Journal of Petrology | Volume 40 | Number 4 | Pages 653-678 | 1999
© Oxford University Press 1999
The Chemical and Isotopic Differentiation of an Epizonal Magma Body: Organ Needle Pluton, New Mexico
1 Cooperative Institute for Research in Environmental Sciences (Cires) and Department of Geological Sciences Campus Box 216, University of Colorado, Boulder, CO 80309, USA
2 Department of Geology, Idaho State University Pocatello, Id 83209, USA
3 Department of Geosciences, Franklin and Marshall College Lancaster, PA 17604, USA
Received March 21, 1997; Revised typescript accepted September 30, 1998
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Major and trace element, and Nd and Sr isotopic compositions of whole rocks and mineral separates from the Oligocene, alkaline Organ Needle pluton (ONP), southern New Mexico, constrain models for the differentiation of the magma body parental to this compositionally zoned and layered epizonal intrusive body. The data reveal that the pluton is rimmed by lower
Nd (
–5) and higher 87Sr/86Sr (0.7085) syenitic rocks than those in its interior (Nd –2, 87Sr/86Sr 0.7060) and that the bulk compositions of the marginal rocks become more felsic with decreasing structural depth. At the deepest exposed levels of the pluton, the Nd –5 lithology is a compositionally heterogeneous inequigranular syenite. Modal, compositional and isotopic data from separates of rare earth element (REE)-bearing major and accessory mineral phases (hornblende, titanite, apatite, zircon) demonstrate that this decoupling of trace and major elements in the inequigranular syenite results from accumulation of light REE (LREE)-bearing minerals that were evidently separated from silicic magmas as the latter rose along the sides of the magma chamber. Chemical and isotopic data for microgranular mafic enclaves, as well as for restite xenoliths of Precambrian granite wall rock, indicate that the isotopic distinction between the marginal and interior facies of the ONP probably reflects assimilation of the wall rock by Nd –2 mafic magmas near the base of the magma system. Fractional crystallization and crystal–liquid separation of the crustally contaminated magma at the base and along the margins of the chamber generated the highly silicic magmas that ultimately pooled at the chamber top. KEY WORDS: assimilation; enclave; epizonal pluton; isotope geochemistry; trace element geochemistry
|
Introduction |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
A fundamental problem of igneous petrogenesis is the origin of the high-silica magmas that are tapped by large-volume ash-flow tuffs (Hildreth, 1981
). Many workers agree that high-silica magmas are products of fractional crystallization and crystal–liquid separation along the side walls of large epizonal magma bodies, which together create residual, differentiated magmas that buoyantly rise along the side walls and pool at the chamber roof (e.g. Sparks et al., 1984; McBirney et al., 1985; Feeley & Davidson, 1994; Spera et al., 1995). However, except for the studies of Sawka et al. (1990) and Mahood & Cornejo (1992), little direct evidence of these processes has been described from ash-flow tuffs or exposed epizonal plutons. The paucity of evidence from ash-flow tuffs reflects their selective and possibly spatially biased sampling of the entire parental magma body (Wilson & Hildreth, 1997). Also, although epizonal plutons may expose the crystallized remnants of an original magma body, these commonly show evidence of subsolidus chemical re-equilibration (Hildreth, 1981).
A rare example of an epizonal intrusive rock that has preserved evidence of the processes involved in the differentiation of shallow-level silicic magmas is the reversely zoned Organ Needle pluton (ONP), one phase of the Oligocene, alkalic Organ Mountain batholith of southern New Mexico (Fig. 1; Seager, 1981). The ONP displays chemical variations comparable with those inferred for epizonal magma bodies from studies of silicic composition ash-flow tuffs (Seager & McCurry, 1988). Verplanck et al. (1995) showed that it is isotopically heterogeneous, with a highly silicic cap of different Sr and Nd isotopic compositions from the underlying syenite but of identical isotope characteristics with compositionally heterogeneous syenite exposed along the pluton's margins at its deepest exposed level. This apparent link between silicic rocks at the top of the pluton and more mafic lithologies at depth led to the suggestion that the silicic cap was produced by side-wall migration of differentiated magma generated at depth in the magma system (Verplanck et al., 1995). However, we did not fully test this hypothesis.
|
Here, we present new major and trace element and Nd and Sr isotopic data from both whole rocks and mineral separates from various facies of the ONP to further assess the origin and evolution of the highly silicic rocks. We conclude that the high-silica magmas in this system were, in fact, generated by fractional crystallization and separation of both major and accessory minerals from the differentiated liquids along the margins of an epizonal magma body. The differentiated magmas apparently originated from a mafic parental magma at or near the base of the chamber which (1) assimilated Precambrian granite wall rock, (2) differentiated to more silicic compositions, and (3) migrated along the side walls of the chamber and pooled at its roof.
|
Geological Setting |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
The ONP is the largest and oldest of three epizonal alkalic plutons of the Oligocene Organ Mountain batholith. The batholith is exposed over an area of
300 km2 in south–central New Mexico (Fig. 1). It intrudes 1.7–1.4 Ga Precambrian granite basement (Condie & Budding, 1979), as well as Paleozoic carbonate and siliciclastic sedimentary rocks (Seager, 1981). Faulting tilted the batholith 20° to the west during mid- to late-Tertiary extension and exposed 6 km of its structural depth (Seager, 1981).
The general petrographic and chemical characteristics of the ONP have been described elsewhere and so are only briefly reviewed here (see Seager & McCurry, 1988; Verplanck et al., 1995). Most of the pluton is an equigranular syenite, vertically zoned from syenite to quartz syenite (57–66% SiO2; Seager & McCurry, 1988). The innermost portions of the pluton, however, consist of monzodiorite that probably was a layer of relatively mafic magma drawn upwards into the syenitic magma body during venting of the chamber (Beyer, 1986). The highest structural levels of the pluton are cupolas of porphyritic alkali feldspar granite (SiO2 74–77%; Seager & McCurry, 1988). The cupolas are separated from underlying quartz syenite by a 1–100 m thick transition zone, the existence of which suggests that a steep compositional gradient existed in the parental magma body. The preservation of the transition zone indicates that the pluton crystallized rapidly, probably when the chamber erupted the Squaw Mountain tuff (Seager & McCurry, 1988). The latter is one of several small-volume ash-flow tuffs (100 km3) that form the volcanic cover of the ONP and is the extrusive equivalent of the pluton's alkali feldspar granite cupolas (Seager & McCurry, 1988).
The alkali feldspar granite has lower Nd and higher 87Sr/86Sr ratios than the equigranular syenite (–4.9 to –5.2 and 0.7099–0.7153 vs –2.3 to –2.8 and 0.7054–0.7056; Verplanck et al., 1995). The isotopic compositions of the granite are, however, identical to those of an inequigranular lithology of the ONP identified at the southeastern, and structurally deepest, margin of the pluton (Verplanck et al., 1995). The latter (exposed over an area of km2; Fig. 1) ranges from inequigranular syenite to quartz syenite (57–66% SiO2; Verplanck et al., 1995). Unlike the interior syenitic portion of the ONP, major and trace element abundances in the inequigranular rocks are uncorrelated.
|
Samples |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
There were three main objectives of our study, the first of which was to further assess whether differentiated magmas were generated at depth and then migrated along the magma chamber margins. To address this issue we collected samples of the ONP from closely spaced (20–50 m) intervals along transects at the pluton margin. The three transect localities are (1) Ash Canyon (samples 1391, 1491, and 1591), (2) North Texas Canyon (1891, 2791, and 2891), and (3) Blair Canyon along the pluton's western margin (491, 591, and 691; Fig. 1). In addition, a sample (4292) was obtained of a near-vertical quartz syenite dike that cross-cuts the eastern margin of the pluton and contains abundant xenoliths of Precambrian granite basement.
Second, we tested the assertion of Verplanck et al. (1995) that the decoupling of major and trace element abundances in the inequigranular syenite (IEQ) was controlled by the modal abundances of cumulate major and accessory minerals. For this purpose we obtained trace element abundance data (by instrumental neutron activation analysis) from each major lithology of the pluton. The 46 samples analyzed included many of those studied by Butcher (1990) and Verplanck et al. (1995), and 12 new syenitic samples and two new samples of the monzodiorite (Tables 1 and 2; Fig. 1). From a subset of these samples we separated major and accessory minerals for detailed chemical and isotopic studies. Apatite and, where present, hornblende, zircon, and titanite were obtained from monzodiorite (sample 592) and from 6–9, 6591, and 1692 of the equigranular syenite and 5492, 5292, and 5092 of the inequigranular syenite.
|
|
Our final objective was to investigate the proposed involvement of mafic parental magmas and crustal assimilation in the development of the ONP (Verplanck et al., 1995
). For this objective, we sampled equigranular to porphyritic Precambrian granite wall rock from positions directly adjacent to the pluton margin (samples 5591, 5792, 5491, 2991, and 3891) and 3 km east of the pluton margin (4291). In addition, three mafic enclaves (samples 492I, 6191 and 5592I) and two xenoliths (6091 and 5991) were obtained from the pluton. Ovoid mafic inclusions [or mafic microgranular enclaves (MME), using the terminology of Didier & Barbarin (1991)], ranging from 2 to 25 cm in diameter occur throughout the interior portions of the ONP at low (<1 vol. %) abundances. Sample 492I is from a typical MME that had a sharp contact with host syenite and no obvious change in grain size across the enclave (see Seager & McCurry, 1988). Pillow-like mafic enclaves up to 1 m in diameter are also present within the inequigranular syenite, and samples 6191 and 5592I represent these bodies. Angular xenoliths up to 0.5 m in diameter that are present in the IEQ facies of the pluton include dark and fine-grained (sample 6091) and light and coarse-grained (sample 5991) varieties. |
Analytical Methods |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Modes of major rock-forming minerals in selected whole-rock samples (Table 1) were determined using stained slabs with a surface area of 200–700 cm2 and grid spacing of 0.5–1.0 cm, depending on grain size. Rocks were pulverized in a ceramic container at the University of Colorado. Major and selected trace element data from these powders (Table 1) were obtained using a Philips 2400 X-ray fluorescence vacuum spectrometer at Franklin and Marshall College following procedure outlined by Boyd & Mertzman (1987)
. Trace element data for whole-rock powders (Table 2) were obtained by instrumental neutron activation at the Radiation Center, Oregon State University, following the methods ofLaul (1979).
Feldspar compositions of wall-rock and xenolith samples (Table 3), and chevkinite compositions from sample 5292 from the IEQ (Table 4), were determined by electron microprobe analyses using a Cameca SX-50 microprobe at the Department of Geology and Geophysics, University of Utah, Salt Lake City. Analytical techniques are described in Table 3, and have been described by Nash (1992).
|
|
Weight fractions for hornblende, titanite, and chevkinite were determined by countin between 1500 and 2500 points and converting the volume percent to weight percent using literature values of mineral densities (Table 5). Apatite and zircon weight percents were estimated by assigning all P and Zr in each whole rock to apatite and zircon, respectively. Chevkinite rare earth element (REE) abundances (Table 4) were obtained by electron microprobe. Concentrations of Rb, Sr, Nd and Sm in hornblende titanite, apatite, and zircon mineral separates were determined by isotope dilution (Table 5).
|
The mineral separates were obtained by crushing whole-rock samples and using a Wilfley table to concentrate heavy minerals. This concentrate was sieved to remove the >60 mesh fraction and separated using heavy liquids and a magnetic separator. The separates were hand-picked, washed with reagent grade acetone, and rinsed with distilled water. Apatite grains were dissolved in 4 N HCl. Hornblende grains were dissolved in concentrated HF and 6 N HNO3 in screw-cap TeflonTM containers on a hot plate for 3–5–days; titanite and zircon separates were dissolved in the same acids but in TeflonTM capsules placed in stainless steel ParrTM bombs in a 200°C oven. Whole-rock samples were dissolved in open containers in HF and HClO4. Any remaining residue was separated by centrifuging, dissolved by methods described for mineral separates, and returned to the mother solution.
The Sr and Nd isotopic compositions of mineral separates and whole rocks (Tables 5 and 6 were obtained at the University of Colorado, Boulder, using a Finnigan-MAT 261, six-collector solid source mass spectrometer (see Farmer et al., 1991). The Nd and 87Sr/86Sr values reported below are calculated at 36 Ma.
|
|
Results |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Whole-rock samples
Rocks from the ONP margin
Our samples from traverses at Ash, Texas, and Blair Canyons show that the ONP is evidently rimmed by syenitic rocks of
Nd –5 at all structural levels (Fig. 2). Lower Nd rocks are spatially restricted to the outermost 20–100–m of the pluton (Fig. 2a). This veneer of Nd –5 is substantially narrower than the apparent thickness of the IEQ. The rocks with Nd –5 increase in SiO2 with decreasing structural depth (Fig. 2b). Other major elements covary with SiO2 as they do in the ONP (Table 1). At the deepest structural level the rocks resemble adjacent equigranular syenite in their bulk compositions (Table 1). However, at the shallower structural level represented by the Texas Canyon samples, the Nd –5 rocks display higher SiO2 contents than neighboring syenite of the pluton's interior (Fig. 2b). Rocks from the ONP margins lack obvious systematic variations of their trace element contents with structural depth (Fig. 2c), except for an increase in the magnitude of the Eu anomaly at progressively shallower depths (Fig. 3g–i). Although the 87Sr/86Sr ratios (0.7075–0.7191) of the rocks from the ONP's margins are consistently more radiogenic than those of the pluton's interior, Sr isotopic compositions for rocks of Nd –5 do not vary regularly with depth (Fig. 4). Samples with the most radiogenic Sr occur at Texas Canyon, where the whole-rock Sr contents of these rocks (75–110 ppm) are also lower than those of Nd 5 rocks at either Ash Canyon (400–908 ppm) or Blair Canyon (199–315 ppm).
|
, 
, equigranular syenite samples. (b) Weight percent SiO2 vs approximate structural height for samples shown in (a). (c) Nd abundances (ppm) vs approximate structural height for samples shown in (a).
|
|
The quartz syenite dike displays major (SiO2 68.3 wt %), trace element, and Sr and Nd isotopic compositions that overlap those of the IEQ (87Sr/86Sr 0.7108,
Nd = –4.7; Tables 1 and 2; Fig. 4).
Monzodiorite of the ONP
The monzodiorite consists of plagioclase, pyroxene, amphibole and biotite, with minor potassium feldspar and quartz, and accessory apatite, Fe–Ti oxide minerals and zircon, as expected from previous observations of this lithology (Beyer 1986; Fig. 5). Chemically, the monzodiorite extends the major element trends defined by the remainder of the pluton towards lower SiO2 (56.6–59.9%) and K2O (3.3–4.1%), and higher Al2O3 (16.4–16.6%, Table 1; see also Beyer, 1986) and has the highest Sc, Sr and light REE (LREE), Th, and U abundances of any of the major facies of the pluton (Table 2). Accordingly, the lowest SiO2 sample of this rock type (1092) is used for normalization values for other ONP samples (Fig. 3). Monzodiorite Nd values range from –3.5 to –3.6 and 87Sr/86Sr from 0.7056 to 0.7059 (Fig. 4; see alsoButcher, 1990).
|
Equigranular syenite, inequigranular syenite (IEQ), and alkali feldspar granite (AFG) of the ONP
Major element compositions, modes and Sr and Nd isotopic compositions for the new samples of both the equigranular syenite and inequigranular syenites are consistent with the values previously established for these lithologies by Verplanck et al. (1995)
and are not tabulated here. These data are available from the authors upon request. Trace element data (Table 2) from the equigranular syenite reveal that REE, Rb, and Ba abundances increase with SiO2, up to about 70 wt % of the latter, but decrease in rocks with 70–73 wt % SiO2 (Fig. 3a). In contrast, Sr contents consistently decrease, and Eu/Eu* increases, with SiO2 throughout the entire range of syenite bulk compositions (Fig. 6a). Syenite and quartz syenite samples all display lower (La/Yb)cn (10–12) than observed for the monzodiorites (13–14; Fig. 6b). However, one quartz syenite (sample 3892) from the southern lobe not only has unusually high 87Sr/86Sr (0.7095) and low Nd (–4.7; Fig. 4), but also displays much higher REE contents than other syenite samples (Fig. 3c). In addition, syenite samples 4392, 6591, 3592, and 1391, all collected from eastern exposures of the equigranular rocks (Fig. 1), contain up to 2800 ppm Ba and display positive Eu anomalies (Fig. 6a) and (La/Yb)cn of 16–19 (Fig. 6b) despite having 87Sr/86Sr and Nd values (0.7065 and 2.5 to –3.2, respectively) like those of other equigranular syenites (Table 6).
|
Trace element data from the IEQ clearly illustrate that REE, Ba and Sr contents do not correlate with major elements. The trace elements generally are enriched compared with syenite samples from the interior of the ONP (Fig. 3e). In addition, Eu/Eu* (0.6–2) and (La/Yb)cn (
15–18; Fig. 6) are consistently higher for IEQ samples than for equigranular syenite samples of similar SiO2. The AFG is richer in Rb, Th and heavy REE (HREE), poorer in LREE, Sr and Ba contents, and displays low (La/Yb)cn (5–12) compared with other portions of the ONP (Figs 3f and 6b). Samples of the AFG show more pronounced negative Eu anomalies (Eu/Eu* = 0.37–0.14) than other ONP rock types (Fig. 6a).
Precambrian granite wall rock
The wall rock of the ONP is the White Sands pluton, which consists of (1) coarse-grained, porphyritic granite that contains K-feldspar megacrysts up to 3 cm long, and (2) a medium-grained, equigranular granite (Condie & Budding, 1979; Seager, 1981). The major element compositions of the two overlap and SiO2 ranges from 64 to 73 wt % (Table 1). Both granite types display high Th (19–39 ppm) and LREE abundances (Nd 85–140 ppm; Table 2). At 36 Ma, these rocks would have had 87Sr86/Sr of 0.7427–0.8303 and Nd of –10.4 to –13.7 (Table 6; Fig. 7).
|
) with fields for major lithologic units of the ONP.Xenoliths and mafic enclaves in the ONP
Samples of two mafic enclaves in the IEQ are of trachybasalt and andesite compositions (Table 1; Le Bas et al., 1968
). The more mafic enclave (6191) consists of subhedral, weakly zoned laths of plagioclase, sub- to anhedral augite, andw biotite, with minor hornblende, and Fe–Ti oxide minerals and accessory apatite. The enclave is large ion lithophile element (LILE) and LREE poor, and rich in HREE and Sc compared with the most mafic portions of the ONP (Fig. 3j). The Nd (–2.0) of the enclave is similar to that of equigranular syenite samples (Table 6) and the 87Sr/86Sr ratio (0.7154) is the highest we obtained for any enclave sample (Fig. 7). The composition of the andesitic enclave sample (5592I) resembles that of monzodiorite (Table 1) but the enclave's trace element characteristics are like those of IEQ samples (Table 2; Fig. 3j). The Nd and Sr isotopic compositions (Nd –4.2, 87Sr/86Sr 0.7087) of the enclave are also within the range of inequigranular syenite samples (Fig. 7).
Mafic enclave sample 492I is from the interior of the pluton and consists of subhedral plagioclase phenocrysts set in a finer-grained matrix of clinopyroxene, hornblende, biotite, and Fe–Ti oxide minerals with trace amounts of apatite and zircon. The bulk composition of this enclave is similar to that of the monzodiorite facies of the ONP (Table 1), but the enclave is richer in Sr and Ba (Fig. 3j), and displays higher Nd (–1.3) and lower 87Sr/86Sr (0.7058) values (Fig. 7).
Xenolith sample 6091 consists of abundant titanite grains up to 5 mm in length set in or partially enclosing a fine-grained, equigranular matrix of equant and weakly zoned plagioclase, hornblende, Fe–Ti oxide minerals, and quartz along with trace apatite, epidote, and zircon. Xenolith sample 5991 consists of feldspar megacrysts up to 3 cm in length, set in an inequigranular matrix of plagioclase, hornblende, biotite, Fe–Ti oxide minerals, clinopyroxene, and quartz with trace titanite, apatite, and zircon. The megacrysts consist of K-feldspar cores rimmed by fine-grained, subhedral plagioclase and quartz that appear to have replaced pre-existing K-feldspar. Compositions of K-feldspar cores are similar to those of large K-feldspar grains found in adjacent Precambrian granite, although the former are slightly more sodic (Ab17–23) than the latter (Ab8—12; Table 3). Compositions of the plagioclase rims of megacrysts (An21–26) also resemble those of plagioclase grains in the Precambrian wall rock, but are considerably less calcic than small plagioclase grains that form the matrix of the xenolith (Table 3).
Xenolith sample 6091 is poorer in SiO2, Na2O and K2O, but richer in CaO, TiO2, Al2O3, FeO and P2O5 compared with sample 5991 (Table 1). Sample 6091 is also enriched in U, Th, Sr, REE (particularly Eu), Sc and Hf, but depleted in Rb and Cs relative to sample5991 (Fig. 7k). Despite their different petrographic and chemical characteristics, the two xenoliths display nearly identical whole-rock Nd values ( –12.4; Fig. 7). The whole-rock 87Sr/86Sr value of sample 6091 (0.7174) is lower than that of sample 5991 (0.7241; Fig. 7).
Major and accessory mineral separates
Euhedral grains of zircon and apatite are present throughout the equigranular syenite. Apatite abundance decreases from 1.4 to 0.3 wt % from monzodiorite to quartz syenite (Table 5). Titanite is present in equigranular syenite with SiO2 >62%, in which it occurs as 1 mm long, euhedral to subhedral grains. Titanite, zircon and apatite are present in all samples of the IEQ. In these rocks, titanite grains are larger than in the equigranular syenite (up to 3 mm long) and are generally subhedral to anhedral with corroded margins. They are substantially larger than titanite grains found in the equigranular syenite. Apatite in the IEQ varies from 0.4 to 1.3 wt % of sample modes (Table 5) and occurs both as discrete needles with equant hexagonal cross-sections and as inclusions in plagioclase. Zircon is a euhedral interstitial phase that represents <1 wt % of sample modes (Table 5). Chevkinite, a common accessory phase in syenitic rocks (Imaoka & Nakashima, 1994), is present as discrete grains in several samples of IEQ (0.1 wt % of mode) and is extremely enriched in REE (Table 4). This phase was not encountered in any samples of the equigranular syenite.
Titanite is the mineral richest in Sm and Nd (2810–6345 ppm Nd), followed by apatite and hornblende (Table 5). Zircon has the lowest Nd contents (51–105 ppm) and highest 147Sm/144Nd ratios (0.106–0.1730). Titanite separates from the IEQ samples display higher Sm and Nd contents than those from equigranular syenite. In the interior of the pluton, the hornblende separate from the titanite-bearing equigranular syenite sample (6591) has lower Nd contents than those from themonzodiorite or lower wt % SiO2 equigranular syenite, titanite-absent,samples (Table 5).
Mineral separates from IEQ samples display lower Nd values (–4.6 to–5.4) than their counterparts from equigranular syenite (Nd –2.1 to –3.3) or monzodiorite (Nd –3.4 to –3.8; Fig. 8) samples. This isotopic difference is identical to that observed for the corresponding whole-rock data. Apatite separates from the IEQ samples exhibit a range of Rb (2–7 ppm) and Sr (208–513 ppm) contents and have uniformly higher 87Sr/86Sr (0.70944–0.70995) than apatite separates from monzodiorite or equigranular syenite samples (87Sr/86Sr 0.70539–0.70650; Table 5).
|
, titanite. |
Discussion |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Constraints on production of high-silica magma
The whole-rock chemical and isotopic data reported here support the hypothesis of Verplanck et al. (1995)
that the interior of the ONP is vertically zoned from monzodiorite to quartz syenite, with all lithologies having Nd values from –2 to –4, whereas the margins of this pluton have uniformly lower Nd –5 and range from monzodiorite to alkali feldspar granite in composition.
Within this basic framework we can now ask: what process created high silica (>75 wt % SiO2) magmas parental to the AFG in the ONP magma system? Like many metaluminous high-silica rhyolites, the AFG is poor in Ba, Sr and LREE, rich in U, Th and HREE, and displays a moderate negative Eu anomaly (Hildreth, 1981). In metaluminous high-silica rhyolites, these trace element characteristics are thought to reflect the removal, at depth along the chamber side walls, of major (e.g. alkali feldspar, plagioclase, quartz), minor (hornblende, biotite, ilmenite, magnetite) and accessory minerals (such as zircon, allanite) from the magmas parental to the high-silica magmas (Michael, 1983). Did crystal separation produce the magmas parental to the AFG, and did silicic migrate upwards along the chamber side walls of the ONP?
Evidence for side-wall migration of silicic magmas
Although Verplanck et al. (1995) suggested that the IEQ and AFG might be genetically related, they could not demonstrate a physical link between the two The data presented here, however, show that a narrow ‘channel’ of Nd –5 rock is present along much of the side wall of the ONP (Fig. 2). Furthermore, the rocks in the ‘channel’ show evidence for increasing differentiation with decreasing structural depth (Fig. 2). The bulk compositions of the channel rocks converge on those of the AFG but increasingly diverge from the compositions of adjacent syenite with decreasing depth. These data indicate that the magmas parental to the AFG were delivered to the top of the magma chamber by migrating upwards along the chamber side walls. The spatial variation of chemical compositions of Nd – 5 rocks along the margins of the ONP further suggests that magmas that migrated along the chamber side walls were segregated by density and that the lowest density, Si-richest, magmas rose to the uppermost parts of the chamber(see Mahood & Cornejo, 1992). However, the Nd values ( –5) of the border lithologies of the ONP are invariably lower than those for adjacent, more interior, parts of the pluton, which display Nd –2 to –3. Thus, these data support the hypothesis of Verplanck et al., (1995) that neither the AFG nor any of the Nd border lithologies (including the IEQ) formed solely by side-wall fractional crystallization of the magma that was parental to interior parts of the pluton, a scenario common to many physicochemical models of the production of highly silicic magmas (e.g. McBirney et al., 1985; Spera et al., 1995).
Evidence for crystal separation from Nd –5 magmas
Although side-wall fractional crystallization of the magma directly underlying the AFG could not have produced this lithology, the similarity between the chemical compositions of the AFG and coeval metaluminous ash-flows implies that the AFG preserves magma compositions (Hildreth, 1979) and that these magmas were products of crystal–liquid separation (Michael, 1983). The decoupling observed between major an trace element compositions in the IEQ indicates that these rocks could contain at leas part of the crystalline residue that was removed by fractional crystallization, thus creating the highly silicic magmas parental to the AFG. The presence of cumulate plagioclas ± alkali feldspar in the IEQ is recorded by its less negative Eu anomalies (at comparable SiO2 contents) than interior parts of the pluton. Many samples of IEQ have Eu/Eu* >1 (Fig. 6a), which indicates the preferential partitioning of Eu2+ into plagioclase (Henderson, 1984). The high Ba and Sr contents of the IEQ samples also signal the presence of cumulate feldspar (Fig. 9), because both elements are commonly concentrated in feldspar minerals (e.g. Blundy & Wood, 1991; Icenhower & London, 1996). Finally, the inequigranular texture of the IEQ, and its subhedral plagioclase phenocrysts, both hint at a cumulate origin for this lithology.
|
The high LREE abundances and (La/Yb)cn (Fig. 6b) of the IEQ could also result from accumulation of REE-bearing minerals, a possibility that might be substantiated with mass balance calculations based on chemical and isotopic data for the mineral separates from the IEQ. Unfortunately, a full LREE budget for these rocks cannot be calculated from the available data, because of lack of information about (1) REE abundances in the opaque oxide minerals or in REE-bearing mineral inclusions (Bea, 1996
) and (2) REE zoning within the various REE-bearing phases (Sawka, 1988), and (3) a lack of accurate modal or compositional data on chevkinite. The differences in the REE contents of two titanite separates from sample 5092 emphasize the fact that considerable variation occurs in the REE abundances of the accessory minerals in the IEQ samples (Table 5). Nevertheless, a first-order calculation of the fractional contributions of each of the various REE-bearing phases to the total Sm and Nd budget in each of the IEQ samples investigated can be made (Fig. 10). Within the IEQ, titanite (the modal abundance of which ranges from 0.5 to 2 vol. %; Table 5) is the dominant residence site of LREE, followed by apatite, hornblende (if present), and, to a much lesser degree, zircon. Where present, chevkinite contains a significant fraction of the available Sm and Nd (Fig. 10), but it is still of secondary importance compared with titanite. The modal abundance of titanite and total LREE abundances are correlated in the IEQ (Fig. 11). The wide range of modal abundances of titanite, the ubiquitous presence of titanite in all IEQ samples, the lack of titanite in samples with <62 wt % SiO2 from the interior of the ONP, and the fact that titanite as well as apatite grains in IEQ samples are relatively large, anhedral and partially resorbed, all point to a cumulate origin for many accessory minerals in the IEQ. The presence of both major and accessory cumulate minerals in the IEQ accounts for decoupling of its major and trace element abundances. Unlike the AFQ, the IEQ does not represent original magma compositions.
|
|
Apatite, hornblende, titanite, and zircon separates in the IEQ all display essentially identical Nd isotopic compositions (
Nd=–5.1 ± 0.2), regardless of whole-rock bulk composition (Fig. 8). This implies that these phases crystallized from magmas with identical Nd isotopic characteristics (see Fig. 14, below). Similarly, apatite separates from IEQ samples have the same Sr isotopic compositions as the whole rocks (0.7095; Table 5). Both the Sr and Nd isotopic compositions of mineral separates from IEQ samples overlap the isotopic compositions of the AFG. This observation is strong evidence that many of the accessory and major minerals present in the IEQ represent the crystalline residue removed from the magma(s) parental to the AFG. The removal of apatite, plagioclase, and titanite from a residual magma parental to the AFG is also consistent with its low P2O5 and Sr contents (Tables 1 and 2), negative Eu anomalies, and, assuming that titanite and apatite are LREE enriched (Bea, 1996), its generally low LREE abundances and (La/Yb)cn (Fig. 6b; Michael, 1983).
|
In summary, the data presented here provide direct evidence that highly silicic epizonal magmas can evolve by early crystallization of minerals deep in a magma system, followed by separation of residual magma and early formed crystals presumably during vertical migration of the residual magma along the side walls of a magma chamber.
Origin of Nd –5 magma
The presence of Nd –5 rocks along the margins of the ONP suggests that the silicic magmas that rose along the chamber side walls had this Nd isotopic signature before reaching the exposed structural levels. Although the exact age of the quartz syenite dike that cuts the eastern margin of the ONP is unknown, its Nd and Sr isotopic compositions are similar to the samples from the margins of the ONP (Table 6). This observation suggests that magmas with appropriate isotopic characteristics were present beneath the exposed portions of the ONP. Do the low Nd and high 87Sr/86Sr of the margins of the ONP represent the isotopic compositions of the parental magmas that rose into the mid- to upper crust, or were these isotopic compositions inherited from open-system processes occurring at these crustal levels? The mafic enclaves in the IEQ can constrain both the isotopic and bulk compositions of the magmas parental to the border lithologies of the ONP. Most mafic enclaves are products of mixing of mafic magma into a felsic magma host (Didier & Barbarin, 1991). The presence of such enclaves within the IEQ suggests that intermediate to mafic magmas were contemporaneous with silicic syenitic magma preserved along the margins of the ONP. However, enclaves are unlikely to represent parental magmatic compositions, because of in situ differentiation processes (Eberz & Nicholls, 1990) and diffusional exchange of elements between enclaves and the surrounding host magma (Holden et al., 1991; Van der Laan & Wyllie, 1993). For example, the trace element abundances and Nd and Sr isotopic characteristics of enclave sample 5592I, with 55.5 wt % SiO2, resemble those of the IEQ (Tables 1, 2 and 6). This enclave may be evidence that Nd –5 andesitic magmas were present at depth in the magma system and available for differentiation to more silicic compositions. Alternatively, the enclave may represent mafic magma that had chemically and isotopically re-equilibrated with more silicic Nd –5 melts and so provides little insight into the origin of the Nd –5 magmas themselves.
Enclave 6191, in contrast, is a trachybasalt with Nd and Sr isotopic compositions that differ from the surrounding IEQ and so is probably not a product of re-equilibration with intermediate to silicic composition, Nd –5 magmas. Although its low total REE contents (Fig. 3j) and high Sc suggest that the xenolith is at least in part of a cumulate origin (i.e. of clinopyroxene; Rollinson, 1993), it is none the less evidence that mafic magmas of Nd –2 were present at depth in the magma system. This conclusion is supported by the presence of other mafic enclaves with similar Nd isotopic compositions within the equigranular syenite (sample 492I; Fig. 7), and with the conclusion of Verplanck et al., (1995) that the equigranular syenites evolved from mafic, Nd –2, parental magmas. However, enclave 6191 also has a high 87Sr/86Sr of 0.7153, despite its high Nd value. Decoupling of Sr and Nd isotopic compositions is common in mafic enclaves, a phenomenon generally interpreted to reflect faster self-diffusion rates of Sr relative to Nd between injected mafic magmas and a felsic magmatic host (Holden et al., 1991). This cannot be the case for the ONP, because the IEQ samples have lower 87Sr/86Sr ratios than does the enclave. The only rock near the ONP that could have imparted a high 87Sr/86Sr to this enclave is the Precambrian granitic wall rock. This observation suggests that low Nd magmas were injected into the mid- to upper crust where they interacted with felsic melts of the Precambrian wall rock.
Evidence for partial melting of Precambrian wall rock
Direct evidence of anatexis of Precambrian granite along the margins of the ONP is provided by the angular xenoliths found in the IEQ. Two lines of evidence indicate that these xenoliths were derived from the Precambrian wall rock of the ONP. First, the cores of partially recrystallized K-feldspar megacrysts within the xenolith are similar in composition to the cores of K-feldspar megacrysts from the Precambrian granite wall rock (Table 3). Second, the xenoliths exhibit Nd isotopic compositions identical to those of the wall rock (Fig. 7).
The compositions of these xenoliths have been modified from their initial state; sample 6091, which has 39 wt % SiO2, cannot represent an original rock composition. Compared with a typical wall-rock sample (5792), both xenoliths not only have lower LILE and Si contents, but also higher Ca, Fe, U, Fe, Ta, Th and, in the case of sample 6091, Eu contents (Fig. 12). However, the high specific gravities of both xenoliths (2.8 and 3.2 g/cm3) relative to unmodified wall rock (2.7 g/cm3) indicate that some of these concentration variations may be due to volume loss.
|
To assess mass gains or losses of a given element in the xenoliths relative to their (likely) Precambrian granite protoliths, we calculated composition–volume relationships (Gresens, 1967
) using the expression 
Wi)/Wo = (Vd/Vp)(
d/p)Wid – Wip. In this expression, Wi is the change in mass of component i, Wo is the initial mass of the parent rock (chosen to be 100 g of wall-rock sample 5792), Vd and Vp are the volumes of daughter and parent, d and p are their densities, and Wid and Wip are the mass fractions of component i in the daughter and parent. Following Glazner (1988), we plot log(Vd/Vp) for each element with the assumption of immobility Wi = 0). In such a plot, a cluster of log(Vd/Vp) is used to identify elements that were immobile during alteration of parent rock, with the mean log(Vd/Vp) of these elements then representing a measure of the volume ratio of ‘daughter’ to ‘parent’ rock.
For sample 6091, the LREE, Zr, Hf, Mg, and Hf were apparently immobile, clustering at a log(Vd/Vp) of 0.45, indicating a volume loss of 60% (Fig. 13). Elements with high log(Vd/Vp) probably represent components removed from the parent rock (in this case Al, Si, Ba, Na, Rb, K), whereas lower values represent added components (Eu, U, Ca, Sr, Fe). For sample 5991, clustering of LREE, Zr, and Hf occurs at log(Vd/Vp) of –0.2, implying a volume loss of 36%. The calculation suggests that Cs, K, Rb, and Si were lost and Fe, Sr, Ta, U, and Ca were added (Fig. 13).
|
The apparent loss of Si, K, Rb, and Cs from the xenoliths compared with the Precambrian granite indicates that the xenoliths are probably the residua from dehydration melting of the Precambrian granite (Patiño Douce & Beard, 1995
). Reactions involving biotite, plagioclase and quartz generate not only a high-silica magma (>70% SiO2), but a residua that includes Fe-oxide minerals (Patiño Douce & Beard, 1995). The latter minerals are a major component of the xenolith matrices. However, the xenoliths are not simply residua of partial melting. They must have either have been infused or chemically exchanged with a source of Ca and additional Fe. One possible source of such elements might be mafic magma. The addition of Ca to the xenoliths is consistent with the presence of more calcic matrix plagioclase in the xenolith than in the Precambrian granite (Table 3). The low 87Sr/86Sr of the xenoliths relative to the Precambrian wall rock may also reflect the interaction of the xenoliths with mafic magmas containing low 87Sr/86Sr strontium (Lesher, 1990), although the loss of radiogenic Sr from Rb-rich mineral phases such as biotite during dehydration melting could also account for their Sr isotopic characteristics (Knesel & Davidson, 1996).
Assimilation of wall-rock melts
Can the assimilation of wall rock or partial melts derived from it account for any of the chemical or isotopic variability observed in the ONP? The apparent lack of xenocrysts in IEQ suggests that assimilation of wall rock occurred by incorporation of partial melts rather than the bulk assimilation of wall rock (Green, 1994). Isoenthalpic melt assimilation decreases the liquidus temperature of magma and the volume of crystallized material associated with assimilation (Reiners et al., 1995, 1996). The assimilation of melt will significantly change the isotopic, but not major element, compositions of resultant mixed magmas. Therefore, Nd –5 rocks of the ONP may be products of incorporation of wall-rock melts into a mafic magma of Nd –2.
Using xenolith sample 5991 as the restite remaining after partial melting of the wall rock, it is possible to roughly estimate from mass balance considerations the composition of silicic melts that may have been removed during anatexis. Assuming a volume loss of 36% and no addition of Si, or loss of Fe, Mg, or Ca to the xenolith, a silicic magma extracted from this xenolith should contain 30 ppm Nd and have 78 wt % SiO2. The latter value is compatible with experimental studies of partial melting of biotite gneiss (Patiño Douce & Beard, 1995). Simple mixing of 20 wt % of this silicic magma (with Nd –12.5 typical of the Precambrian basement) with magma compositionally similar to enclave 6191 produces a hybrid magma with Nd –5, and with an SiO2 content of only 53 wt %. Less crust is required to produce an Nd –5 mixed magma through bulk assimilation, because of the high Nd content of the wall rock (Table 6). None of these calculations realistically depict the wall-rock assimilation that may have taken place, but they show that wall assimilation could reproduce the isotopic compositions of IEQ without grossly affecting the bulk composition of the original mafic magma.
Constraints on the evolution of the equigranular syenite
Previous studies of the ONP concluded that the equigranular syenite is compositionally zoned from 60 to 70 wt % SiO2 and probably evolved by closed-system fractional crystallization (Seager & McCurry, 1988; Yanicak, 1992; Verplanck et al., 1995). These conclusion are supported by the generally uniform Nd and Sr isotopic compositions of equigranular syenite and by the correlations of major element abundances with model mineral variations in trends that follow predictions of fractional crystallization models.
The trace element data of this study support these conclusions. With increasing silica contents, Ba and Sr decrease and the calculated Eu anomaly becomes increasingly negative. These features suggest that crystallization and isolation of K-feldspar and plagioclase from a remaining magma formed the equigranular syenite. The Nd and Sm budget of the equigranular syenite shows that at the onset of titanite crystallization (in rocks that contain 62 wt % SiO2), the Nd content of hornblende decreases by 85% (Table 5). This indicates that LREE preferentially entered titanite instead of hornblende as the magma continued to fractionate. The coupling of major element, modal, and mineral chemistry variations of the equigranular syenite contrasts with observations of the inequigranular syenite. Many equigranular syenite samples probably approximate liquid compositions (see also Gromet & Silver, 1983). An exception to this generalization is the suite of samples (4392, 6591, 3592) adjacent to the lower margin of the ONP. These rocks have Nd and Sr isotopic compositions similar to the remainder of the equigranular syenite, but are relatively enriched in both Ba and Sr (Fig. 9), and also display positive Eu anomalies (Figs 3 and 6a). The rocks may preserve a region of accumulation of K-feldspar that was physically isolated from the magma. A second exception is quartz syenite sample 3892. This rock was obtained from a position surrounded by Nd –2 equigranular syenite (Fig. 1) but has Sr and Nd isotopic compositions like those of the IEQ and AFG (0.7095 and –4.7, respectively). The REE contents of sample 3892 (Table 2) are comparable with those of the IEQ samples. This sample may represent Nd –5 magma that migrated up along the margin and was back mixed into the interior portion of the pluton, as was proposed by Mahood & Cornejo, (1992) for rocks of the La Gloria pluton. This mixing relationship implies that the interior and margins of the ONP were comagmatic. Because sample 3892 evidently remained chemically and isotopically distinct from surrounding rocks, the back mixing probably occurred just before solidification.
The isotopic data from the more mafic portion of the ONP do not support the conclusion by Seager & McCurry (1988) that the underlying monzodioritic facies of the ONP is simply a less differentiated part of the syenitic facies. Although the Sr isotopic composition of the monzodiorite core of the pluton is within the range of values defined for equigranular syenite, the Nd isotopic composition (Nd –3.5) of the monzodiorite is distinctly lower. The monzodiorite may, instead, represent a mixture of Nd –2 and Nd –5 mafic magmas near the base of the magma chamber.
Summary
The four basic elements of a simplified model for the evolution of the ONP are (Fig. 14) as follows:
- An 48% SiO2 magma, with uniform Nd of –1 to –2, enters a mid- to upper-crustal magma chamber from a greater depth in the continental lithosphere. Because the quartz syenite dike that intrudes the eastern margin of the ONP (sample 4292; Figs 1 and 14) contains xenoliths of Precambrian basement rocks, the chamber probably necked substantially at depths just below the current level of exposure.
- A magma of Nd –5 was underneath a higher Nd ( –2) during the early evolution of the ONP magma body. This situation could have arisen by (a) the sequential injection of separate magma batches with distinctly different Nd isotopic compositions into a magma chamber (Mills et al., 1997), or (b) interaction between an Nd –2 parental magma and low Nd Precambrian granitic basement. We cannot rule out the former possibility, which would also require that the separate magma batches did not intermix as the Nd –5 magma differentiated and rose along the chamber side walls. Given the evidence for partial melting of the Precambrian granitic rocks, the latter possibility seems more likely. In this case, wall-rock xenoliths would be added mechanically to deep-seated portions of the magma body, where they would partially melt and possibly be assimilated. Assimilation would generate an SiO2 50 wt % magma with Nd –5. Such a magma might contrast sufficiently in composition and viscosity with the overlying Nd –2 magma so that mixing between the two magma ‘types’ would be impeded.
- The silicic magmas of the ONP were products of fractional crystallization and crystal–liquid separation. The felsic dikes from the eastern portion of the pluton all display Nd –5, which implies that differentiated magmas were present at depth in the ONP magmatic system. This suggests that silicic magmas that rose along the chamber walls probably did not differentiate at the current levels of exposure, but had acquired silicic compositions at greater depths in the system. If the Nd –5 magma was convecting in the neck of the chamber, it could have developed a steep thermal gradient that intersected the liquidus at depth in the system (Jackson, 1961; Mahood & Cornejo, 1992). The trace element differences and Sr isotopic heterogeneities observed in the Nd –5 rocks of the ONP margins indicate that their parental magmas rose as separate batches along the pluton's side walls. Each batch probably had a slightly different history of differentiation and wall-rock assimilation. The magma parental to the equigranular syenite probably evolved independently from the enveloping Nd –5 magma. The differentiation of the former probably occurred by side-wall fractional crystallization and magma migration (Spera et al., 1995). Interaction between the Nd –5 and –2 magma could account for the isotopic characteristics of the monzodiorite, and the presence of low Nd quartz syenite in the ONP.
- The ONP probably formed by rapid cooling of the magma body during the eruption of the Squaw Mountain tuff (Seager & McCurry, 1988). This accounts for the widely distributed cupolas of the AFG, as well as the presence of the monzodiorite in the pluton's interior. Even the mafic enclaves, restite xenoliths, and cumulate rocks in the IEQ may be ‘rip-ups’ that were carried to shallower depths during rapid magma movement associated with this final eruption of the ONP magma chamber.
|
Conclusions |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Differentiated magmas not only migrated upward along the margins of the magma body parental to the ONP, but their fractional crystallization and crystal–liquid separation produced highly silicic magmas with characteristic trace element abundances and isotope signatures. The differentiated magmas that evidently pooled at the top of the ONP magma body to form the AFG were not derived from the magma that produced the adjacent, underlying quartz syenites. Instead, the highly silicic magma parental to the AFG was derived from mafic magma(s) at or near the base of the ONP magma chamber. The latter underwent fractional crystallization, magma migration and, probably, significant wall-rock assimilation, in processes distinct from those that formed the bulk of the exposed ONP.
|
Acknowledgements |
|---|
We thank Anita Grunder, Calvin Miller, Paul Tomascak and an anonymous reviewer for their insightful comments on the manuscript. Particular thanks go to Sorena Sorensen for her patient editorial guidance. Access to the field area was arranged by Sam Seek (White Sands Missile Range) and Kevin Von Finger (Fort Bliss Military Reservation). This work was supported in part by NSF Grants EAR 91–05705 (G.L.F.) and EAR 91–06169 (M.M.). Instrumental neutron activation analyses were obtained from the Oregon State University Radiation Center through the OSU Reactor Use Sharing Program.
|
FOOTNOTES |
|---|
* Present address: US Geological Survey, 3215 Marine Street, Boulder, CO 80303, USA.
Corresponding author. Telephone: 303-492-6534. Fax: 303-492-1149. e-mail: farmer{at}terra.colorado.edu
|
References |
|---|
|
TOPABSTRACTIntroductionGeological SettingSamplesAnalytical MethodsResultsDiscussionConclusionsReferences |
|---|
Anders E., Ebihara M. Solar system abundances of the elements. Geochimica et Cosmochimica Acta (1982) 46:2363–2380.[Web of Science]
Bea F. Residence of REE, Y, Th and U in granites and crustal protoliths; implications for the chemistry of crustal melts. Journal of Petrology (1996) 37:521–552.
Beyer J. N. B. Petrology of a mafic facies in the Organ Needle Pluton, Dona Ana County, New Mexico. (1986) New Mexico State University. 89. M.S. Thesis.
Blundy J. D., Wood B. J. Crystal-chemical controls on the partitioning of Sr and Ba between plagioclase feldspar, silicate melts, and hydrothermal solutions. Geochimica et Cosmochimica Acta (1991) 55:193–209.[Web of Science]
Boyd F. R., Mertzman S. A. Composition and structure of the Kaapvaal lithosphere. In: Magmatic Processes: Physicochemical Principles; a Volume in Honor of Hatten S. Yoder, Jr—Mysen B. O., ed. (1987) University Park, PA: Geochemical Society. 13–24.
Butcher D. P. Geochemistry and Nd/Sr systematics of selected lithologic units of the Oligocene Organ Cauldron and Batholith, south central New Mexico. (1990) New Mexico State University. 145. M.S. Thesis.
Condie K. C., Budding A. J. Geology and geochemistry of Precambrian rocks, central and south–central New Mexico. New Mexico Bureau of Mines and Mineral Resources, Memoir (1979) 35:58.
Didier J., Barbarin B. The different types of enclaves in granites—nomenclature. In: Enclaves and Granite Petrology, Developments in Petrology 13—Didier J., Barbarin B., eds. (1991) Amsterdam: Elsevier. 19–24.
Eberz G. W., Nicholls I. A. Chemical modification of enclave magma by post-emplacement crystal fractionation, diffusion and metasomatism. Contributions to Mineralogy and Petrology (1990) 104:47–55.[Web of Science]
Farmer G. L., Broxton D. E., Warren R. G., Pickthorn W. Nd, Sr, and O isotopic variations in metaluminous ash-flow tuffs and related volcanic rocks at the Timber Mountain/Oasis Valley Caldera Complex, SW Nevada: implications for the origin and evolution of of large-volume silicic magma bodies. Contributions to Mineralogy and Petrology (1991) 109:53–68.[Web of Science]
Feeley T. C., Davidson J. P. Petrology of calc-alkaline lavas at Volcan Ollague and the origin of compositional diversity at Central Andean stratovolcanoes. Journal of Petrology (1994) 35:1295–1340.
Feklichev V. G. Diagnostic Constants of Minerals (1992) Moscow: Mir. 678.
Glazner A. F. Stratigraphy, structure, and potassic alteration of Miocene volcanic rocks in the Sleeping Beauty area, central Mojave Desert, California. Geological Society of America Bulletin (1988) 100:424–435.
Green N. L. Mechanism for middle to upper crustal contamination: evidence from continental-margin magmas. Geology (1994) 22:231–234.
Gresens R. L. Composition-volume relationships of metasomatism. Chemical Geology (1967) 2:47–55.
Gromet L. P., Silver L. T. Rare earth element distributions among minerals in a granodiorite and their petrogenetic implications. Geochimica et Cosmochimica Acta (1983) 47:925–939.[Web of Science]
Henderson P. General geochemical properties and abundances of the rare earth elements. In: Rare Earth Element Geochemistry, Developments in Geochemistry 2—Henderson P., ed. (1984) Amsterdam: Elsevier. 1–32.
Hildreth W. The Bishop Tuff: evidence for the origin of compositional zonation in silicic magma chambers. Geological Society of America, Special Paper (1979) 180:43–75.
Hildreth W. Gradients in silicic magma chambers: implications for lithospheric magmatism. Journal of Geophysical Research (1981) 86:10153–10192.
Holden P., Halliday A. N., Stephens W. E., Henney P. J. Chemical and isotopic evidence for major mass transfer between mafic enclaves and felsic magma. Chemical Geology (1991) 92:135–152.[Web of Science]
Icenhower J., London D. Experimental partitioning of Rb, Cs, Sr, and Ba between alkali feldspar and peraluminous melt. American Mineralogist (1996) 81:719–734.[Abstract]
Imaoka T., Nakashima K. Chevkinite in syenites from Cape Ashizuri, Shikoku Island, Japan. Neues Jahrbuch für Mineralogie, Monatshefte (1994) 8:358–366.
Jackson E. D. Primary textures and mineral associations in the ultramafic zone of the Stillwater Complex, Montana. US Geological Survey Professional Paper (1961) 358:106.
Klein K., Hurlbut C. S. Jr. Manual of Mineralogy (1993) New York: John Wiley. 681.
Knesel K. M., Davidson J. P. Isotopic disequilibrium during melting of granite and implications for crustal contamination of magmas. Geology (1996) 24:243–246.
Laul J. C. Neutron activation analysis of geological material. Atomic Energy Review (1979) 17:603–695.
Le Bas M. J., Le Maitre R. W., Streckeisen A., Zanettin B. A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology (1968) 27:745–750.
Lesher C. E. Decoupling of chemical and isotopic exchange during magma mixing. Nature (1990) 344:235–237.
Mahood G. A., Cornejo P. C. Evidence for ascent of differentiated liquids in a silicic magma chamber found in a granitic pluton. Transactions of the Royal Society of Edinburgh, Earth Science (1992) 83:63–69.
McBirney A. R., Baker B. H., Nilson R. H. Liquid fractionation. Part I: Basic principles and experimental simulations. Journal of Volcanology and Geothermal Research (1985) 24:1–24.[Web of Science]
Michael P. J. Chemical differentiation of the Bishop tuff and other high-silica magmas through crystallization processes. Geology (1983) 11:31–34.
Mills J. G. Jr, Saltoun B. W., Vogel T. A. Magma batches in the Timber Mountain magmatic system, Southwestern Nevada Volcanic Field, Nevada, USA. Journal of Volcanology and Geothermal Research (1997) 78:185–208.[Web of Science]
Nash W. P. Analysis of oxygen with the electron microprobe: applications to hydrous glass and minerals. American Mineralogist (1992) 77:453–457.[Abstract]
Patiño Douce A. E., Beard J. S. Dehydration-melting of biotite gneiss and quartz amphibolite from 3 to 15 kbar. Journal of Petrology (1995) 36:707–738.
Reiners P. W., Nelson B. K., Ghiorso M. S. Assimilation of felsic crust by basaltic magma: thermal limits and extents of crustal contamination of mantle-derived magmas. Geology (1995) 23:563–566.
Reiners P. W., Nelson B. K., Nelson S. W. Evidence for multiple mechanisms of crustal contamination of magma from compositionally zoned plutons and associated ultramafic intrusions of the Alaska Range. Journal of Petrology (1996) 37:261–292.
Rollinson H. R. Using Geochemical Data: Evaluation, Presentation, Interpretation (1993) Harlow, UK: Longman. 352.
Sawka W. N. REE and trace element variations in accessory minerals and hornblende from the strongly zoned McMurry Meadows pluton, California. Transactions of the Royal Society of Edinburgh, Earth Science (1988) 79:157–168.
Sawka W. N., Chappell B. W., Kistler R. W. Granitoid compositional zoning by side wall boundary layer differentiation: evidence from the Palisade Crest intrusive suite, central Sierra Nevada. Journal of Petrology (1990) 31:519–553.
Seager W. R. Geology of the Organ Mountains and southern San Andres Mountains, New Mexico. New Mexico Bureau of Mines and Mineral Resources Memoir (1981) 36:97.
Seager W. R., McCurry M. The cogenetic Organ cauldron and batholith, south–central New Mexico: evolution of a large-volume ash flow cauldron and its source magma chamber. Journal of Geophysical Research (1988) 93:4421–4433.
Sparks R. S. J., Huppert H. E., Turner J. S. The fluid dynamics of evolving magma chambers. Philosophical Transactions of the Royal Society of London, Series A (1984) 310:511–534.
Spera F. J., Oldenburg C. M., Christiansen C., Todesco M. Simulations of convection with crystallization in the system KAlSi2O6–CaMgSi2O6: implications for compositionally zoned magma bodies. American Mineralogist (1995) 80:1188–1207.[Abstract]
Van der Laan S. R., Wyllie P. J. Experimental interaction of granitic and basaltic magmas and implications for mafic enclaves. Journal of Petrology (1993) 34:491–517.
Verplanck P. L., Farmer G. L., McCurry M., Mertzman S., Snee L. W. Isotopic evidence on the origin of compositional layering in an epizonal magma body. Earth and Planetary Science Letters (1995) 136:31–41.[Web of Science]
Wilson C. J. N., Hildreth W. The Bishop tuff: new insights from eruptive stratigraphy. Journal of Geology (1997) 105:407–439.[Web of Science]
Yanicak S. Petrology and major and trace element geochemistry of the quartz syenite facies of the Organ Needle pluton, New Mexico. (1992) New Mexico State University. 165. M.S. Thesis.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||