard shale
Red hard shale
"Micaceous" red shale
"Micaceous" sandstone
Glassy tuff
Tuffaceous sandstone
Green colored sandstone
Breccia
Paleozoic Basemel
9 Sample location
Figure 1. Schematic section through the Iwaizumi clay deposits (Fujii 1970). Sample Number
Iw-9, – 10, – 11
Iw-8
Iw-7, -6
Iw-4
Iw-5, -3
Iw-2
Iw-1
GEOLOGIC SETTING AND
SAMPLING
Basement rocks surrounding the Iwaizumi clay
mine consist of Paleozoic sandstone, chert, shale and
granodiorite, and are overlain unconformably by late
Cretaceous and Paleogene volcanic rocks, which have
a glassy texture and are highly permeable (Fujii 1970).
The detailed stratigraphy, lithology and mineralogy of
the kaolin deposits in the vicinity of the Iwaizumi clay
mine (39~ 141~ 350 m above sea level) have been described previously by Fujii (1970), Iijima
(1972) and Tanai et al. (1978). Kaolinitic clays occur
in 3 units, from oldest to youngest, the Yokomichi,
Komatsu and Nameiri formations (Figure 1). An un-
conformity exists between the Yokomichi and Komat-
su formations (Tanai et al. 1978). A fission-track zir-
con date of 71.2 + 4.4 Ma has been reported for the
Upper Yokomichi Formation (Kato et al. 1986).
Eleven samples representing the stratigraphic range
of kaolinite in the Iwaizurni clay mine, plus a ground-

410 Mizota and Longstaffe Clays and Clay Minerals
Table 1. Sample horizons and stable-isotope results for kaolinites from the Iwaizurni Clay Mine, northeast Japan.
Color
Sample Stratigraphy I Age I Lithology 2 (moist)
Iw-11 Nameiri Fm. Oligocene No. 35 fire clay 10YR3/I
Iw-10 Nameiri Fro. Oligocene No. 34 fire clay 5Y1.7/1
Iw-9 Nameiri Fro. Oligocene No. 2 fire clay 5Y5/2
Iw-8 Nameiri Fro. Oligocene No. 1 fire clay 2.5Y4/2
Iw-7 Komatsu Fin, Oligocene Flint clay (upper unit) 7.5YR4/2
Iw-6 Komatsu Fm. Oligocene Flint clay (lower unit) 7.5YR6/2
Iw-5 Yokomichi Fro. Late Cretaceous Red shale (upper unit) 7.5R3/4
Iw-4 Yokomichi Fro. Late Cretaceous White shale 5Y7/1
Iw-3 Yokomichi Fm. Late Cretaceous Red shale (lower unit) 7.5R4/6
Iw-2 Yokomichi Fm. Late Cretaceous 3"Micaceous" red shale 7.5R4/6
Iw-1 Yokomichi Fro. Late Cretaceous 3"Micaceous" sandstone 5Y7/1
1 After Tanai et al. (1978).
2 After Fuji (1970).
3 Vermicular aggregation of kaolinite.
i Calculated isotopic composition of meteoric water in equilibrium with kaolinite at 20 ~ and
~: 15 ~ determined using the geot_bermometer of Lambert and Epstein (1980) for hydrogen: 10001na(k~or~n~ ….. ) =
-4.53(106)T -z + 19.4, and Land and Dutton (1978) for oxygen: 10001na(k~l~,it …… 0 = 2.50(106)(T-2) – 2.87.
water sample draining from the base of the Yokomichi
Formation, which was collected in March 1993, were
obtained for this study (Table 1, Figure 1). All samples
are composed almost entirely of highly crystalline ka-
olinite. Contamination by other minerals is normally
<–5%. Dark grey to red, "micaceous" units (Yoko-
michi Formation) at the base of the section (samples
Iw-1 and -2) consist almost entirely of granular and
booklet-like mineral aggregates with a diameter of 0.5
to 1.0 ram, giving the material the texture of a sand-
stone. XRD analysis shows that the "micaceous'" min-
eral is kaolinite. The overlying, hard, massive shale
(samples Iw-3 to -5) comprises about 50 m of the sec-
tion and is also rich in kaolinite. Lower portions of
the shale are normally reddish, grading upward into
bluish-grey to white material. About 1 to 2 m of dark
grey to dark brown flint clays (Komatsu Formation,
samples Iw-6 and -7) have developed directly on the
shale. The shale and flint clays have similar textures,
with the uppermost shale grading into the lowermost
clays. The fire-clay deposits of the Nameiri Formation
(samples Iw-8 to -11), located higher in the section,
are associated with coal. Fujii (1970) suggested that
the fire clays from seams 1, 2 and 34 are of sedimen-
tary origin, whereas fire clay from seam 35 formed
directly through alteration of dacitic or rhyolitic tuff
(Figure 1).
ANALYTICAL METHODS
Two to 5 g of moist sample were disaggregated
lightly using a wooden pestle in an iron mortar, and
then treated with hydrogen peroxide (15%) at 80 to 95
~ The clay suspension was agitated ultrasonically (28
Khz) for 15 min. The <2 p,m size-fraction was sepa-
rated by repeated centrifugation. To remove Fe oxides,
the clay fraction was treated with a mixture of 0.3 M
Na-citrate and dithionite at 25 ~ and shaken gently overnight. Excess salts were removed by dialysis and
the clay suspension dried at 50 ~ A portion (10 to
20 mg) was air-dried on a glass slide and then exam-
ined by XRD using Co Ks radiation.
Stable isotope results are reported using the normal
~-notation relative to Vienna Standard Mean Ocean
Water (V-SMOW). Hydrogen for isotopic analysis was
obtained from 50 mg samples of pure kaolinite using
a procedure modified after Bigeleisen et al. (1952) and
Godfrey (1962). Samples were degassed at 150 ~ for
2 h under vacuum, and then maintained under vacuum
at room temperature for 12 h. The samples were then
reheated under vacuum for 30 min at 200 ~ prior to
extraction of the hydroxyl group and hydrogen. During
extraction, the samples were heated to 1000 ~ for 30
rain and the resulting gases passed first over Cu oxide
at 500 ~ and then U metal at 800 ~ The hydrogen
gas was analyzed using a PRISM II dual inlet, gas-
source mass-spectrometer. Precision of the kaolinite
~D measurements was _-+2%o.
Oxygen for isotopic analysis was liberated quanti-
tatively from dried 20 mg samples of kaolinite by re-
action with bromine pentafluoride at 600 ~ (Clayton
and Mayeda 1963). Prior to reaction, samples were
degassed for 2 h under vacuum at 150 ~ Oxygen was
converted to CO 2 by reaction with an incandescent car-
bon rod, and the 81so value of this gas was measured
using an OPTIMA dual inlet, gas-source mass-spec-
trometer. An internal quartz standard calibrated to
NBS-28 gave a blsO value of +11.47%o, compared to
its average value of +11.5%o. A value of +9.66 -+
0.13%o for NBS-28 is obtained routinely in our labo-
ratory.
The 81sO value of the water sample was determined
using the conventional CO2-H:O equilibration method
(Epstein and Mayeda 1953). The ~D value of the water

Vol. 44, No. 3, 1996 Stable isotope geochemistry of kaolinite, Iwaizumi clay deposit, Japan 411
Table 1. Extended.
Kaolinil~ (%e) Water (%0)
~D ~80 8D 81sO
-97 +16.6 -66t -9.7?
-98 + 16.2 -67 – 10.1
-103 +15.9 -74 -10.4
-101 +15.7 -71 -10.6
-106 +15.2 -76 -11.1
-100 +15.9 -70 -10.4
-100 +14.8 -70 (-68):~ -11.5 (-12.5):~
-100 +15.5 -70 (-68) -10.8 (-11.8)
-102 +14.7 -72 (-70) -11.6 (-12.6)
-103 +14.0 -73 (-71) -12.3 (-13.3)
-103 +14.2 -73 (-71) -12.1 (-13.1)
was obtained by reduction over hot metallic Zn (Cole-
man et al. 1982).
RESULTS
X-ray diffraction of clay separates treated to remove
non-crystalline constituents, for example, ferrihydrite,
organic matter, etc., showed that kaolinite was the only
mineral constituent. Other crystalline matter was not
observed, despite the <1% sensitivity to phases such
as quartz and mica. We consider all kaolinite samples
analyzed to be essentially free of contaminants. The
purity of these separates rules out contamination as the
major cause of variation in their isotopic compositions.
The 8D values of the kaolinite range from -106 to
-97%~, and the 8180 values, from +14.0 to +16.6%~
(Table 1). The 8180 values show a systematic increase
from the base to the top of the sampled section. The
water sample from the mine has a 8D value of
-66.5%~ and a 8t80 value of -10.2%o.
DISCUSSION
To interpret the hydrogen- and oxygen-isotope com-
positions of the Iwalzumi kaolinites, it is first neces-
sary to choose the most appropriate kaolinite-water
isotopic fractionations from those that are currently
available. The hydrogen-isotope fractionation between
kaolinite and water is not particularly well-known. The
divergent results of Savin and Epstein (1970), Taylor
(1974), Lambert and Epstein (1980), and Liu and Ep-
stein (1984) have been summarized by Kyser (1987)
and Savin and Lee (1988). Fractionation factors for
surficial to hydrothermal temperatures have also been
reported by Lawrence and Taylor (1971, 1972), Su-
zuoki and Epstein (1976) and Marumo et al. (1979,
1980) among others. Fortunately for the purposes of
this study, the spread among the reported values for
the mineral-water fractionation is lowest at surficial
temperatures, ~–-3%~ at 15 to 25 ~ (Kyser 1987, his
Figure 18). Because of its widespread use, we have
chosen the hydrogen-isotope geothermometer of Lam-
bert and Epstein (1980) to calculate water composi-
tions, but we note that other curves are unlikely to produce significantly different values at surficial tem-
peratures.
The oxygen-isotope fractionation between kaolinite
and water has been reviewed by Kyser (1987) and
Savin and Lee (1988). Consistent fractionations have
been reported by Savin and Epstein (1970) and Law-
rence and Taylor (1971, 1972) for natural kaolinites
formed at surficial temperatures (–20 ~ Kaolinite-
water fractionations for natural hydrothermal systems
have been described by Eslinger (1971) and Marnmo
et al. (1982). In addition, empirical and semi-empirical
derivations of the kaolinite-water fractionation equa-
tion have been provided by Savin and Lee (1988) and
Zheng (1993). Land and Dutton (1978) combined the
data of Eslinger (1971) and Savin and Epstein (1970)
to produce an empirical kaolinite-water geothermom-
eter that has been particularly useful for studies of
lower temperature systems (Longstaffe 1983, 1989).
We have adopted this equation because it utilizes rea-
sonably well-constrained, low temperature fractiona-
tions obtained for natural kaolinite-water systems. For
studies of kaolinite weathering, we place less confi-
dence in equations produced by extrapolation of high
temperature experiments, although the latter results
may be more suitable for the study of hydrothermal
systems (Marumo et al. 1995).
Isotopic Evidence for Cool-Temperature Weathering
The 8D and 8180 values of the Iwaizumi kaolinite
are illustrated in Figure 2. The distribution of data do
not support a hydrothermal origin for these clays. In-
stead, the samples plot on, or close to, the weathering
(–20 ~ Kaolinite Line (Sheppard et al. 1969 after
Savin and Epstein 1970). Detailed inspection exhibits
that the Oligocene samples plot within error on the
Kaolinite Line, whereas Cretaceous samples are slight-
ly but systematically displaced to the left (Figure 3).
All samples plot far to the right of the line that dis-
criminates between clays of hydrothermal versus
weathering origins (Supergene-Hypogene Line of
Sheppard et al. 1969).
The Iwalzumi samples occupy a position along the
Kaolinite Line that is well below the range known for
kaolinite formed in tropical or warm-temperate cli-
mates. Kaolinite formed in hot climates generally has
higher 8D (-80 to -40%o) and 8180 (+17 to +23%o)
values (Savin and Epstein 1970; Lawrence and Taylor
1971, 1972; Sheppard 1977; Marumo et al. 1982; Has-
sanipak and Eslinger 1985). Lower 8-values, like those
reported here, are acquired by kaolinites formed dur-
ing weathering at high latitudes or altitudes, or at a
large distance from the coast, where meteoric water
was more depleted of D and 180, and temperatures
were generally lower. Previously reported examples
include Permo-Triassic kaolinites from the Raniganj
Basin of northeastern India (8180 = +7 to +11%o;
Dutta 1985) and the Bowen (+3.5 to +6.5%o; Botz et

412 Mizota and Longstaffe Clays and Clay Minerals
Figure 2. Hydrogen- versus oxygen-isotope plot for kaolinite and associated meteoric water from the lwaizumi clay mine.
The Meteoric Water Line is taken from Craig (1961). The Kaolinite and Supergene-Hypogene Lines are taken from Sheppard
et al. (1969). The field (heavy stippled pattern) for kaolinite formed during weathering in tropical or warna-temperate climates
and water in equilibrium with these kaotinites has been compiled from Savin and Epstein (1970), Lawrence and Taylor (1971,
1972), Sheppard (1977), Marumo et al. (1982) and Hassanipak and Eslinger (1985). Key: <~ = kaolinite samples from
Oligocene units. [] = kaolinite from upper Cretaceous units. D (73) = kaolinite reported by Marumo et al. (1982). Area A
(x) = water compositions in equilibrium with Oligocene kaolinites at 20 ~ Area B (~) = water compositions in equilibrium
with Cretaceous kaolinites at 15 ~ assuming no post-forrnational hydrogen-isotope exchange between the kaolinite and
younger, downward percolating meteoric water. Area C (+) = calculated position on the Meteoric Water Line for water at
oxygen-isotope equilibrium with Cretaceous kaolinites at 15 ~ the hydrogen-isotope composition for this water has been
calculated using the equation for the Meteoric Water Line (Craig 1961). Area C' (1~) = Cretaceous kaolinites in equilibrium
with water shown in Area C at 15 ~
Most kaolinites formed in a weathering (supergene) environment plot along a trend close to the Kaolinite Line, with the
Supergene-Hypogene line serving to separate kaolinite of supergene origin, on the right, from kaolinite of probable hypogene
origin, on the left (Sheppard et al. 1969). Kaolinite from the Iwaizumi clay mine plots on or close to the kaolinite line, but
below the range typical for tropical to warm-temperate climates. The heavy arrow indicates the proposed change within the
gD values of the Cretaceous kaolinite resulting from low temperature, post-formational hydrogen-isotope exchange with
meteoric water.
al. 1986) and Gunnedah (+6 to +9%o; Bird and Chivas
1988) basins of eastern Australia. The depleted com-
positions reflect the lower gD and g180 values of me-
teoric water at the higher latitude of the Gondwana
landmass in the southem Hemisphere during weath-
ering.
We suggest that the Iwaizumi data are also typical
of a cooler climatic regime, with the small increase in
g~80 values from the Cretaceous to the Oligocene por-
tions of the section (Table 1) indicating that the Oli- gocene kaolinites (+ 15.2 to + 16.6%0, av. = + 15.9%0)
formed under slightly different conditions than the
Cretaceous kaolinites (+14.0 to +15.5%o, av. =
+14.6%o). The gD values calculated for the meteoric
water in equilibrium with the Iwaizumi kaolinites at
20 ~ range from -76 to -66%0. At 15 ~ these val-
ues are 2%~ higher, and were calculated for Cretaceous
samples only (Table 1). The water g~80 values range
from -12.3 to -9.7%0 at 20 ~ 1%o lower at 15 ~
(Table 1). The compositions (20 ~ calculated for wa-

Vol. 44, No. 3, 1996 Stable isotope geochemistry of kaolinite, Iwaizumi clay deposit, Japan 413
-95
7 #
11
[] 100
. ~-~i ~ ~~/I '~
-105 ~'~
-110 '"'-"~ …. Z- ~'~'~' 9 ' ' …. ' …. ' …. I 13 13.5 14 14.5 15 15.5 16 16.5 17
B 180 %o (SMOW)
Figure 3. Hydrogen- versus oxygen-isotope plot for kaolin-
ite from the Iwaizumi clay mine (magnified view of Figure
2). Key" ~ = kaolinite samples from Oligocene units; and
[] = kaolinite from upper Cretaceous units. Sample numbers
(IW-1 to -11) are given next to each data point, corresponding
to the information provided in Table 1 and Figure 1. Regres-
sion Line A describes the Oligocene kaolinite samples and
closely matches the Kaolinite Line of Sheppard et al. (1969).
Regression Line B describes the Cretaceous kaolinite samples
and has a much shallower slope (~5) than Line A or the
Kaolinite Line.
ter in equilibrium with Oligocene kaolinites (Figure 2,
area A) plot directly on the meteoric water line of
Craig (1961), overlapping measured values of ground-
water from the Iwaizumi clay mine. Waters in equilib-
rium with the Cretaceous kaolinites at 20 ~ not
shown in Figure 2, plot just to the left of the meteoric
water line. This displacement increases if a lower tem-
perature (15 ~ is used in the calculation (Figure 2,
area B). Regardless of the exact temperature assumed
for formation of these kaolinites, calculated water
compositions remain typical of cool to cool-temperate
climatic regions (Dansgaard 1964; Yurtsever and Gat
1981). Meteoric water with such B-values is quite
common in high latitude and/or altitude regions of
central and northeastern Japan, as well as mid-latitude
regions of eastern, inland China (Mizota and Kusakabe
1994). Cooler climates are not incompatible with ka-
olinite formation during weathering. Bird and Chivas
(1988) emphasized that permeability of the parent
rock, rather than temperature, is the most critical pa-
rameter in the development of thick kaolinite deposits.
The highly permeable character of the parent materials
at Iwaizumi (Figure 1) undoubtedly allowed effective
leaching of silica and other cations.
Post-Formational Hydrogen-Isotope Exchange?
The BD-B180 relationships among the Iwaizumi ka-
olinites are illustrated in more detail in Figure 3. Oli- gocene samples regress along a trend (Line A) that
matches within error the Kaolinite Line of Savin and
Epstein (1970). The small variation in the isotopic
compositions of the Oligocene kaolinites probably re-
flects fluctuations in the isotopic composition of me-
teoric water as it percolated downward during Oligo-
cene kaolinite formation. At the base of the Oligocene
section (sample Iw-6), admixture of kaolinite derived
from the underlying Cretaceous shale is also possible
and consistent with the results illustrated in Figure 3.
As mentioned earlier, the lower B~80 values of the
Cretaceous kaolinites probably indicate that weather-
ing occurred at lower average temperatures (<20 ~
than in Oligocene or younger times, assuming that the
kaolinite oxygen-isotope compositions have remained
unchanged since formation. Compared to the Oligo-
cene kaolinites, the Cretaceous samples also exhibit a
much lower slope (–5) on the BD-B~80 plot (Figure 3,
Line B). We do not believe that this difference arises
from diagenetic alteration of the Cretaceous material,
as overburden thickness probably never exceeded 300
m and maximum burial temperatures were -<30 ~
Instead, we suggest that the lower slope reflects post-
formational, hydrogen-isotope exchange between the
Cretaceous kaolinite and downward percolating Oli-
gocene and younger meteoric water. An alternative ex-
planation is formation from evaporating meteoric wa-
ter. However, this possibility seems unlikely for it re-
quires the thick kaolinitic weathering and associated
enrichment of hematite and gibbsite (Iijima 1972) to
have occurred under very dry conditions, rather than
the wet climates that typically promote the formation
of such weathering products. Still other scenarios, such
as Cretaceous kaolinite formation at higher tempera-
tures than the younger equivalents, yet from meteoric
waters with isotopic compositions typical of a cooler
rather than warmer climate, require so many special
circumstances that they are not considered further.
Figures 2 and 3 illustrate that both Oligocene and
modern meteoric water have appropriate hydrogen-
isotope compositions to produce the necessary, post-
formational enrichment of D in the Cretaceous kaolin-
ites. Bird and Chivas (1988) and Longstaffe and Ay-
alon (1990) have shown that post-formational hydro-
gen-isotope exchange between kaolinite and water can
occur to temperatures -<40 ~ without disturbing the
original oxygen-isotope compositions of these clays.
Bird and Chivas (1988) documented a much larger
shift to the left of the Kaolinite Line for surficial ka-
olinites of Permian to Tertiary age from eastern Aus-
tralia. Their results extend the trend observed here for
the Cretaceous Iwate kaolinites and convincingly dem-
onstrated that such a pattern is best explained by hy-
drogen-isotope exchange with later, isotopically heavi-
er meteoric waters.
For purposes of illustrating the effect of such hy-
drogen-isotope exchange, we have adopted a temper-

414 Mizota and Longstaffe Clays and Clay Minerals
ature of 15 ~ for development of the Cretaceous
weathering profile. Using this value, the isotope frac-
tionation equations for kaolinite-water and the isotopic
expression for the meteoric water line, we can con-
strain the hydrogen-isotope composition of the Creta-
ceous meteoric water to area C on Figure 2. From that,
we have calculated the original, and substantially low-
er, 8/) values of the Cretaceous kaolinite, prior to post-
formational exchange (Figure 2, area C'). The result-
ing isotopic compositions define a 15 ~ surficial ka-
olinite line, which trends to the right of, but parallel
to, the 20 ~ line. Such results lie within the scatter
known for surficial kaolinites about the Kaolinite Line.
Most of this variation probably arises from the range
of temperatures at which surficial kaolinite formation
has occurred throughout the world.
Our model does not require that 15 ~ be exactly
the temperature of Cretaceous weathering at Iwate, or
that this temperature was constant. But it does seem
that interaction with younger, meteoric waters has
preferentially reset the hydrogen-isotope compositions
of these Cretaceous kaolinites. That the 8D values of
the Cretaceous kaolinites increase upward in the sec-
tion may indicate that the lowermost samples have ex-
changed less completely or less continuously with
downward percolating water, particularly later in the
evolution of the profile. The Cretaceous samples may
have possibly exchanged hydrogen primarily with old-
er meteoric waters (Paleocene-Eocene precipitation on
the Cretaceous unconformity?) that were less rich in
D than their younger equivalents. Progressive devel-
opment of the overlying Oligocene kaolinite led to de-
creased permeability, and lower water/rock ratios
within the lower portions of the Iwate section. The
Oligocene kaolinite samples show little effects of post-
Oligocene hydrogen isotopic exchange, perhaps be-
cause less time was available, but more likely because
Oligocene and modern meteoric waters apparently dif-
fered little in isotopic composition, making hydrogen-
isotope exchange very difficult to discern.
Some minor but systematic variation in the condi-
tions of Cretaceous kaolinite formation are indicated
by the small but gradual increase in oxygen-isotope
compositions strafigraphically upward through the
Cretaceous kaolinite (Table 1). One explanation is a
small but systematic change in temperature and hence
the oxygen-isotope composition of meteoric water, as
weathering progressed at the Cretaceous surface. An-
other possibility is progressive depletion of 180 in the
downward percolating groundwater because of water-
rock interaction and neoformation of (1sO-rich) kaolin-
ite. As noted by Clauer (personal communication
1995), our small data set points to the need for thor-
ough study of water/rock ratios, and their implications
for kaolin stable isotope compositions, during devel-
opment of weathering profiles. Associated Hematite and Gibbsite
In the Yokomichi Formation, samples Iw-3 and -5
are characterized by a red coloration (Table 1) that
reflects abundant hematite. Upper horizons also con-
tain up to 11% gibbsite (Iijima 1972). This mineral
association has been interpreted as evidence for ka-
olinite development in a humid, subtropical climate
(Iijima 1972; Tanai et al. 1978). However, in a study
of glacial gravels from the European Alps, Schwert-
mann et al. (1982) showed that reddening (hematite
formation) was possible in temperate climates with
mean annual temperatures as low as 7 ~ and that
warmer, interglacial intervals are not required to ex-
plain rubification of soils. Rather, the prerequisite for
hematite formation is coarse, highly permeable, parent
materials.
Formation of gibbsite during weathering is also not
restricted to tropical or subtropical climates. Excellent
drainage and low-silica parent rocks are more impor-
tant than climate in promoting the formation and per-
sistence of gibbsite in natural environments (Jut 1980;
Macias-Vasquez 1981). Gibbsite is known from Scot-
tish soils, where cool climatic conditions prevail (Wil-
son 1969; Mellor and Wilson 1989; Hall et al. 1989),
from alpine glacial environments of the northwestern
USA (Reynolds 1971; Reynolds and Johnson 1972)
and from coal measures flushed by glacial meltwaters
in northern Canada (Van der Flier-Keller and Fyfe
1987). We conclude that the occurrence of gibbsite and
hematite in the Iwaizumi area is not inconsistent with
the cool-temperate conditions proposed here for
weathering to kaolinite.
Our argument for a cooler climate rests upon the
fact that meteoric waters with the isotopic composi-
tions calculated here are atypical of tropical regions.
But low-180 and low-D precipitation can occur in trop-
ical climates during monsoons (Yurtsever and Gat
1981; Aharon 1983; Bird 1988). Such an explanation
for our results cannot be discounted out of hand. How-
ever, other observations also imply cooler climatic
conditions during formation of the Iwaizumi clay de-
posits. First, isotopically depleted meteoric waters (SD
= -90 to -70%~) have been proposed as the source
of hydrothermal fluids responsible for late Cretaceous
sericite deposits around the Kamioka mine in central
Japan (Shimazaki and Kusakabe 1990). The calculated
~D values are characteristic of cool-temperate to cool
regions (Dansgaard 1964). Second, some paleogeo-
graphic reconstructions indicate that during the late
Cretaceous to Paleogene, the Japanese Islands were
located along the eastern margin of the mid-latitude
region of the east Asian continent (Minato et al. 1965).
Further paleomagnetic studies are needed to confirm
this suggestion and to test our hypothesis of a cooler
climate during formation of the Iwaizumi kaolinite de-
posits.

Vol. 44, No. 3, 1996 Stable isotope geochemistry of kaolinite, Iwaizumi clay deposit, Japan 415
SUMMARY
The ~D and 8180 values of kaolinite from late Cre-
taceous and Oligocene deposits at Iwaizumi, north-
eastern Japan, suggest that these clays formed during
weathering, rather than hydrothermal alteration. The
kaolinites from the Oligocene deposits are sufficiently
depleted of D and 180 to suggest that weathering oc-
cur-red under a cool to cool-temperate climate rather
than during warm-temperate or tropical conditions.
This observation adds to the growing body of thought
that other variables such as permeability of parent ma-
terials are more critical than temperature to the devel-
opment of thick kaolinite deposits. The kaolinites from
the Cretaceous deposits are even more depleted of ~80,
suggesting formation under even cooler climatic con-
ditions than the overlying Oligocene equivalents.
However, the hydrogen-isotope compositions of these
older kaolinites have a distribution that suggests ex-
change with younger, downward percolating meteoric
water. These data provide further support for the view
that hydrogen-isotope exchange between kaolinite and
water can occur in sedimentary environments at quite
low temperatures, independent of oxygen-isotope ex-
change. Any interpretation of hydrogen- and oxygen-
isotope compositions for clay minerals from such sys-
tems must recognize such a possibility.
ACKNOWLEDGMENTS
We are grateful for the field assistance of K. Yagishita,
Iwate University and M. Orihashi, Iwate Clay Mine Corpo-
ration, and the laboratory help of E Middlestead and D. La-
cina (University of Western Ontario). FJL also acknowledges
financial support from the Natural Sciences and Engineering
Research Council of Canada. We thank J. Lawrence and N.
Clauer for thought-provoking reviews of the original manu-
script.
REFERENCES
Aharon P. 1983. Analysis of the anomalous 180]160 and D/H
isotope ratios in the tropical rainfall over the western Pa-
cific Ocean. Trans Am Geophys Union 64:196.
Bigeleisen J, Perlman ML, Prosser HC. 1952. Conversion of
hydrogenic materials to hydrogen for isotopic analysis.
Anal Chem 24:1356-1357.
Bird MI. 1988. Isotopically depleted rainfall and E1 Nino.
Nature 331:489-490.
Bird MI, Chivas AR. 1988. Stable-isotope evidence for low-
temperature kaolinitic weathering and post-formational hy-
drogen-isotope exchange in Permian kaolinites. Chem Geol
(Isotope Geosci Sec) 72:249-265.
Biscaye PE. 1965. Mineralogy and sedimentation of recent
marine deep sea clay in the Atlantic Ocean and adjacent
seas and oceans. Geol Soc Am Bull 76:802-832.
Botz RW, Hunt JW, Smith JW. 1986. Isotope geochemistry
of minerals in Australian bituminous coal. J Sediment Pet-
rol 56:99-111.
Clauer NE 1995. Centre de Geoch. de la Surface 1, rue Bles-
sig, 67084 Strasbourg, France.
Clayton RN, Mayeda TK. 1963. The use of bromine penta-
fluoride in the extraction of oxygen from oxides and sili-
cates for isotopic analysis. Geochim Cosmochim Acta 27:
43-52. Coleman ML, Sheppard T J, Durham J J, Rouse JE, Moore
GR. 1982. Reduction of water with zinc for hydrogen iso-
tope analysis. Anal Chem 54:993-995.
Craig H. 1961. Isotopic variations in meteoric waters. Sci-
ence 133:1702-1703.
Dansgaard W. 1964. Stable isotopes in precipitation. Tellus
16:436–468.
Davey BG, Russell JD, Wilson MJ. 1975. Iron oxide and
clay minerals and their relation to colours of Red and Yel-
low Podzolic Soils near Sydney, Australia. Geoderma 14:
125-138.
Dutta PK. 1985. In search of the origin of cement in sili-
clastic sandstones: An isotopic approach. Chem Geol (Iso-
tope Geosci Sec) 52:337-348.
Epstein S, Mayeda TK. 1953. Variation of O t8 content of
waters from natural sources. Geochim Cosmochim Acta 4:
213-224.
Eslinger EV. 1971. Mineralogy and oxygen isotope ratios of
hydrothermal and low-grade metamorphic argillaceous
rocks [Ph.D. Dissertation]. Cleveland, OH: Case Western
Reserve University. 205 p.
Fujii N. 1970. Genesis of the flint clay deposit at the Iwate
mine, northeast Japan. J Geol Soc Japan 76:623-636 (in
Japanese).
Godfrey JD. 1962. The deuterium content of hydrous min-
erals from the East-Central Sierra Nevada and Yosemite
National Park. Geochim Cosmochim Acta 26:1215-1245.
Hall AM, Mellor A, Wilson MJ. 1989. The clay mineralogy
and age of deeply weathered rock in north-east Scotland.
Z Geomorph 72:97-108.
Hassanipak AA, Eslinger EV. 1985. Mineralogy, crystallin-
ity, 180/160 and D/H of Georgia kaolins. Clays & Clay
Miner 33:99-106.
Hu A, Zhang R. 1988. Genesis of Iwate clay deposit, Japan.
In:Report of International Research and Development Co-
operation, Project No. 8314. Geological Survey of Japan,
Agency of Industrial Science and Technology, Ministry of
International Trade and Industry, Tsukuba, Ibaraki. p 117-
137 (in Japanese).
Iijima A. 1972. Latest Cretaceous-Early Tertiary lateritic
profile in northeastern Kitakami Massif, northeast Honshu,
Japan. J Fac Sci Univ Tokyo, II 18:325-370.
Juo ASR. 1980. Mineralogical characteristics of Alfisols and
Ultisols. In: Theng BKG, editor. Soils with variable charge.
New Zealand Soc Soil Sci. p 69-86.
Kato M, Fujiwara Y, Minoura N, Koshimizu S, Saito M.
1986. Fission track age of zircons in the Upper Cretaceous
Yokomichi Formation, northern Kitakami Mountains, Ja-
pan. J Geol Soc Japan 92:821-822. (in Japanese)
Kyser TK. 1987. Equilibrium fractionation factors for stable
isotopes. In: Kyser TK, editor. Stable isotope geochemistry
of low temperature processes. Mineralog Assoc Canada,
Short Course 13:1-84.
Lambert SJ, Epstein S. 1980. Stable isotope investigations
of an active geothermal system in Valles Caldera, Jemez
Mountains, New Mexico. J Volcan Geotherm Research 8:
111-129.
Land LS, Dutton SP. 1978. Cementation of a Pennsylvanian
deltaic sandstone: Isotopic data. J Sediment Petrol 48:
1167-1176.
Lawrence JR, Taylor HP, Jr. 1971. Deuterium and oxygen-18
correlation: Clay minerals and hydroxides in Quaternary
soils compared to meteoric waters. Geochim Cosmochim
Acta 35:993-1003.
Lawrence JR, Taylor Jr HR. 1972. Hydrogen and oxygen
isotope systematics in weathering profiles. Geochim Cos-
mochim Acta 36:1377-1393.

416 Mizota and Longstaffe Clays and Clay Minerals
Liu K-K, Epstein S. 1984. The hydrogen isotope fraction-
ation between kaolinite and water. Isotope Geosci 2:335-
350 and Chem Geol 46:335-350.
Longstaffe FJ. 1983. Diagenesis, IV. Stable isotope studies
of diagenesis in clastic rocks. Geosci Can 10:44-58.
Longstaffe FJ. 1989. Stable isotopes as tracers in clastic dia-
genesis. In: Hutcheon IE, editor. Burial diagenesis. Miner-
alog Assoc Canada, Short Course 15:201-277.
Longstaffe FJ, Ayalon A. 1990. Hydrogen-isotope geochem-
istry of diagenetic clay minerals from Cretaceous sand-
stones, Alberta, Canada: evidence for exchange. Appl Geo-
chem 5:657-668.
Macias-Vasquez E 1981. Formation of gibbsite in soils and
saprolites of humid temperate zones. Clays & Clay Miner
16:43-52.
Marumo K, Nagasawa K, Kuroda Y. 1979. The hydrogen
isotopic composition of kaolin minerals in Japan. J Jpn As-
soc Miner Petrol Econ Geol 74:294-300 (in Japanese).
Marumo K, Nagasawa K, Kuroda Y. 1980. Mineralogy and
hydrogen isotope geochemistry of clay minerals in the
Ohnuma geothermal area, northeastern Japan. Earth Planet
Sci Lett 47:255-262.
Marumo K, Matsuhisa Y, Nagasawa K. 1982. Hydrogen and
oxygen isotopic compositions of kaolin minerals in Japan.
In: Van Olphen H, Veniale E editors. Developments in sed-
irnentology. International Clay Conference 1981. Elsevier
Sci Pub 35:315-320.
Marumo K, Longstaffe FJ, Matsubaya O. 1995. Stable iso-
tope geochemistry of clay minerals from fossil and active
hydrothermal systems, southwestern Hokkaido, Japan.
Geochim Cosmochim Acta 59:2545-2559.
Mellor A, Wilson MJ. 1989. Origin and significance of gibbsitic
montane soils in Scotland, U.K. Arctic Alpine Res 21:417-
424.
Minato M, Gorai M, Funahashi M. 1965. The geological
development of the Japanese Islands, Tsukiji Shokan.
442 p.
Mizota C, Kusakabe M. 1994. Spatial distribution of gD and
8180 values of surface and shallow groundwaters from Ja-
pan, south Korea and east China. Geochem J 28:387-410.
Oilier CD. 1969. Weathering. New York: American Elsevier.
1-270.
O'Neil JR, Kharaka YK. 1976. Hydrogen and oxygen iso-
tope exchange reactions between clay minerals and water.
Geochim Cosmochim Acta 40:241-246.
Reynolds RC. 1971. Clay mineral formation in an alpine en-
vironment. Clays & Clay Miner 19:361-374. Reynolds RC, Johnson NM. 1972. Chemical weathering in
the temperate glacial environment of the Northern Cascade
Mountains. Geochim Cosmochirn Acta 36:537-554.
Ruiz Cruz MD, Moreno Real L. 1993. Diagenetic kaolin-
ite/dickite (Betic Cordilleras, Spain). Clays & Clay Miner
41:570-579.
Savin SM, Epstein S. 1970. Oxygen and hydrogen isotope
geochemistry of clay minerals. Geochim Cosmochim Acta
34:25-42.
Savin SM, Lee M. 1988. Isotopic studies of phyllosilicates.
In: Bailey SW, editor. Hydrous phyllosilicates (exclusive of
micas). Mineralog Soc Am, Reviews in Mineralogy 19:
189-223.
Schwertmann U, Murad E, Schulze DG. 1982. Is there Ho-
locene reddening (hematite formation) in soils of axeric
temperature areas? Geoderma 27:209-223.
Sheppard SME Nielson RL, Taylor HE Jr. 1969. Oxygen and
hydrogen isotope ratios of clay minerals from porphyry
copper deposits. Econ Geol 64:755-777.
Sheppard SME 1977. The Cornubian batholith, SW England:
D/H and ~sO/160 studies of kaolinite and other alteration
minerals. J Geol Soc (London) 133:573-591.
Shimazaki H, Kusakabe M. 1990. D/I-I ratios of sericites
from the Kamioka mining area. Mining Geol 40:385-388.
Suzuoki T, Epstein S. 1976. Hydrogen isotope fractionation
between OH-bearing minerals and water. Geochim Cos-
mochim Acta 40:1229-1240.
Tanai T, Iijima A, Agatsuma T. 1978. Late Cretaceous-Pa-
leogene stratigraphy in the environs of the lwate clay mine,
northern Kitakami Massif, northeast Honshu. J Geol Soc
Japan 84:459-473 (in Japanese).
Taylor HE Jr. 1974. The application of oxygen and hydrogen
isotope studies to problems of hydrothermal alteration and
ore deposition. Econ Geol 69:843-883.
Van der Flier-Keller E, Fyfe WS. 1987. Geochemistry of two
Cretaceous coal-bearing sequences: James Bay lowlands,
northern Ontario, and Peace River basin, northeast British
Columbia. Can J Earth Sci 24:1038-1052.
Wilson MJ. 1969. A gibbsitic soil derived from the weath-
ering of an ultrabasic rock on the island of Rhum. Scott J
Geol 5:81-89.
Yurtsever Y, Gat JR. 1981. Atmospheric waters. In: Gat JR,
Gonfiantini R, editors. Stable isotope hydrology: Deuterium
and oxygen-18 in the water cycle. Vienna: International
Atomic Energy Agency. p 103-142.
Zheng Y-E 1993. Calculation of oxygen isotope fraction-
ation in hydroxyl-bearing silicates. Earth Planet Sci Lett
120:247-263.
(Received 10 July 1994; accepted 18 September 1995; Ms.
2539)