Petrography and origin of metasedimentary xenoliths in lavas from Tongariro Volcanic Centre

This article was downloaded by: [University of Birmingham]On: 13 November 2014, At: 09:57Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office:Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

New Zealand Journal of Geology andGeophysicsPublication details, including instructions for authors and subscriptioninformation:http://www.tandfonline.com/loi/tnzg20

Petrography and origin of metasedimentaryxenoliths in lavas from Tongariro VolcanicCentreI.J. Graham a ba Research School of Earth Sciences , Victoria University of Wellington ,Private Bag, Wellington , New Zealandb Institute of Nuclear Sciences , Private Bag, Lower Hutt , New ZealandPublished online: 16 Jan 2012.

To cite this article: I.J. Graham (1987) Petrography and origin of metasedimentary xenoliths in lavasfrom Tongariro Volcanic Centre, New Zealand Journal of Geology and Geophysics, 30:2, 139-157, DOI:10.1080/00288306.1987.10422179

To link to this article: http://dx.doi.org/10.1080/00288306.1987.10422179

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)contained in the publications on our platform. However, Taylor & Francis, our agents, and ourlicensors make no representations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressed in this publicationare the opinions and views of the authors, and are not the views of or endorsed by Taylor &Francis. The accuracy of the Content should not be relied upon and should be independentlyverified with primary sources of information. Taylor and Francis shall not be liable for anylosses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantialor systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, ordistribution in any form to anyone is expressly forbidden. Terms & Conditions of access and usecan be found at http://www.tandfonline.com/page/terms-and-conditions

http://www.tandfonline.com/loi/tnzg20

http://www.tandfonline.com/action/showCitFormats?doi=10.1080/00288306.1987.10422179

http://dx.doi.org/10.1080/00288306.1987.10422179

http://www.tandfonline.com/page/terms-and-conditions

New Zealand Journal of Geology and Geophysics, 1987, Vol. 30: 139-157 139 0028-8306/87/3002-0139$2.50/0 © Crown copyright 1987

Petrography and origin of metasedimentary xenoliths in lavas from T ongariro Volcanic Centre

I.J.GRAHAM Research School of Earth Sciences Victoria University of Wellington Private Bag Wellington, New Zealand*

Abstract Abundant xenoliths of metasedimen-tary origin in lavas of Tongariro Volcanic Centre (New Zealand) are of three main types. Quartz-rich xenoliths are conspicuous in most lavas and include rare garnet-bearing schists and more abundant metaquartzites. The latter represent restites of gar-net-schist modified by extraction of granitic partial melt. Feldspar-rich xenoliths occur in subequal abundance to quartz-rich xenoliths and include biotite-schists and pyroxene-hornfelses. The latter represent restites after extraction of partial melt from biotite-schist. Uncommon (but widespread) spinel-rich xenoliths also occur. All of the above xenolith types were derived from Torlesse terrane greywacke, regionally metamorphosed to gneiss. They represent the quartzose, feldspathic, and micaceous layers of gneiss, thermally disaggregated following immersion in host lava. The dominance of restite assemblages (i.e., metaquartzites and pyroxene-hornfelses) testifies to the widespread production, extraction, and assimilation of granitic partial melts of xenoliths with host magmas.

Keywords xenoliths; metasedimentary rocks; petrography; geochemistry; Sr-87/Sr-86; genesis

INTRODUCTION Tongariro Volcanic Centre (TVC) is the southern part of a young «250 000 years) andesite-dacite volcanic arc extending along the eastern side of Taupo Volcanic Zone, central North Island, New Zealand (Cole 1984). TVC lavas are dominantly calcalkalinet and contain abundant xenoliths of a wide variety of types. In some flows they make up

*Present address: Institute of Nuclear Sciences, Private Bag, Lower Hutt, New Zealand.

Received 14 January 1986, accepted 30 April 1987

more than 1 % of the mode, but usually only a few millimetre to centimetre-sized fragments occur per cubic metre of lava (W. R. Hackett pers. comm. 1984). Steiner (1958) described the petrography of vitrified, quartzose, and feldspathic xenoliths from Ngauruhoe 1954 lava, but other studies of TVC volcanics include only brief references to xenoliths and their significance to andesite petrogenesis (e.g., Ewart & Stipp 1968; Cole 1978; Blattner & Reid 1982). This paper describes in detail the petrog-raphy, mineral chemistry, bulk-rock chemistry, and isotopic composition of metasedimentary xeno-liths and discusses possible origins.

Sample preparation and analytical methods are fully discussed in Graham (1985a). Mineral ana-lyses were made using a Jeol 733 Superprobe, and bulk-rock analyses were determined by XRF at the Analytical Facility, Victoria University of Welling-ton. Sr was analysed isotopically on a VG MM30B mass spectrometer at the Institute of Nuclear Sciences, Lower Hutt, New Zealand. Precision is given as the standard error of the mean for the 95% confidence interval, and analyses are standardised against 87Sr/86Sr for NBS987 = 0.71015.

PETROGRAPHY OF XENOLITHS Metasedimentary xenoliths included in TVC lavas are of three main types and collectively comprise more than 80% of all recorded metamorphic xen-oliths (see Graham & Hackett 1987, table 6). In the following descriptions, use will be made of a pri-mary classification which reflects dominant mineralogy: quartz-rich = TYPE QRX; feldspar-rich = TYPE FRX; and spinel-rich = TYPE SRX. This is done because, as will be shown, each of these xenolith lithologies are genetically linked to a common proto lith, a fact which would be obscured by exclusive use of rock descriptors. Full petrographic descriptions of all cited examples are contained in Appendix 1, and additional chemical or mineralogical data are available on request from the author.

tMinor volumes of andesite of a more tholeiitic nature occur at several peripheral vents (e.g., Hauhungatahi, Ohakune).

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

140 New Zealand Journal of Geology and Geophysics, 1987, Vol. 30

Quartz-rich (TYPE QRX) xenoliths Quartz-rich xenoliths are ubiquitous in TVC lavas and are of two main types. Gamet-bearing schists (TYPE QRXa) are relatively rare but are locally abundant in Iwikau Member (Whakapapa For-mation) pyroclastics (Hackett 1985). Metaquartz-ites (TYPE QRXb), containing more than 90% modal quartz, occur in lavas of all ages and are the most conspicuous and probably most abundant xenolith type.

TYPE QRXa xenoliths are typically foliated and exhibit centimetre-wide banding with contrasting mineral assemblages: (1) quartz + calcic-plagio-clase + clinopyroxene ± titanite ± ilmenite; (2) sodic-plagioclase + garnet ± orthopyroxene ± biotite ± glass. EPMA analyses of phases in each assemblage are given in Table 1. In assemblage (1), plagioclase ranges in composition from bytownite to anorthite, and pyroxene ranges from salite to ferroaugite. Textures are typically granoblastic (Fig. lA). In assemblage (2), plagioclase compositions are more sodic, ranging from an30 to an40• In 17492, anhedral, unzoned plagioclase and quartz grains are surrounded by clear granitic glass (Fig. IB) and

some almandine-rich garnet mantled by aluminous hypersthene also occurs. A 1 mm hexagonal inclu-sion in one garnet grain (Fig. 1 C) consists of a ver-micular intergrowth of hercynite and gedrite. The hexagonal crystal shape and chemical composition of the constituent minerals (Table 1) suggest the inclusion was probably originally cordierite. Schreyer (1965) reported that at high temperatures and pressures of 10 kbar, Fe-rich cordierite breaks down to assemblages containing an ortho-amphi-bole (ferrogedrite). However, he was not certain whether these assemblages represent stable equilib-rium, or whether they are metastable substitutes for other parageneses such as almandine + silli-manite + quartz. Grieve & Fawcett (1974) inves-tigated the stability ofchloritoid below 10 kbar PH o.

2 They showed that the assemblage aluminous ferro-anthophyllite + staurolite + hercynite was stable with respect to chloritoid only at pressures greater than 5.5 kbar and temperatures greater than 600°C. They further showed that Fe-cordierite breaks down to ferro-anthophyllite + staurolite + quartz under similar P-T conditions. These studies suggest the breakdown of cordierite to gedrite + pleonaste was

Table 1 EPMA analyses of gamet-schist (TYPE QRXa) xenoliths from TVC lavas.

Cat. no.: 17490 17490 17491 17491 17491 17491 17491 17491 17485 17485 Phase: PL CPX GLASS PL OPX GNT PLEON GED BIO GNT

Assemblage: (1) (1) (2) (2) (2) (2) (2) (2) (2) (2)

Si02 44.82 50.28 73.21 61.01 49.45 37.76 0.11 43.77 34.92 37.96 Ti02 0.00 0.13 0.31 0.00 0.18 0.00 0.34 0.00 6.27 0.00 AI20 3 35.58 0.63 13.67 24.86 4.52 21.87 52.01 9.56 17.10 21.12 FeO 0.28 16.23 2.40 0.26 27.59 29.90 40.55* 32.37 17.18 27.98 MnO 0.00 1.11 0.00 0.00 0.28 0.84 0.36 0.92 0.00 5.32 MgO 0.00 6.77 0;16 0.00 17.27 7.78 3.73 11.77 10.94 5.84 Cao 18.68 22.25 0.32 6.20 0.22 0.90 0.00 0.20 0.00 1.90 Na20 0.94 0.10 3.32 7.47 0.00 0.00 0.00 0.00 0.53 0.00 K20 0.07 0.00 6.32 1.05 0.00 0.00 0.00 0.00 9.21 0.00 Total 100.37 99.50 99.80t 100.85 99.51 99.05 97.10 98.59t 96.15t 100.12

Oxygens 8 6 8 6 12 4 22 22 12 Si 2.06 1.97 2.70 1.90 5.94 0.003 7.00 5.23 3.00 Ti 0.00 0.00 0.00 0.01 0.00 0.008 0.00 0.71 0.00 Al 1.93 0.03 1.30 0.20 4.06 1.813 1.80 3.04 1.97 Fe 0.01 0.61 O.Ot 0.89 3.94 1.002 4.33 2.15 1.85 Mn 0.00 0.04 0.00 om 0.11 0.009 0.13 0.00 0.36 Mg 0.00 0.40 0.00 0.99 1.83 1.164 2.81 2.44 0.69 Ca 0.92 0.94 0.29 0.01 0.15 0.000 0.03 0.00 0.16 Na 0.08 om 0.64 0.00 .0.00 0.000 0.00 0.15 0.00 K 0.00 0.00 0.06 0.00 0.00 0.000 0.00 1.76 0.00 Total 5.00 4.00 5.00 4.01 16.03 3.000 16.10 15.48 8.03

Catalogue numbers (Cat. no.) are for samples lodged in the petrological collection of Victoria University of Wellington, New Zealand.

PL = plagioclase; CPX = clinopyroxene; OPX = orthopyroxene; PLEON = pleonaste; GED = gedrite' BIO = biotite; GNT = garnet. '

*Some iron is Fe20 3• tAlso contains combined volatiles.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

Graham-Petrography of xenoliths, Tongariro l4l

Fig. 1 Back-scattered electron image (BEl) photographs of assemblages in quartz-rich xenoliths. Abbreviations are as follows: cpx = clinopyroxene, qz = quartz, an = anorthite, pi = plagioclase, gnt = garnet, ilm = ilmenite, gl = glass. A Assemblage (I) of TYPE QRXa xenolith 17492 (i.e., 17490). B Assemblage (2) of 17492 (i.e. , 17491). C Garnet inclusion in 17491. Association of pleonaste (light-grey blebs) and gedrite (darker grey) probably results from prograde alteration of cordierite. D Reaction at the interface between TYPE QRXb xenolith 17885 and host lava. Zones are described in text. Scale bars for A, B, D = 100 11m; for C = 10 11m.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


Qz QRXa .-·-0 17491

Fig. 2 Normativ~ glass compo-sitions in some quartz-rich and feldspar-rich xenoliths, plotted in the system quartz-albite-ortho-clase (Tuttle & Bowen 1958) . Filled symbols are whole-rocks. Asterisk = S-type granite mini-mum melt composition (White & Chappell 1977). Cotectic lines are for obsidian-anorthite mixtures with ab/an ratios of 3.8 to infinity at PH 0 = 2 kbar (von Platen 1965).2 Circled cross = ternary minimum.

QRXb .---~ 17463 QRXb ••..... -0 17885

FRXa ._c> 17483 FRXb .··_··v 17458 FRXb [] 17444

Ab

prograde and occurred in response to the increase in temperature following incorporation of the xen-olith in host magma. This sudden change in P-T conditions may also have triggered partial melting in 17491 which, from the K-rich, Ca-poor com-position of the glass (Table 1), appears to have involved mainly the destruction of alkali-feldspar and biotite (+quartz). This effectively forced the composition away from a granitic minimum melt (Fig. 2), towards the albite-orthoclase join. No melt is present in 17485 where assemblage (2) is quartz + biotite + plagioclase + garnet. Biotite-garnet exchange equilibria (Ferry & Spear 1978) yields a temperature for the garnet rim of 946°C (Table 2). This is higher than the temperature of 808°C at 10

Or

kbar for gamet-orthopyroxene equilibria in 17491 (Hariey 1984a). Pressure estimates based on assem-blage (I) (plagioclase + clinopyroxene + quartz) (Ellis 1980) give low or negative values, suggesting the equilibria is not appropriate to the composi-tions observed in these assemblages (the low alu-mina contents of clinopyroxene produce large errors in the estimation of the Ca-Tchermak's compo-nent). Gamet-orthopyroxene geobarometry (Har-ley 1984b) requires an aluminosilicate to be present in the assemblage and, since none was found, the calculated pressure of 7.75 kbar (at 800T) is suspect.

Nearly monomineralic metaquartzite xenoliths (TYPE QRXb) are abundant and widespread, par-

Table 2 Metamorphic temperatures ofxenoliths from TVC lavas, based on cation exchange equilibria.

Temperature Cat. no. Type Assemblage ("C) Method

17485 QRXa GNT-BIO 946 Ferry & Spear (1978) 17491 QRXa GNT -OPX 808 Harley (1984a) 17425 FRXa PL-KS 800 Brown & Parsons (1981) 17483 FRXa PL-KS 825 Brown & Parsons (1981) 17415 FRXb TMT-ILM 898 Stormer (1983) 17443 FRXb TMT -ILM 953 Stormer (1983) 17458 FRXb ~TMT-ILM 954 Stormer (1983) 17497 FRXb TMT -ILM 932 Stormer (1983) 17489 SRX TMT-ILM 979 Stormer (1983)

QRX = quartz-rich; FRX = feldspar-rich; SRX = spinel-rich; KS alkali feldspar; TMT = titanomagnetite; ILM = ilmenite.

Other abbreviations as for Table 1.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

Graham-Petrography of xenoliths, Tongariro

ticularly in Ohakune and Ngauruhoe 1954 lavas. Mineralogically, these consist either entirely of quartz (17436, 17468), or contain small pockets « 5% by volume) of plagioclase, pyroxene, and/or spinel (17885, 17476, 17463). Textures are coarsely granoblastic (17885) and some examples have a directional fabric (17436). Quartz varies in grain size from 0.05 to 1 mm and usually shows undu-lose extinction; plagioclase is anorthite, and pyrox-ene is typically salite or ferrosalite (Table 3). A 5 X 5 mm area in 17476 contains an intergrowth of an or-thite and wollastonite with minute inclusions of a more Fe rich wollastonite. Mn-rich hypersthene occurs in 17463. Some examples contain small, irregular areas of plagioclase and pyroxene sur-rounded by silica-rich glass, but this association is by no means common (as was suggested by Steiner 1958). Glass compositions are variable but are more calcic and less potassic than those in TYPE QRXa xenoliths (e.g., 17491) and plot above the granite minimum melt (Fig. 2).

Xenolith-host contact relationships The interface between TYPE QRXb xenoliths and host lava is of two types. Commonly, the xenolith edge meets host lava in a sharp regular line and mineral grains are smoothly broken across it

143

(17476). In other cases, a narrow reaction zone sur-rounds the xenolith (Fig. 1 D). This is made up of an irregular zone of clear, homogeneous, silica-rich glass and a zone next to host lava containing 0.5-1 mm clinopyroxene microlites. EPMA analyses of each phase are given in Table 3. Holgate (1954) reviewed the behaviour of quartzose xenoliths immersed in basaltic magma and concluded that diffusion into the xenolith margins of alkalis and water from the magma causes early production of a granitic melt at the contact. During cooling, the melt composition is progressively changed through further diffusion until finally it has a composition similar to that oflate-stage residua of the host lava. Sato (1975) experimentally examined the phenom-ena of glass-clinopyroxene coronas around quart-zose xenoliths and concluded that alkalis and Al

. diffused against their concentration gradients, pro-ducing a low-melting point liquid from the quartz. This behaviour was also reported by Watson & Jurewicz (1984) from experiments with oceanic tholeiite and granite at l250°C and 10 kbar. In the contact zone, interdiffusion of elements took place resulting in considerable uptake of potassium by the basaltic melt and eventual loss of Na from the basalt to the granite. Sato (1975) also explained the occurrence of clinopyroxene coronas; these result

Table 3 EPMA analyses of metaquartzite (TYPE QRXb) xenoliths from TVC lavas.

Cat. no.: 17463 17463 17463 17476 17476 17885 17885 17885 17885 17885 Phase: GLASS PL OPX PL WOLL CPX:j: GLASS:j: GLASS§ CPX§ MESII

SiO, 76.37 44.82 52.02 43.47 50.49 50.83 68.26 72.01 52.97 65.01 TiO, 0.15 0.00 0.00 0.00 0.00 0.17 0.63 0.63 0.17 1.14 AI,O, 12.39 35.85 1.06 35.12 0.00 0.68 10.64 10.01 0.68 12.22 FeO 1.65 0.00 26.02 0.33 9.62 14.66 4.64 4.04 9.65 7.33 MnO 0.00 0.00 1.07 0.00 0.83 0.79 0.00 0.00 0.33 0.20 MgO 0.56 0.00 18.91 0.00 0.23 7.86 0.74 0.61 16.41 1.26 CaO 3.52 18.20 1.01 20.35 38.36 23.75 2.38 1.71 19.59 3.65 Na,O 2.62 0.56 0.00 0.13 0.00 0.26 2.91 2.89 0.30 4.84 K,O 2.01 0.00 0.00 0.00 0.00 0.09 2.97 3.25 0.00 1.36 Total 99.27t 99.43 100.09 99.50 99.53 99.08 93.24t 95.14t 99.81 97.11 t

Oxygens 8 6 8 6 6 6 Si 2.07 1.97 2.03 2.00 1.98 1.96 Ti 0.00 0.00 0.00 0.00 0.01 0.01 Al 1.95 0.05 1.93 0.00 0.03 0.02 Fe 0.00 0.83 0.01 0.32 0.48 0.29 Mn 0.00 0.03 0.00 0.03 0.03 0.01 Mg 0.00 1.07 0.00 0.01 0.46 0.91 Ca 0.90 0.04 1.02 1.63 0.99 0.78 Na 0.05 0.00 0.01 0.00 0.02 0.02 K 0.00 0.00 0.00 0.00 0.00 0.00 Total 4.97 3.99 5.00 3.99 4.00 4.00

Description and abbreviations as for Tables 1 and 2. WOLL = wollastonite; MES = mesostasis. :j:xenolith; §reaction rim; II host lava.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


from high (Na + K)/ Al ratios of corona glass which increases the chemical potential of CaO causing cli-nopyroxene rather than orthopyroxene to crystal-lise. It is also possible that the xenolith acts merely as a reactive interface at which host meso stasis crystallises clinopyroxene, so being enriched in Si and K and depleted in Fe, Mg, and Ca. However, there are difficulties with this interpretation. Firstly, both Al and Na are lower in the xenolithic melt than in the host meso stasis, yet pyroxene is also low in these elements. Concomitant crystallisation of plagioclase is a solution to this paradox but no plagioclase-rich zones occur. Secondly, there is no observed change in glass composition between the xenolith margin and the pyroxene zone, yet pyrox-ene is confined to the region immediately adjacent to the host lava. It is more likely, therefore, that zoned contact relationships between quartz-rich . xenoliths and host lava result from extraction of partial melts from the xenolith which interact dif-fusively with host mesostasis.

Origin of TYPE QRX xenoliths The occurrence of quartzose and quartz-anorthite-salite assemblages in both TYPE QRXa and QRXb xenoliths suggests a common genesis. However, the widespread distribution of TYPE QRXb and the relative rarity of TYPE QRXa remains to be explained. Battey (1949) suggested that quartzose xenoliths in Ngauruhoe 1949 lava were thermally altered Tertiary sandy limestones (some large xen-oliths of Tertiary grit have recently been described from Mount Egmont; J. Collen pers. comm. 1984). The abundance of quartzose xenoliths (to the exclusion of most other types) in Ohakune lava make such an origin attractive, because the Ohak-une vents pass up through the thickest part of the Tertiary sequence in the vicinity of the TVC. How-ever, there is evidence to suggest that those xen-oliths were probably not derived from this near-surface source: (1) Luminescence petrography shows no differentiation of the quartz, as would be expected from diagenetic overgrowths on a detrital core derived from a metamorphic or igneous source; luminescence effects are those typical of high-tem-perature quartz (J. Collen pers. comm. 1984); (2) No significant quartz-sandstone or grit horizons occur in the North Island Tertiary sequence (Sug-gate et al. 1978); (3) Even if a suitable sandstone horizon did exist beneath TVC, we might expect that the derivative metaquartzite xenoliths would be accompanied by many metasiltstones and meta-calcarenites, both of which occur in the Tertiary sequence but are rare as xenoliths. A second pos-sibility is that quartz-rich xenoliths represent cherts from within the Mesozoic basement. Although dif-ficult to deny on chemical or petrographic grounds

(Roser 1983), the relative abundance of these xen-oliths is inconsistent with their rarity in the source terrane. A more probable origin is that proposed by Steiner (1958) who considered them to repre-sent relict bands of quartzofeldspathic gneiss, sepa-rated from denser micaceous layers by thermal expansion during rapid increase in temperature. The parental gneiss might be equivalent to lower parts of the Mesozoic basement where greywacke has changed to gneiss during high-grade regional metamorphism.

Feldspar-rich (TYPE FRX) xenoliths Feldspar-rich xenoliths occur in most lavas but only pyroxene-hornfels (TYPE FRXb) are abundant and these are ubiquitous as small grey-black inclusions. Biotite-schists (TYPE FRXa) are described only from Iwikau Member pyroclastics. One unusually large (12X8X4 cm) example, 17425, is strongly foliated and finely segregated into layers of con-trasting mineralogy: in the more felsic layers, pla-gioclase (an33ab6oor7) and sanidine (an3ab2sor69) coexist; in the more mafic layers, biotite, alumi-no us hypersthene (mfs6)' ilmenite, and pleonaste occur (Table 4). Sillimanite, frequently surrounded by sanidine, is always orientated parallel to the fol-iation (Fig. 3A, B). In 17483, sanidine contains needles of sillimanite or mullite and larger anhed-ral crystals of corundum (Fig. 3C). Elsewhere, interstitial areas of potassic glass occur (Fig. 3D). These mineralogical relationships are of the type: muscovite + quartz = mullite + sanidine/per-aluminous melt (Brindley & Maroney 1960). In 17425, alteration of muscovite has produced san-idine with sillimanite whereas in 17483 this trans-formation has proceeded further to produce a silica-poor, potassic melt (Fig. 2). Biotite is absent from the mineral assemblage of 17484, and orthopyrox-ene, sometimes as 1 mm long porphyroblasts, occurs with plagioclase and pleonaste. This, there-fore, probably represents an assemblage interme-diate between TYPES FRXa and FRXb (described below) and provides a genetic link between them.

Pyroxene-hornfels xenoliths (TYPE FRXb) are ubiquitous in TVC lavas. Steiner (1958), referred to dark grey feldspathic xenoliths in Ngauruhoe 1954 lava, which " ... represent re-melted, modified and recrystallised bands of quartzo-feldspathic gneiss.". Hacket (1985) indicated such xenoliths were probably amongst the most abundant types in Ruapehu lavas but, because their grey colour matches that of host lavas, they are less conspic-uous than other types. Most examples are elon-gated, angular, and small (2-50 mm X 1-10 mm). All exhibit layering parallel to their long axis and many show partings in that direction. Textures range from hypidiomorphic granular (17477) to

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


granoblastic (17419). Contacts are often sharp but irregular due to interpenetration of mineral grains (Fig. 4A) and margins are sometimes intensely deformed and recrystallised. Mineral assemblages are dominated by unzoned plagioclase (60-80%), whose compositions range from andesine to labra-dorite. In 17440 and 17419 (in particular), plagio-clase becomes progressively more clouded near the host-xenolith contact (Fig. 4A). Such clouding (Poldevaart & Gilkey 1954) often results from dif-fusion of Fe into plagioclase through channels pro-duced by unmixing and is most common in plagioclase of intermediate composition. It occurs when temperatures are held high for long periods of time in the presence of water (Smith 1974). In 17419, clouding is probably related to thermal alteration near the contact caused by incorporation of the xenolith in the lava. EPMA analysis of the clouded marginal plagioclase reveals fine-scale variation in composition from anS3 to an7b but no significant change in FeO content. Layering in TYPE FRXb xenoliths consists of 1-2 mm wide segregations of subhedral, subophitic, green-pink pleochroic hypersthene, titanomagnetite, and ilmenite (Fig. 4B). Almandine garnet and cordierite occur in 17873 (Table 5), but not in other examples. Minor minerals are quartz, biotite (low-pressure alteration of hypersthene?), and zircon. Pleonaste is ubiquitous and occurs as rounded, 0.5-1 mm

145

diameter porphyroblasts sometimes with a corun-dum (sapphire) core (Fig. 4C). Ilmenite and titan-omagnetite usually coexist giving equilibration temperatures near 950°C (Table 2) and equivalent oxygen fugacities of about 10- 1 t, close to the Ni-NiO buffer (Stormer 1983). Several TYPE FRXb xenoliths contain brown, interstitial glass. Some-times, as in 17458, this originates from host lava (Steiner 1958) having a dacitic composition and containing numerous rounded 11m-sized pyroxene microlites (as inferred from chemistry). Otherwise, glass of a more siliceous composition occurs in fractures (17444) and in small (lOX 10 mm) vugs (17415) containing euhedral zoned plagioclase and hypersthene up to I mm long (Fig. 4D). These pla-gioclases have cloudy, glass-filled interiors and the orthopyroxenes are more Fe rich than those of the surrounding xenolith. Such assemblages probably· crystallised from melt-rich pockets prior to incor-poration of the xenolith in host lava and, subse-quently, re-equilibrated with the last remaining liquid.

Textures, mineral habits, and geothermometry suggest that TYPE FRXb xenoliths were recrystal-ised under pyroxene hornfels or granulite facies conditions. Steiner (1958) noted that, occasionally, hypersthene phenocrysts protruded from host lava into xenolith interiors and that the crystal margins were markedly corroded, indicating that the xen-

Table 4 EPMA analyses of biotite-schist (TYPE FRXb) xenoliths from TVC lavas.

Cat. no.: 17425 17425 17425 17425 17425 17425 17483 17483 17483 17483 Phase: PL KS OPX BIO ILM PLEON PL KS GLASS BIO

Si02 59.97 64.69 45.74 34.72 0.11 0.Q7 61.66 65.88 64.41 34.07 Ti02 0.00 0.00 0.20 5.08 51.42 0.29 0.00 0.00 0.25 3.47 A120 J 25.40 19.41 8.06 17.83 0.15 55.63 25.10 19.49 18.32 18.91 FeO 0.15 0.00 29.29 17.13 43.26* 35.50* 0.19 0.19 4.05 19.32 MnO 0.00 0.00 0.93 0.16 1.34 0.34 0.00 0.00 0.00 0.00 MgO 0.00 0.00 14.31 11.21 2.08 6.55 0.00 0.00 0.61 9.29 CaO 7.01 0.50 0.15 0.00 0.00 0.00 6.02 0.36 0.39 0.10 Na20 6.92 3.12 0.00 0.64 0.00 0.00 7.38 3.69 3.70 0.83 K20 1.26 11.69 0.00 9.04 0.00 0.00 1.19 10.92 6.27 8.60 Total 100.71 99.41 98.68 95.61 t 98.36 98.38 101.54 100.53t 98.00t 94.59

Oxygens 8 8 6 22 3 4 8 8 22 Si 2.67 2.96 1.78 5.22 0.00 0.00 2.71 2.97 5.23 Ti 0.00 0.00 0.01 0.58 0.98 0.0l 0.00 0.00 0.40 Al 1.33 1.05 0.37 3.16 0.01 1.85 1.30 1.04 3.43 Fe 0.01 0.00 0.96 2.16 0.91 . 0.85 0.01 0.0l 2.48 Mn 0.00 0.00 0.03 0.02 0.03 0.01 0.00 0.00 0.00 Mg 0.00 0.00 0.84 2.51 0.08 0.28 0.00 0.00 2.13 Ca 0.33 0.03 0.0l 0.00 0.00 0.00 0.28 0.02 0.02 Na 0.60 0.28 0.00 0.19 0.00 0.00 0.63 0.32 0.25 K 0.Q7 0.68 0.00 1.74 0.00 0.00 0.Q7 0.63 1.69 Total 5.00 5.00 4.00 15.58 2.00 3.00 5.00 4.99 15.63

Description and abbreviations as for Tables I, 2, and 3.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


Fig.3 BEl photographs of assemblages in feldspar-rich xenoliths. Abbreviations are as follows: ks = alkali feldspar (sanidine), bi = biotite, sill = sillimanite, pi = plagioclase, gl = glass. A Assemblage of TYPE FRXa xenolith 17425. Central area is sillimanite (dark grey), sanidine (light grey), and biotite (white); outer areas are plagioclase (medium grey) and biotite. B Close-up of central area of (A). C Assemblage of TYPE FRXa xenolith 17483. San-idine contains needles of mullite or sillimanite and patches of corundum (dark grey). D Glassy pocket in 17483. Scale bars for A, B, D = 100 ~m; for C = 10 ~m.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

Graham-Petrography of xenoliths, Tongariro 147

Fig. 4 BEl photographs of assemblages in feldspar-rich xenoliths. Abbreviations are as follows: xen = xenolith, opx = orthopyroxene, pi = plagioclase, op = opaques. A Contact between TYPE FRXb xenolith 17419 and host lava. Plagioclase in the xenolith is mildly zoned (darker = Na rich) giving a "clouded" appearance; crystal edges penetrating host lava are euhedral and compositionally similar to host microphenocrysts. B Assemblage of TYPE FRXb xenolith 17443 showing segregation of hypersthene, magnetite, ilmenite, and pleonaste in a matrix of granoblastic plagio-clase. C Pleonaste porphyroblast in TYPE FRXb xenolith 17458. Core is corundum (sapphire) with inclusions of plagioclase and pleonaste; granitic glass is everywhere interstitial to plagioclase. D Vug assemblage of TYPE FRXb xenolith 17415. Scale bars for A, C, D = 100 ~m; for B = 1000 ~m.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


olith was still liquid at the time the lava was crys-tallising. Such a relationship was rarely observed in the examples described here, although the pres-ence of interstitial glass, granulitic textures, and relict layering all suggest that these xenoliths were at some stage subjected to temperatures at or near the melting point of the assemblages observed.

Origin of TYPE FRX xenoliths Pyroxene-hornfels xenoliths (TYPE FRXb) could be fine-grained plagioclase cumulates resulting from fractional crystallisation of andesitic magma. How-ever, the typically metamorphic textures, relict fol-iations, and high P-T mineral assemblages do not support such an origin. In addition, most exampies have higher 87Sr/86Sr ratios (Table 6) than their host lavas (range = 0.70480-0.70620; Graham & Hack-ett 1987) indicating that they must be derived from a different source.

An alternative origin linking TYPE FRX xen-oliths to gneissic basement seems more plausible and certainly more attractive. Read (1935) described feldspathic xenoliths ("orthonorites") in the Haddo House district of Aberdeenshire, Scotland. These were fine-grained plagioclase-rich rocks containing hypersthene as spongy, sometimes poikilitic, masses, and were identical to "micronorite" xen- . oliths described by Read (1966) at Mill of Bod dam.

The descriptions and published bulk-rock chem-istries are strikingly similar to some TYPE FRXb xenoliths. Read interpreted the mineral assem-blages as resulting from assimilation of pelitic country rock (andalusite"cordierite schist) by oli-vine gabbro. The pelitic rocks were first desilicated and then the aluminous "restites" reacted with the contaminated magma to reprecipitate plagioclase and hypersthene (producing orthonorite zones). Micronorite zones were interpreted as relict alu-minous restites which have re-equilibrated, in situ, with the host lava. Steiner (1958) considered that feldspathic xenoliths in Ngauruhoe 1949 and 1954 lavas represented remelted feldspathic bands of gneiss. When such bands are broken up and scat-tered throughout a magma, they melt and interact with it. The resulting syntectic melt remains essen-tially feldspathic in composition and on crystallis-ation gives rise mainly to feldspar. This is supported by (1) granulitic texture, (2) occurrence of relict high P-T minerals (garnet, cordierite), and (3) high equi-libration temperatures of coexisting Fe-Ti oxides (900-1000°C). However, the dominantly metamor-phic mineralogy and presence of relict foliation suggest a slightly different genesis, more in line with Read's interpretation, involving progressive alter-ation of dominantly feldspathic layers of gneiss. TYPE FRXb xenoliths are thus considered restites

Table 5 EPMA analyses of pyroxene-hornfels (TYPE FRXb) xenoliths from TVC lavas.

Cat. no.: 17419 17419 17443 17443 17443 17443 17443 17458 17458 17458 17458 17458 Phast:: PL OPX PL OPX ILM TXT PLEON PL OPX GLASS PLEON CO

Si02 52.59 49.08 56.67 51.00 0.00 0.00 0.00 52.75 52.52 65.85 0.00 0.00 Ti02 0.00 0.65 0.00 .0.18 48.69 15.59 0.36 0.00 0.30 1.38 0.62 0.89 A120 3 29.44 7.22 26.96 3.23 0.40 8.04 58.41 29.20 3.18 13.53 56.78 99.39 FeO 0.43 16.29* 0.37 20.71* 45.08* 69.47 29.72 0.24 15.63* 6.40 31.77 0.39 MnO 0.00 0.42 0.00 0.55 0.54 0.37 0.46 0.00 0.38 0.00 0.27 0.00 MgO 0.08 25.32 0.00 23.20 4.00 3.17 9.92 0.07 27.76 2.00 11.49 0.00 Cao . 12.33 0.41 9.44 0.26 0.00 0.00 0.00 12.35 0.53 1.56 O.{)O 0.00 Na20 4.39 0.00 5.84 0.00 0.00 0.00 0.00 4.71 0.00 2.46 0.00 0.00 K20 0.23 0.00 0.52 0.00 0.00 0.00 0.00 0.27 0.00 2.96 0.00 0.00 Total 99.49 99.39 99.80 99.73 98.71 96.64 98.87 99.59 100.30 96.30t 100.92 100.67

Oxygens 8 6 8 6 3 4 4 8 6 4 Si 2.40 1.78 2.55 1.90 0.00 0.00 0.00 2.40 1.88 0.00 Ti 0.00 0.02 0.00 0.01 0.90 0.42 0.01 0.00 0.01 0.00 Al 1.58 0.31 1.43 0.14 0.01 0.34 1.89 1.57 0.13 1.80 Fe 0.02 0.49 0.01 0.65 0.93 2.07 0.68 0.01 0.47 0.72 Mn 0.00 0.01 0.00 0.02 0.01 0.01 0.01 0.00 0.01 0.01 Mg 0.01 1.37 0.00 1.29 0.15 0.17 0.41 0.01 1.48 0.00 Ca 0.60 0.02 0.46 0;01 0.00 0.00 0.00 0.60 0.02 0.00 Na 0.39 0.00 0.51 0.00 0.00 0.00 0.00 0.42 0.00 0.00 K 0.01 0.00 0.03 0.00 0.00 0.00 0.00 0.02 0.00 0.00 Total 5.01 4.00 4.99 4.00 2.00 3.00 3.00 5.03 4.00 3.00

Description and abbreviations as for Tables 1-4. CO = corundum.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


after extraction of melt from biotite schist (TYPE FRXa) but are not totally remelted as suggested by Steiner (1958).

Spinel-rich (TYPE SRX) xenoliths Spinel-rich xenoliths, although relatively low in total abundance, occur in most lava formations. Typical examples are small (20-30 mm), strongly foliated, and are themselves sometimes included in cognate cumulate inclusions (as in 17888). Mineral assem-blages are characterised by high abundances of Fe-oxides, particularly titanomagnetite, ilmenite, and/or pleonaste (Table 7). Narrow millimetre-wide) layers rich in titaniferous biotite, poikilitic aluminous hypersthene, and spinel alternate with layers rich in calcic plagioclase (an7Can96)' Olivine is a minor component in several examples and

149

ranges in composItIon from f061 (17410) to f073 (17888). In 17489, coexisting titanomagnetite and ilmenite give an equilibration temperature of 979Q C (Table 2). Overall grain size is small (av. <0.5 mm) and textures are typically granoblastic. Xenolith-host contacts are sometimes marked by growth of clinopyroxene on the host side but are otherwise regular and· sharp.

Origin of TYPE SRX xenoliths The origin of TYPE SRX xenoliths is unclear owing to their somewhat unusual composition, small size, and comparative rarity. One possible source are Torlesse terrane metabasites which have similar mineralogies and bulk-rock chemistries (Roser 1983) but these are relatively rare. Alternatively, the xenoliths could be finely laminated mafic

Table 6 Bulk-rock chemical compositions of xenoliths from TVC lavas.

VUW: 17492 17485 17463 17436 17425 17483 17419 17443 17497 17410 TYPE: QRXa QRXa QRXb QRXb FRXa FRXa FRXb FRXb FRXb SRX

Major elements (wt %) Si02 70.9 72.5 85.3 98.0 50.0 49.4 49.3 50.8 51.6 36.3 Ti02 0.5 0.6 0.0 0.0 l.l 1.8 1.5 1.2 1.0 4.3 Al20 3 13.8 13.9 8.9 0.8 25.6 2.59 25.6 24.6 25.1 20.3 Fe20 3 0.6 0.7 0.1 0.1 3.7 3.5 3.9 4.2 3.5 10.8 FeO 2.9 3.3 0.5 0.3 3.7 3.5 3.9 4.2 3.5 10.8 MnO 0.2 0.1 0.0 0.0 0.1 0.1 0.1 0.1 0.1 0.5 MgO l.l 1.3 0.1 0.2 3.0 2.7 2.5 2.9 2.5 9.0 CaO 9.3 4.5 3.8 0.3 4.3 7.7 8.9 6.7 8.6 5.7 Nail 0.6 2.2 l.l 0.2 4.6 3.6 3.6 4.4 3.6 1.3 K20 0.1 0.9 0.2 0.1 3.8 1.5 0.6 0.9 0.2 0.6 P2O, 0.2 0.2 0.0 0.0 0.3 0.4 0.2 0.2 0.1 0.3 LOI 0.9 1.5 0.2 0.0 0.7 1.2 0.8 0.2 1.2 1.4 C.1.P. W. norm Q 46.6 43.1 69.5 95.6 2.8 3.2 2.8 1.3 co 1.6 6.9 5.1 3.4 4.8 3.2 7.8 or 0.4 5.2 1.0 0.4 22.3 8.9 3.4 5.4 1.4 3.7 ab 5.0 18.9 9.4 1.6 34.0 30.7 30.6 37.1 32.2 11.0 an 34.8 21.1 18.7 l.l 19.4 35.9 42.6 31.5 41.9 26.4 hy 2.6 7.8 1.0 0.9 7.5 8.0 9.7 11.2 5.1 Other 10.8 2.3 0.3 0.2 17.2 9.2 8.4 8.7 7.2 26.1 Trace elements (ppm) Ba 31 174 13 3 1709 587 783 878 292 360 Cr 31 33 6 4 59 106 73 87 73 163 Rb <2 42 4 3 144 48 16 18 2 40 Sr 605 260 204 26 708 604 896 545 448 342 V 69 82 5 16 179 248 221 208 208 564 Zr 223 135 8 <2 289 361 473 220 126 236 Rb/Sr 0.002 0.163 0.Dl8 ; 0.100 0.203 0.079 0.050 0.098 0.014 0.115 I 0.70700 0.70890 0.70801 0.70611 0.70662 0.70616 0.70570 0.70800 0.70702 0.70830

Major element analyses normalised to 100% volatile-free (volatile loss (LOI) is given for comparison). Q = quartz, co = corundum, or = orthoclase, ab = albite, an = anorthite, hy = hypersthene. Others include diopside, nepheline, olivine, magnetite, ilmenite, and apatite. I = 87Sr/86Sr, Fe20JFeO = 0.2 (TYPE QRX), = 0.5 (TYPE FRX).

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


cumulates, but the high Sr isotopic ratios of some samples (e.g., 17410 = 0.70800) precludes a cog-nate origin (see previous discussion). Although strong supportive evidence is presently lacking, the presence of foliation, granulitic texture, and the mineralogy suggest that TYPE SRX xenoliths might represent disaggregated micaceous layers of gneiss. The single, high, ilmenite-magnetite equilibration temperature (979°C) indicates high-grade reconsti-tution, but the occurrence of abundant biotite and the lack orany partial melt is difficult to explain. Tindle & Pearce (1983) described metagreywacke xenoliths from the Loch Doon granitic intrusion of Scotland comprising recrystallised biotite, actino-litic amphibole, and green and brown spinel. Traces of brown glass occurred along cleavage planes of biotite. In this association, biotite is considered to be residual (with amphibole and plagioclase) after extraction of melt. Alternatively, the biotite may be caused by high Ti02 contents (6-7%) which tend to stabilise the mineral at high temperatures (Grapes 1986). If TYPE SRX xenoliths are so derived from gneiss, their relative rarity is probably due to two factors: (1) micaceous layers of gneiss constitute only 10-20% of the bulk-rock, and (2) the xenoliths have relatively high specific gravity (compared to quartz-rich and feldspar-rich xenoliths) causing them to sink preferentially. This would also explain

the association of some TYPE SRX xenoliths with cumulate nodules.

METAMORPHIC CONDITIONS AS DEDUCED FROM MINERAL ASSEMBLAGES Some xenoliths contain mineral assemblages which allow P-T conditions of metamorphism to be esti-mated (Table 2). Temperatures are high, ranging from 800 to 1000°C, being consistent with the pres-ence of sillimanite (and the absence of low-tem-perature Al2SiOs polymorphs), high Ti02 contents of biotite, absence of muscovite, and intermediate plagioclase compositions. Although accurate pres-sure estimates are not possible because of inappro-priate mineral assemblages, the widespread occurrence of orthopyroxene and relative rarity of garnet (though often composition related) indicate low to moderate pressures. Cordierite plus alman-dine in one TYPE FRXb xenolith restricts that assemblage to regional hypersthene zone, granulite facies (Winkler 1979) and suggests pressure limits of 4-7 kbar (Currie 1971). The absence of amphi-bole indicates largely anhydrous metamorphic con-ditions, and this is further supported by low-volatile contents of partial granitic melts.

Table 7 EPMA analyses of spinel-schist (TYPE SRX) xenoliths from TVC lavas.

Cat. no.: 17410 17410 17410 17410 174lO 174lO 17489 17489 17489 17489 Phase: PL PL OL OPX BIO PLEON OPX BIO ILM TMT

Si02 44.38 48.71 35.84 50.09 38.14 0.08 52.94 36.12 0.00 0.14 Ti02 0.00 0.00 0.00 0.42 6.99 1.24 0.27 6.94 46.39 13.08 Al20 3 35.77 31.87 0.00 3.60 13.68 51.39 2.77 16.03 0.37 6.92 FeO 0.20 0.43 32.91 23.86 8.67 34.91* 16.31 12.13 45.57* 70.91* MnO 0.00 0.00 1.16 1.09 0.00 0.66 0.74 0.00 0.49 0.51 MgO 0.06 0.00 29.61 19.78 18.01 lO.04 26.11 15.48 4.98 3.56 CaO 19.26 15.16 0.00 0.56 0.00 0.00 0.43 0.12 0.00 . 0.00 Na20 0.49 2.76 0.00 0.00 0.85 0.00 0.00 0.93 0.00 0.00 K20 0.00 0.27 0.00 0.00 9.34 0.00 0.00 8.88 0.00 0.00 Total 100.16 99.20 99.52 99.40 95.68t 98.33 99.57 96.63t 97.80 95.11

Oxygens 8 8 4 6 22 4 6 22 3 4 Si 2.05 2.25 0.99 1.89 5.53 0.00 1.92 5.26 0.00 om Ti 0.00 0.00 0.00 om 0.76 0.03 om 0.75 0.86 0.36 Al 1.94 1.73 0.00 0.16 2.34 1.71 0.12 2.76 om 0.29 Fe 0.01 0.02 0.76 0.76 1.05 0.82 0.50 1.48 0.94 2.13 Mn 0.00 0.00 0.03 0.04 0.00 0.02 0.02 0.00 0.01 0.02 Mg om 0.00 .1.22 1.12 3.89 0.42 1.41 3.36 0.18 0.19 Ca 0.95 0.75 0.00 0.02 0.00 0.00 0.02 0.02 0.00 0.00 Na 0.04 0.25 0.00 0.00 0.24 0.00 0.00 0.26 0.00 0.00 K 0.00 0.02 0.00 0.00 1.73 0.00 0.00 1.65 0.00 0.00 Total 5.00 5.02 3.00 4.01 15.54 3.00 4.00 15.54 2.00 3.00

Description and abbreviations as for Tables 1-5. OL = olivine.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


DISCUSSION It has been suggested in the earlier discussions on probable origins that quartz-rich, feldspar-rich, and spinel-rich xenoliths ~rom TVC l~vas a~e each derived from metasedImentary gneISS. It IS prob-able from certain aspects of geochemistry and iso-topi~ composition and by association, that Mesozoic greywacke, in particular Torlesse te~ane greywacke, is the ultimate source of the xenohths, but the possibility of other protoliths cannot be positively· dismissed. However, geophysical evi-dence (see Graham & Hackett (1987) for a sum-mary) indicates thin .crust «20 .km) be~eath t~e TVC and does not pomt to an anCIent granIte-gneIss terrane underlying the region. Indeed, granitic tex-tures or compositions are absent from the xenolith suite and Sr isotopic ratios (Table 6) are much lowe~ than would be expected for Paleozoic (or older) granites or gneiss (Aronson 1965; Adams 1975).

Sedimentary basement in the central North Island is either Waipapa terrane or Torlesse terrane Mesozoic greywacke (Graham 1985a, b). Because of the mineralogical and chemical transformations which must accompany high-grade metamorphism of greywacke, it is difficult, if not impossible, to unequivocally determine which of these basement types is the main source of the xenoliths. However, Torlesse is favoured over Waipapa for the follow-ing reasons: (1) Unmodified xenoliths oflow-grade Torlesse greywacke occur in some TVC lavas .b~t no Waipapa equivalents have been noted; (2) vltn-fied metagreywacke xenoliths from Ngauruhoe and Pukeonake have bulk-rock chemistries and Sr iso-topic compositions similar to Torlesse terrane greywacke (Graham & Gra~s in press); (3) T~PE QRXa xenoliths (garnet SChISt) show geochemIcal similarities with Torlesse greywacke, at least for the more immobile elements (Fig. 5 and 6); and (4) 87Sr/86Sr of most of the xenoliths is greater than 0.70600 (Fig. 7) which, given that that is probably a minimum value because of partial equilibration with host magma, is higher than most Waipapa ter-rane rocks·.

A preferred genetic model for the derivation of TYPES QRX, FRX, and SRX xenoliths is as follows. The gneissic precursor is a metamor-phosed Torlesse terrane greywacke protolith ~nd comprises quartz-rich, feldspar-rich, and mIca-ceous segregations which are mutually incoherent when subjected to high temperatures and which separate because of thermal expansion. Partial melts

*A single (1180 value for 17485, the xenolith whose che~istry most closely resembles that of a greywacke, IS 12.13%0 and falls well within the Torlesse field (Blattner & Reid 1982).

151

are produced and extracted from them, caus~ng contamination of host magmas and altenng (slightly) their bulk chemical compositions. Because of density contrasts, the quartz-rich and, to a less~r extent, feldspar-rich restites remain "floating" m the magma whereas the heavier micaceous restites "sink".

Gribble & O'Hara (1967) discussed the interac-tion of pelitic materials with basic magma, par-ticularly with reference to the Haddo House xenolith suite (Read 1935), and pointed out that the apparent enrichment in alumina and depletion in silica of residual xenoliths does not necessarily imply metasomatic interchange with the magma but is a natural result of the formation and separation of a partial melt fraction from an original schist at an appropriate temperature. This conclusion was also arrived at by McRae & Nesbitt (1980) who presented mass balance calculations documenting progressive bulk chemical changes in metagrey-wacke and metapelite after separation of incre-ments of granite minimum melt. They noted. that, during partial melting, enrichment of Fe relative to Mg, and strong absorption of water i~ the melt, leaves the restite with increasing proportIons ofMg-rich minerals such as cordierite. Extraction of a granitic melt from metapelite results in an increase in the Al content of the restite, whereas, for grey-wacke, extraction of an alkali granite melt is required to have the same effect. These results are similar to those observed here. However, because the "real" starting compositions of each compo-nent in the model are unknown, it is not possible to mass balance the process. The present data show that predicted chemical changes (i.e., increases in AI, Fe, Mg, and Ca and decreases in Si02 and K) have occurred between felsic and micaceous parents and their restites (Fig. 5 and 6). For highly quart-zose parents, however, Si02 is likely to increase,as mica and feldspar are melted and extracted (Gnb-ble & O'Hara 1967). All xenoliths show a greater than expected increase in Ca and Sr concentration, perhaps as a result of back-diffusion during melt extraction. If so, then this may have lead to at least partial equilibration of Sr isotopes, explaining the discrepancy between some xenolith ratios and those of the probable source rocks (Fig. 7).

The preceding discussion has dealt only with the question of progenesis of metasedimentary xeno-liths in TVC lavas and has not attempted to address the question of crustal assimilation (of host lavas). As previously noted, xenolithic glass compositions are highly variable, and, whereas some do approx-imate to a minimum granitic melt (Fig. 2), others do not and appear to reflect disequilibrium partial melting on a scale smaller than a typieal ~andspecimen (see Grapes 1986). However, despIte these uncertainties, such melts provide important clues

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


40 SO 60 70 80 90 10 40 60 70 80 90

4 Ti02 At 203 32 6 28 •

--0)0 24 6 6

~ 0 6 66 20 ..... Cl 6 16 ~ • 0 14

. ~

~ 0 8 00

4

-0 0-

16 19·5 FeO r 66 MgO 8

14 6 7

12 6

10 66 S

8 • 4

6 0

'f..~ 0 4

2

-0 0-

16 CoO No2O 8

14

12 6

10 5 ~ 0 0 a 6 0 • ·0 ..... 00 • 4 C7l 8 . 0 tl °O ~

6 0 • 0 •

CD • 4 • 0 ~6 2 0

-0 0-K20 40 50 60 70 80 90

8 wgt % Si 0 2

6 XENOUTHS MESOZOIC BASEMENT

• TYPE ORXO ~ TORLESSE 5 o TYPE ORXb

• TYPE ~RXo ffi] WA I PAPA 4 0 TYPE ~RX • • 6 TYPE SRX

6

0 • • • -0 • 40 50 60 70 80 90 100

wgt % Si O2 Fig.S Major element versus Si02 variation diagrams for TVC metasedimentary xenoliths. Also shown are fields for Waipapa and Torlesse terrane greywackes (Graham 1985 a, b).

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

Graham-Petrography of xenoliths, Tongariro 153

100 40 50 60 70 80 90 100

1600 Ba Rb 400

1400

1200 300

E 1000 0. 0 200 0. 800 0

600 6 t • • 400 6 100

6 0 0 200 6 0 • • 6 • • • 66 ~ -0 0-

0 400 1°) Zr Y 80

70

300 6 60 • 50 6 00

200 6 40 0 66. 6 " 0 30

0 0 100 ~ 20

6 00 10

-0 0 0-0

800 Sr Sc 80

• 6 70 700

600 •

" • 6 60

$0 0 • 500 6 50

• 0 6 400 40

6 6 30 300 66 0 •• 20 200 •

~ 100 10 0 -0 0-

6 6 400 56' 6 852 6 Cr 200 V

6 6

300 150

E 6 0. 200 • 100 0.

• ~o 00

100 • 0 50

-0 0-40 50 60 70 80 90 100 40 50 60 70 80 90 100

wgt % Si0 2 wgt%S i02

Fig. 6 Trace element versus SiO, variation diagrams for TVC metasedimentary xenoliths. Fields and symbols as for Fig. 5.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


·71200

• 11100

·71000

.. "" .70900 '" co

i:-... "" t::. co ·70800

o

•

•

o o

••

•

o

B

' 70~0 ~~~ __ ~~ __ ~~ __ ~~ __ L--L~~~~ ____ ~ __ ~~~G-~ ____ -L ____ ~ ____ ~ __ ~

o 1 ·000 '007

87Rb /,6Sr

Fig.7 87Sr/86Sr versus 87Rb/86Sr (A) and '/Sr (B) for TVC metasedimentary xenoliths. Fields and symbols as for Fig. 5.

as ' to the composltlOn of the crustal assimilant involved in the generation of TVC lavas.

ACKNOWLEDGMENTS Many thanks to Drs P. Blattner, D. Seward (Institute of Nuclear Sciences), J. W. Cole, and J. A. Gamble (Victoria University of Wellington) for helpful comments and dis-cussions and to Professor D. S. Coombs (University of Otago) and Dr C. J. Adams (Institute of Nuclear Sciences) for constructive criticism of the text.

REFERENCES Adams, C. J. D. 1975: Discovery of Precambrian rocks

in New Zealand: Age relations . of the Greenland Group and Constant Gneiss, West Coast, South Island. Earth and planetary science letters 28 : 98-104.

Aronson, J. L. 1965: Reconnaissance rubidium-strontium geochronology of New Zealand plutonic and meta-morphic rocks. New Zealand journal of geology and geophysics 8 : 401-423.

Battey, M. H. 1949: The recent eruption of Ngauruhoe. Records of the Auckland (N.z.) Institute 3(6): 387-395.

Blattner, P.; Reid, F. W. 1982: The origin of lavas and ignimbrites of the Taupo Volcanic Zone, New ' Zealand, in the light of oxygen isotope data. Geo-chimica et cosmochimica acta 46 : 1417-1429.

Brindley, G. W.; Maroney, D. M. 1960: High temperature reactions of clay mineral mixtures and their ceramic properties, II. Journal of the American Ceramic Society 43: 511-516.

Brown, W. L.; Parsons, I. 1981: Towards a more practical two-feldspar geothermometer. Contributions to mineralogy and petrology 76: 369-377.

Cole, J. W. 1978: Andesites of the Tongariro Volcanic Centre, North Island, New Zealand. Journal ofvol-canology and geothermal research 3: 121-153.

---- 1984: Taupo-Rotorua depression: an ensialic marginal basin of North Island, New Zealand. In: Kokelaar, B. P.; Howells, M. F. ed. Marginal basin geology: volcanic and associated sedimentary and tectonic processes in modem and ancient marginal basins. Special publication of the Geological Society of London.

Currie, K. L. 1971: Coexisting cordierite and garnet. Con-tributions to mineralogy and petrology 33: 215-226.

Ellis, D. J. 1980: Osumillite-sapphirine-quartz granulites from Enderby Land, Antarctica: P-T conditions of metamorphism, implications for garnet-cordierite equilibria and the evolution of the deep crust. Contributions to mineralogy and petrology 74: 201-210.

Ewart, A.; St;pp, J. J. 1968: Petrogenesis of the volcanic rocks of the central North Island, New Zealand, as indicated by a study of87Sr/86Sr ratios, and Sr, Rb, K, U, and Th abundances. Geochimica et cosmo-chimca acta 32: 699-736.

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


Ferry, J. M.; Spear, F. S. 1978: Experimental calibration of the partitioning of Fe and Mg between biotite and garnet. Contributions to mineralogy and petrology 66: 113-117.

Graham, I. J. 1985a: Rb-Sr geochronology and geochem-istry of Torlesse metasediments from the central North Island, New Zealand. Chemical geology (Isotope Geoscience Section) 52: 317-331.

---- 1985b: Petrochemical and Sr isotopic studies of lavas and xenoliths from Tongariro Volcanic Centre - Implications for crustal contamination of calc-alkaline magmas. Unpublished Ph.D. the-sis, lodged in the Library, Victoria University of Wellington, New Zealand.

Graham, I. J.; Grapes, R. H. in press: The origin and petrogenesis of vitrified metagreywacke xenoliths, Tongariro Volcanic Centre, New Zealand.

Graham, I. J.; Hackett, W. R 1987: Petrology of calc-alkaline lavas from Ruapehu volcano and related vents, Taupo Volcanic Zone, New Zealand. Jour-nal of petrology 28(3): 531-567.

Grapes, R. H. 1986: Melting and thermal reconstitution of pelitic xenoliths, Wehr volcano, East Eifel, West Germany. Journal of petrology 27(3) : 43-96.

Gribble, C. D.; O'Hara, M. J. 1967: Interaction of basic magma with pelitic materials. Nature 214 (5094): 1198-1201.

Grieve, R. A. F.; Fawcett, J. J. 1974: The stability of chloritoid below 10 kb PH o. Journal of petrology 15(1): 113-139. 2

Hackett, W. R. 1985: Geology of Ruapehu Volcano. Unpublished Ph.D. thesis, lodged in the Library, Victoria University of Wellington, New Zealand.

Harley, S. L. 1984a: An experimental study of the par-titioning of Fe and Mg between garnet and ortho-pyroxene. Contributions to mineralogy and petrology 86: 359-373.

---- 1984b: The solubility of alumina in orthopy-roxene coexisting with garnet in FeO-MgO-AI20r Si02 and CaO-FeO-MgO-AI,orSi02. Journal of petrology 25(3) : 665-696.

Holgate, N. 1954: The role of liquid immiscibility in igneous petrogenesis. Journal of geology 62 : 439-480.

McRae, N.; Nesbitt, H. 1980: Partial melting of common metasedimentary rocks: a mass balance approach. Contributions to mineralogy and petrology 75: 21-26.

Poldevaart, A.; Gilkey, A. K. 1954: On clouded plagio-clase. American mineralogist 39: 75-91.

Read, H. H. 1935: The gabbros and associated xenolithic complexes of the Haddo House district, Aber-deenshire. Quarterly journal of the Geological Society of London 91: 591-639.

155

---- 1966: An orthonorite containing spinel xen-oliths with late diaspore at Mill of Boddam, Insch, Aberdeenshire. Proceedings of the Geological Asso-ciation 77: 65-78.

Roser, B. P. 1983: Comparative studies of copper and manganese mineralisation in the Torlesse, Wai-papa and Haast Schist terranes, N.Z. Unpublished Ph.D. thesis, lodged in the Library, Victoria Uni-versity of Wellington, New Zealand.

Sato, H. 1975: Diffusion coronas around quartz xeno-crysts in andesite and basalt from Tertiary Vol-canic Region in northeastern Shikoku, Japan. Contributions to mineralogy and petrology 50 : 49-64.

Schreyer, W. 1965: Zur stabilitat des ferrocordierits. Con-tributions to mineralogy and petrology 11: 297-322.

Smith, J. V. 1974: Feldspar minerals. I. Crystal structure and physical properties. Heidelberg, Springer-Ver-lag. 627 p.

Steiner, A. 1958: Petrogenetic implications of the 1954 Ngauruhoe lava and its xenoliths. New Zealand journal of geology and geophysics 1 : 325-363.

Stormer, J. C. Jun. 1983: The effects of recalculation on estimates of temperature and oxygen fugacity from analyses of multicomponent iron-titanium oxides. American mineralogist 68: 586-594.

Suggate, R. P.; Stevens, G. R; Te Punga, M. T. ed. 1978: The geology of New Zealand. Vol. I. Wellington, Government Printer. 343 p.

Tindle, A. G.; Pearce, J. A. 1983: Assimilation and partial melting of continental crust: evidence from the mineralogy and geochemistry of autoliths and xen-oliths. Lithos 16: 185-202.

Tuttle, 0. F.; Bowen, N. L. 1958: Origin of granite in the light of experimental studies in the system NaAlSiJO.-KAlSi30.-Si02-H20. Memoirs of the Geological Society of America 74.

von Platen, H. V. 1965: Experimental anatexis and gen-esis of migmatites. Pp. 203-216 in: Pitcher, W. S.; Flinn, G. W. ed. Controls of metamorphism. Edin-burgh and London, Oliver and Boyd.

Watson, E. B.; Jurewicz, S. R 1984: Behaviour of alkalies during diffusive interaction of granitic xenoliths with basaltic magma. Journal of geology 92(2): 121-131.

White, A. J. R.; Chappell, B. W. 1977: Ultrametamorph-ism and granitoid gneisses. Tectonophysics 43 : 7-22.

Winkler, H. F. G. 1979: Petrogenesis of metamorphic rocks. New York, Springer Verlag.

(Appendix lover page)

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


APPENDIX 1 Petrographic descriptions of metasedimentary xenoliths from Tongariro Volcanic Centre lavas.

VUW 17410* Wahianoa Formation lava from Wahi-anoa Valley, Mount Ruapehut (T20/335072);. Grey, fine-grained schist (TYPE SRX) (60X40X 15)§ containing bytownite-anorthite + hypersthene + olivine + biotite + pleonaste (CHEMISTRY, EPMA, 87Sr/86Srll).

VUW 17415 Te Herenga Formation lava from Whak-apapaiiti Valley, Mount Ruapehu (S20/294147). Grey, vesicular pyroxene-hornfels (TYPE FRXb) (lOX lOX 10). Andesine + hypersthene-bronzite + titanomagnetite + pleonaste. Vugs with andesine + hypersthene-bronzite + rhyolitic glass (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17419 Whakapapa Formation lava froin the road to Meads Wall Ski Tow, Mount Ruapehu (T20/311154). Grey, vesicular, nematoblastic pyroxene-hornfels (TYPE FRXb) (15X60X30). Labradorite (clouded near contact) + hypersthene-bronzite + titanomagnetite + pleonaste (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17425 Iwikau Member (Whakapapa Formation) airfall, Mount Ruapehu (T20/32S139). Grey, finely seg-regated and foliated gneiss (TYPE FRXa) (120XSOX25). (I) Andesine + sanidine + sillimanite; (2) Biotite + alu-minous-hypersthene + pleonaste + ilmenite (CHEM-ISTRY, EPMA, 87Sr/86Sr).

VUW 17436 Ohakune andesite from Ohakune Railway Station (S20/1719SI). White, coarse-grained quartzite with stained, nematoblastic texture (TYPE QRXb) (25 X 15 X 15). Reactive contact similar to 178S5 (CHEM-ISTRY, 87Sr/86Sr).

VUW 17440 Whakapapa Formation lava from Top 0' Bruce, Mount Ruapehu (S20/293161). Light-grey vesi-cular hornfels with highly clouded plagioclase (TYPE FRXb) (15X 15X5). Labradorite + hypersthene ± pleonaste ± titanomagnetite ± ilmenite (CHEMISTRY, 87Sr/86Sr).

VUW 17443 Whakapapa Formation lava from Lower Tama Lake, Mount Ruapehu (T20/345 195). Grey, banded hornfels with I mm pyroxenitic segregations (TYPE FRXb) (35X20X 10). Andesine + hypersthene + pleon-aste + titanomagnetite + ilmenite (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17444 Whakapapa Formation lava from South Ruapehu (T20/301064). Light-grey, vesicular hornfels (TYPE FRXb) (lOX lOX 5). Labradorite + hypersthene + titanomagnetite + apatite + rhyolitic glass (EPMA).

*Catalogue number of Geology Department petrological collection, Victoria University of Wellington.

tHost lava formation and locality details. ;Metric grid reference (NZMS 270).; §Approximate dimensions of xenolith in millimetres. II Chemistry, microprobe analyses or Sr isotopic compo-sition available from author on request.

VUW 17458 Whakapapa Formation lava from South Ruapehu (precise. locality unknown). Grey, vesicular hornfels (TYPE FRXb) (30X lOX 10). Andesine-labra-dorite + bronzite + pleonaste (corundum core) ± ilmenite ± titanomagnetite ± dacitic glass. (CHEMIS-TRY, EPMA, 87Sr/86Sr).

VUW 17463 Ngauruhoe 1954 andesite (Tl9/355252). White, coarse-grained quartzite with unstrained granular quartz (TYPE QRXb) (70X45X40). Oxidised contact. Quartz ± anorthite. ± hypersthene ± rhyolitic glass. (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17468 Ngauruhoe 1954 andesite (Tl9/355252). White, massive saccharoidal quartzite (TYPE QRXb) (120XSOX60). 99% quartz. (CHEMISTRY).

VUW 17476 Whakapapa Formation lava from the road to Mead's Wall Ski Tow, Mount Ruapehu (S20/311 154). White, coarse-grained quartzite of mildly strained, inter-lobate quartz (TYPE QRXb) (50X30X25). Quartz ± anorthite ± wollastonite ± ferrowollastonite. (EPMA).

VUW 17482 Iwikau Member (Whakapapa Formation) airfall, Mount Ruapehu (T20/32S139). Red-green-white banded gneiss (TYPE QRXa) (45X20X 10). Quartz + bytownite-labradorite + ferroaugite + ilmenite + hema-tite. (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17483 Iwikau Member (Whakapapa Formation) airfall, Mount Ruapehu (T20/32S139). Grey, finely seg-regated and foliated gneiss (TYPE FRXa) (lOX 5 X 5). (I) Oligoclase + sanidine + corundum; (2) Biotite + pleon-aste + titanomagnetite + syenitic glass. (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17484 Iwikau Member (Whakapapa Formation) airfall, Mount Ruapehu (T20/32S139). Grey, finely seg-regated gneiss (TYPE FRXa) (25 X lOX 10). Plagioclase + hypersthene + pleonaste. (CHEMISTRY, 87Sr/86Sr).

VUW 17485 Iwikau Member (Whakapapa Formation)' airfall, Mount Ruapehu (T20/32S139). Brown-white, con-torted, banded gneiss (TYPE QRXa) (60X50X40). (I) Quartz + anorthite + salite ± ilmenite ± hematite ± titanite. (2) Andesine + biotite + almandine. (3) Quartz. (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17489 Iwikau Member (Whakapapa Formation) airfall, Mount Ruapehu (T20/32S139). Grey, fine-grained schist (TYPE SRX) (20 X 10 X 5). Labradorite-bytownite + bronzite + biotite ± titanomagnetite ± ilmenite. (CHEMISTRY, EPMA).

VUW 17492 Iwikau Member (Whakapapa Formation) ·airfall, Mount Ruapehu (T20/32S I 39). Green-grey-white saccharoidal gneiss (TYPE QRXa) (50X50X50). (I) Quartz + anorthite + ferrosa1ite + jtanite (VUWI7490). (2) Quartz + andesine + hypersthene + almandine + rhyolitic glass + ilmenite ± gedrite ± pleonaste (VUW 17491). (CHEMISTRY, EPMA, 87Sr/86Sr).

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14


VUW 17497 Te Herenga Formation lava from Whak-apapaiiti Valley, Mount Ruapehu (S20/294147). Grey, vesicular pyroxene-hornfels (TYPE FRXb) (30X25X 15). Andesine-labradorite + hypersthene + titanomagnetite + ilmenite. (CHEMISTRY, EPMA, 87Sr/86Sr).

VUW 17885 Pukeonake andesite (Tl9/318256). White, coarse-grained quartzite (TYPE QRXb) (80X30X25). (1) Quartz ± anorthite ± ferrosalite ± rhyodacitic glass. (2) Augite + rhyolitic glass (reaction rim). (EPMA).

VUW 17888 Wahianoa Formation lava from south-western rim of Whangaehu Gorge, Mount Ruapehu (T20/359096). Grey, fine-grained, granular schist (TYPE SRX) (25X lOX5). Bytownite + olivine + hypersthene + titanomagnetite. (EPMA).

VUW 17892 Te Herenga Formation lava from Mount Ruapehu (precise locality unknown). Grey, fine-grained schist (TYPE SRX) (25 X lOX 10). Plagioclase + olivine + hypersthene + opaques. Augite-rich reaction rim.

VUW 17894 Wahianoa Formation lava from Whan-gaehu Gorge, Mount Ruapehu (precise locality unknown). Light-grey pyroxene hornfels (TYPE FRXb) (lOX5X5). Oligoclase + ferrohypersthene + almandine + cordier-ite + ilmenite + pleonaste. (EPMA).

157

Dow

nloa

ded

by [

Uni

vers

ity o

f B

irm

ingh

am]

at 0

9:57

13

Nov

embe

r 20

14

Documents

Petrography and origin of metasedimentary xenoliths in lavas from Tongariro Volcanic Centre