Content uploaded by Alexander U Falster
Author content
All content in this area was uploaded by Alexander U Falster on Nov 17, 2017
Content may be subject to copyright.
The Canadian Mineralogist
Vol. 54, pp. 1053-1070 (2016)
DOI: 10.3749/canmin.1600017
BULK COMPOSITION OF MT. MICA PEGMATITE, MAINE, USA:
IMPLICATIONS FOR THE ORIGIN OF AN LCT TYPE PEGMATITE BY ANATEXIS
WILLIAM SIMMONS
§
,ALEXANDER FALSTER, AND KAREN WEBBER
Maine Mineral and Gem Museum, 99 Main St, Bethel, Maine 04217, U.S.A.
ENCARNACI ´
ON RODA-ROBLES
Dept. de Mineralog´
ıa y Petrolog´
ıa, Univ. del Pa´
ıs Vasco UPV/EHU, P.O. Box. 644, E-48080 Bilbao, Spain
ANDREW P. BOUDREAUX AND LEAH RAE GRASSI
Gulf of Mexico Business Unit, Chevron Corporation, Covington, Louisiana 70435, U.S.A
GARY FREEMAN
Coromoto Minerals, LLC, 48 Lovejoy Road, South Paris, Maine 04281, U.S.A.
ABSTRACT
The Mt. Mica pegmatite is famous for producing gem tourmaline for nearly 200 years. The dike, ranging in thickness from 1 to
8 m and dipping 208SE, has a simple zonal structure consisting of a wall zone and core zone. The wall zone is essentially devoid
of K-feldspar. The outer portion of the pegmatite consists of quartz, muscovite, albite (An 1.8), and schorl. Muscovite is the
dominant K-bearing species in the outer portion of the pegmatite. Potassium feldspar only appears in the core zone adjacent to
pockets. The pegmatite is subparallel to the foliation of the enclosing migmatite, and leucosomes show a gradational contact with
the pegmatite where juxtaposed. Texturally, the pegmatite and leucosomes appear to be in equilibrium with no change in grain size
or composition where the two are in contact. Garnet-biotite thermometry of the migmatite at the contact yields an average
temperature of 630 8C, which is consistent with the pressure-temperature conditions inferred for a Sebago Migmatite Domain
(SMD) assemblage of sillimanite, quartz, muscovite, biotite, and alkali feldspar formed at 650 8C and 3 kb. Gradational contact
between the leucosomes and pegmatite suggests that the pegmatitic melt was at the same temperature. Coromoto Minerals LLC
began mining in 2003 and the mine now extends down-dip for over 100 m to a depth of 33 m. A very detailed and accurately
surveyed geologic map produced by owner/operator Gary Freeman during mining shows the total area of pegmatite removed, the
spatial distribution and aerial extent of pockets, massive lepidolite (compositions near trilithionite) pods, microcline, and xenoliths.
The map was analyzed using image analysis and thickness values of the units to calculate the total volumes of pegmatite mined,
lepidolite pods, and all pockets found. Forty-five drill cores were taken across the pegmatite from the hanging wall to foot wall
contacts, along a transect intentionally avoiding lepidolite pods and miaroles. Cores were pulverized, thoroughly mixed and
homogenized, and the percent Li content calculated from the mapped volume was added to produce a sample that was
representative of the bulk composition of Mt. Mica. The sample was then analyzed by fusion ICP spectroscopy for major and trace
elements and DCP spectroscopy for B and Li. Structural water was determined by LOI. Water content was calculated using the
calculated volume of open space (pocket volumes), assuming that the pockets were filled with water-rich fluid. This fluid content
was added to LOI water (above 500 8C) to estimate a maximum H
2
O content of 1.16 wt.% of the pegmatite melt. REE plots of
bulk pegmatite versus leucosomes from the migmatite are strikingly similar. Chondrite-normalized REE patterns of leucosomes
and pegmatite are very flat with no Eu anomaly, whereas the Sebago granite is more strongly LREE-enriched and displays a
pronounced negative Eu anomaly. Spider diagrams of leucosomes and pegmatite versus average crust show very similar patterns.
These results suggest that the Mt. Mica pegmatitic melt did not form by fractional crystallization of the older Sebago pluton, but
instead was derived directly from partial melting of the metapelitic rocks of the SMD. Batches of anatectic melt accumulated and
coalesced into a larger volume that subsequently formed the pegmatite. This is the first chemical evidence presented for the
formation of an LCT type pegmatite by direct anatexis.
Keywords: Mt. Mica, pegmatite, pegmatite bulk composition, anatexis, Sebago Migmatite Domain, Maine.
§
Corresponding author e-mail address:wsimmons@uno.edu
1053
INTRODUCTION
The Mt. Mica pegmatite is the site of the first
reported occurrence of tourmaline in North America
(Hamlin 1873, 1895) and is famous for producing gem
tourmaline for nearly 200 years. It is experiencing a
remarkable new chapter in its long-lived and histor-
ically important mining history. Coromoto Minerals
LLC acquired the property and began mining in 2003.
Recent mining by owner Gary Freeman has produced a
large amount of quantitative information about the
pegmatite. Careful mapping of the location and sizes
of lepidolite (used as the series name for trioctahedral
micas near trilithionite in composition) masses,
pockets, and the volume of pegmatite mined under-
ground has provided an unprecedented data base
which we utilized in this study. Over the last dozen
years of mining, several hundred pockets have been
carefully documented. Their sizes and contents have
been carefully measured and recorded. These pockets
have yielded large gem-quality crystals of green and
pink tourmaline rivaling the best material ever
produced from Mt. Mica in its 195-year history. The
new phase of mining activity by Coromoto Minerals
LLC, which began as an open trench mine in 2003,
now continues down-dip underground for over 100 m
to a depth of about 30 m. The pockets range in size
from a few cm
3
to one in excess of 500 m
3
. Several
dozens of the intermediate to larger pockets have
produced thousands of carats of gem-quality tourma-
line and lesser quantities of morganite. Pocket density
averages about one every 3 m with larger pockets
having greater spacing and small ones having less,
making this one of the most pocket-rich pegmatites in
North America. In addition to the gem material,
thousands of high-quality mineral specimens including
tourmaline, beryl, apatite, lepidolite, rose and smoky
quartz, hydroxylherderite, cassiterite, pollucite, and
kosnarite have been recovered.
GENERAL GEOLOGY
Host rocks
Most of the exposed bedrock of Maine is Paleozoic
in age. Western Maine has been affected by several
orogenic events, alternating with periods of sedimen-
tation. The tectonic events began with the Cambrian
Penobscottian orogeny which affected NW to N-
central Maine. The Ordovician Taconic orogeny which
followed is probably related to collisions of volcanic
island arcs with North America. Subsequently, the
Devonian Acadian orogeny was caused by the
collision of North America with the Avalon micro-
continent.
The Mt. Mica pegmatite occurs near Paris, Maine,
USA in rocks of the Central Maine Belt (CMB). The
CMB, in the northern Appalachians, is a prominent
NE–SW-trending unit that is composed of a Lower
Paleozoic sedimentary sequence that was intruded by
Devonian to Permian igneous rocks (Solar & Brown
2001a, 2001b, Solar & Tomascak 2009, Tomascak et
al. 1996). In this belt, the metamorphic facies change
from greenschist in the northern to upper amphibolite
facies and migmatite in the southern portion (Guidotti
1989, 1993). Rare-element-bearing pegmatites are
relatively common in the CMB. They are typical
LCT-type pegmatites ( ˇ
Cern´
yet al. 2012) and occur
mainly in the southwestern portion of the CMB in an
area known as the Oxford pegmatite field (Wise &
Brown 2010). The best-studied pegmatites in this field
occur inside or at the limits of the Sebago Migmatite
Domain (SMD). This domain surrounds the re-defined
and now much smaller Sebago pluton (,400 km
2
) and
consists of stromatic migmatites and diatexites,
permeated by smaller bodies of more geochemically
heterogeneous granite (Solar & Tomascak 2009).
Migmatites and upper amphibolite metasandstones,
metapelites, and metacarbonates of Silurian age are
common host lithologies of the pegmatites (Wise &
Francis 1992).
Uranium-Pb data indicate that the age of the
Sebago pluton is 296 63 Ma (Foord et al. 1995) to
293 Ma (Tomascak et al. 1996). Thus, the granitic
melts associated with anatexis of the metasedimentary
sequence are inferred to have intruded during the
Alleghanian orogeny (Bradley et al. 2016). Two-mica
granites are the most common plutonic rock of the
Sebago pluton. Less common are biotite granites and
megacrystic leucogranites (Wise & Brown 2010).
Geochemical characteristics of the southwestern
Maine pegmatites indicate that they should be
classified tectonically as syn-collisional (Simmons et
al. 2003).
Pegmatites
Overall, pegmatite bodies in the SMD are
concordant with the foliation of the host rocks,
although some display irregular and locally discordant
contacts. Generally, pegmatites in the SMD exhibit
internal mineralogical and textural zonation. The
outermost wall zone is typically less than 1 m thick,
with a homogeneous pegmatitic texture. Potassium
feldspar, albite, quartz, biotite, muscovite, and garnet
6schorl are the most common minerals in this zone,
except in Mt. Mica, which has only albite in the wall
zone with microcline appearing only in the core zone.
Many pegmatites exhibit a comb structure of wedge-
shaped schorl crystals, some up to 50 cm in length,
1054 THE CANADIAN MINERALOGIST
that grow directly from the contact with the country
rock in the hanging wall, inside a homogeneous
pegmatite matrix (this is particularly well developed
in the Emmons pegmatite). The wall zone is
gradational into the intermediate zone of graphically
intergrown feldspar and quartz, with accessory garnet,
biotite, muscovite, and black tourmaline. In some
pegmatites, the intermediate zone is asymmetric and
thicker under the core zone, e.g., in the Havey
pegmatite. The core zones consist of meter-sized
masses of blocky K-feldspar and/or quartz with a
more evolved assemblage of finer-grained irregular
pods consisting of albite, muscovite, lepidolite, Li-
tourmaline, Li-Al and Fe-Mn-phosphates, Sn-, Nb-,
and Ta-oxides, beryl and Cs-beryl, spodumene,
petalite, and pollucite. Within the core-zone pods,
pockets are relatively common. The pockets contain
gemmy elbaite associated with quartz, lepidolite,
albite, cassiterite, and clay minerals. Another notable
feature in the foot wall of CMB pegmatites is the
existence of a garnet layer or line under the core zone
and roughly paralleling the contact of the pegmatite.
The garnet line helps the miners to determine the
limits of the core and pockets, as pockets never occur
below this layer. Prismatic schorl crystals intergrown
with or close to the garnet line are oriented
perpendicular to the contact and indicate the sense
of upward crystal growth. In some cases, a second line
of schorl occurs below the garnet line. Overall, the
composition of CMB pegmatites is not significantly
enriched in rare elements, as these are concentrated
only in the innermost parts of the bodies and
constitute a low percentage of the volume of the
pegmatite. Textural and mineralogical criteria indicate
that pegmatites in the CMB crystallized from the
borders inward. Fractionation processes were highly
effective during the crystallization of these pegma-
tites. In general, the composition of the primary
minerals changes progressively from the contact to
the core zone, where compositions change more
sharply. The Li/(Fe þMg) ratio of tourmaline
increases, the K/Rb of micas and K-feldspar and the
Fe/(Fe þMn) of phosphates decreases, Li and F in
mica and Cs in beryl increases as well as a general
increase in the proportion of Li-, F-, and Cs-bearing
minerals, parallel to a decrease in Fe-Mn-Mg-bearing
phases (Roda-Robles et al. 2015, Wise & Brown
2010).
METHODS
Whole rock sampling
About 10 m beneath the surface, a Stihl model
B261C hand-held diamond core drill was used to
collect 45 2.5 315 cm cores of the pegmatite from the
hanging-wall contact to the foot-wall contact. The
spacing between the cores was about 10 cm and the
transect intentionally avoided lepidolite masses and
miaroles. The cores were sliced longitudinally and half
of each was pulverized. To achieve homogeneity, the
powders were mixed by vigorous shaking in a
ceramic-lined Spec ball mill for 30 min. Equal aliquots
of 20 g of the resulting powders were thoroughly
mixed and homogenized to produce a composite
powder from the 45 core samples that was represen-
tative of the bulk composition of the pegmatite sans
lepidolite masses.
Whole rock fusion ICP analyses
Whole-rock analyses were performed at the
ACTLABS analytical facility in Ontario, Canada.
The analytical package uses fusion ICP-OES for the
major elements and ICP-MS for trace elements.
Scanning electron microscopy
An AMRAY 1820 digital SEM was used for
imaging and qualitative analyses and was operated at
an acceleration potential of 20 kV, a final aperture of
400 lm, 08sample tilt, and a working distance of 18
mm. Images were acquired at a resolution of 2048 3
2048 pixels using Iridium Ultra, part of an integrated
software package by IXRF SYSTEMS, Inc.
Electron microprobe analyses
Quantitative chemical analyses of these samples
were obtained using an ARL-SEMQ electron micro-
probe in the wavelength-dispersive mode with an
accelerating potential of 15 kV, 15 nA beam current,
and 2 lm beam diameter. The following elemental
standards were used: adularia-St. Gotthard (K, Si);
albite-Tiburon (Na, Al); An50-Nain (Ca, Al); Cpx-26
(Fe, Mg); rhodonite-Broken Hill (Mn); montebrasite
(Al, P); triphylite (P, Fe); lithiophilite (Mn); TiO
2
synthetic (Ti); pollucite (Cs); Rb-leucite (Rb); fluo-
rapatite-Cerro de Mercado (P); and fluorphlogopite
synthetic (F). Five spots per sample were analyzed
with count times of 45 s per spot. Backgrounds were
determined using the MAN method (Donovan &
Tingle 1996), utilizing any applicable standard listed
above and the following additional standards: andalu-
site, hematite-Elba, CaWO
4
, PbO, V
2
O
5
, ZrO
2
, MgO,
PbO, ZnO, and Al
2
O
3
. Matrix effects were corrected
using the UqZ correction procedure (Pouchou &
Pichoir 1991). Detection limits for Fe, Mn, Ti, Ca:
0.007 wt.%; for Nb, Ta, U, Pb, K, Si, Al, Sc: 0.009–
0.011 wt.%, Mg: 0.012 wt.%, Na: 0.022 wt.%, and F:
0.020 wt.%.
BULK COMPOSITION OF MT. MICA PEGMATITE 1055
Direct-coupled plasma spectroscopy
A Spectronics Spectroscan V instrument was used
to analyze for Li and B. The instrument was operated
at a photomultiplier tube voltage of 10000 V and with
count times of 10 s at 670.784 nm for Li and 249.733
nm for B. Standards were prepared from spectro-
graphic-grade pure lithium carbonate and boric acid
that was dissolved in diluted nitric acid to produce the
concentrations of the standards used. Samples of
known concentration were analyzed as check stan-
dards. High standards ranged from 10 to 100 ppm, and
low standards ranged from 0.001 to 0.1 ppm of the
element analyzed. Samples were prepared by dissolv-
ing ~200 mg of each in 5–10 mL of 51% hydrofluoric
acid, later diluted with distilled water to 0.035 L.
Fluorine determination of bulk pegmatite
Approximately 50 mg of powder from sample 45b
was fused to a glass bead on a platinum strip attached
to the electrodes of the carbon coater in a vacuum of 1
310
–5
torr. The powder was flash-melted on the strip
by applying voltage to the strip, which was instantly
heated to white hot. The resulting glass beads were
then mounted in epoxy, ground down until intercep-
tion, and then polished according to standard proce-
dures and coated with 250 615 ˚
A of carbon in a
vacuum of 1 310
–5
torr. The fused beads were then
analyzed in an ARL-SEMQ electron microprobe for
the major and common minor elements in order to
apply proper ZAF corrections. To assess the results,
the major element values other than F were compared
to the ICP fusion analyses and were found to be
virtually identical to within about 1 wt.% of the ICP
values of the major elements except for Na, which was
only within 10 wt.% of the ICP value, due to it its
volatility during melting.
GENERAL GEOLOGY OF THE PEGMATITE
The Mt. Mica pegmatite intrudes stromatic
migmatite of the SMD (Solar & Brown 2001b,
Solar & Tomascak 2009) (Fig. 1). The migmatite
consists of felsic leucosomes of quartz and feldspar
and melanosomes of biotite-quartz-feldspar schist.
The contact between the pegmatite and melanosomes
is sharp but completely gradational between the
leucosomes and the pegmatite (Fig. 2). In a few
places along the contact, a weakly developed 2–4 cm
comb structure of oriented muscovite crystals is
present. In most places the contact is parallel to
foliation, but in places where the pegmatite cuts
foliation, ductile deformation is clearly evident (Fig.
2). In general, zoning in the Mt. Mica pegmatite is
indistinct. Internal zoning is asymmetric and not well
developed and, basically, in most places no interme-
diate zone can be distinguished (Fig. 3). The wall
zones extend inward from the hanging wall and foot
wall contacts to the core zone. Grain size increases
gradually up to the core zone which is marked by an
abrupt increase in grain size, especially of muscovite
which forms large 4 to 10 cm ‘‘ A’’-shaped twinned
books, and by the appearance of more evolved
minerals and miarolitic cavities. The dike ranges in
thickness from 1 to 8 m and dips about 20 to 258to
the SE. To date, the mining extends underground
down-dip about 100 m to a depth of 33 m beneath the
surface. The outer zones of the pegmatite consist of
FIG. 1. Mt. Mica pegmatite showing the contact with the
overlying stromatic migmatite host rock. Mine opening
2.5 m. FIG. 2. Mt. Mica contact relationship with migmatite showing
ductile deformation of migmatite and gradational nature
of the contact between the pegmatite (upper portion of
image) and the leucosomes of the migmatite. Image width
is 0.5 m (underground workings).
1056 THE CANADIAN MINERALOGIST
nearly endmember albite (An ~0.5), quartz, musco-
vite, and schorl. Visual estimates and point counts of
polished slabs yielded a modal composition of the
wall zone of roughly 40% quartz, 39% albite, and
19% muscovite. The wall zone of Mt. Mica is
unusual, as it is essentially devoid of K-feldspar; the
major K-bearing mineral is muscovite. Potassium
feldspar only appears as large masses in the core zone
of the pegmatite adjacent to some of the larger
pockets, where it serves as a pocket indicator in some
cases. Notably, one, or in some instances two,
distinctive garnet and schorl lines occur along the
somewhat finer-grained footwall portion of the dike
(Fig. 3). The lines are about 0.5 to 1.0 m above the
base of the pegmatite and are roughly parallel to the
pegmatite-country rock contact. The line consists of
an undulating, somewhat sinusoidal to irregular
concentration of 1 to 5 cm garnet crystals in a matrix
of quartz and feldspar. The most evolved portion of
the pegmatite is always above this horizon in the
more incompatible-rich core zone of the pegmatite,
where the pockets occur. Thus, this line is an
important marker that the miners interpret to be the
bottom of the productive zone, below which no
pockets occur.
The core consists mainly of quartz, albite, micro-
cline, and schorl with local pods of white cleavelandite
and, less commonly, pods of deep purple lepidolite
with elbaite, spodumene, pollucite, cassiterite, colum-
bite-group minerals, and rare beryl (Figs. 4–9). Thin
section analyses of the lepidolite masses revealed that
their bulk modal composition is 71% lepidolite and
29% quartz and albite.
Miarolitic cavities or pockets are relatively com-
mon in Mt. Mica and are the source of the gem elbaite
that this pegmatite is so famous for. The pockets are
ovoid in shape and tend to be most elongated in the
horizontal plane and range in size from a few cm to
one chamber in excess of 11 m across.
FIG. 3. Internal structure of the Mt. Mica Pegmatite, showing
the hanging wall contact, wall zone, core zone, and the
positions of the garnet and schorl lines below the core
zone in the lower wall zone. Pegmatite cross section 1.5 3
2.5 m.
FIG. 4. Evolved mineralogy in the core zone with lepidolite,
montebrasite, albite var. cleavelandite, and carbonate
units in the center.
FIG. 5. Lepidolite mass with associated large montebrasite
pods, altered spodumene laths, and green elbaite in the
core zone. About 2 m across.
FIG. 6. Lepidolite, pink and green elbaite, and albite next to a
large pollucite mass.
BULK COMPOSITION OF MT. MICA PEGMATITE 1057
MINERALOGY
Feldspars
The dominant feldspar in Mt. Mica is a white
albite. Potassium feldspar is virtually absent in the
outer zone of the pegmatite. The Ca content of the
albite is uniformly low in the wall zone, averaging
about An
2.0
. Albite in the core zone approaches pure
endmember albite, averaging An
0.2
. Large, beige-
colored K-feldspar crystals, some over a meter in
maximum dimension, occur within the core zone. The
crystals are perthitic and determined by X-ray
diffraction to be near-maximum microcline. Micro-
cline typically occurs close to or extending into
miarolitic cavities. The average K/Rb ratio of 150
for K-feldspar (Marchal et al. 2014) is somewhat high
for a B-rich LCT pegmatite, suggesting that Mt. Mica
is not very evolved. However, the lower K/Rb ratio of
40 for K-feldspar located in and around pockets and
pollucite pods reveals that Mt. Mica is moderately
evolved in these regions (Marchal et al. 2014).
Muscovite and lepidolite
Muscovite and Li-rich muscovite are the principal
micas and are the dominant K-species in the wall zone
of the pegmatite. The micas are small and show little
change in composition from the pegmatite contact up
to the core zone margin, with Li content ranging from
.0.01 to 0.9 apfu (Marchal et al. 2014). There is an
abrupt increase in Li-rich muscovite crystal size at the
core zone where large, euhedral, twinned mica crystals
up to 17 cm across occur. Crystals in close proximity
to or extending into pockets are commonly rimmed
with lepidolite (used as the series name for trioctahe-
dral micas near trilithionite in composition) (Fig. 10).
The rims consist of a mosaic overgrowth of 2 to 5 mm
lepidolite crystals. There is a sharp boundary in
composition between the muscovite and the lepidolite
rim. Pods of lepidolite, up to several m in size, are
intermittently distributed within the core zone. The
FIG. 7. A m-sized pollucite mass with veins of lepidolite.
FIG. 8. A miarolitic cavity, primarily filled with quartz
crystals. About 50 cm field of view.
FIG. 9. A miarolitic cavity with green tourmaline and a large
thick tabular beryl crystal. Field of view about 25 cm
FIG. 10. A-type twinned Li-rich muscovite crystal rimmed by
lepidolite, from a miarolitic cavity.
1058 THE CANADIAN MINERALOGIST
lithium content of all lepidolite from pods and rims
ranges from 2.0 to 3.4 apfu (Marchal et al. 2014). In a
few instances, macroscopic interlayering of muscovite
and lepidolite on a several-micron scale occurs where
a muscovite crystal extends into pockets or lepidolite
pods. The details of the mica compositional ranges are
presented in Marchal et al. (2014).
Tourmaline
Tourmaline is present in all the pegmatite zones. In
the wall zone, tourmaline occurs as fine- to coarse-
grained anhedral to subhedral prisms of black schorl
and is occasionally graphically intergrown with quartz.
In places, a distinct comb structure of tapered or
wedge-shaped crystals fans out from the contact into
the wall zone, pointing towards the interior of the
pegmatite. This texture is similar to that described
from San Diego Co., California pegmatites that has
been attributed to rapid crystal growth (Webber et al.
1997, 1999). Schorl crystals also radiate around some
pockets and the convergence direction of their long
axes serves as a pocket indicator. Crystals that extend
close to or into a pocket are typically color-zoned,
grading from black to green to pink in color (Fig. 11).
In the core zone around and near pockets the
tourmaline is dominantly elbaite that occurs as opaque
to translucent green, mm- to cm-size prisms. In the
pods of fine-grained lepidolite masses, small pinkish
tourmaline crystals are relatively common, but most
are altered to clay minerals. Inside the pockets in the
core zone spectacular, gem-quality prismatic crystals
of blue to green to pink, pink and green color-zoned
prisms, and watermelon tourmaline occur, some as
large as 15 cm in length (Fig. 12). This elbaite is
associated with albite var. cleavelandite, quartz
crystals up to 30 cm, and medium-to-very coarse
crystals or books of Li-rich muscovite that may be
rimmed by lepidolite. The tops of some pink elbaite
crystals are capped with a thin layer of black
tourmaline; the crystals are similar to the Mohrenkopf
crystals from Elba, Italy.
Most pockets are partially filled with a rubble of
broken elbaite crystals lying on the floor of the pocket.
In some cases, the larger fragments can be reassem-
bled into the original crystals. Tourmaline exhibits the
expected trend of Fe-rich schorl in the outer portion of
the pegmatite evolving to greater contents of Al and Li
in the core zone (Simmons et al. 2005a, b). The pink
and green color of elbaite in pockets is mainly a
consequence of Fe content. Colors grade from black to
blue and blue-green to light green to pink as Fe content
drops in the core zone and particularly in pockets
(Simmons et al. 2005a, b).
Columbite-group mineralogy
Columbite-group minerals show only minor to
moderate enrichment in Mn and Ta, reaching only
columbite-(Mn) (Simmons et al. 2013). The Mt. Mica
pegmatite is not very rich in columbite-group
minerals. Among the high field-strength element-
bearing minerals, cassiterite and zircon are the most
abundant. In fact, Mt. Mica seems rather depleted in
Ta relative to other B-rich LCT pegmatites.
Cassiterite
Cassiterite is widespread in the interior zones of the
Mt. Mica pegmatite. In the late 1800s an attempt was
made to mine cassiterite from this pegmatite for tin
(King & Foord 2000). Masses of cassiterite in the
FIG. 11. Color-zoned tourmaline adjacent to a pocket,
showing gradation of color from black to green to pink.
Field of view 1 m.
FIG. 12. Gem-quality green elbaite from a Mt. Mica pocket.
BULK COMPOSITION OF MT. MICA PEGMATITE 1059
decimeter range have been noted occasionally, but
more commonly they are only in the cm range.
Beryl
Beryl is less common in the Mt. Mica pegmatite,
but several decimeter-sized morganites have been
found in large miaroles. Beryl found in the massive
pegmatite is bluish-green to yellowish, whereas in
miaroles the beryl is white, colorless, or pinkish.
Compared to other pegmatites in the Oxford field,
notably the Orchard, Emmons, and Bennett pegma-
tites, the Mt. Mica pegmatite appears to have a paucity
of common beryl. It seems to be more abundant in the
miarolitic cavities (Fig. 9), suggesting that Be was
retained until the pocket stage.
Spodumene
Spodumene is restricted in its occurrence to the
inner zones and has been found with lepidolite and
pollucite. The maximum size reaches about 0.5 m.
Most spodumene has undergone alteration to phyllo-
silicates.
Pollucite
Pollucite has been found in m-sized masses in the
inner zones of the pegmatite. Cleavelandite, micro-
cline, montebrasite, lepidolite, and spodumene are
commonly associated with pollucite at Mt. Mica.
Recently, gem-quality pollucite has also been found.
The gem-quality pollucite occurs in miarolitic cavities
and is associated with crystals of cleavelandite, quartz,
lepidolite, and fluorapatite, and gem-quality elbaite
and beryl. Gem-quality pollucite is rare, but one
pocket found in 2014 contained a large number of 2–3
cm-sized strongly etched crystals.
Mn and Fe carbonate pods
The presence of vuggy masses of Mn and Fe
carbonates associated with abundant fibrous fluorapa-
tite (Figs. 13–14) in the core zone indicates an
important role of CO
2
in the late stages of the
consolidation of the pegmatite. Compositionally, these
carbonates are just barely Mn-dominant, whereas
carbonates from the more evolved Bennett and
Emmons pegmatites are essentially rhodochrosite
(Johnson et al. 2013). These carbonate masses are
very rich in mineral species: blue tourmaline, fluo-
rapatite, muscovite, quartz, l¨
ollingite, pyrite, uraninite,
childrenite–eosphorite, moraesite, hydroxylherderite,
cassiterite, pollucite, and hureaulite (Johnson et al.
2013). This represents a unique, late-stage micro-
chemical environment with a distinct carbonate-rich
chemical signature.
Apatite
Minerals of the apatite group, essentially only
represented by fluorapatite, are not very abundant in
the Mt. Mica pegmatite, except in miaroles. However,
in the interior of one very large miarolitic cavity,
fluorapatite was abundant in crystals up to 2 cm in
maximum dimension. These fluorapatites are mainly
colorless to light greenish.
Other phosphate minerals
Aside from apatite, the Mt. Mica pegmatite has
abundant montebrasite in masses to several decime-
ters. Triphylite has only been found very rarely in
centimeter-sized masses. In the carbonate pods,
accessory hureaulite, childrenite, and eosphorite have
been found. Moraesite has been noted from phosphate
pods as well as in small miaroles.
FIG. 13. Photograph of a carbonate pod. Note the skeletal
habit of the tan-colored carbonate and the intimate
association with blue tourmaline. Field of view: 25 cm. FIG. 14. Fluorapatite in a small miarole in a carbonate mass.
Field of view: 2 mm, crossed nicols.
1060 THE CANADIAN MINERALOGIST
Zircon and other zirconium minerals
Zircon is a common accessory mineral, with
crystals attaining 1 cm in maximum dimension. They
are commonly zoned with chaotic interiors and more
Hf-rich rims. Inclusions of tiny uraninite grains are
very common. Mt. Mica is the type location for two
unusual zirconium phosphate minerals: kosnarite and
mccrillisite (Brownfield et al. 1993, Foord et al. 1994).
Both of these species have been found associated with
the carbonate masses as well as with the late-stage rose
quartz crystals. Pristine zircon crystals without any
evidence of dissolution or replacement occur in close
proximity to kosnarite, thus it appears that the Zr in the
kosnarite was not derived from dissolution of zircon
and may have been present in the melt in some other
form that crystallized kosnarite and mccrillisite. There
is no evidence of increased alkalinity that could have
destabilized zircon. The close association with car-
bonate masses may hint at a carbonate chelate
associated with high phosphate activity that may have
played a role in keeping Zr available for crystallizing
these phosphate minerals.
HYDROTHERMAL ALTERATION
Late-stage hydrothermal activity and mineral
alteration is minimal and is typically most evident in
the core zone and particularly in and near pockets.
Albitization of K-feldspar in and around the pockets is
widespread, and the abundance of cleavelandite is
likely related to fluid activity. Etched beryl, replaced
spodumene, veined pollucite, selectively corroded
tourmaline, and Fe-stained K-feldspar appear to be
related to hydrothermal activity. Interestingly, the
primary phosphate minerals triphylite and montebra-
site in Mt. Mica show only incipient alteration.
Almost all of the spodumene has undergone
essentially complete replacement by a fine-grained
mixture of albite and muscovite (cymatolite). No
eucryptite has been seen among the alteration products
of the spodumene. The pollucite is typically veined
with millimeter-thick veins of lepidolite. Near the
veins, the pollucite is much less Cs-rich and instead
Cs-rich analcime is adjacent to the lepidolite veins.
The morganite variety of beryl from pockets may show
incipient etching and corrosion as a result of
hydrothermal action. Moraesite and hydroxylherderite
are the common secondary Be-minerals that formed
when the phosphorus activity was elevated. The
feldspars are generally not intensely altered except in
some miarolitic cavities where slight alteration is
observed.
Tourmaline in the core zone and in the miarolitic
cavities also shows the effects of hydrothermal
alteration. Color-zoned elbaite with pink cores and
green rims has undergone alteration or dissolution of
the pink tourmaline cores where it was exposed to late-
stage corrosive fluids. Typically, the green rim is
completely unaffected, whereas the pink core may
have been almost entirely removed or replaced by
muscovite or lepidolite. In some of the miarolitic
cavities, hydrothermal activity is also likely responsi-
ble for the mobilization of boron from dissolving
elbaitic tourmaline. The boron reacts with the
tourmaline and reprecipitates a more Fe-rich, but X-
site vacant-dominant foitite, typically as a termination
on both elbaite and rossmanite. Alteration of schorl in
the wall zone is not evident, suggesting that the water
activity was too low to affect the wall-zone schorl, as
fluid activity wasn’t sufficient for alteration until the
latest stages of crystallization.
Hydrothermal effects on the primary phosphate
minerals, such as triphylite and montebrasite, are very
minor. Triphylite may show a thin rim of ferrisickler-
ite, but no further alteration is evident. In the Mt.
Marie and Emmons pegmatites alteration is more
extensive, yielding heterosite–purpurite and other
secondary phosphate minerals. Montebrasite exhibits
only a thin rim of dark alteration product of Mn-rich
phosphates. There is an interesting relationship
between secondary zirconium phosphates and zircon.
In small miarolitic cavities, typically associated with
siderite–rhodochrosite and blue elbaite–schorl series
tourmaline, the Zr-phosphates kosnarite and mccrilli-
site are associated with nearby zircon. This is likely a
result of elevated phosphorus activity and the remain-
ing Zr in the melt. Zircon itself does not appear to have
undergone any alteration or corrosion.
In the miarolitic cavities, the last hydrothermal
minerals to form appear to be a mixture consisting of
cookeite and clay minerals such as kaolinite or illite.
There is evidence of montmorillonite, but this may be
due to surficial weathering. The kaolinite found within
the miarolitic cavities appears to mainly be the result
of late-stage alteration of albite in the cavities. Thus,
the kaolinite that encases tourmaline crystals appears
to be the result of alteration of original albite crystals
intergrown with tourmaline, as seen in unaltered
pockets.
GARNET-BIOTITE THERMOMETRY OF HOST ROCK
Garnet-biotite thermometry was conducted on
country rock samples containing the mineral pairs.
Calculations and the non-ideal mixing parameters of
Bhattacharya et al. (1992) were used. Garnet-biotite
thermometry from sample pairs at the pegmatite-
country rock contact yielded a temperature range of
600 to 660 8C, with an average of 630 8C (Clark et al.
BULK COMPOSITION OF MT. MICA PEGMATITE 1061
2013). Considering the textural evidence of the
gradational contact between the leucosomes and the
pegmatite, we infer that the pegmatitic melt and the
leucosomes were at the same temperature. This
temperature is consistent with the P-Tconditions
inferred for rocks of the Sebago Migmatite Domain
from an assemblage of sillimanite, quartz, muscovite,
biotite, and alkali feldspar (650 8C and 3 kbar) by
Guidry et al. (2013).
BULK CHEMICAL COMPOSITION
The very large grain size and internal zonation of
Mt. Mica makes it virtually impossible to determine its
bulk chemical composition via traditional XRF or ICP
analysis of whole-rock samples. A new method to
estimate the bulk composition was developed that
combines accurate mapping of the pegmatite as it was
mined with chemical analyses of a series of selected
core samples across the pegmatite. Forty-five drill
cores were taken across the pegmatite from the
hanging-wall contact to the foot-wall contact along a
transect that intentionally avoided lepidolite masses
and miaroles. The cores were sliced longitudinally and
half of each was pulverized. Equal aliquots of 20 g of
the resulting powders were thoroughly mixed and
homogenized to produce a sample that was represen-
tative of the bulk composition of the pegmatite sans
lepidolite masses. The sample was then analyzed by
fusion ICP-OES for the major elements and ICP-MS
for the trace elements, and by DCP spectroscopy for
boron and lithium. Structural water was determined by
loss on ignition above 500 8C to be 0.94 wt.%. The
results are presented in Table 1.
A detailed geologic map (Fig. 15), produced by
mine owner Gary Freeman during the mining process,
shows the spatial distribution and aerial extent of the
pockets, massive lepidolite, microcline, and xenoliths
superimposed on a shaded layer that represents the
area of the mined pegmatite. The 2011 map was
vectorized in ArcGIS, imported into ImageJ software,
and set to the appropriate scale (Boudreaux et al.
2013). The image was color-thresholded to highlight
masses of lepidolite. The program provided area
calculations of the lepidolite masses, yielding 5.07%.
The lepidolite mass and pegmatite areas were
multiplied by the measured thicknesses of the
individual masses to convert to volume percent. This
yielded a volume of 1.22% lepidolite pods in the
pegmatite, within a total mined volume of 7516 m
3
.
The density calculated from the whole-rock chemical
composition is 2.67 g/cm
3
. Thus, the weight of the
whole mass of mined pegmatite equals 2.007 310
10
g.
The volume of lepidolite calculated from the map and
thicknesses of each mass equals 91.87 m
3
. The density
of the lepidolite masses (71% lepidolite and 29% albite
and quartz) is 2.77 g/cm
3
. This yields 2.545 310
8
g
lepidolite mass, which equals 1.27 wt.% of the bulk
pegmatite composition. The sparsely mapped micro-
cline areas were judged to be too small to accurately
quantify their volume.
The proper proportion of milled bulk lepidolite
needed to produce a new powder containing 1.27 wt.%
lepidolite was then added to a second aliquot of
powder from the original 45 whole-rock cores. The
adjusted mixture was rehomogenized and reanalyzed
by fusion ICP and DCP. In addition, fluorine was
quantified by fusing the whole-rock powders with the
added lepidolite into glass beads and analyzing them
via electron microprobe for major and common minor
elements. Fluorine values were averaged and incorpo-
rated into the new data set. Lithium and boron were
analyzed by DCP and were also added to the data set.
These procedures provided a new whole-rock bulk
chemical composition (45B) that accounts for the
measured pods of lepidolite material (Table 2).
Estimating pocket volume and total water content
The area of each pocket measured by the ImageJ
software was converted to volume percent by calcu-
lating a best-fit ellipsoid, given two perpendicular,
horizontal axes from the map and an estimated vertical
dimension from the cross-section and from measure-
ments of individual pockets. The calculated volumes
were summed and normalized to total pegmatite
volume, giving an estimated volume of 1.2%. The
TABLE 1. BULK GEOCHEMISTRY OF THE MT. MICA
PEGMATITE BASED ON FUSION ICP ANALYSES OF
45 WHOLE-ROCK CORES, EXCLUDING LEPIDOLITE
MASSES: (MM-45A)
MM-45A
oxide wt.%element ppm
SiO
2
71.82 Sr 44
TiO
2
0.07 Ba 34
Al
2
O
3
16.74 Nb 18.8
FeO 1.11 Ta 3.61
MnO 0.02 Zr 24
MgO 0.15 Hf 1.5
CaO 0.49 Sn 59
Na
2
O 5.45 Cs 47
K
2
O 1.86 Rb 326
P
2
O
5
0.21 B* 240
Li
2
O* 0.08
LOI 0.94
Total 98.94 wt.%
* Analyzed via Direct-Coupled Plasma Spectrometry
1062 THE CANADIAN MINERALOGIST
value was adjusted for the amount of fill in the
pockets, which was estimated visually and measured in
two pockets to be about 40%. The filling in the pockets
consists of fractured quartz and tourmaline with small
amounts of muscovite, lepidolite, and rare apatite. The
contents of the two pockets had a measured porosity of
39.7%. This porosity value was used to refine the total
volume of open space in the pockets within the
mapped portion of Mt. Mica to approximately 0.90%.
The water content of the pockets was estimated by
assuming the calculated open space filled with a water-
rich fluid. The exsolved fluid filling the open space is
certainly not pure water, but a solute-rich, hydro-
silicate fluid. However, in order to estimate the
maximum water content contributed by the pockets
to the bulk composition of the pegmatite, it was
assumed that the open space was filled with H
2
O.
Based on the temperature and pressure of the country
rocks determined from phase equilibria by Guidry et
al. (2013) and the garnet-biotite thermometry of Clark
et al. (2013), the density of the water was calculated at
625 8C and 3 kbar to be 0.6709 g/cm
3
.
Weight percent water was estimated as follows: the
calculated open space from the map of 3.129 310
3
ft
3
is 60% open and 40% filled with material having a
porosity of 39.75%, which gives a total pocket volume
of 2.375 310
3
ft
3
or 6.725 310
7
cm
3
, which
TABLE 2. WHOLE-ROCK GEOCHEMISTRY OF THE
MT. MICA PEGMATITE BASED ON ICP ANALYSES OF
45 WHOLE-ROCK CORES, INCLUDING LEPIDOLITE,
CALCULATED POCKET WATER, STRUCTURALLY
BOUND WATER, AND FLUORINE: MM-45B
MM-45B
Oxide wt.%element ppm
SiO
2
72.08 Sr 44
TiO
2
0.07 Ba 35
Al
2
O
3
17.33 Nb 23.5
FeOt 1.21 Ta 6.3
MnO 0.04 Zr 17
MgO 0.15 Hf 1
CaO 0.48 Sn 84
Na
2
O 5.35 Cs 99.9
K
2
O 2.08 Rb 636
P
2
O
5
0.20 B* 287
Li
2
O* 0.11
H
2
O 0.22
LOI 0.94
F
a
0.25
Subtotal 100.51
–O”F 0.11
Total 100.40
Note: (Analyzed via:
a
EMPA, *Direct-Coupled Plasma
Spectrometry – new value, 2016 reanalysis)
FIG. 15. Mining map and cross section of the Mt. Mica pegmatite showing the areal extent of the mined pegmatite, xenoliths,
lepidolite pods, pockets, and microcline (mapped by Gary Freeman).
BULK COMPOSITION OF MT. MICA PEGMATITE 1063
multiplied by 0.6709 g/cm
3
(the density of water at 3
kbar and 625 8C, IRC fluid properties calculator: irc.
wisc.org/properties [date used: Feb. 2016]) gives 4.51
310
7
grams water in the pockets, or 0.22 wt.% water.
This water content was added to the 0.94 wt.% water
determined by LOI (above 500 8C) to estimate a total
H
2
O content of 1.16 wt.% for the bulk pegmatite melt.
The resulting whole rock bulk composition of the Mt.
Mica pegmatite is shown in Table 2.
Estimating the Cs content added by massive pollucite
As massive pollucite is difficult to recognize during
mining, the contribution of Cs from pollucite has not
been quantified, since no record of pollucite content
was documented. We can estimate that between one-
half to one ton of pollucite was encountered during
mining of the mapped area, which will elevate the Cs
content of the bulk pegmatite somewhat. To make a
crude approximation of the overall impact of pollucite
on the Cs content, for the purpose of calculation we
estimated the effect of one ton of pollucite as follows:
one ton of pollucite with 35 wt.% Cs equals 9 310
5
g
Cs divided by 2 310
10
g pegmatite mined or 0.0045
wt.%, or about 15 ppm. This gives us at least an order
of magnitude estimate of the range of additional Cs
content (8 to 15 ppm) for the bulk pegmatite.
Estimating the composition of pocket mineral contents
The bulk composition of the pocket contents is
estimated as follows: pockets constitute 2.3749 10
3
ft
3
or 0.89 vol.% and are about 40% filled with crystals and
fractured mineral fragments. The fractured material that
accumulates at the bottom of the pockets has an overall
porosity of about 40%. The major mineral content of
the pockets consists of about 40% quartz, 35% albite,
15% elbaite, 5% Li-rich muscovite, 5% K-feldspar, and
minor to trace amounts of beryl, pollucite, apatite,
montebrasite, spodumene, and cassiterite. The com-
bined weight % contribution of all elements except Si
(0.11) and Al (0.02) is less than 0.01 wt.%, which is a
virtually negligible addition to the overall major and
minor element bulk chemical composition of the whole
pegmatite as determined above (Table 2).
Calculating ‘‘normative pegmatite mineralogy’’
To evaluate how well this whole rock composition
matches the modal mineral composition of Mt. Mica, a
modified ‘‘normative pegmatite mineralogy’’ was
iteratively calculated from the whole rock composition
(Table 2) using a custom-made spreadsheet to
calculate mineral proportions, using the analyzed Mt.
Mica (non-standard normative) minerals tourmaline,
lepidolite, muscovite, and garnet (Table 3).
TABLE 3. REPRESENTATIVE MINERAL COMPOSITIONS USED IN THE
CALCULATION OF THE ‘‘NORMATIVE’’ AMOUNTS OF TOURMALINE, LEPIDOLITE,
MUSCOVITE, AND GARNET
Mt. Mica analyzed minerals used to calculate a modified
‘‘pegmatite normative composition’’
Tourmaline Lepidolite Muscovite Garnet
Composition Composition Composition composition
SiO
2
36.41 50.57 46.09 36.62
TiO
2
0.21 0.02 0.01 0.02
B
2
O
3
10.46 0.00 0.00 0.00
Al
2
O
3
33.86 26.68 36.98 20.82
FeO 12.96 0.41 0.77 26.93
MnO 0.48 0.24 0.09 15.52
MgO 0.53 0.10 0.19 0.18
CaO 0.03 0.01 0.06 0.31
Li
2
O 0.54 5.34 0.29 0.00
Na
2
O 1.32 0.39 0.46 0.06
K
2
O 0.02 9.88 9.78 0.00
H
2
O 3.16 1.11 4.15 0.11
Rb
2
O 0.00 0.97 0.11 0.00
Cs
2
O 0.00 0.18 0.00 0.00
F 0.95 7.23 0.79 0.00
Subtotal 100.93 103.14 99.74 100.57
O”F 0.40 3.05 0.33 0.00
Total 100.53 100.09 99.41 100.57
1064 THE CANADIAN MINERALOGIST
In this process, all of the boron was used to make
tourmaline (composition 99% schorl, 1% elbaite), 20%
of the remaining fluorine was used to make lepidolite,
94% of the K
2
O remaining after these two phases was
used to make muscovite, and 1% of the FeO remaining
after these three phases was used to make garnet
(which is limited by the amount of remaining
manganese) (Tables 2–4). The final weight percentag-
es of each oxide remaining after tourmaline, lepidolite,
muscovite, and garnet were removed were then
inserted into a standard CIPW normative calculation
(Hollocher 2004). The resulting bulk mineralogy is
given in Table 5. The results match closely with the
observed modal mineral proportions of the wall zone,
confirming the classification of Mt. Mica as an albite-
dominant (An 1.8) granitic pegmatite with only 0.71
wt.% K-feldspar. Muscovite is the dominant K-bearing
mineral phase in the pegmatite, instead of K-feldspar.
The final norm contains 1.68 wt.% hypersthene and
0.571 wt.% corundum, but this is most likely the result
TABLE 4. CHANGES IN BULK CHEMICAL COMPOSITION OF ORIGINAL ANALYZED
MM-45B AFTER THE SUBTRACTION OF TOURMALINE, LEPIDOLITE, MUSCOVITE,
AND GARNET
Original
MM-45B
After
Tourmaline
After
Lepidolite
After
Muscovite
After
Garnet
SiO
2
72.08 71.76 71.42 62.50 62.43
TiO
2
0.07 0.07 0.06 0.06 0.06
Al
2
O
3
17.33 17.03 16.85 9.69 9.66
Fe
2
O
3
0.00 0.00 0.00 0.00 0.00
FeO 1.21 1.10 1.09 0.94 0.90
MnO 0.04 0.03 0.03 0.02 –0.01
MgO 0.15 0.15 0.14 0.11 0.11
CaO 0.48 0.48 0.48 0.47 0.47
Na
2
O 5.35 5.34 5.34 5.25 5.25
K
2
O 2.08 2.08 2.01 0.12 0.12
P
2
O
5
0.20 0.20 0.20 0.20 0.20
Cs
2
O 0.01 0.01 0.01 0.01 0.01
Rb
2
O 0.07 0.07 0.06 0.04 0.04
B
2
O
3
0.09 0.00 0.00 0.00 0.00
Li
2
O 0.11 0.11 0.07 0.01 0.01
H
2
O 1.16 1.13 1.12 0.32 0.32
F 0.25 0.24 0.20 0.04 0.04
Subtotal 100.68 99.79 99.09 79.79 79.61
O”F 0.11 0.10 0.08 0.02 0.02
Total 100.58 99.69 99.01 79.77 79.59
Weight %Removed:
Tourmaline 0.89
Lepidolite 0.68
Muscovite 19.24
Garnet 0.18
Note: The weight percentage of normative tourmaline, lepidolite, muscovite, and garnet
removed at each step.
TABLE 5. ‘‘NORMATIVE’’ MINERAL COMPOSITION
CALCULATED FROM THE BULK CHEMICAL
COMPOSITION GIVEN IN TABLE 2
‘‘normative’’ mineral wt.%
plagioclase (An
1.8
) 45.31
quartz 30.25
muscovite* 19.24
hypersthene 1.68
lepidolite* 0.68
tourmaline* 0.89
orthoclase 0.71
corundum 0.57
apatite 0.48
garnet* 0.18
ilmenite 0.11
fluorite 0.05
magnetite 0.07
TOTAL 100.22
BULK COMPOSITION OF MT. MICA PEGMATITE 1065
of underestimating the amount of garnet, since 3
hypersthene þ1 corundum ¼1 garnet [3(Fe,Mg)SiO
3
þ
Al
2
O
3
(Fe,Mg)
3
Al
2
(SiO
4
)
3
].
BULK PEGMATITE TRACE ELEMENT GEOCHEMISTRY
Whole rock chemical analyses of the Mt. Mica
composite pegmatite, granite, and leucosomes are
shown in Table 6. Trace-element contents of the
pegmatite and leucosomes are notably distinct from
the granite samples from the Sebago granite within the
SMD. Plots of the chemistry of the bulk pegmatite 45-
core composite with leucosomes from the migmatitic
country rock are strikingly similar in REE content and
almost an order of magnitude lower than that of the
granitic samples plotted against chondrite and upper
crust (Figs. 16 and 17). The spider diagram of rock
versus average crust also shows very similar patterns
for the leucosomes and the pegmatite (Fig. 18).
Overall, the granitic samples are close to or slightly
higher than average crustal values. Except for Rb, U,
Ta, and Pb, the pegmatite and leucosome samples are
nearly an order of magnitude lower than the granitic
rock. The enrichment in alkalis and Ta is consistent
with the LCT nature of the pegmatite, however the
spike in U in both the pegmatite and leucosome
samples is notable, and the Pb spike is inferred to be
related to radiogenic Pb.
DISCUSSION AND CONCLUSIONS
Careful mapping by Gary Freeman of the location
and sizes of the lepidolite masses, pockets, and the
volume of pegmatite mined underground has provided
an unprecedented data base which is the foundation for
this study. Over the last 13 years of mining, several
hundred pockets have been carefully documented.
Their sizes and contents have been carefully measured
and recorded. The pockets have yielded large gem-
quality crystals of green and pink tourmaline rivaling
the best material ever produced from Mt. Mica in its
195-year history. The pockets range in size from a few
cm
3
to one in excess of 500 m
3
. Several dozens of the
intermediate to larger pockets have produced thou-
sands of carats of gem-quality tourmaline and lesser
quantities of morganite. Pocket density averages about
one every 3 m, with larger pockets having greater
spacing and small ones having less, making this one of
the most pocket-rich pegmatites in North America.
FIG. 16. Whole rock chondrite-normalized REE plot for the
Mt. Mica pegmatite (red), SMD leucosomes (blue), and
Sebago pluton granite (green). Chondrite values from
McDonough & Sun (1995).
FIG. 17. Whole rock spider diagram sample of Mt. Mica
pegmatite (red), SMD leucosomes (blue), and Sebago
pluton granite (green), relative to upper continental crust.
Normalization values from Taylor & McLennan (1985).
FIG. 18. Whole rock spider diagram of Mt. Mica pegmatite
(red), SMD leucosomes (blue), and Sebago pluton granite
(green) relative to total crust. Normalization values from
Rudnick & Fountain (1995).
1066 THE CANADIAN MINERALOGIST
TABLE 6. WHOLE ROCK FUSION ICP AND MS ICP ANALYSES OF THE MT. MICA PEGMATITE, SMD
LEUCOSOMES, AND SEBAGO
Pegmatite Granitic rocks Leucosomes
MM-45-A MM-45-B 1A.1 S35-M4 29 6 B 1 3B.1 P-6
SiO
2
71.82 72.08 75.00 63.97 67.81 68.63 73.67 71.95 72.9
Al
2
O
3
16.74 17.33 11.64 16.11 13.57 14.16 13.65 16.24 13.64
Fe
2
O
3
(T) 1.21 1.18 4.57 8.42 5.34 4.76 1.57 0.83 0.45
MnO 0.032 0.039 0.238 0.123 0.094 0.075 0.046 0.03 0.014
MgO 0.15 0.15 1.46 2.7 2.36 2.28 0.23 0.09 0.3
CaO 0.49 0.48 2.42 0.51 2.21 2.22 0.31 0.94 3.99
Na
2
O 5.45 5.35 2.16 1.02 2.91 3.19 3.58 4.95 1.25
K
2
O 1.86 2.08 1.37 3.49 1.8 2.37 4.28 1.88 3.94
TiO
2
0.067 0.067 0.677 1.269 0.646 0.667 0.087 0.035 0.046
P
2
O
5
0.21 0.20 0.14 0.08 0.13 0.10 0.15 0.25 0.32
LOI 1.59 1.54 0.89 2.95 2.09 1.77 0.99 1.20 1.52
Total 99.61 100.50 100.60 100.6 98.97 100.2 98.57 98.41 98.37
ppm
Sc 2 2 10 15 14 12 2 ,11
Be 25 24 5 2 2 2 7 12 12
V 14 13 83 179 104 84 ,5811
Zn 30 30 50 110 110 80 130 ,30 30
Ga 33 34 20 28 15 18 26 19 21
Ge 5.6 6.1 20 2.2 2.1 2.6 4.5 4.3 1.8
As 667 594 50 ,5,5,515396
Rb 326 636 60 191 69 234 522 182 193
Sr 44 44 15 83 274 188 29 52 547
Y 2.2 2.2 2.8 30.4 22.2 24.8 2.5 4.1 2.5
Zr 24 17 16 291 195 182 13 15 26
Nb 18.8 23.5 264 23.6 6.8 14.6 6.8 7.2 0.7
Sn 59 81 329 2 1 4 12 17 1
Sb 0.5 0.5 25.4 ,0.2 ,0.2 ,0.2 ,0.2 ,0.2 ,0.2
Cs 47 99.9 177 5.4 1.7 11 79.4 18.4 12
Ba 34 35 14.2 373 397 434 58 25 242
La 3.04 2.7 ,2 55.7 24.8 36.1 1.74 3.51 2.93
Ce 5.6 4.75 0.7 110 50.2 75.9 3.24 6.35 4.74
Pr 0.62 0.52 ,0.1 12.8 5.81 8.54 0.32 0.7 0.59
Nd 2.41 1.78 12 49.1 22.1 30.7 1.07 2.68 2.08
Sm 0.63 0.52 ,0.2 9.57 4.41 6.48 0.35 0.83 0.41
Eu 0.174 0.178 41.6 1.18 1.09 1.44 0.164 0.286 0.253
Gd 0.51 0.39 167 7.19 4 5.17 0.4 0.85 0.31
Tb 0.08 0.08 31 1.13 0.66 0.86 0.09 0.16 0.06
Dy 0.43 0.44 61.5 6.36 4.11 4.71 0.51 0.95 0.38
Ho 0.07 0.06 7.22 1.26 0.82 0.9 0.08 0.14 0.08
Er 0.19 0.18 27.1 3.85 2.45 2.58 0.2 0.35 0.23
Tm 0.033 0.031 5.99 0.575 0.365 0.372 0.031 0.046 0.036
Yb 0.21 0.2 1.03 3.82 2.31 2.4 0.19 0.27 0.23
Lu 0.027 0.026 4.8 0.645 0.369 0.376 0.03 0.041 0.035
Hf 1.5 1 0.77 7.8 5.1 4.9 0.8 0.5 1.4
Ta 3.61 6.3 4.79 1.67 0.52 1.19 0.94 2.17 0.16
W 4.3 5.3 0.94 0.7 1.2 0.9 0.9 2.3 0.6
Tl 1.61 3.1 2.91 1.2 0.53 1.92 3.31 0.81 1.07
Pb 15 14 0.445 15 50 15 30 26 21
Bi 5.3 4.6 2.83 ,0.1 ,0.1 0.2 0.6 0.6 0.4
Th 1.21 0.94 0.457 18.8 7.42 14.9 0.93 1.14 1.09
U 11.1 12.4 4.4 7.33 1.82 3.87 17.4 7.27 3.63
BULK COMPOSITION OF MT. MICA PEGMATITE 1067
The ductile contact relationships of the country
rock and the pegmatite show they were both at
elevated temperature. The gradational nature of the
pegmatite contact with the leucosomes of the migma-
tite suggest that they were in thermal equilibrium. The
assemblage found in the nearby SMD of muscovite,
quartz, sillimanite, K-feldspar, and biotite, together
with the water-saturated granite melting curve, suggest
the conditions were near 650 8C and 3 kb. The garnet-
biotite thermometry of 630 8C from the adjacent
biotite-rich country rock corroborates this estimate.
We therefore conclude that the pegmatite intruded at
about 630 to 650 8C at a pressure of about 3 kb. As for
the few areas along the contact that show weak
development of comb structure muscovite oriented
more or less perpendicular to the contact, we infer that
this represents the result of an initial period of rapid
crystallization related to undercooling. The undercool-
ing is inferred to be the result of a lag in initiation of
crystallization and the result of compositional quench-
ing caused by local loss of H
2
O and F by the
crystallization of muscovite. Both processes combined
to produce undercooling and rapid crystallization once
crystallization commenced. We emphasize that the
undercooling and rapid crystallization implied by the
comb structure muscovite at the contact is not related
to a temperature difference between the pegmatite and
the country rock.
The model for pocket formation employed in the
calculations of pocket volume and total water content
of the pegmatite is that described by Simmons et al.
(2012), which involves exsolution of a second fluid
phase after most of the pegmatitic melt has crystal-
lized. The residual melt accumulates toward the center
of the dike and becomes progressively enriched in
water and other fluxing materials until ultimately a
second aqueous silicate-rich fluid exsolves, forming
more or less spherical segregations or bubbles of a
flux-rich aqueous silicate-rich fluid that constitutes a
protopocket. According to a widely accepted theory of
pocket formation (Jahns & Burnham 1969, Simmons
et al. 2012), once the supercritical aqueous fluid starts
to exsolve, diffusion of ions from the coexisting
silicate melt into the silicate-rich fluid supplies
nutrients to the crystals growing in the protopocket.
Continued rapid diffusion of ions from the silicate melt
to the growing crystal surfaces in the fluid of the
protopocket is proposed to explain the greater volume
of crystals found in the pockets than what could have
grown from the less dense, hydrous, pocket-forming
fluid alone. Thus, the flux-rich aqueous fluid is the
medium through which ions diffuse to the growing
crystal surfaces. We note that this model of pocket
formation contrasts with that of London (2013), but we
feel this model best supports the observations of
pockets found at Mt. Mica.
We believe that the bulk chemical calculations
provide a reasonably accurate estimate of the whole
rock composition of the Mt. Mica pegmatite and
illustrate that the pegmatite was not close to water
saturation at the time of emplacement and that, overall,
the pegmatite is only moderately evolved, and that
only areas within the core zone and the pockets are
highly evolved, illustrating the effectiveness of
fractionation during the crystallization of the pegma-
tite. Also, although Mt. Mica is generally considered
to be a relatively pocket-rich pegmatite, the actual
volume of open space is quite small, less than 1.0%,
and the calculated maximum amount of water
contained in the melt that formed the pegmatite is
about 1.16 wt.%, with less than 0.22 wt.% attributable
to the pockets.
Finally, in contrast to suggestions by previous
authors (Wise & Brown 2010, Wise & Francis 1992)
that this pegmatite was derived by fractional crystal-
lization from the Sebago Granite pluton (or any other
pluton), we believe the chemical evidence presented
here strongly suggests that the Mt. Mica pegmatitic
melt could be derived directly from partial melting of
the metapelitic rocks of the SMD. Although the
textures appear to suggest that Mt. Mica formed at
the same time as the SMD leucosomes, recent zircon
geochronology (Bradley et al. 2016) and our prelim-
inary monazite and zircon geochronology of proximal
SMD leucosomes show that Mt. Mica is younger
(Alleghanian ~260 Ma) than the leucosomes (Neo-
acadian ~340 Ma). Thus, we suggest that Mt. Mica
may be related to a later anatectic melt formed during
a thermal event related to the transition from the
Alleghanian orogeny to the onset of Triassic conti-
nental breakup. We suggest that the Mt. Mica
pegmatitic melt did not form in situ, but that batches
of anatectic melt accumulated and coalesced into a
larger volume that subsequently formed the Mt. Mica
pegmatite. This is the first bulk chemical evidence
supporting the formation of an LCT-type pegmatite by
direct anatexis.
ACKNOWLEDGMENTS
We are indebted to the Maine Mineral and Gem
Museum and the Department of Earth and Environ-
mental Sciences at the University of New Orleans for
their support of the analytical facilities. We gratefully
acknowledge valuable discussions with Frank Perham
of West Paris, Maine and the constructive reviews by
Pietro Vignola and Jan Cemp´
ırek which helped
improve the manuscript.
1068 THE CANADIAN MINERALOGIST
REFERENCES
BHATTACHARYA, A., MOHANTY, L., MAJI, A., SEN, S., & RAITH,
M. (1992) Non-ideal mixing in the phlogopite-annite
binary: constraints from experimental data on Mg–Fe
portioning and a reformulation of the biotite-garnet
geothermometer. Contributions to Mineralogy and Pe-
trology 111, 87–93.
BOUDREAUX,A.,GRASSI,L.,SIMMONS, W., FALSTER, A.,
WEBBER, K., & FREEMAN, G. (2013) Cartography to
chemistry: Estimating the bulk composition of the Mt.
Mica Pegmatite via map analysis, Maine, USA. Abstracts,
PEG 2013: The 6
th
International Symposium on Granitic
Pegmatites, 24–25.
BRADLEY, D.C., SHEA, E., BUCHWALDT, R., BOWRING, S.,
MCCAULEY, A., BENOWITZ, J., & O’SULLIVAN, P.B. (2016)
Geochronology and orogenic context of lithium-cesium-
tantalum pegmatites in the Appalachians. Abstract,
Geological Society of America, Northeastern Section
Meeting, 51-5.
BROWNFIELD, M.E., FOORD, M.E., SUTLEY, S.J., & BOTINELLY,
T. (1993) Kosnarite, KZr
2
(PO
4
)
3
, a new mineral from
Mount Mica and Black Mountain, Oxford Co., Maine.
American Mineralogist 78, 653–656.
ˇ
CERN ´
Y, P., LONDON, D., & NOVAK, M. (2012) Granitic
Pegmatites as Reflections of their Sources. Elements 8,
289–294.
CLARK, K., SIMMONS, W., WEBBER, K., FALSTER, A., RODA-
ROBLES,E.,&FREEMAN, G. (2013) Contact Zone
Mineralogy and Geochemistry of Mount Mica Pegmatite,
Oxford Co., Maine, USA. Abstracts, PEG 2013: The 6
th
International Symposium on Granitic Pegmatites, 30–31.
DONOVAN, J.J. & TINGLE, T.N. (1996) An improved mean
atomic number background correction for quantitative
microanalysis. Journal of the Microscopy Society of
America 2, 1–7.
FOORD, E.E., BROWNFIELD, M.E., LICHTE, F.E., DAVIS, A.M., &
SUTLEY, S.E. (1994) Mccrillisite, NaCs(Be.Li)Zr
2
(PO
4
)
4
*1–2 H
2
O, a new mineral species from Mount
Mica, Oxford Co., Maine , and new data for gainesite.
Canadian Mineralogist 32, 839–842.
FOORD, E.E., SNEE,L.W.,ALEINIKOFF, J.N., & KING,V.T.(1995)
Thermal histories of granitic pegmatites, western Maine,
USA. Geological Society of America, Abstract, A-468.
GUIDOTTI, C.V. (1989) Metamorphism in Maine: An Over-
view. In Studies in Maine Geology 3(R.D. Tucker & R.G.
Marvinney, eds.). Maine Geological Survey, Augusta,
Maine (1–19).
GUIDOTTI, C.V. (1993) H
2
O Solubility in Haplogranitic Melts;
Compositional, Pressure, And Temperature Dependence.
Geological Society of America, Northeastern Section
Meeting, Abstracts with Programs 25, A21.
GUIDRY, J., FALSTER, A., SIMMONS, W., & WEBBER, K. (2013)
Mineralogy and Geochemistry of Pelitic Country Rock
within the Sebago Migmatite Domain, Oxford Co.,
Maine. Abstracts, PEG 2013: The 6
th
International
Symposium on Granitic Pegmatites, 57–58.
HAMLIN, A.C. (1873) The Tourmaline. James R. Osgood &
Company, Boston, Massachusetts, 107 pp.; republished in
2004 by Rubellite Press, New Orleans, Louisana.
HAMLIN, A.C. (1895) The History of Mount Mica of Maine,
U.S.A. and its Wonderful Deposits of Matchless Tourma-
lines. A.C. Hamlin, 72 pp. plus color plates, republished
in 2004 by Rubellite Press, New Orleans, Louisiana.
HOLLOCHER, K. (2004) Calculation of a CIPW Norm From A
Bulk Chemical Analysis. Excel Spreadsheet Program,
Geology Department, Union College, Petrology Home
Page.
JAHNS, R.H. & BURNHAM, C.W. (1969) Experimental Studies
of Pegmatite Genesis: I. A Model for the Derivation of
and Crystallization of Granitic Pegmatites. Economic
Geology 64, 843–864.
JOHNSON, M., SIMMONS, W.B., FALSTER, A., & FREEMAN,G.
(2013) A Study of Heavy Minerals In A Unique
Carbonate Assemblage From The Mt. Mica Pegmatite,
Oxford County, Maine. Abstracts, PEG 2013: The 6
th
International Symposium on Granitic Pegmatites, 65–67.
KING, V.T. & FOORD, E.E. (2000) Mineralogy of Maine 2:
Mining history, gems, and geology. Maine Geological
Survey, Augusta, Maine, 524 pp.
LONDON, D. (2013) Crystal-Filled Cavities in Granitic
Pegmatites: Bursting the Bubble. Rocks & Minerals 88,
527–534.
MARCHAL, K.L., SIMMONS, W.B., FALSTER, A.U., WEBBER,
K.L., & RODA-ROBLES, E. (2014) Geochemistry, Mineral-
ogy, and Evolution of Li-Al Micas and Feldspars from the
Mount Mica Pegmatite, Maine, USA. Canadian Miner-
alogist 52, 221–233.
MCDONOUGH, W.F. & SUN, S.S. (1995) Composition of the
Earth. Chemical Geology 12, 223–253.
POUCHOU, J.L. & PICHOIR, F. (1991) Quantitative analysis of
homogeneous or stratified microvolumes applying the
model ‘‘PAP’’ in microprobe analysis. In Electron Probe
Quantitation (K.F.J. Heinrich & D.E. Newbury, eds.).
Plenum Press, New York City, New York (31–75).
RODA-ROBLES, E., SIMMONS, W.B., PESQUERA, A., GIL-CRESPO,
P.P., NIZAMOFF, J., & TORRES-RUIZ, J. (2015) Tourmaline
as a petrogenetic monitor of the origin and evolution of
the Berry-Havey pegmatite (Maine, U.S.A.). American
Mineralogist 100, 95–109.
RUDNICK,R.L.&FOUNTAIN, D.M. (1995) Nature and
composition of the continental crust: a lower crustal
perspective. Reviews in Geophysics 33, 267–309.
SIMMONS, W.B., WEBBER, K.L., FALSTER, A.U., & NIZAMOFF,
J.W. (2003) Pegmatology: Pegmatite mineralogy petrol-
BULK COMPOSITION OF MT. MICA PEGMATITE 1069
ogy and petrogenesis. Rubellite Press, New Orleans,
Louisiana, 176 pp.
SIMMONS, W.B., FREEMAN, G., FALSTER, A.U., LAURS, B.M., &
WEBBER, K.L. (2005a) New Tourmaline Production from
Mt. Mica, Maine. Rocks & Minerals 80, 396–408.
SIMMONS, W.B., LAURS, B.M., FALSTER, A.U., KOIVULA, J.I., &
WEBBER, K.L. (2005b) Mt. Mica: A renaissance in
Maine’s gem tourmaline production. Gems & Gemology
41, 2–15.
SIMMONS, W.B., PEZZOTTA, F., SHIGLEY, J.E., & BEURLEN,H.
(2012) Granitic pegmatites as sources of colored gem-
stones. Elements 8, 281–287.
SIMMONS, W.B., FALSTER, A.U., WEBBER, K.L., & RODA-
ROBLES, E. (2013) Mount Mica pegmatite, Maine, USA.
Abstracts, PEG 2013: The 6
th
International Symposium
on Granitic Pegmatites, 135–136.
SOLAR, G.S. & BROWN, M. (2001a) Deformation partitioning
during transpression in response to Early Devonian
oblique convergence, northern Appalachian orogeny,
USA. Journal of Structural Geology 23, 1043–1065.
SOLAR, G.S. & BROWN, M. (2001b) Petrogenesis of migmatites
in Maine, USA: Possible source of peraluminous
leucogranite in plutons? Journal of Petrology 42(4),
789–823.
SOLAR,G.S.&TOMASCAK, P.B. (2009) The Sebago pluton and
the Sebago Migmatite Domain, Southern Maine: results
from new studies. 2009 Annual Meeting of Northeastern
Section, Geological Society of America, Field Trip 2, 1–24.
TAYLOR, S.R. & MCLENNAN, S.M. (1985) The Continental
Crust; Its composition and evolution; an examination of
the geochemical record preserved in sedimentary rocks.
Blackwell Scientific Publications, Oxford, England, 312
pp.
TOMASCAK, P.B., KROGSTAD, E.J., & WALKER, R.J. (1996) U-Pb
monazite geochronology of granitic rocks from Maine:
implications for late Paleozoic tectonics in the Northern
Appalachians. The Journal of Geology 104, 185–195.
WEBBER, K.L., FALSTER, A.U., SIMMONS, W.B., & FOORD, E.E.
(1997) The role of diffusion-controlled oscillatory
nucleation in the formation of line rock in pegmatite-
aplite dikes. Journal of Petrology 38, 1777–1791.
WEBBER, K.L., SIMMONS, W.B., FALSTER, A.U., & FOORD, E.E.
(1999) Cooling rates and crystallization dynamics of
shallow level pegmatite-aplite dikes, San Diego County,
California. American Mineralogist 84, 708–717.
WISE, M.A. & BROWN, C.D. (2010) Mineral chemistry,
petrology and geochemistry of the Sebago granite-
pegmatite system, southern Maine, USA. Journal of
Geosciences 55, 3–26.
WISE, M.A. & FRANCIS, C.A. (1992) Distribution, classifica-
tion and geological setting of granitic pegmatites in
Maine. Northeastern Geology 14, 82–93.
Received February 2, 2016. Revised manuscript accepted
July 5, 2016.
1070 THE CANADIAN MINERALOGIST
