Harry Et Al 2018

Contributions to Mineralogy and Petrology (2018) 173:77
https://doi.org/10.1007/s00410-018-1504-z
ORIGINAL PAPER
Magma evolution beneath Bequia, Lesser Antilles, deduced

from petrology of lavas and plutonic xenoliths
Michal Camejo‑Harry1,2 · Elena Melekhova2 · Jon Blundy2 · William Attridge2 · Richard Robertson1 ·
Thomas Christopher1,3
Received: 28 March 2018 / Accepted: 11 July 2018 / Published online: 5 September 2018
© The Author(s) 2018
Abstract
Extrusive and intrusive igneous rocks represent different parts of a magmatic system and ultimately provide complementary
information about the processes operating beneath volcanoes. To shed light on such processes, we have examined and quanti-
fied the textures and mineral compositions of plutonic and cumulate xenoliths and lavas from Bequia, Lesser Antilles arc.
Both suites contain assemblages of iddingsitized olivine, plagioclase, clinopyroxene and spinel with rare orthopyroxene and
ilmenite. Mineral zoning is widespread, but more protracted in lavas than xenoliths. Plagioclase cores and olivine have high
anorthite (An ≤ 98) and low forsterite (Fo ≤ 84) compositions respectively, implying crystallisation from a hydrous mafic melt
that was already fractionated. Xenolith textures range from adcumulate to orthocumulate with variable mineral crystallisa-
tion sequences. Textural criteria are used to organize the xenoliths into six groups. Amphibole, notably absent from lavas,
is a common feature of xenoliths, together with minor biotite and apatite. Bulk compositions of xenoliths deviate from the
liquid line of descent of lavas supporting a cumulate origin with varying degrees of reactive infiltration by evolved hydrous
melts, preserved as melt inclusions in xenolith crystals. Volatile saturation pressures in melt inclusions indicate cumulate
crystallization over a 162–571 MPa pressure range under conditions of high dissolved water contents (up to 7.8 wt% H 2O),
consistent with a variety of other thermobarometric estimates. Phase assemblages of xenoliths are consistent with published
experimental data on volatile-saturated low-magnesium and high-alumina basalts and basaltic andesite from the Lesser Antil-
les at pressures of 200–1000 MPa, temperatures of 950–1050 °C and dissolved H2O contents of 4–7 wt%. Once extracted
from mid-crustal mushes, residual melts ascend to higher levels and undergo H2O-saturated crystallization in shallow, pre-
eruptive reservoirs to form phenocrysts and glomerocrysts. The absence of amphibole from lavas reflects instability at low
pressures, whereas its abundance in xenoliths testifies to its importance in mid-crustal differentiation processes. A complex,
vertically extensive (6 to at least 21 km depth) magmatic system is inferred beneath Bequia. Xenoliths represent fragments
of the mush incorporated into ascending magmas. The widespread occurrence of evolved melts in the mush, but the absence
of erupted evolved magmas, in contrast to islands in the northern Lesser Antilles, may reflect the relative immaturity of the
Bequia magmatic system.
Keywords Lesser Antilles · Bequia · Grenadines · Xenoliths · Cumulates · Amphibole
Communicated by Othmar Müntener.

Introduction
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s00410-018-1504-z) contains
Magma generation at volcanic arcs is driven by interaction
supplementary material, which is available to authorized users.
between slab-derived materials and the mantle. Conven-
* Michal Camejo‑Harry tional models of subduction zone magmagenesis emphasise
[email protected] percolation of sediment-derived partial melts and hydrous
1 fluids from the slab into the mantle wedge, driving partial
The University of the West Indies Seismic Research Centre,
St. Augustine, Trinidad and Tobago melting (Arculus 1994; Pearce and Peate 1995). An alter-
2 native view of magma formation suggests that a mélange
School of Earth Sciences, University of Bristol, Wills
Memorial Building, Bristol BS8 1RJ, UK of sediments and fluid already exists at the top of the slab,
3 rising as diapirs into the mantle, whereupon they undergo
Montserrat Volcano Observatory, Flemmings, Montserrat
13
Vol.:(0123456789)
77 Page 2 of 26 Contributions to Mineralogy and Petrology (2018) 173:77
partial melting to form arc magmas (Nielsen and Marschall We exploit these characteristics through a detailed petro-
2017). Regardless of the ultimate origin of arc magmas, sub- logical, mineralogical and geochemical study of lavas and
sequent intra-crustal differentiation significantly modifies their xenoliths from the island of Bequia, Grenadines. This
parental melts on their way to the surface. Reconstructing follows similar studies conducted on neighbouring islands,
magmatic processes from the erupted volcanic rock record Grenada (Stamper et al. 2014a) and St. Vincent (Tollan et al.
alone can be problematic since eruptions tend to sample (and 2011), that highlighted the need to establish magma storage
blend) discrete parts of the magmatic system. conditions beneath the intervening Grenadines archipelago.
The more accessible volcanic rock record is heavily
biased towards melt-rich extrusive rocks. While useful for
providing information on pre- and syn-eruptive processes, Geological setting
their melt-dominated character limits what they can tell us
about the differentiation history of parental primary, man- The Lesser Antilles is an intra-oceanic arc resulting from the
tle-derived melts. High-pressure phase assemblages within westward subduction (~ 2 cm a−1) of the North and South
cumulate-textured plutonic xenolith blocks, however, have a American plates beneath the Caribbean plate (Fig. 1a). The
greater preservation potential than phenocrysts in lavas and current arc is approximately 750 km long and comprises
are more likely to provide insight into the true fractionation islands Saba to Grenada that contain 21 potentially active
history of the magmas (Arculus and Wills 1980; Macdonald volcanoes (Lindsay et al. 2005). Various criteria (seismic-
et al. 2000). Studying these complimentary extrusive and ity, magma production rates, rock types) have been used to
intrusive (xenolith) rock archives in conjunction provides a segment the arc, with lava chemistry proving the most dif-
powerful means for reconstruction of the processes operat- ficult to categorize systematically (Macdonald et al. 2000).
ing in arc magmatic systems and the driving forces behind Brown, et al. (1977) found a chemical variation along the arc
eruptions. that required the presence of more than one magma series.
Remarkable features of the Lesser Antilles arc are both However, more recent studies on individual islands have
its diversity of magma compositions within a close spatial identified a number of local magmatic lineages suggesting
context (Brown et al. 1977) and the abundance of erupted that grouping Lesser Antilles rocks based solely on major
plutonic xenoliths (Arculus and Wills 1980; Lewis 1973). element chemistry may mask complexities in magmagenesis
Fig. 1 a Lesser Antilles arc region showing the location of the main tions (after Westercamp et al. 1985) and sampling locations (orange
volcanic centres considered active or potentially active (shown as yel- stars—lava; blue stars—xenoliths)
low triangles). b Geological map of Bequia showing identified forma-
13
Contributions to Mineralogy and Petrology (2018) 173:77 Page 3 of 26 77
(Heath et al. 1998; Howe et al. 2015; Stamper et al. 2014a; stoichiometric methods of Droop (1987) for spinel and clino-
Toothill et al. 2007). Geophysical studies (Arnaiz-Rodriguez pyroxene, and Holland and Blundy (1994) for hornblende.
and Orihuela 2013; Boynton et al. 1979; Christeson et al. MI analytical conditions comprised a 5 µm-defocused spot,
2008; Kopp et al. 2011) reveal a strongly layered crust with 4 nA beam current and 20 kV accelerating voltage, to mini-
a mean thickness of 29 km and pronounced mid-crustal dis- mize the migration of alkalis during analysis (Humphreys
continuities at depths of 2–20 km that vary along strike of et al. 2006). Both techniques were calibrated using oxide,
the arc. The scale of compositional and structural variation mineral and metal standards.
is clearly seen towards the south of the arc from St. Vincent Volatile (H2O and C O2) concentrations in MI were ana-
to Grenada, including the Grenadines archipelago. These lyzed on gold-coated polished thin sections by secondary
islands span a distance of only 140 km, yet significant dif- ion mass spectrometry (SIMS) using the Cameca IMS4f
ferences have been documented in their magma source and instrument at the NERC ion-microprobe facility, University
storage conditions. Experiments on primitive basalts from of Edinburgh. On any one MI, C O2 (as 12C+) was measured
1 +
Grenada (Stamper et al. 2014a) and St. Vincent (Melek- first, followed by H2O (as H ). Instrument conditions com-
hova et al. 2015; Pichavant et al. 2002b) have determined prised a nominal 10 kV 16O− 2–5 nA primary beam focused
the source in the mantle wedge is located at a greater depth to an ~ 10 µm spot on the sample surface. Positive secondary
beneath Grenada than St. Vincent (Stamper et al. 2014a). ions were extracted at 4.5 kV with an energy offset of 50 eV
Little is known about the middlemost Grenadines, prompting for CO2 (75 eV for H2O) and a 25 µm image field. 12C+ data
a need to determine whether along-arc variations in this arc were obtained at high mass resolution to eliminate 24Mg2+
section are discrete or gradational. interference. Volatile species were calibrated against basaltic
Bequia is the northernmost and second largest island of glass standards using data analysis procedures of Blundy and
the Grenadines. Volcanic activity occurred during the Plio- Cashman (2005).
cene with the extrusion of basaltic to andesitic lavas cut by Whole rock major and trace elements were analysed by
dykes (5.0–3.5 Ma Briden et al. 1979). Three formations X-ray fluorescence spectrometry (XRF) using a Bruker AXS
have been identified, corresponding to sequential phases of S4 Pioneer instrument at the University of East Anglia. Sam-
sometimes concurrent volcanism (Fig. 1b): Southern Penin- ples were crushed and powdered using the rock splitter, jaw
sular (≈ 4–5 Ma), Northern Peninsular (≈ 4.5–3.5 Ma) and crusher and planetary ball mill at the University of Bristol.
Mount Pleasant (≈ 5 Ma) formations (Westercamp et al. Pressed pellets and glass beads for analysis were prepared
1985). These are interbedded with heterolithic debris flows at the University of East Anglia.
throughout the island. Isotopic work on Bequia lavas has
distinguished two suites: isotopically homogeneous suite
(IHS) and isotopically diverse suite (IDS), the former inter- Xenolith petrology
preted to have undergone no crustal assimilation and the
latter abundant sediment interaction from the Guyana Shield The studied plutonic xenoliths were obtained as ex situ
(Smith et al. 1996). The two suites are, however, petrograph- deposits from heterolithic debris flows in the northern part
ically very similar and show no systematic geographical or of the island (Fig. 1b) during a 2011 field campaign. The
temporal distribution on the island. Plutonic xenoliths have mineralogy of the xenoliths consists mainly of olivine, pla-
also been studied in some detail (Attridge 2013; Hasanie gioclase, clinopyroxene, calcic amphibole, spinel, ilmenite
2000), with their isotopic variation successfully linked to and rare orthopyroxene with apatite and biotite as acces-
the IHS and IDS lavas (Munday et al. 2000). It has not yet sory phases. Olivine is commonly partially to completely
been shown how, or if at all, the textural and compositional iddingsitized with few instances of unaltered grains in any
features of cumulate xenoliths are related to their extrusive one sample. Other phases are generally alteration-free. Pla-
counterparts and how similar/different the Bequia magmatic gioclase is the only ubiquitous phase in every sample.
system is to St Vincent and Grenada. A total of 50 xenoliths were collected and studied in thin
section. Most samples comprise grains ranging from 2 to
5 mm. Some samples are very coarse grained and/or porphy-
Analytical techniques ritic (6–20 mm grains) while 5% of the sample set are wholly
microcrystalline (< 500 µm). A variety of cumulate textures
Major element concentrations of minerals and melt inclu- are observed from adcumulate, mesocumlate to orthocumu-
sions (MI) were analysed on polished, carbon-coated thin late according to the nomenclature of Wager, et al. (1960).
sections with a Cameca SX100 electron microprobe at the Zoning and resorption textures are abundant.
University of Bristol. Mineral analytical conditions com- The crystallization sequence of each xenolith was
prised a 20 kV accelerating voltage, 10 nA beam current and determined by textural observations of included and
1 µm spot size. Ferric iron contents were estimated using the interstitial phases. The order of phase crystallization is
13
variable throughout the xenolith suite with some samples Group 3 prismatic hornblende
having more than one generation of plagioclase and spinel (ol + plag + cpx + hbl + spl ± ilm)
(Table 1). A notable feature is the late appearance of olivine
in some sequences (Fig. 3a–c). The only consistent feature The majority (45–61%) of grains are prismatic adcumulate
is the crystallisation of plagioclase before clinopyroxene amphiboles exhibiting 120° grain boundaries (Fig. 4c).
with the exception of a single plagioclase-free clinopyrox- Interstitial phases are rare, occurring in only one sample
ene + amphibole xenolith variety. as a second generation of spinel. Grain size differences
Samples were point-counted (900–1800 points per sam- are observed with samples being completely microcrystal-
ple) using a mechanical stage and the volume modes were line (ø ≤ 500 µm) or extremely coarse (ø ≤ 20 mm). Some
used to classify the xenoliths using the scheme of Steckeisen samples are porphyritic. Olivine (< 2%), when present, is
(1976) (Table 1). Volume modes were converted to mass heavily iddingsitized, anhedral and always microcrystalline.
modes using mineral densities (Deer et al. 2013) (Fig. 2a). Plagioclase (29–46%) is subhedral with sieve-textured cores.
Textural features were used to organize the xenoliths into six Clinopyroxene (up to 12%) grains are usually subhedral and
groups (1–6) that are further subdivided (a, b, c, etc.) on the only present as completely rounded inclusions in hornblende.
basis of the observed mineral assemblages and crystallisa- Ilmenite is only observed in microcrystalline samples. Zon-
tion sequences (Fig. 4; Table 2). ing is prevalent. The principal crystallisation sequences
are: 3a—plag > ol > cpx > hbl; 3b—plag > cpx > hbl; 3c—
ol > plag > cpx > hbl; 3d—ol > plag > hbl.
Group 1 hornblende‑free (ol + plag + cpx + spl + ilm)
This group is distinguished by mesocumulate/orthcumulate

Group 4 subhedral poikilitic hornblende
textures lacking amphibole (Fig. 4a). Olivine is partially to
(plag + cpx + hbl + spl ± ol ± ilm ± bt ± ap ± opx)
completely iddingsitized, with few instances of iddingsitized
cores within unaltered rims. Olivine (ø ≤ 4 mm) and clino-
Subhedral poikilitic hornblende with inclusions of one or
pyroxene (ø ≤ 5 mm) grains are anhedral making up 13 and
more minerals (particularly rounded plagioclase and spi-
30% by mass respectively. Plagioclase grains (57–78%) are
nel) is dominant (up to 56%) (Fig. 4d). Plagioclase (48%)
generally subhedral and sieve-textured, but may be anhedral
is subhedral (ø ≤ 2 mm) with sieve textures; clinopyrox-
when observed as inclusions. Plagioclase grains are mostly
ene (< 15%) is subhedral or anhedral (ø ≤500 µm). Idding-
2 mm in size, except where there is more than one genera-
sitized olivine, biotite, apatite, orthopyroxene and ilmenite
tion, early crystallising plagioclase occurs as small rounded
occur in one sample (BQA10); spinel is the opaque oxide
inclusions (ø < 500 µm) in other minerals while the later
component of the other members of this group. Zoning is
generation may reach up to 8 mm, with sutured grain con-
widespread, but most pronounced in sample BQA10. The
tacts. Spinel and ilmenite generally occur as inclusions, but
principal crystallisation sequences are: 4a—plag > cpx > hbl;
spinel may in some cases occur interstitially. Zoning occurs
4b—plag > ol/opx > hbl.
in all phases, however, in some olivine gabbros plagioclase
is unzoned. The principal crystallisation sequences are: 1a—
plag > ol > cpx; 1b—ol > plag > cpx.
Group 5 interstitial hornblende
(ol + plag + cpx + opx + hbl + spl + ilm)
Group 2 euhedral clinopyroxene
(ol + plag + cpx ± hbl ± spl) This group exhibits one of two textural patterns: either (a)
minor interstitial amphibole displaying alteration of clinopy-
This group is characterised by euhedral (ø ≤ 5 mm) clino- roxene to amphibole along grain edges and as blebs along
pyroxene (comprising up to 40%) that is absent from all cleavage planes (Fig. 4e) or (b) large expanses of poikilitic
other xenolith groups (Fig. 4b). Textures are adcumulate amphibole plates (ø ≤ 12 mm). Textures are orthocumu-
and mesocumulate. Olivine (13%) is anhedral (ø ≤ 3 mm) late. Interstitial glass is observed, albeit generally partially
and partially to completely iddingsitized. Plagioclase crystallised. Olivine (< 10%) is anhedral (ø ≤3 mm), with a
(47%) is subhedral with sieve-textured cores. Plagioclase combination of iddingsitized and unaltered grains, the former
grain sizes are chiefly 2 mm, but where there is more than sometimes existing as iddingsitized patches within unaltered
one generation, the later generation may reach up to 5 mm. olivine grains. Plagioclase (38–80%) is subhedral (ø ≤ 2 mm)
Amphibole (3%), if present, has subhedral poikilitic textures. with sieve-textured cores. Clinopyroxene (5–36%) is anhedral
Spinel when present occurs as minor inclusions. Zoning is (ø ≤ 2 mm) and poikilitic. Of all xenolith groups, orthopyrox-
pervasive. The principal crystallisation sequences are: 2a— ene is most significant here (< 11%) with grain sizes < 500 µm.
plag > ol > cpx > hbl; 2b—ol > plag > cpx. Zoning is prevalent. The principal crystallisation sequences
13
Table 1 Xenolith classification, mineral mass modes and crystallization sequences
Sample Name Cumulate texture Textural designation ol plag cpx opx hbl spl ilm ap bt Crystallization sequence
BQ4 ol gabbro Mesocumulate hbl free 11 63 18 − − 8 − − − spl, plag, ol, cpx

BQ12 ol troctolite Mesocumulate hbl free 15 78 7 − − ± − − − spl, ol, plag, cpx
BQA12 leuco ol gabbro Mesocumulate hbl free 12 70 17 − − 0.5 0.5 − − ol/plag, cpx, spl/ilm
BQA35 ol gabbro Orthocumulate hbl free 13 57 30 − − − − − − plag, ol/cpx
BQB9 leuco ol gabbro Mesocumulate hbl free 10 70 20 − − − − − − plag, ol/cpx, plag
[An93] [An93]
BQA11 ol gabbro Adcumulate Euhedral cpx 12 52 36 − − − − − − ol, plag, cpx
BQB11 ol–hbl gabbro Mesocumulate Euhedral cpx 14 43 40 − 3 ± − − − plag, ol, spl, cpx, hbl, plag
[An83] [An88]
BQ7 hbl gabbro Adcumulate Prismatic hbl − 42 1 − 53 4 − − − plag, spl, cpx, hbl, plag/spl
[An86] [An86]
BQ11 ol–hbl gabbro Mesocumulate Prismatic hbl 1 46 7 − 45 1 − − − plag, ol, spl, cpx, hbl, spl
BQA15 hbl gabbro Adcumulate Prismatic hbl − 42 3 − 55 − − − − plag, cpx, hbl
BQA22 mela ol–hbl gabbro Adcumulate Prismatic hbl 1 38 − − 61 ± − − − ol/spl, plag, hbl, plag
[An88] [An87]
BQA24 mela ol–hbl gabbro Adcumulate Prismatic hbl 1 29 12 − 58 − ± − − ol, plag, cpx, ilm, hbl
BQ8 hbl gabbro Mesocumulate Subhedral poikilitic hbl − 41 1 − 56 2 − − − spl, plag, cpx, hbl
BQA10 leuco ol–hbl gabbronorite Mesocumulate Subhedral poikiltic hbl ± 74 − ± 24 − 0.5 1 0.5 plag, ilm, bt/ol/opx, hbl, plag, ap
[An55] [An84]
BQA18 pyroxene–hbl gabbro Orthocumulate Subhedral poikilitic hbl − 55 15 − 29 1 − − − spl, plag, cpx, hbl
BQA6 ol–hbl gabbronorite Orthocumulate Interstitial hbl 7 38 36 11 7 1 − − − ol/plag, spl, cpx, opx, hbl
BQA7 leuco ol–hbl gabbro Mesocumulate Interstitial hbl 7 81 9 − 3 ± ± − − spl, ol, plag, ilm, cpx, hbl
BQA16 ol–hbl gabbronorite Orthocumulate Interstitial hbl 10 51 30 2 6 ± 1 − − spl, plag, opx, cpx, ol, ilm/hbl, plag
BQA33 ol–hbl gabbro Orthocumulate Interstitial hbl 11 52 5 ± 31 1 − − − spl/plag, opx, ol/cpx, hbl, plag/cpx
BQA40 ol–hbl gabbronorite Orthocumulate Interstitial hbl 2 58 9 8 23 ± ± − − ol, plag, spl, cpx, opx, hbl, plag
BQA3 ol gabbro and anorthosite layers Mesocumulate Layered 29 47 23 ± − 0.5 0.5 - - plag, spl, ilm, cpx, ol
1 99 − − − − ± − − plag/ilm, ol, plag
BQA20 hbl pyroxenite and gabbro layers Adcumulate Layered − − 90 − 10 − − − − cpx, hbl
− 65 30 − − 5 − − − spl, plag, cpx
Crystallisation histories constructed through petrographic observation of included mineral phases

Average An content (mineral cores) of samples containing two generations of plagioclase are highlighted in brackets
− mineral phase absent, ± minor mineral phase, ol olivine, plag plagioclase, cpx clinopyroxene, opx orthopyroxene, hbl hornblende, spl spinel, ilm ilmenite, ap apatite, bt biotite
Page 5 of 26
13
77
Fig. 2 Modal phase proportions (by mass) of phases in Bequia rocks: shown for each sample. Fo compositions are not listed where olivine
a xenoliths b phenocrysts and c glomerocrysts. Note that glomero- is highly altered. Samples are listed in order of decreasing Fo con-
crysts phase proportions and assemblages are averages per sample. tent of olivine followed by An content of plagioclase within a given
Sample numbers and textural group/rock type designations are given. group. For lavas, compositional ranges are provided for both phe-
Average olivine Fo contents and core/rim plagioclase An contents are nocrysts and glomerocrysts where possible
are: 5a—plag > ol > cpx > hbl; 5b—ol > plag > cpx > hbl; 5c— Hornfelsed xenoliths
ol > plag > cpx > opx > hbl; 5d—plag > opx > cpx > ol > hbl.
A few xenolith samples (BQA20, BQA40), irrespective of
their textural group designation, exhibit mortar or granoblas-
Group 6 layered tic textures with some instances of mineral alteration (Figure
(ol + plag + cpx + opx + hbl + spl + ilm) A1, Online Resource 1). Closely interlocking, fine-grained
crystals comprise gabbroic lithologies. Mineral alteration is
This group is unique in that it consists of distinct layering on a mainly observed in pyroxene grains by their unusual deep
microscopic scale (Fig. 4f). Layers differing by texture, grain green appearance in plane-polarized light. Such textures may
size and/or mineral assemblage are juxtaposed in any one be related to high-temperature metamorphic reactions as a ves-
thin section. Unaltered poikilitic olivine (28%) is observed in tige of direct wall–rock interaction. There is no mineralogical
one sample (BQA3, Fig. 3a), which is not observed in any or chemical evidence to warrant the hornfelsed xenoliths being
other textural group. Members of this group contain at least placed in a separate group.
one monomineralic layer (anorthosite, BQA3 or pyroxenite,
BQA20), which is again absent from other textural groups.
Sieve textured plagioclase is a recurring feature. Zoning is Lava petrology
prevalent. Sample BQA20 (Group 6c) is unique in containing
disrupted layers of a plagioclase-free clinopyroxene + amphi- Sixteen lava samples, collected during fieldwork in 2011
bole assemblage surrounded by a gabbroic matrix of plagio- and 2015, were studied (Fig. 1b). The lava suite consists
clase, clinopyroxene and spinel (Fig. 4f). The principal crys- of basalts, basaltic andesites and a single andesite with
tallisation sequences are: 6a—plag > cpx > ol; 6b—plag > ol; glomeroporphyritic textures. Phenocryst assemblages con-
6c—cpx > hbl; 6d—plag > cpx. sist of olivine, plagioclase and clinopyroxene with minor
13
Table 2 Summary of xenolith groups and subdivisions monomineralic and polymineralic clusters with two or
Group Textural designation Main assemblages
more grains, the latter occurring only in basalts and basaltic
and crystallisation andesites (Fig. 2c). Melt inclusions (MI) are lacking in phe-
sequences nocrysts. Secondary minerals prehnite and calcite occupy
veins and vesicles, suggesting host lava flows were affected
1 Hornblende-free a. plag, ol, cpx
by low-temperature metamorphism after emplacement. The
b. ol, plag, cpx
groundmass consists of microlites of the phenocryst and
2 Euhedral clinopyroxene a. plag, ol, cpx, hbl
microphenocryst minerals. The basalts and basaltic andesites
b. ol, plag, cpx
are generally phenocryst- and microphenocryst-rich, with
3 Prismatic hornblende a. plag, ol, cpx, hbl
the groundmass making up < 50% vol. of the rocks. Con-
b. plag, cpx, hbl
versely, the andesite is phenocryst-poor with > 60% vol
c. ol, plag, cpx, hbl
groundmass.
d. ol, plag, hbl
4 Subhedral poikilitic hornblende a. plag, cpx, hbl
b. plag, ol/opx, hbl
Olivine
5 Interstitial hornblende a. plag, ol, cpx, hbl
Almost all olivine phenocrysts and microphenocrysts are
b. ol, plag, cpx, hbl
partially to completely replaced by iddingsite. In some
c. ol, plag, cpx, opx, hbl
cases, groundmass olivine shows iddingsitization. Olivine
d. plag, opx, cpx, ol, hbl
can be both subhedral and anhedral, and occurs in sizes
6 Layered a. plag, cpx, ol
ranging from a few hundreds of micrometres to 2.5 mm.
b. plag, ol
Phenocrysts exhibit various stages of corrosion from
c. cpx, hbl
being partially embayed to complete dissolution of inte-
d. plag, cpx
riors. Olivine is generally the first phase to crystallize,
however, rounded inclusions of other phases (plagioclase
(Fig. 5d) and clinopyroxene) within poikilitic olivine
orthopyroxene, spinel and ilmenite. Amphibole is notably were observed in some samples suggesting an additional,
absent. Phenocryst mass modes for the lavas were calcu- later generation of olivine.
lated on a vesicle-free basis (Fig. 2b). Phenocryst aggre-
gates (glomerocrysts) are ubiquitous. They occur as both
Fig. 3 Backscattered elec-

tron images of crystallisation
sequences in xenoliths. a
Poikilitic olivine with inclusions
of plagioclase and clinopyrox-
ene (BQA3, ol gabbro layer).
b Olivine with inclusions of
plagioclase, together with a
later generation of plagioclase
imposing growth on olivine
(BQA3, anorthosite layer). c
Olivine with orthopyroxene
and plagioclase inclusions and
clinopyroxene with plagio-
clase inclusions (BQA16). d
Large plagioclase grains with
sutured contacts, together with
clinopyroxene with plagioclase
inclusions (BQB9)
13
Fig. 4 Photomicrographs (plane polarized light, ppl) of representative inated by early crystallising prismatic amphibole (BQ11). d Group
xenolith groups. a Group 1 containing anhedral, interlocking idding- 4 containing subhedral poikilitic amphibole with rounded inclusions
sitized olivine, plagioclase and clinopyroxene with two generations of of plagioclase and spinel (BQ8). e Group 5 containing intercumulus
plagioclase observed as inclusions in clinopyroxene and large grains amphibole with unaltered olivine (BQA8). f Group 6 with clinopy-
with sutured contacts (BQB9). b Group 2 with zoned euhedral clino- roxene–amphibole cluster surrounded by a microcrystalline adcumu-
pyroxene, anhedral iddingsitized olivine, subhedral sieve-textured late of plagioclase, clinopyroxene and spinel (BQA20)
plagioclase and minor poikilitic amphibole (BQB11). c Group 3 dom-
Plagioclase of the interior and exterior of grains is a common fea-

ture with grains having sieve textures and being partially
Plagioclase grain sizes range from a few hundred micro- embayed respectively. Two texturally dominant plagioclase
metres up to 3 mm. Individual grains are euhedral to sub- types are distinguished: zoned phenocrysts and unzoned
hedral. Twinning and zoning are widespread. Corrosion microphenocrysts.
13
Fig. 5 Photomicrographs of
representative lava textures. a
Basalt (cross polarized light,
xpl) containing a small glom-
erocryst with grains protruding
into groundmass (BQ10). b
Basalt (xpl) with a small glom-
erocryst containing grains a few
hundreds of micrometres or less
(BQ01). c Andesite (ppl) with
abundant plagioclase grains
with vesicles and trachytic
textured groundmass (BQBL2).
d Basalt (ppl) containing olivine
grains with plagioclase inclu-
sions (BQ03)
Pyroxene absent. Monomineralic (plagioclase) clusters dominate the

andesitic sample (Fig. 5c). Plagioclase-only (ø ≤ 2 mm) and
Clinopyroxene sizes range from several hundreds of micro- clinopyroxene-only (ø ≤ 3 mm) clots are also observed in
metres to ≤ 1 cm, with grains being euhedral to subhedral. basaltic samples. Minerals exhibit similar disequilibrium
Occasionally, one or more grain boundaries are fractured. textures, such as zoning and resorption to those in phe-
Twinning and zoning (concentric and sector) are common. nocrysts. Crystals from all cluster types protrude out into
Plagioclase inclusions are observed suggesting crystallisa- the groundmass indicating that they were not rock fragments
tion before clinopyroxene. Three textural types of pyrox- but loose aggregates of crystals when entrained into the host
ene are distinguished: large, isolated zoned megacrysts; magma (Seaman 2000). Interstitial glass is absent.
zoned and unzoned phenocrysts; and unzoned microphe-
nocrysts. Subhedral orthopyroxene (< 2 mm) is a minor,
unzoned phenocryst (< 2%) in a single sample (BQ11). Xenolith mineral chemistry
Xenolith mineral compositions are presented in Table A1

Opaque oxides (Online Resource 2). Mg# is always expressed as 100 Mg/
(Mg + Fe2+) with F e2+ determined by stoichiometry or via
Spinel tends to occur as a minor component of glomero- the method of Holland and Blundy (1994) for amphibole.
crysts and in the groundmass. Ilmenite is rare, confined Olivine is present in the majority of the xenoliths studied
to two basalts (BQ04 & BQ07). Both types of oxide have (≈ 75%). Forsterite contents range from Fo71–55. NiO con-
homogenous textures. centrations are generally below detection, with only the lay-
ered group displaying detectable concentrations from 0.08 to
0.14 wt%. CaO concentrations vary from 0.09 to 0.31 wt%
Glomerocrysts with no clear correlation with Fo (Fig. 6). Conversely, Fo
contents correlate negatively with MnO (0.7 to 0.4 wt%).
Polymineralic glomerocryst sizes range from a few Zoning could not be reliably measured due to iddingsite
hundred micrometres to 1 cm. Assemblages includes alteration along grain rims.
spl–ol–plag–cpx, ol–plag, ol–plag–cpx, plag–cpx and ol–cpx Plagioclase (An98–49) is a ubiquitous xenolith phase. Nor-
(Fig. 5a, b). Amphibole and orthopyroxene are notably mal zoning is dominant with rimward decreases in An from
13
Fig. 7 Variation in clinopyroxene compositions for xenoliths and

Fig. 6 Variation in olivine composition for xenoliths and lavas. Xeno- lavas in terms of Ca atoms per formula unit versus Mg#. Two
liths are denoted by circles and lavas by squares. Note the preferen- divergent trends of decreasing Ca content with decreasing Mg# are
tial clustering of xenolith groups and the similarity in glomerocryst observed for clinopyroxene phenocrysts and glomerocryts. Note high
and xenolith Fo compositions. CaO content lower in xenoliths than Mg# (≥ 90) clinopyroxene in xenoliths from Group 1 and some lavas
lavas, denoting calcium diffusing out of olivine during cooling. Iso-
therms calculated from the olivine–clinopyroxene thermometer of
Shejwalkar and Coogan (2013) for orthopyroxene-bearing xenoliths
and lavas Calcic amphibole (Mg#98−60) with pargasite composition
(Leake et al. 2003) is present in just over half of the xeno-
liths analyzed. Normal zoning is observed (≤ 5% Mg#). Tex-
1 to 34%. BQA10 shows an extreme rimward decrease of turally, three types of amphibole were identified: (a) early
An83–An49. Reverse zoning is less prevalent with rimward crystallizing prismatic grains and late crystallizing (b) poiki-
increases in An ≤ 2% (e.g. A n91 to A
n93 in BQA35). Where litic and (c) interstitial grains (Fig. 4c, d, e). Amphiboles
there is more than one generation of xenolith plagioclase, corresponding to the different textural groups define distinct
the average An content either does not vary considerably compositions clustering in terms of K, AlIV and Ti versus
between the two generations or shows higher An contents in Mg# (Fig. 8). Within each group, core and rim compositions
the larger, later grains (e.g. BQA10 and BQB11, Table 1). often overlap. Low Mg# amphiboles from Group 4 (BQA10)
Clinopyroxene (Mg# 97−70) is present in all xenoliths contain unusually high K contents (> 0.14 pfu) compared to
except BQA10, that contains a single orthopyroxene grain, other textural groups. In fact, Group 4 contains two distinct
and BQA22. All grains are zoned showing normal (≤ 12% compositions distinguished by Mg#. The lower Mg# cluster
Mg# decrease), reverse (≤ 5% Mg# increase) or oscillatory belongs to a single sample (BQA10), whereas the remaining
zoning. Compositions consist mainly of diopside and salite. samples exhibit higher Mg#. Group 5 exhibits two clusters
With increasing Mg#, A l(IV) and Ca (pfu) (Fig. 7) contents separated by K contents: higher K concentrations belong
increase while Ti (pfu) decreases. Group 1 contains a sub- to BQA33 and BQA16, while the remainder of the sam-
set of highly magnesian (Mg# >90) clinopyroxenes (from ples in this group contain lower K concentrations. A sub-
sample BQB9) that plot separately from the rest of the suite. set of amphiboles has notably high AlIV contents (> 2 pfu).
These grains also exhibit normal and reverse zoning. Cr contents are generally below detection except in some
Orthopyroxene (Mg#68−56) is present in 5% of the sam- amphiboles from Groups 2 and 6 (0.03–0.05 pfu). Group 2
ples analyzed. When occurring with amphibole, orthopy- contains subhedral poikilitic-textured hornblende with com-
roxene compositions are En64–55. The clinopyroxene-free positions matching the rest of those in Group 4. Extremely
sample (BQA10) has an orthopyroxene composition (En55) high Mg# amphibole, distinct from the other textural groups,
distinct from the other clinopyroxene-bearing xenoliths occurs in Group 6.
(En64–61). Orthopyroxene in the amphibole-free xenolith Spinel is present in more than 75% of the xenoliths stud-
(Group 1) is more Fe-rich (En66), lying instead closer to ied, though never exceeding > 8% by mode. In terms of
lava compositions. Cr# = Cr/Cr + Al, compositions range from Cr-rich spinel
to titaniferous magnetite. The most Cr-rich spinels (Cr#52−44)
13
are observed in Groups 1b, 2a, 6a and 6b (Figure A2, Online

Resource 3), occurring generally as inclusions in olivine
or plagioclase. Interestingly, the rest of Group 6 contains
Cr-poor spinels (Cr# = < 10) from the hornfelsed sample
BQA20.
Lava phenocryst chemistry
Phenocryst mineral compositions are presented in Table A1

(Online Resource 2).
Olivine crystals exhibit a broader range of compositions
(Fo84–45) than the xenoliths. Olivine in glomerocryst clus-
ters has a Fo range that closely mirrors the xenoliths. The
extent of normal zoning (when measurable) is at most a 7%
Fo decrease towards the rim (Fo65 to F o58, BQL1). Reverse
zoning is also observed, e.g. Fo55 to F o57 in BQL1. All phe-
nocrysts contain > 0.15 wt% CaO (Fig. 6). MnO steadily
decreases with increasing Fo from 1 to 0.2 wt%. NiO con-
tents are generally below detection except at F o≥ 80, where
they reach up to 0.13 wt%.
Plagioclase shows a broad range of compositions
(An96–11). However, the lowest An compositions (< An30) are
rare (Table A1, Online Resource 2); the majority of analyses
have a range of A n96–40. Plagioclases exhibit normal and
reverse zoning with oscillatory zoning common. The extent
of normal zoning is up to a 46% An decrease towards the
rim (An89 to An43, BQL1). The extent of reverse zoning
was found to be at most a 15% An increase rimward (An63
to An74, BQ11). The most calcic plagioclase compositions
were found in the cores of zoned crystals; lowest An con-
tents were found in rims.
Clinopyroxene (Mg#98−60) phenocrysts comprise diop-
side, salite and high-Ca augite compositions. Normal zoning
occurs in all textural types, with core and rim compositions
averaging Mg#85 and Mg#77 respectively. Reverse zoning is
restricted to glomerocrysts, isolated megacrysts and large
phenocrysts. Clinopyroxene Ca content increases with
increasing Mg# reaching a maximum of 0.96 pfu at Mg#98.
Phenocrysts and glomerocrysts display two clear trends:
low-Ca and high-Ca (Fig. 7). Clinopyroxene shows a trend
of increasing A lIV and decreasing Ti (pfu) with increasing
Mg#, with glomerocrysts following the same narrow range
in Mg# displayed by the xenoliths.
Orthopyroxene (Mg# 70−69) was found only in BQ11
(Fig. 2b) and lacks obvious zoning.
Spinel is predominantly titaniferous magnetite (≤ 2% by
mode). Anomalous Cr-rich spinels (Cr#77−58) are found in
two samples as rounded inclusions in olivine and as anhedral
microphenocrysts (Figure A2, Online Resource 3).
Fig. 8 Variation in hornblende composition for plutonic xenoliths in terms of

a K (pfu), b Ti (pfu) and c AlIV (pfu) versus Mg#. Note the wide range in Mg#
and compositional grouping of xenolith groups
13
Xenolith melt inclusions conducting a mass balance calculation to determine the

strength of the relationship between the measured XRF
Melt inclusions (MI), ranging in size from < 10 to 50 µm, major element data and the averaged major element compo-
occur in most phases except orthopyroxene, biotite and apa- sitions of the constituent minerals: olivine, plagioclase and
tite. Those in plagioclase are often devitrified. MI have a clinopyroxene. The smallest residuals were computed when
pale brown colour in transmitted light and vary in shape the cores and rims for plagioclase and clinopyroxene were
from rounded to oval. While many contain a vapour bubble, considered separately to compute the whole rock composi-
some are bubble-free. Such bubbles may contain substan- tion. XRF whole rock data for BQA35 were used in prefer-
tial concentrations of C O2 which may result in underesti- ence to calculated bulk composition in all plots.
mates of the original dissolved CO2 (Moore et al. 2015). Lava whole rock compositions range from medium-K
We selected for analysis twelve large (≥ 30 µm in diameter) basalts to basaltic andesites (50–58 wt% SiO2, Table 4).
glassy, daughter mineral-free MI (Table 3). They show typical fractionation trends with decreasing
MI EPMA totals are ≥ 90.5 wt%. MI record low K dFe−Mg MgO, i.e. increasing SiO2, Na2O and P2O5 and decreas-
values when compared to expected Kd values (Table 3) ing CaO and FeO (Fig. 9). There is a fractionation peak in
for olivine (Roeder and Emslie 1970), clinopyroxene and Al2O3 (22 wt%) corresponding to the onset of plagioclase
hornblende (calculated using St. Kitts experiment Run 3–1, crystallisation. With the exception of BQA35, all xenoliths
Melekhova et al. 2017) suggestive of post-entrapment crys- show compositions that are distinct from the liquid line of
tallization (PEC). Corrections for PEC, made using experi- descent described by the lavas (Fig. 9) confirming that they
mental Kd values from Melekhova et al. (2017), ranged from are indeed true extractive assemblages (or cumulates) and
1 to 8%. Corrected and uncorrected MI compositions are not frozen aliquots of erupted magma. In contrast, BQA35
reported in Table 3. Corrected MI compositions, normalised lies along the liquid line of descent of the lavas. This sam-
to 100% anhydrous, vary from 63.6 to 69.8 wt% S iO2 and ple is denoted “non-cumulate gabbro” in Fig. 9 and likely
0.6 to 2 wt% MgO (Fig. 9). Their distribution describes a represents magma that has solidified in situ.
trend that extends to beyond the highest SiO2 contents of
the bulk lavas. MI are depleted in FeO and TiO2 (Figure
A3, Online Resource 4), as expected for greater degrees of Comparison of xenoliths and lavas
fractionation. There is no relationship between MI composi-
tion and xenolith group or MI host. The most evolved melt Several textural and compositional likenesses were observed
is a clinopyroxene-hosted MI from Group 3. between the plutonic xenoliths and lavas. Plagioclase is a
MI H2O and CO2 concentrations, as measured by SIMS, dominant phase in both lava phenocrysts and plutonic sam-
range from 2.8 to 7.8 wt.% and 55 to 2405 ppm, respectively ples but not in glomerocrysts (Fig. 2). Similar disequilib-
(Table 3). Volatile contents were not affected significantly rium textures (e.g. sieve textured cores and resorption) were
by PEC. Again, there is no obvious correlation between MI revealed in all suites. Zoning is ubiquitous, but generally
H2O content and host mineral or xenolith group. The MI with more protracted in lava phenocrysts and glomerocrysts
the highest H 2O content was found in a clinopyroxene grain compared to xenoliths. The most striking difference is the
from sample BQA35. Chlorine contents range from 1000 to ubiquity of amphibole in xenoliths and its absence in lavas.
3900 ppm; sulphur contents are generally below detection Compositional ranges in all phases are generally broader in
(< 382 ppm). lavas and more restricted in xenoliths. A complex growth
history is, therefore, inferred for phases in both lavas and
Whole rock compositions xenoliths.
Bulk rock major element compositions of the plutonic xeno- Olivine

liths were calculated using averaged mineral compositions
and mass fraction modes. Where olivine compositions could Bequia lava olivines have a wider Fo range (Fo84–45) than
not be accurately measured due to alteration, calculations their xenolith counterparts ( Fo71–55), which have very simi-
were not attempted. Five xenoliths were analysed using XRF lar Fo range to glomerocrysts. Higher Fo contents in lava
(as for lavas) to check the accuracy of xenolith bulk rock olivines (Fo ≥ 80) hint toward either a more primitive par-
calculations (Table 4). Generally, good agreement (≥ 97% ent or the incorporation of another population of olivine
relative agreement) is observed for the two methods with crystallising under different conditions just before eruption.
the exception of sample BQA35. In this case we infer that Olivine NiO contents were generally below the detection
the effects of mineral zoning and the sample’s heteroge- limit and no relationship is observed between CaO and Fo
neity contributed to this mismatch. This was checked by (Fig. 6). However, xenoliths show lower CaO contents com-
pared to phenocrysts and glomerocrysts suggestive of Ca
13
Table 3 Xenolith melt inclusion compositions
Group: Sample: 1 1 1 1 2 2 3 3 3 3 5 5
Name: BQA12 BQA12 BQA35 BQA35 BQA1 BQA1 BQ7 BQA13 BQA13 BQA13 BQA8 BQA8
Host: bqa12mi7 bqa12mi20 bqa35mi4 bqa35mi3a bqa1mi4a bqa1mi4b bq7mi1a bqa13mi9 bqa13mi8a bqa13mi8b bqa8mi44a bqa8mi44b
ol ol cpx cpx ol ol hbl ol cpx cpx ol ol
SiO2 63.43 64.81 59.11 60.64 64.98 64.97 62.35 60.20 67.40 65.79 60.52 61.73
TiO2 0.55 0.56 0.07 0.02 0.40 0.35 0.59 0.17 0.32 0.34 0.67 0.53
Al2O3 16.98 16.47 19.44 20.11 17.07 17.25 18.53 16.70 14.95 16.20 17.32 17.09
FeO 2.85 3.03 0.64 0.52 2.36 1.99 3.69 4.25 2.70 3.74 4.08 3.22
MnO 0.00 0.04 0.10 0.09 0.04 0.08 0.24 0.07 0.02 0.06 0.07 0.09
MgO 0.31 0.39 0.04 0.01 0.27 0.08 0.40 0.49 0.27 0.37 0.50 0.32
CaO 4.13 3.38 6.51 5.53 3.29 3.95 4.48 4.25 1.74 2.42 4.83 4.76
Na2O 4.03 4.06 3.21 3.80 3.87 4.44 4.24 3.38 3.68 4.08 4.41 3.76
K2O 1.60 1.67 1.07 0.79 1.90 1.46 0.98 3.42 3.22 2.59 1.39 1.56
P2O5 0.21 0.45 0.22 0.22 0.28 0.26 0.20 0.17 0.12 0.19 0.24 0.24
SO2 0.05 0.02 0.00 0.02 0.02 0.00 0.08 0.05 0.00 0.02 0.06 0.04
Cl 0.26 0.39 0.10 0.17 0.25 0.36 0.29 0.44 0.18 0.25 0.34 0.25
Cr2O3 0.02 0.00 0.00 0.03 0.00 0.01 0.00 0.00 0.03 0.00 0.02 0.02
H2O 5.80 6.78 6.61 7.84 5.15 4.57 4.72 5.33 2.78 5.81 5.22 6.08
sd 1.43 0.17 0.41 0.18 0.19 0.14 0.11 0.14 0.08 0.13 0.15 0.13
CO2 236 1786 bdl 55 2405 187 97 1365 944 106 340 1059
sd 77 77 - 24 178 26 18 218 86 12 24 77
Psat 220 527 241 316 571 172 162 407 251 200 232 365
Total 100.20 102.05 97.12 99.79 99.88 99.77 100.79 98.92 97.41 101.86 99.67 99.69
Kdmeasured 0.133 0.156 0.026 0.009 0.167 0.064 0.115 0.123 0.062 0.062 0.162 0.132
Kdpredicted 0.300 0.300 0.280 0.280 0.300 0.300 0.310 0.300 0.280 0.280 0.300 0.300
Restored compositions
SiO2 62.91 64.30 58.79 60.31 64.64 64.47 60.82 59.42 66.13 64.27 59.98 61.19
TiO2 0.54 0.55 0.11 0.06 0.39 0.34 0.74 0.16 0.31 0.32 0.65 0.52
Al2O3 16.66 16.18 18.91 19.66 16.88 16.95 18.11 16.17 13.77 14.43 16.94 16.73
FeO 3.46 3.59 0.84 0.70 2.78 2.62 4.39 5.16 3.22 4.32 4.79 3.91
MnO 0.01 0.05 0.10 0.09 0.05 0.09 0.24 0.08 0.05 0.09 0.08 0.10
MgO 0.86 0.90 0.56 0.42 0.56 0.53 1.29 1.46 1.43 1.93 1.09 0.89
CaO 4.05 3.32 7.10 6.04 3.25 3.88 4.96 4.12 3.42 4.67 4.72 4.66
Na2O 3.95 3.99 3.10 3.70 3.82 4.36 4.12 3.27 3.39 3.63 4.32 3.68
K2O 1.57 1.64 1.03 0.77 1.88 1.44 0.92 3.31 2.95 2.30 1.36 1.53
Cr2O3 0.02 0.01 0.00 0.04 0.00 0.01 0.00 0.00 0.02 0.00 0.02 0.02
Page 13 of 26
13
77
diffusion out of olivine during cooling (Kohler and Brey

1990; Shejwalkar and Coogan 2013). Isotherms calculated
for orthopyroxene-bearing xenoliths using the Shejwalkar
and Coogan (2013) olivine–pyroxene thermometer indicate
crystallisation between 850 and 900 °C, slightly lower than
orthopyroxene-bearing lava olivines (900–950 °C). Higher
Ca contents in phenocryst olivines may also suggest that the
magmas from which they crystallised contained lower H2O
concentrations than those of cumulates (Gavrilenko et al.
2016), which may in turn reflect differences in the pressures
of crystallisation.
Plagioclase
Anorthite ranges are notably broad in both lava (phenocrysts:

predominantly An96–40 and glomerocrysts: An96–41) and
xenolith (An98–49) plagioclase. Lava phenocryst and glom-
sd (standard deviation) calculated by full error propagation of counting statistics, instrument background and calibration working curve
Kdpredicted are experimental Kd values from Roeder and Emslie (1970) (ol-liquid) and Melekhova et al. (2017) (cpx-liquid, hbl-liquid)
erocryst cores lie in the range A

n96–86 and An96–76, respec-
tively, whereas xenolith cores are An98–84. K2O concentra-
tions decrease with increasing An, however, the majority of
lava concentrations range up to 0.5 wt% (with < 2% of rim
analyses range 1.7–5 wt%) while xenolith concentrations
range up to only 0.3 wt%.
Clinopyroxene
Clinopyroxene zoning occurs in lavas and xenoliths, how-

ever, it is more common in glomerocrysts. While clinopy-
roxene grains tend to be euhedral/subhedral in lavas, xeno-
lith clinopyroxene is generally anhedral except for those in
Group 2. Lava clinopyroxene has diopside, salite and high-
Ca augite compositions, whereas xenoliths mainly cluster
Restored melt inclusion major element compositions calculated using Kdpredicted
Psat (saturation pressure) in MPa using method of Ghiorso and Gualda (2015)
around diopside and salite. The Mg# ranges for phenocrysts,

glomerocrysts and xenoliths are Mg#98−60, Mg#93−66 and
Mg#97−70 respectively. The variation in Ca (pfu) with Mg#
(Fig. 7) highlights two trends likely reflecting the control
of temperature on Ca. The first trend displayed by lavas
O2 (in ppm) measured by ion-microprobe
Major elements (in wt%) determined by electron microprobe
and xenoliths shows higher Ca contents suggesting lower

crystallisation temperatures. The second trend, denoted by
a second population of lavas and fewer xenoliths, contains
lower Ca suggestive of higher crystallisation temperatures.
ol olivine, cpx clinopyroxene, hbl hornblende
Lower temperatures in xenoliths are consistent with intratel-

luric cooling prior to entrainment. It is also conceivable that
the diverging Ca trends indicate small variations in the H2O
contents of the magmas from which different clinopyrox-
ene populations crystallised. At depth, higher H 2O contents
decrease crystallisation temperatures (and silica activity)
leading to clinopyroxene compositions with higher Ca con-
Table 3 (continued)
H2O (in wt%) and C
tents (Müntener et al. 2001). Phenocrysts from lavas and

glomerocrysts show similar distribution and variation in Ca
and Mg#.
13
Fig. 9 Bulk major element variations in Bequia xenoliths (estimated ies (031-22b, STV315, GW40) from Pichavant et al. 2002a, Pichavant
from mineral modes and compositions), xenolith MI (EPMA, cor- and Macdonald 2007 and Pichavant et al. 2018 are shown by big dia-
rected for PEC) and lavas (XRF). Melt compositions in equilibrium monds. Chemical composition of experimental melts from these stud-
with cumulate calcic amphibole are also displayed (with associated ies in equilibrium with solid residues shown on Fig. 12 are plotted as
error bars highlighted), calculated using chemometric equations of small diamonds for comparison with Bequia liquid line of decent. See
Zhang, et al. (2017). Note overlap between calculated melts and MI. text for discussion
The starting compositions of three Lesser Antilles experimental stud-
13
Table 4 Normalized (to 100%) Group Sample SiO2 TiO2 Al2O3 FeO MnO MgO CaO Na2O K2O P2O5 Cr2O3
xenolith and lava compositions
(wt% oxides) estimated from 1 BQ4 42.40 0.96 22.67 11.34 0.14 6.15 15.40 0.91 0.01 – 0.01
mineral modes and XRF
BQ12 44.36 0.06 27.30 4.99 0.10 6.00 16.49 0.71 0.01 – 0.00
respectively
BQ12a 45.26 0.09 28.34 4.39 0.09 5.17 15.78 0.86 0.01 0.01 –
BQA12 46.18 0.50 23.87 6.32 0.12 5.76 15.80 1.36 0.03 – 0.05
BQA35 45.81 0.28 21.21 5.68 0.11 8.91 17.28 0.67 0.02 – 0.03
BQA35a 53.76 0.87 19.18 7.95 0.17 4.95 9.30 3.14 0.58 0.10 –
BQA35MB 53.37 1.14 19.31 8.19 0.11 5.18 9.89 4.94 0.00 0.00 –
BQB9 45.27 0.14 25.46 4.25 0.08 6.58 17.57 0.63 0.01 – 0.01
2 BQA11 46.40 0.23 19.43 5.73 0.12 9.21 18.31 0.50 0.01 – 0.05
BQB11 46.80 0.37 15.85 7.14 0.15 12.04 16.70 0.82 0.03 0.10
3 BQ7 42.64 1.96 20.89 10.88 0.15 7.49 13.79 2.03 0.15 – 0.02
BQA15 44.18 1.57 22.23 6.00 0.08 8.25 15.57 1.73 0.39 – 0.01
4 BQ8 42.80 1.60 22.15 9.28 0.12 7.79 14.16 1.92 0.15 – 0.02
BQA18 45.00 0.94 24.06 4.83 0.08 6.34 17.27 1.34 0.13 – 0.00
5 BQA6 47.94 0.45 15.57 8.14 0.18 10.40 16.39 0.77 0.04 – 0.12
BQA7 46.31 0.43 26.73 5.06 0.08 4.01 15.39 1.93 0.05 – 0.01
BQA7a 46.82 0.85 25.03 6.10 0.13 5.14 14.39 1.50 0.03 0.01 –
BQA16 45.61 0.99 19.02 8.14 0.18 8.22 16.91 0.77 0.12 – 0.03
–
BQA33 43.17 1.78 18.04 11.50 0.20 10.17 12.92 1.90 0.32 0.01
BQA33a 46.01 1.04 20.80 9.56 0.18 8.29 12.29 1.57 0.24 0.02 –
BQA40 46.37 0.51 22.84 6.83 0.13 6.82 14.99 1.38 0.11 – 0.00
BQA40a 48.42 0.47 21.98 6.15 0.13 6.76 14.39 1.54 0.12 0.02 –
6 BQA3 45.03 0.39 15.26 12.24 0.24 12.43 13.44 0.86 0.02 – 0.10
–
BQA3 45.65 0.00 34.10 0.60 0.01 0.27 17.98 1.36 0.02 0.00
BQA3 44.04 0.60 16.46 12.55 0.23 12.10 12.95 0.95 0.01 – 0.11
Basalt BQ04 50.08 0.60 14.81 8.39 0.18 10.57 13.71 1.44 0.16 0.05 –
BQ05 50.85 0.93 22.58 7.86 0.18 2.93 9.89 3.40 1.21 0.18 –
BQ06 50.93 0.65 15.52 9.23 0.19 9.08 12.24 1.96 0.16 0.06 –
BQ07 51.41 0.80 17.92 8.47 0.18 7.06 11.31 2.31 0.42 0.09 –
BQ10 51.91 0.95 19.30 9.15 0.19 5.11 8.94 3.18 1.12 0.15 –
BQ14 51.25 0.86 19.78 8.15 0.17 5.82 10.73 2.55 0.59 0.09 –
BQ17 50.77 0.96 20.93 8.07 0.17 4.76 10.35 2.99 0.86 0.12 –
BQ19 51.53 0.82 19.43 7.90 0.18 5.89 11.11 2.66 0.37 0.11 –
Basaltic BQ01 53.73 1.02 18.22 8.60 0.21 4.65 9.58 3.12 0.73 0.15 –
Andesite BQ08 52.87 0.88 19.55 7.92 0.18 5.18 9.68 2.89 0.73 0.12 –
BQ09 52.22 0.95 19.63 8.20 0.18 5.18 9.91 2.81 0.80 0.12 –
BQ15 52.91 0.94 19.33 8.14 0.17 5.19 9.53 2.83 0.84 0.12 –
BQ16 53.02 1.12 20.61 9.02 0.19 2.66 8.43 3.74 1.05 0.16 –
Andesite BQ18 58.10 0.71 20.05 6.67 0.20 1.19 6.91 4.69 1.29 0.20 –
MB mass balance calculation, – not analysed

a
XRF data from selected xenoliths for comparison to calculations
Orthopyroxene Amphibole
Orthopyroxenes are generally distinct between lava phe- The presence of amphibole in xenoliths and its absence in
nocrysts and xenoliths. Orthopyroxenes in lavas span A lIV lavas may reflect its instability at low pressures and during
0.04–0.06, Ca (pfu) 0.06–0.10 and Mg#70−69, whereas in magma ascent. The evidence of cooling of xenolith assem-
lIV 0.01–0.04, Ca (pfu) 0.04–0.06 and
xenoliths they span A blages to temperatures below lavas may also have enhanced
Mg#68−56. Lower Ca content of xenolith orthopyroxene is amphibole stability. Where present, amphibole is a late
consistent with lower temperatures. phase in xenoliths, suggestive of reaction of hydrous evolved
13
Table 5 Estimates of temperature, pressure and oxygen fugacity for the plutonic xenoliths
Sample Group Method Phases T (°C) P (MPa) log fO2
relative to ΔNNO
BQA3 6 Ballhaus et al. (1991) ol–spl 1008–1055 Assumed 500 MPa 1.31 to 1.36
BQA3 6 Andersen and Lindsley (1985)a spl–ilm 812–828 – 1.15 to 1.33
BQB11 2 849 – 1.86
BQA40 5 Wells (1977) cpx–opx 939–957 – –
Putirka (2008) 920–930 800–880 –
BQ7 3 Holland and Blundy (1994)b hbl–plag 1002–1044 Assumed 500 MPa –
BQ8 4 1005–1043 –
BQ11 4 1017 –
BQA7 5 1008–1021 –
BQA10 4 623–979 –
BQA18 4 968 –
BQA22 3 957–1003 –
BQA24 3 965–1008 –
BQA40 5 938–995 –
BQB11 2 1028–1046 –
BQA3 6 Ziberna et al. (2017)c cpx–ol–plag 926–945 300, 303, 370 –
BQA35 1 982–1051 221, 315, 347 –
399, 409
BQB9 1 995 437 –
BQ1 5 936–938 322, 416 –
BQ1 5 spl–cpx–ol–plag – 224, 328 –
a
Lindsley and Spencer (1982) oxide formula recalculation
b
[edenite + albite = richterite + anorthite]
c
Multiple reaction barometer
melts with anhydrous crystals, possibly via a percolative

infiltration process (e.g. Smith 2014). Such a process would
not have occurred in lavas.
Spinel
Lava and xenolith spinels exhibit chromite and titaniferous

magnetite compositions, the latter being more abundant.
Spinel is modally more significant in xenoliths (Fig. 2) pos-
sibly a result of density-related concentration (Arculus and
Wills 1980). The variable proportions of spinel in xenoliths
are reflected in the wide spread in bulk FeO (Fig. 9f) and
TiO2 contents (Figure A3, Online Resource 4). Cr-rich spi-
nel occurs in both lavas and xenoliths with maximum Cr#77
and Cr#54 respectively (Figure A2, Online Resource 3).
Discussion
Geothermobarometry of plutonic xenoliths
The storage conditions (P–T–H2O–ƒO2) under which the

xenoliths crystallised prior to entrainment in erupted lavas Fig. 10 Temperature estimates of xenoliths. Temperature ranges for
each geothermometer utilized are distinguished by xenolith group
13
were calculated to establish the configuration of the sub- crystallization. If this sample is omitted, there is an appar-
volcanic system beneath Bequia. Mineral assemblages for ent progressive reduction in plagioclase–amphibole tem-
providing reliable estimates are severely limited in lavas. peratures from Group 2 to 5 (Fig. 10). Similarly, using
Phenocrysts lack melt inclusions (MI) and significant pro- the COlP thermometer, Group 1 has higher crystallisation
portions of Cr-rich spinel, ilmenite, orthopyroxene and horn- temperatures than Groups 5 and 6. Spinel–ilmenite phases
blende to allow for the use of many thermobarometers. The exhibit low temperatures (812–849 °C) due to their ready
iddingsitized nature of olivine is also a major hindrance. For re-equilibration with the melt.
these reasons we present estimates of the intensive param- MI were trapped over a pressure range of 160–570 MPa
eters for xenolith crystallisation only, for which a wide range (Fig. 11; Table 3) equivalent to 6–21 km vertical depth within
of methods is available. the crust (assuming a density of 2750 kg/m3). Because the
A combination of geothermometers was applied to xeno- presence of vapour bubbles in MI precludes the unequivo-
liths (Table 5): olivine–spinel (± 92 °Ballhaus et al. 1991), cal establishment of original CO2 contents, these are mini-
magnetite–ilmenite (± 36 °C; Andersen and Lindsley 1985), mum magma crystallisation pressures. However, pressure
hornblende–plagioclase (± 40 °C, Holland and Blundy calculations using Ziberna, et al. (2017) are in the range
1994), clinopyroxene–orthopyroxene (± 40 °C, Putirka 2008; 220–440 MPa (± 50 MPa), in agreement with MI pressures
± 70 °C; Wells 1977) and clinopyroxene–olivine–plagio- and supportive of volatile saturation with limited C
O2 seques-
clase (COlP) (± 20 °C, Ziberna et al. 2017). Post entrapment tration into bubbles. We believe that values obtained using
crystallisation (PEC) effects preclude the use of mineral- the clinopyroxene-orthopyroxene barometer of Putirka (2008)
melt thermometers. Major element data for touching rims overestimates crystallisation pressure largely because its large
and/or included phases and their adjacent hosts were used in error (± 280 MPa) embraces the pressure range being inves-
all thermometer calculations to ensure textural equilibrium. tigated (Ziberna et al. 2017). The preservation of MI with
Blundy and Cashman (2008) demonstrated that pressure has elevated water contents (4.6–7.8 wt%) suggests that xenoliths
a modest dependence on thermometry (± 72 °C GPa−1) at were entrained and erupted rather quickly. Oxygen fugacity
sub-volcanic pressures. Therefore, where a pressure estimate estimates (Andersen and Lindsley 1985; Ballhaus et al. 1991)
was necessary for thermometry, a nominal value of 500 MPa range from 1.2 to 1.9 log units above the NNO buffer.
was applied. The olivine–spinel Fe–Mg exchange thermom-
eter is only applicable at high Cr#. Due to the infrequency Constraints from phase petrology
of Cr-spinel, appropriate co-existing phases were available
in only one sample (BQA3). Temperature estimates for co- Additional constraints on P–T–H2O conditions of xeno-
existing oxide pairs were made using the ILMAT program lith crystallization can be obtained from equilibrium
of LePage (2003) with Lindsley and Spencer (1982) method
for recalculation. The hornblende–plagioclase thermometer
is only valid if plagioclase is less calcic than A
n90 (Holland
and Blundy 1994). Five samples (BQ8, BQA6, BQA18,
BQA40, BQB11) contained plagioclase with higher An, so
their temperatures are underestimates.
Magma storage pressures were estimated by applying the
solubility model of Ghiorso and Gualda (2015) to volatile
saturation contents ( H2O and C O2) from xenolith MI at an
assumed temperature of 950 °C. If melts are undersaturated,
or have lost CO2 to shrinkage bubbles, then these estimates are
minima. Pressures were also calculated by the multiple reac-
tion approach of Ziberna, et al. (2017) using COlP and spi-
nel–clinopyroxene–olivine–plagioclase (SCOlP) assemblages.
Two oxybarometers were evaluated in relation to the NNO
buffer (Andersen and Lindsley 1985; Ballhaus et al. 1991).
Xenolith equilibration temperatures from all available
methods are in the range 623–1055 °C, with the majority
in the range 950–1050 °C (Fig. 10). Plagioclase–amphi- Fig. 11 Volatile concentrations of CO2 and H
2O in xenolith MI. Iso-
bole temperatures show good agreement between samples bars and isopleths calculated for a representative MI using the Mag-
maSat solubility model (Ghiorso and Gualda 2015) at an assumed
(938–1044 °C) with the exception of BQA10 (Table 5). The
temperature of 950 °C. MI host minerals denoted as follows: dia-
broad temperature range (spanning 356 °C) for this more mond, olivine; square, clinopyroxene; triangle, amphibole. Xenolith
evolved sample is interpreted to result from protracted groups are distinguished by colour as shown in legend
13
mineral assemblages and crystallisation sequences. This that is stable only at elevated pressures, and a single low-
approach requires some insight into the nature of the paren- H2O MI (omitted from Fig. 12), the combination of experi-
tal magma(s) responsible for the observed xenoliths. The ments and xenoliths indicates crystallisation at pressures
olivine Fo content (55–71) is inconsistent with a mantle- of 180–570 MPa with H 2Omelt of 4–8 wt%. Assemblages
derived parent, requiring instead some prior olivine ± cpx containing hornblende correspond to higher H2Omelt con-
fractionation, whereas the elevated An content of plagioclase sistent with the higher volatile contents of MI from horn-
is suggestive of a hydrous, high-alumina basalt magma (Pan- blende-bearing samples (Table 3). Orthopyroxene-bearing
jasawatwong et al. 1995; Sisson and Grove 1993). Inspection assemblages are only stable at pressures of ~ 400 MPa.
of available experimental data on starting materials similar Note that the observed mineral assemblages do not require
to Bequia lavas reveals the closest match to mineral assem- H 2O-saturated melts, consistent with the presence of
blages and compositions is provided by three experimental
studies using Lesser Antilles starting materials: high-alu-
mina basalt GW40 from Guadeloupe (Pichavant et al. 2018),
basaltic andesite 031–22b from Martinique (Pichavant et al.
2002a) and low-magnesium basalt STV315 from St. Vin-
cent (Pichavant and Macdonald 2007). In Fig. 9, these three
starting compositions are compared to Bequia whole rock
lava compositions, showing that they represent reasonable
candidates for parental magmas.
The selected experiments were performed in the pres-
ence of a mixed H 2O–CO2 fluid over a pressure range of
200–1000 MPa and fO2 around NNO + 2 log units. Experi-
mental temperatures (950–1050 °C) match closely those
obtained by geothermometry (Table 5). Experimental melts
have H2Omelt contents of 4 to 10 wt%, spanning the range
measured in MI (Table 3). Experimental olivine compo-
sitions are Fo62–85, bracketing the range observed in the
xenoliths (Fig. 2). Experimental plagioclase composition is
An68–89 covering much of the range observed in the xenoliths
with the exception of the very calcic plagioclase ( An> 90).
The conditions required for crystallisation of such calcic
plagioclase were evidently not attained in the experiments,
either because the starting materials had insufficient Ca/Na
ratio (Hamada and Fujii 2007; Panjasawatwong et al. 1995)
or additional phases, such as Cl-rich aqueous fluids played Fig. 12 Crystallisation conditions of Bequia xenoliths constrained
by MI and experimental petrology in terms of wt% H2O in melt and
a role in removing Na from the system (Johannes and Holtz pressure at a temperature of 1000 ± 50 °C. Experimental run products
1991; Orville 1972). for three Lesser Antilles starting materials that show the same min-
Each Bequia xenolith assemblage can be matched to spe- eral assemblages as Bequia xenoliths are plotted along with MI from
cific experimental conditions (Table A2, Online Resource 5) Table 2 (a single low H 2O MI is omitted for clarity). Symbols are col-
our-coded according to the relevant experimental mineral assemblage
indicating that the xenoliths are consistent with crystallisa- or that of the xenolith from which the MI derive. For experiments,
tion of hydrous low-magnesium and high-alumina basalts, the symbols correspond to the three different starting compositions
and basaltic andesite. The restricted experimental conditions (Fig. 9): low-magnesium basalt (LMB) STV315 (Pichavant and
that match each assemblage (Table A2, Online Resource Macdonald 2007); high-alumina basalt (HAB) 031-22b (Pichavant
et al. 2002a); basaltic andesite (BA) GW40 (Pichavant et al. 2018).
5), suggests that there is a limited range of crustal condi- H2O–CO2 fluids and oxide minerals are present across the diagram,
tions under which the xenolith assemblages can crystallise although the stability of individual oxides (magnetite, Cr-spinel,
from reasonable parent magmas. In Fig. 12 we show the ilmenite) is sensitive to fO2, which varies from NNO to NNO + 4 log
experimental phase assemblages on a pressure–H2Omelt dia- units for the plotted experiments. The H2O solubility curve is calcu-
lated using MagmaSat (Ghiorso and Gualda 2015) for a representa-
gram constructed at 1000 ± 50 °C together with MI (Table 3) tive Bequia magma composition. The pressure range estimated from
colour-coded according to the host xenolith mineral assem- multiple reaction barometry (Table 5) is shown as a yellow bar. The
blage. The agreement between experiments, xenolith assem- two blue curves show degassing trends for the same representative
blages, multiple-reaction barometry (Table 5) and MI is Bequia magma with initial volatile contents of: (1) 7.5 wt% H2O and
1 wt% CO2; and (2) 10 wt% H2O and 1 wt% C O2, calculated using
striking. With the exception of the Group 6c xenoliths with MagmaSat. The depth range on the upper horizontal axis was calcu-
the plagioclase-free assemblage clinopyroxene + amphibole, lated for a density of 2750 kg m−3
13
significant CO2 in Bequia MI. We suggest instead that the in the strict sense that they did not form by protracted, in situ
xenoliths were formed by ascending basaltic magmas satu- crystallization.
rated in a mixed H2O–CO2 fluid and crystallising at depths The melt compositions in equilibrium with calcic amphi-
of 8–17 km. Degassing during ascent will progressively bole can be quantified using the chemometry method of
deplete the melts in CO2 relative to H2O as observed in the Zhang, et al. (2017). Figure 9 shows that a wide diversity
MI (Fig. 11). Thus, the spread of H2Omelt in Fig. 12 may of melt compositions (60–75 wt% S iO2) would have been
reflect different degassing paths of a single basaltic parent, present in amphibole-bearing xenoliths from Bequia. The
rather than batches with variable initial H 2O contents. Two most evolved xenolith, BQA10, yields as expected, the
representative degassing trends are illustrated with blue highest melt SiO2. Amphiboles in Group 2 give calculated
curves (labelled 1 and 2) on Fig. 12. Note that both start at melts with 61–65 wt% S iO2 and 1.8–2.5 wt% MgO; Group
elevated CO2 contents (1 wt%), to maintain fluid saturation 3 amphiboles give 61–72 wt% S iO2 and 0.9–3.5 wt% MgO;
throughout ascent. The residual melts produced during poly- Group 4 amphiboles give calculated melts spanning the
baric xenolith crystallisation is consistent with the observed entire SiO2 range for all groups and 0.4–3.8 wt% MgO;
Bequia liquid line of descent (Fig. 9). The rare clinopyrox- Group 5 amphiboles give calculated melts with 67–73 wt%
ene + amphibole assemblage (BQA20) may testify to even SiO2 and 0.7–1.8 wt% MgO; and Group 6 amphiboles give
deeper (1000 MPa ≈ 34 km) crystallisation, plausibly close calculated melts with 63–64 wt% SiO2 and 2.7–3.0 wt%
to the Moho. The texture of this xenolith (Fig. 4f) is sugges- MgO. The calculated melt compositions overlap with xeno-
tive of entrainment of a fragment deeper (plagioclase-free) lith MI, but are not represented by any erupted magmas on
xenolith material into melts that subsequently crystallise to Bequia (Fig. 9). The evolved nature of the trapped and cal-
a shallower (gabbroic) assemblage. culated melts suggests that they are residual interstitial melts
Xenolith crystallisation sequences have been shown to left over from the precipitation of hornblende.
be quite variable (Table 1), notably the early or late appear- One may question why amphibole is so glaringly absent
ance hornblende, olivine and plagioclase, sometimes in the from the lavas, yet so prevalent in deeper crustal xenoliths.
same sample (Figs. 3, 4). This variability can be reconciled Barclay and Carmichael (2004) suggest that the onset of
with the phase petrology constraints (Fig. 12) by consid- rapid hornblende crystallisation in hydrous basaltic mag-
ering different magma trajectories through the mid-crustal mas will impede magma ascent by its associated increase in
mush system. The late appearance of olivine and plagioclase crystallinity. The eruptibility of such magmas is, therefore,
can be produced in evolved basaltic magmas that ascend reduced and may explain the paucity of amphibole-bearing
from sufficient depths that the first assemblage to crystal- erupted magmas at the surface. The tendency for amphibole
lise is olivine- and plagioclase-free. In magmas that start to form reaction rims upon decompression may also increase
to crystallise at lower pressures, olivine and/or plagioclase its bulk density and impede its continued incorporation into
will be the first phase to crystallise. The later appearance ascending magmas at lower depths (Davidson et al. 2007).
of amphibole in many samples is consistent with isobaric It is, therefore, unsurprising that lavas may show geochemi-
addition of H 2O-rich fluids, moving the assemblage into the cal evidence for amphibole fractionation, if they represent
amphibole field in Fig. 12. The ability of different ascent and melts extracted from amphibole-bearing mushes at depths
degassing paths through the crust to produce different crys- of > 8 km. We see no evidence for entrainment of individ-
tallisation sequences, support the idea of extensive vertical ual amphibole crystals and their subsequent breakdown at
and/or horizontal reach of volatile-rich melts through mush lower pressures (Rutherford and Devine 1988; Rutherford
systems beneath Bequia. and Hill 1993). The only evidence for amphibole-rich mush
comes from the xenoliths that are presumably entrained and
The role of amphibole discharged sufficiently rapidly by explosive eruptions that
amphibole is preserved. The preservation of high H 2O in MI
The lack of amphibole from lavas, but its abundance in xeno- also testifies to rapid xenolith ascent and eruption.
liths, has led to amphibole being termed a ‘cryptic’ fraction-
ating phase (Davidson et al. 2007). The textural and compo- Xenocrysts, antecrysts or phenocrysts?
sitional variability of xenolith amphibole suggests that open
system reactive flow is an important process facilitating its Compositional and textural parallels between xenolith and
formation in crustal reservoirs. The percolation of external lava mineral assemblages suggest that the entrainment of
hydrous fluids (or melts) through a H
2O-poor mush may trig- crustal mush fragments may be a significant process operat-
ger the formation of amphibole, changing the composition ing during lava extrusion on Bequia. One manifestation of
of the melt as it flows upward. As for other systems (Smith mush entrainment could be the abundance of glomerocrysts
2014), it is assumed here that the fluids reacting with exist- in lavas. Glomerocrysts may form by synneusis: the drift-
ing phases to form amphibole are not necessarily cogenetic ing together and attachment of crystals suspended in the
13
melt (Vance 1969; Vogt 1921). Alternatively, polymineralic from Grenada to St. Vincent. Grenada contains the only
glomerocrysts may have been sourced from cumulate piles extensive suite of plagioclase-free cumulates in the entire
underlying mush regions. On Bequia, mineral modes and Lesser Antilles arc (Arculus and Wills 1980), while Bequia
chemistry suggest significant differences between glom- and St. Vincent cumulates are dominated modally by pla-
erocrysts and xenoliths. Glomerocrysts have assemblages gioclase. Similarly, ultramafic cumulates such as wehrlites
dominated by olivine and clinopyroxene with no sign of and pyroxenites are present in Grenada, but only present as
reacted amphiboles or their pseudomorphs (Fig. 2c). The rare plagioclase-bearing pyroxenites in Bequia and St. Vin-
abundance of olivine and clinopyroxene and the absence of cent (Figure A4, Online Resource 7). Troctolites are absent
amphibole and orthopyroxene in glomerocrysts suggest that in Grenada, but present in Bequia and (abundantly) in St.
they crystallized close to the liquid of melts with less than Vincent. Cumulate amphibole is dominant in Grenada and
4 wt% H2O (Fig. 12). These are lower pressure conditions substantial in Bequia, but less prevalent in St. Vincent. Gre-
than those inferred for the xenoliths. We suggest that glom- nadian olivine compositions are the most magnesian with
erocrysts are accumulations of near-liquidus minerals from higher Ni contents than Bequia and St. Vincent. Bequia
these H2O-saturated magmas (Fig. 12) en route to shallow xenoliths, however, show the lowest Fo contents amongst
crustal storage regions, where phenocrysts grow. the three islands, suggesting significant depletion of MgO
prior to crystallization of the xenoliths. Cr-rich spinels are
Primitive mantle source absent from St. Vincent xenoliths, but present in Bequia and
Grenada, with the latter having a higher Cr# and greater pro-
Both lavas and xenoliths contain mineralogical evidence for portion. The iddingsitized nature of olivine in Grenada and
primitive melts influencing differentiation processes beneath Bequia versus its unaltered appearance in St. Vincent may
Bequia. Although less magnesian than Grenadian olivine be a reflection of the northward decrease in water content in
phenocrysts (Fo91–73 Stamper et al. 2014a), some olivine the mantle source from Grenada to St. Vincent, as alluded
phenocrysts in Bequia lavas exhibit primitive compositions to by Stamper, et al. (2014a).
of Fo≤ 84 which are more magnesian than xenolith olivines Clinopyroxene Ca compositions in the lavas highlight
(Fig. 6), but contain significantly less Ni. Comparatively two series distinguished by the rate of Ca depletion with
high Cr contents are found in spinel and amphibole analy- decreasing Mg#, suggesting differences in the onset of
ses from specific cumulate groups providing evidence for a clinopyroxene crystallisation (Fig. 7). This is reminiscent of
primitive mantle source. Xenolith Groups 1, 2 and 6 along the M- and C-series lavas of Grenada whose CaO versus
with some phenocrysts contain Cr-rich spinels (Cr# >40) MgO whole rock contents show two separate liquid lines of
(Figure A2, Online Resource 3). Amphibole from Groups 2 descent (Stamper et al. 2014a), attributed to variations in the
and 6 exhibits comparatively high Cr contents with corre- differentiation depth of a common picritic parent (Stamper
spondingly high Mg#. Highly magnesian cores (Mg#100−90) et al. 2014b). A geochemical boundary between St. Vincent
are also observed in augite phenocrysts, glomerocrysts and and the Grenadines has been suggested by Macdonald, et al.
Group 1 xenoliths (Fig. 7). These have been proposed by (2000) on the basis of the restriction of M- and C- series
Thirlwall and Graham (1984) for Grenada C-series basalts as rocks to the southern part of this arc segment. However,
remnants of lavas generated from picritic basalt precursors an observation scarcely mentioned in the literature is the
not seen at the surface. In Bequia, these examples may be presence of M- and C- series lavas in St. Vincent, which is
relicts of primitive melts emplaced into the deeper magmatic clearly observed when CaO versus MgO whole rock data
system. Early precipitation of olivine ± clinopyroxene will from GEOROC are plotted (Fig. 13c). For the most part,
drive the residual melts to low-MgO and high-alumina basalt however, bulk rock lava and MI geochemical data show that
compositions responsible for some of the observed xenolith Bequia exhibits transitional affinities between St. Vincent
assemblages. There are no wehrlitic Bequia xenoliths that and Grenada, with a few exceptions (Fig. 13). For example,
record this early, presumably deep, crystallisation. We sug- the Al2O3 versus MgO trend on Bequia describes a more
gest that they are confined to the lower crust. pronounced peak in Al2O3, while the MgO content of Bequia
lavas are < 11 wt%, whereas St. Vincent and Grenada lavas
Southern arc segment comparisons extend beyond 15 wt% MgO. In fact, Bequia data appear to
fill the gap in MgO content that distinguishes the M- and
Differences in melt compositions and conditions of crys- C- series (Fig. 13a, c).
tallisation in the southern segment of the Lesser Antilles When cumulate textures are compared amongst the
arc are clearly demonstrated in cumulate compositions three islands, Bequia is markedly different from Grenada
and textures from constituent volcanic islands (Table A3, and St. Vincent where cumulates display chiefly adcumu-
Online Resource 6). There is an apparent northward reduc- late textures, consistent crystallisation sequences (phases
tion in compositional primitiveness (Fo, An) of xenoliths always crystallising in the same order), minimal zoning and
13
Fig. 13 Bulk major element variations in lavas and MI from Grenada, St. Vincent and Bequia. Data are obtained from GEOROC database
narrow mineral compositional ranges (Table A3, Online broader range of P–T–H2O conditions (Fig. 12). Although
Resource 6). Bequia cumulates, in contrast, show evidence all three islands contain gabbro xenoliths (Figure A4, Online
for persistent disequilibria and crystallization over a much Resource 7), on St. Vincent and Grenada these belong to just
13
two and one textural sub-divisions (according to Steckeisen

1976), respectively; Bequia contains eleven varieties of gab-
bros as well as gabbronorites (Table 1). These observations,
combined with widespread mineral zoning, substantial pro-
portions of xenoliths with orthocumulate textures, variable
crystallisation sequences, pervasive sieve textured plagio-
clase and broad compositional phase ranges, emphasize dif-
ferent conditions in the mush pile beneath the three islands.
Textural similarities between Grenada and St. Vincent xeno-
liths may be related to the stability of the magmatic systems
beneath these two islands facilitating the establishment of
long-lived volcanism. Conversely, the relatively small size of
Bequia may not have supported the protracted development
of a large magmatic reservoir, such that cumulates form
and melts are extracted over a wider range of conditions
between which there appear to be periods of re-equilibration
and exchange.
Bequia plumbing system
Magmatic processes in Bequia operate over a wide depth Fig. 14 Schematic diagram to illustrate the plumbing system beneath
range (6 to at least 21 km) in the crust (Fig. 14) in what Bequia based on the agreement between phase petrology, thermo-
has been termed a transcrustal magmatic system (TCMS) barometry and melt inclusions. Labelled fields denote P regions from
by Cashman, et al. (2017). Xenolith crystallisation pres- which xenoliths (mush fragments) are sourced and low-magnesium
basalts (LMB), high-alumina basalts (HAB) and basaltic andesites
sures indicate that a vertically-extensive crystal mush (BA) stall and differentiate. These basaltic compositions are the
(crystals + melt) enables polybaric differentiation to occur products of prior differentiation of mantle-derived high-magnesium
in a hot zone extending from near the Moho to the upper basalts (HMB). Colours indicate the SiO2 content of the melts.
crust (Melekhova et al. 2015). This hot zone also contains Arrows show the ascent and differentiation of melts from different
regions within the crust
elevated H2O contents in trapped melts, lending support to
the idea of a hydrous mush dominated by amphibole (David-
son et al. 2007). Several lines of evidence suggest that vol- may have been H 2O-poor, lending support to the idea of
atile-charged low-magnesium and high-alumina basalt and substantial horizontal variability within the TCMS. Such
basaltic andesite magmas stall at different depths through lateral variability is consistent with along-arc geophysi-
the TCMS (Fig. 14) resulting in reaction to produce new cal studies of the Antilles that show considerable vari-
phases: e.g. early and late crystallising hornblende, variable ability in P-wave structure on wavelengths of several tens
crystallisation sequences, zoning, layered lithologies, broad of kilometres (Boynton et al. 1979). It is no surprise that
mineral compositional ranges and extensive crystallisation a layered cumulate group was identified with texturally
temperature ranges. The source of such melts may be from different lithologies juxtaposed next to each other at the
different parent magmas and/or the same melt at different microscopic scale, possibly due to entrainment of earlier
stages of evolution, percolating the mush and effectively assemblages into later magmas. Similar textures have been
modulating the conditions of the system. Similar open sys- observed elsewhere in plutonic xenoliths (Tollan et al.
tem processes were noted from plutonic xenoliths in Marti- 2011).
nique (Cooper et al. 2016). These basaltic compositions are Although Bequia xenoliths have assemblages and crystal-
products of prior crystallisation of mantle-derived magmas. lization sequences consistent with differentiation of hydrous
The olivine ± clinopyroxene cumulates formed during this mafic magmas over a broad depth range, melt compositions
differentiation step are not preserved on Bequia, yet plausi- in these samples are significantly more evolved than erupted
bly constitute much of its lower crust. lavas. The failure of such evolved melts to segregate and
A wide, mid-crustal depth range of crystallisation is erupt on Bequia suggests that the mush never attained suf-
supported by the calculated pressures from melt inclu- ficient size or that melt extraction was inefficient, in com-
sions, thermobarometry and phase petrology (Figs. 11, parison to islands north of St. Vincent, where evolved melts
12 and Table A2, Online Resource 5). The presence of are commonplace (Macdonald et al. 2000). These obser-
an amphibole-free group implies that some mush regions vations support the idea of a TCMS containing vertically
13
connected lenses of fractionated melt segregated from dense Pressure-Temperature Window of Eruptibility. J Petrol 45(3):485–
mafic cumulates, which migrate upwards finally combining 506 (0.1093/petrology/egg091)
Blundy J, Cashman K (2005) Rapid decompression-driven crystalliza-
to form magma reservoirs at physically favourable depths tion recorded by melt inclusions from Mount St. Helens volcano.
(Christopher et al. 2015). Beneath Bequia, these lenses may Geology 33(10):793–796. https://doi.org/10.1130/G21668.1
never have attained significant thicknesses. Eruption, which Blundy J, Cashman K (2008) Petrologic Reconstruction of Magmatic
may be triggered by destabilization of the TCMS, allows System Variables and Processes. Rev Mineral Geochem 69:179–
239. https://doi.org/10.2138/rmg.2008.69.6
melts from different depths to mix together and incorpo- Boynton CH, Westbrook GK, Bott MHP, Long RE (1979) A seismic
rate fragments (xenoliths) of surrounding crystal-dominated refraction investigation of crustal structure beneath the Lesser
mush with different crystallisation histories. During ascent, Antilles island arc. Geophys J Roy Astron Soc 58:371–393
H2O-saturated crystallization of extracted melts leads to the Briden JC, Rex DC, Faller AM, Tomblin JF (1979) K-Ar geochronol-
ogy and palaeomagnetism of volcanic rocks in the lesser Antil-
formation of liquidus phenocryst assemblages and glomero- les Island Arc. Philos Trans R Soc Lond Ser A Math Phys Sci
crysts that differ from those of the mush from which they 291(1383):485–528. https://doi.org/10.2307/75166
were extracted, notably in the absence of amphibole and Brown GM, Holland JG, Sigurdsson H, Tomblin JF, Arculus RJ (1977)
predominance of olivine. The tendency of melts extracted Geochemistry of the lesser Antilles Volcanic Island Arc. Geochim
Cosmochim Acta 41:785–801
from mushes to produce characteristic, low-pressure phe- Cashman KV, Sparks RSJ, Blundy JD (2017) Vertically extensive and
nocryst assemblages effectively obscures conditions within unstable magmatic systems: a unified view of igneous processes.
the mid-crustal mush. The study of both extrusive and plu- Science 355:3055. https://doi.org/10.1126/0.1126/science.aag30
tonic (xenolith) samples helps to reconcile this discrepancy. 55
Christeson GL, Mann P, Escalona A, Aitken TJ (2008) Crustal struc-
ture of the Caribbean–northeastern South America arc-conti-
Acknowledgements This research was supported by a Commonwealth nent collision zone. J Geophys Res 113:B08104. https://doi.
Scholarship to MC-H, funded by the UK government and a NERC org/10.1029/2007JB005373
grant NE/N001966/1 to EM and JB. We thank Stuart Kearns for help Christopher TE, Blundy J, Cashman K, Cole P, Edmonds M, Smith
with EPMA, Richard Hinton for assistance with SIMS and Jenni Bar- PJ, Sparks RSJ, Stinton A (2015) Crustal-scale degassing due to
clay for facilitating XRF analysis. We thank Richard Arculus, Susanne magma system destabilization and magma-gas decoupling at Sou-
Skora, Charlotte Stamper and Mikel Diez for help during sample col- friere Hills Volcano, Montserrat. Geochemisty, Geophys Geosyst
lection. This work has benefitted from useful discussions with Kathy 16:2797–2811 https://doi.org/10.1002/2015GC005791
Cashman about glomerocrysts. Constructive reviews by Saskia Erd- Cooper GF, Davidson JP, Blundy JD (2016) Plutonic xenoliths from
mann and Peter Ulmer, and the editorial handling of Othmar Müntener Martinique, Lesser Antilles: evidence for open system processes
are gratefully acknowledged. and reactive melt flow in island arc crust. Contrib Miner Petrol
https://doi.org/10.1007/s00410-016-1299-8
Open Access This article is distributed under the terms of the Crea- Davidson JP, Turner S, Handley H, Macpherson C, Dosseto A (2007)
tive Commons Attribution 4.0 International License (http://creativeco Amphibole “sponge” in arc crust? Geology 35:787–790. https://
mmons.org/licenses/by/4.0/), which permits unrestricted use, distribu- doi.org/10.1130/G23637A.1
tion, and reproduction in any medium, provided you give appropriate Deer WA, Howie RA, Zussman J (2013) An Introduction to the Rock-
credit to the original author(s) and the source, provide a link to the Forming Minerals. The Mineralogical Society, London
Creative Commons license, and indicate if changes were made. Droop GTR (1987) A general equation for estimating Fe 3 + concen-
trations in ferromagnesian silicates and oxides from microprobe
analyses, using stoichiometric criteria. Mineral Mag 51:431–435
Gavrilenko M, Herzberg C, Vidito C, Carr MJ, Tenner T, Ozerov A
References (2016) A calcium-in-olivine geohygrometer and its application
to subduction zone magmatism. J Petrol 57(9):1811–1832. https
Andersen DJ, Lindsley DH (1985) New (and final!) models for the Ti- ://doi.org/10.1093/petrology/egw062
magnetite/ilmenite geothermometer and oxygen barometer. Spring Ghiorso MS, Gualda GAR (2015) An H2O–CO2 mixed fluid satura-
Meeting EoS Transactions American Geophysical Union 66(18) tion model compatible with rhyolite-MELTS. Contrib Miner
Arculus RJ (1994) Aspects of magma genesis in arcs. Lithos Petrol 169:1–30
33:189–208 Hamada M, Fujii N (2007) H 2O-rich island arc low-K tholeiite
Arculus RJ, Wills KJA (1980) The petrology of plutonic blocks magma inferred from Ca-rich plagioclase-melt inclusion equi-
and inclusions from the Lesser Antilles island arc. J Petrol libria. Geochem J 41:437–461
21(4):743–799 Hasanie RA (2000) Crystallization porcesses, geothermometry and
Arnaiz-Rodriguez MS, Orihuela N (2013) Curie point depth in Ven- geobarometry of the volcanic blocks of Bequia, Lesser Antilles.
ezuela and the Eastern Caribbean. Tectonophysics 590:38–51. University of Windsor, Windsor
https://doi.org/10.1016/j.tecto.2013.01.004 Heath E, Macdonald R, Belkin H, Hawkesworth CJ, Sigurdsson H
Attridge W (2013) The magmatic history of the Bequia Cumulates, (1998) Magmagenesis at Soufriere Volcano, St Vincent, Lesser
Lesser Antilles, West Indies. Geology MSci Thesis. Univeristy Antilles Arc. J Petrol 39(10):1721–1764
of Bristol Holland T, Blundy J (1994) Non-ideal interactions in calcic amphi-
Ballhaus C, Berry RF, Green DH (1991) High pressure experimental boles and their bearing on amphibole-plagioclase thermometry.
calibration of the olivine-orthopyroxene-spinel oxygen geoba- Contrib Miner Petrol 116(4):433–447
rometer: implications for the oxidation state of the upper mantle. Howe TM, Lindsay JM, Shane P (2015) Evolution of young
Contrib Miner Petrol 107:27–40 andesitic–dacitic magmatic systems beneath Dominica,
Barclay J, Carmichael ISE (2004) A Hornblende Basalt from West- Lesser Antilles. J Volcanol Geoth Res 297:69–88. https://doi.
ern Mexico: Water-saturated Phase Relations Constrain a org/10.1016/j.jvolgeores.2015.02.009
13
Humphreys MCS, Kearns S, Blundy JD (2006) SIMS investigation origin of magmatic high-An plagioclase. Contrib Mineral Petrol
of electron-beam damage to hydrous, rhyolitic glasses: implica- 118(4):420–432 doi. https://doi.org/10.1007/s004100050024
tions for melt inclusion analysis. Am Miner 91:667–679 Pearce JA, Peate DW (1995) Tectonic implications of the composition
Johannes W, Holtz F (1991) Melting of plagioclase in granite and of volcanic arc magmas. Annu Rev Earth Planet Sci 24:251–285
related systems—composition of coexisting phases and kinetic Pichavant M, Macdonald R (2007) Crystallization of primitive basal-
observations. In: Brown PE, Chappell BW (eds) The Second tic magmas at crustal pressures and genesis of the calc-alkaline
Hutton Symposium on the Origin of Granites and Related Rocks igneous suite: experimental evidence from St Vincent, Lesser
Kohler TP, Brey GP (1990) Calcium exchange between olivine and Antilles arc. Contrib Miner Petrol 154(5):535–558. https://doi.
clinopyroxene calibrated as a geothermobarometer for natural org/10.1007/s00410-007-0208-6
peridotites from 2 to 60 kb with applications. Geochim Cosmo- Pichavant M, Martel C, Bourdier J-L, Scaillet B (2002a) Physical
chim Acta 54:2375–2388 conditions, structure, and dynamics of a zoned magma chamber:
Kopp H, Weinzierl W, Becel A, Charvis P, Evain M, Flueh ER, Gail- Mount Pelee (Martinique, Lesser Antilles Arc). J Geophys Res
ler A, Galve A, Hirn A, Kandilarov A, Klaeschen D, Laigle M, 107(B5):2093. https://doi.org/10.1029/2001JB000315
Papenberg C, Planert L, Roux E (2011) Deep structure of the Pichavant M, Mysen BO, Macdonald R (2002b) Source and H2O con-
central Lesser Antilles Island Arc: relevance for the formation tent of high-MgO magmas in island arc settings: an experimental
of continental crust. Earth Planet Sci Lett 304(1–2):121–134. study of a primitive calc-alkaline basalt from St. Vincent, Lesser
https://doi.org/10.1016/j.epsl.2011.01.024 Antilles arc. Geochim Cosmochim Acta 66(12):2193–2209
Leake BE, Woolley AR, Birch WD, Burke EAJ, Ferraris G, Grice Pichavant M, Poussineau S, Lesne P, Solaro C, Bourdier J-L (2018)
JD, Hawthorne FC, Kisch HJ, Krivovichev VG, Schumacher JC, Experimental parametrization of magma mixing: application to
Stephenson NCN, Wittaker EJW (2003) Nomenclature of amphi- the AD 1530 Eruption of La Soufriere, Guadeloupe (Lesser Antil-
boles: additions and revisions to the international mineralogical les). J Petrol 59(2):257–282. https://doi.org/10.1093/petrology/
association’s 1997 recommendations. Can Mineral 41:1355–1362 egy030
LePage LD (2003) ILMAT: an excel worksheet for ilmenite–magnetite Putirka KD (2008) Thermometers and barometers for volcanic systems.
geothermometry and geobarometry. Comput Geosi 29:673–678 Rev Mineral Geochem 69(1):61–120. https://doi.org/10.2138/
Lewis JF (1973) Mineralogy of the ejected plutonic blocks of the Sou- rmg.2008.69.3
friere volcano St. Vincent: olivine, pyroxene, amphibole and mag- Roeder PL, Emslie RF (1970) Olivine-liquid equilibrium. Contrib
netite paragenesis. Contrib Miner Petrol 38:197–220 Miner Petrol 29:275–289
Lindsay JM, Robertson REA, Shepherd JB, Ali S (2005) Volcanic Haz- Rutherford MJ, Devine JD (1988) The May 18, 1980. eruption of
ard Atlas of the Lesser Antilles. In, vol. Seismic Research Unit, Mount St. Helens. 3. Stability and chemistry of amphibole in the
The University of the West Indies, Trinidad magma chamber. J Geophys Res 93(B10):11949–11959
Lindsley DH, Spencer KJ (1982) Fe-Ti oxide geothermometry: Reduc- Rutherford MJ, Hill PM (1993) Magma ascent rates from amphibole
ing analyses of coexisting Ti-magnetite (Mt) and ilmenite (Ilm). breakdown: An experimental study applied to the 1980–1986
EoS Transactions, American Geophysical Union 63:471 Mount St. Helens eruptions. J Geophys Res 98(B11):19667–19685
Macdonald R, Hawkesworth C, Heath E (2000) The Lesser Antilles Seaman SJ (2000) Crystal clusters, feldspar glomerocrysts, and magma
volcanic chain: a study in arc magmatism. Earth Sci Rev 49(1– envelopes in the Atascosa Lookout lava flow, Southern Arizona,
4):1–76. https://doi.org/10.1016/S0012-8252(99)00069-0 USA: Recorders of magmatic events. J Petrol 41(5):693–716
Melekhova E, Blundy J, Robertson R, Humphreys MCS (2015) Exper- Shejwalkar A, Coogan LA (2013) Experimental calibration of the roles
imental evidence for polybaric differentiation of primitive Arc of temperature and composition in the Ca-in-olivine geothermom-
Basalt beneath St. Vincent, Lesser Antilles. J Petrol 56(1):161– eter at 0.1 MPa. Lithos 177:54–60
192. https://doi.org/10.1093/petrology/egu074 Sisson TW, Grove TL (1993) Experimental investigations of the role
Melekhova E, Blundy J, Martin R, Arculus R, Pichavant M (2017) Pet- of H2O in calc-alkaline differentiation and subduction zone
rological and experimental evidence for differentiation of water- magmatism. Contrib Miner Petrol 113(2):143–166. https://doi.
rich magmas beneath St. Kitts, Lesser Antilles. Contrib Miner org/10.1007/BF00283225
Petrol 172(98) https://doi.org/10.1007/s00410-017-1416-3 Smith DJ (2014) Clinopyroxene precursors to amphibole sponge in arc
Moore LR, Gazel E, Tuohy R, Lloyd AS, Esposito R, Steele-MacInnis crust. Nat Commun 5:4329. https: //doi.org/10.1038/ncomms 5329
M, Hauri EH, Wallace PJ, Plank T, Bodnar J (2015) Bubbles mat- Smith TE, Thirlwall MF, Macpherson C (1996) Trace element and
ter: An assessment of the contribution of vapor bubbles to melt isotope geochemistry of the volcanic rocks of Bequia, Grenadine
inclusion volatile budgets. Am Miner 100:806–823 doi. https:// Islands, Lesser Antilles Arc: a study of subduction enrichment
doi.org/10.2138/am-2015-5036 and intra-crustal contamination. J Petrol 37:117–143. https://doi.
Munday D, Thirlwall M, Smith T (2000) Isotopic Variation in Cumu- org/10.1093/petrology/37.1.117
late Xenoliths from Bequia, Lesser Antilles. In: Goldschmidt Stamper CC, Blundy JD, Arculus RJ, Melekhova E (2014a) Petrology
2000, vol 5. Cambridge Publications, Oxford, p 730 of plutonic xenoliths and volcanic rocks from Grenada, Lesser
Müntener O, Kelemen PB, Grove TL (2001) The role of H2O during Antilles. J Petrol 55(7):1353–1387. https://doi.org/10.1093/petro
crystallization of primitive arc magmas under uppermost mantle logy/egu027
conditions and genesis of igneous pyroxenites: an experimental Stamper CC, Melekhova E, Blundy JD, Arculus RJ, Humphreys MCS,
study. Contrib Miner Petrol 141:643–658. https: //doi.org/10.1007/ Brooker RA (2014b) Oxidised phase relations of a primitive
s004100100266 basalt from Grenada, Lesser Antilles. Mineral Petrol https://doi.
Nielsen SG, Marschall HR (2017) Geochemical evidence for mélange org/10.1007/s00410-013-0954-6
melting in global arcs. Sci Adv 3(4):e1602402 Steckeisen A (1976) To each plutonic rock its proper name. Earth-Sci
Orville PM (1972) Plagioclase cation exchange equilibria with aqueous Rev 12:1–33
chloride solution: results at 700 °C and 2000bars in the presences Thirlwall MF, Graham AM (1984) Evolution of high-Ca, high-Sr
of quartz. Am J Sci 272:234–272 C-series basalts from Grenada, Lesser Antilles: the effects of
Panjasawatwong Y, Danyushevsky LV, Crawford AJ, Harris KL (1995) intra-crustal contamination. J Geol Soc London 141:427–445
An experimental study of the effects of melt composition on pla- Tollan PME, Bindeman I, Blundy JD (2011) Cumulate xenoliths
gioclase-melt equilibria at 5 and 10 kbar: implications for the from St. Vincent, Lesser Antilles Island Arc: a window into
13
upper crustal differentiation of mantle-derived basalts. Contrib Westercamp D, Andreieff P, Bouysse P, Mascle A, Baubron JC (1985)
Miner Petrol 163(2):189–208. https://doi.org/10.1007/s0041 Geologie de l’archipel des Grenadines (Petites Antilles meridi-
0-011-0665-9 onales). Etude monographique, vol 92. Documents du Bureau de
Toothill J, William CA, Macdonald R, Turner SP, Rogers NW, Hawkes- Recherches Geologiques et Minieres, Orleans, p 198
worth CJ, Jerram DA, Ottley CJ, Tindle AG (2007) A Complex Zhang J, Humphreys MCS, Cooper GF, Davidson JP, Macpherson CG
Petrogenesis for an Arc Magmatic Suite, St Kitts, Lesser Antilles. (2017) Magma mush chemistry at subduction zones, revealed by
J Petrol 48(1):3–42 doi. https://doi.org/10.1093/petrology/egl052 new melt major element inversion from calcic amphiboles. Am
Vance JA (1969) On synneusis. Contrib Miner Petrol 24:7–29 Mineral 102:1353–1367. https://doi.org/10.2138/am-2017-5928
Vogt JHL (1921) The physical chemistry of the crystallization and Ziberna L, Green ECR, Blundy JD (2017) Multiple-reaction geobarom-
magmatic differentiation of the igneous rocks. J Geol 29:318–350 etry for olivine-bearing igneous rocks. Am Mineral https://doi.
Wager LR, Brown GM, Wadsworth WJ (1960) Types of igneous cumu- org/10.2138/am-2017-6154
lates. J Petrol 1(1):73–85. https: //doi.org/10.1093/petrol ogy/1.1.73
Wells PRA (1977) Pyroxene thermometry in simple and complex sys-
tems. Contrib Miner Petrol 62:129–139
13

Harry Et Al 2018

Uploaded by

Copyright:

Available Formats

Harry Et Al 2018

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Harry Et Al 2018

Uploaded by

Copyright:

Available Formats

Contributions to Mineralogy and Petrology (2018) 173:77

Magma evolution beneath Bequia, Lesser Antilles, deduced

Keywords Lesser Antilles · Bequia · Grenadines · Xenoliths · Cumulates · Amphibole

Communicated by Othmar Müntener.

This group is distinguished by mesocumulate/orthcumulate

BQ4 ol gabbro Mesocumulate hbl free 11 63 18 − − 8 − − − spl, plag, ol, cpx

Crystallisation histories constructed through petrographic observation of included mineral phases

Fig. 3 Backscattered elec-

Plagioclase of the interior and exterior of grains is a common fea-

Pyroxene absent. Monomineralic (plagioclase) clusters dominate the

Xenolith mineral compositions are presented in Table A1

Fig. 7 Variation in clinopyroxene compositions for xenoliths and

are observed in Groups 1b, 2a, 6a and 6b (Figure A2, Online

Lava phenocryst chemistry

Phenocryst mineral compositions are presented in Table A1

Fig. 8 Variation in hornblende composition for plutonic xenoliths in terms of

Xenolith melt inclusions conducting a mass balance calculation to determine the

Bulk rock major element compositions of the plutonic xeno- Olivine

diffusion out of olivine during cooling (Kohler and Brey

Anorthite ranges are notably broad in both lava (phenocrysts:

erocryst cores lie in the range A

Clinopyroxene zoning occurs in lavas and xenoliths, how-

around diopside and salite. The Mg# ranges for phenocrysts,

and xenoliths shows higher Ca contents suggesting lower

Lower temperatures in xenoliths are consistent with intratel-

H2O (in wt%) and C

tents (Müntener et al. 2001). Phenocrysts from lavas and

MB mass balance calculation, – not analysed

melts with anhydrous crystals, possibly via a percolative

Lava and xenolith spinels exhibit chromite and titaniferous

Geothermobarometry of plutonic xenoliths

The storage conditions (P–T–H2O–ƒO2) under which the

two and one textural sub-divisions (according to Steckeisen

Bequia plumbing system

You might also like