Provenance of the sedimentary Rakaia sub-terrane, Torlesse Terrane, South Island, New Zealand: the...

www.elsevier.com/locate/sedgeo

Sedimentary Geology 168 (2004) 193–226

Provenance of the sedimentary Rakaia sub-terrane, Torlesse

Terrane, South Island, New Zealand: the use of igneous

clast compositions to define the source

Anekant M. Wandresa,*, John D. Bradshawa, Steve Weavera, Roland Maasb,Trevor Irelandc, Nelson Ebyd

aDepartment of Geological Sciences, University of Canterbury, Private Bag 4800, Christchurch, New Zealandb Isotope Geochemistry Laboratory, University of Melbourne, Parkville, Victoria 3010, Australia

cGeochronology and Isotope Geochemistry, Research School of Earth Sciences, The Australian National University,

Canberra, ACT 0200, AustraliadDepartment of Environmental, Earth and Atmospheric Sciences, University of Massachusetts, Lowell, MA 01854, USA

Received 17 September 2003; received in revised form 4 February 2004; accepted 5 March 2004

Abstract

The Permian to Late Triassic Rakaia sub-terrane—part of the Torlesse superterrane of the New Zealand Eastern Province—is

an accretionary complex that comprises an enormous volume of quartzofeldspathic sandstones and mudstones with subsidiary

conglomerates plus minor oceanic assemblages. The field of provenance analysis has undergone a revolution with the

development of single-crystal isotope dating techniques using mainly silt- to sand-sized single mineral grains. Fine sand and

mud may be transported over thousands of kilometres with potential problems for provenance studies, reinforcing the need for

caution when interpreting provenance evidence from single heavy mineral grains. This study focuses on coarser-grained rocks,

notably conglomerates, which involve much shorter transport distances and therefore may be used to trace proximal sources. A

detailed rock sampling programme and geochronological, geochemical and Sr–Nd isotope analysis of igneous clasts from four

conglomerates (Boundary Creek, Te Moana, McKenzie Pass and Lake Hill) have broadly characterised the igneous source for

the Rakaia sub-terrane.

SHRIMP U–Pb zircon ages of Rakaia sub-terrane igneous clasts define three distinct periods of magmatic crystallisation.

The first period ranges in age from 292 to 243 Ma (Permian to Middle Triassic) with two clusters recognisable: a minor Early

Permian one ranging in age from 292 to 277 Ma, and a major Late Permian to Middle Triassic one from 258 to 243 Ma. All

these clasts are confined to the Kazanian (Permian) Te Moana, the Dorashamian (Permian) McKenzie Pass and the Carnian

(Late Triassic) Lake Hill conglomerates. The clasts, which are subduction-related calc-alkaline to high-K calc-alkaline, and

metaluminous to peraluminous, range in lithology from andesite to rhyolite and their plutonic equivalents. Many clasts at Lake

Hill have chemical compositions characteristic of partial melts of a source of basaltic composition in equilibrium with

amphiboleF pyroxeneF garnet (adakites), indicative of a thick crust in the clast source area. The second period comprises

Carboniferous, calc-alkaline, metaluminous to weakly peraluminous clasts of the Permian Boundary Creek conglomerate,

ranging in age from 356 to 325 Ma. The third group consists of two Cambrian clasts, a monzogranite from Te Moana

0037-0738/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.sedgeo.2004.03.003

* Corresponding author. Tel.: +64-3-3642700.

E-mail addresses: [email protected] (A.M. Wandres), [email protected] (J.D. Bradshaw),

[email protected] (S. Weaver), [email protected] (R. Maas), [email protected] (T. Ireland),

[email protected] (N. Eby).

A.M. Wandres et al. / Sedimentary Geology 168 (2004) 193–226194

(c. 497F 8 Ma) and a dacite from Lake Hill (c. 517 Ma). Geochronology, geochemistry and Sr–Nd isotopes of Rakaia

igneous clasts correlate broadly with those of Permian to Triassic plutons and volcanics from the Amundsen and Ross

provinces of Marie Byrd Land. Thus, the Antarctic sector of the Panthalassan margin of Gondwana is the probable source for

the Rakaia sub-terranes as opposed to the other postulated sources in Eastern Australia.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Sandstone provenance; Igneous clasts; Zircon geochronology; Isotope chemistry; Rakaia sub-terrane; Torlesse terrane; New Zealand;

Gondwana margin

1. Introduction The provenance of the vast volumes of Paleozoic–

The Torlesse sedimentary rocks constitute the

Permian to Late Triassic Rakaia and the Late Jurassic

to Early Cretaceous Pahau sub-terranes (Fig. 1) which

make up a large part of the New Zealand micro-

continent (Fig. 2). The two sub-terranes are separated

by the Esk Head Melange (Bradshaw, 1973; Silberling

et al., 1988). Studies of Torlesse depositional and

deformational environments have recognized similar-

ities to other large sandstone-dominated accretionary

complexes in the circum-Pacific region, notably the

Franciscan Complex in California and sandstone belts

in Alaska, Japan and Antarctica. Consequently, a

submarine fan depositional model and an accretionary

wedge deformational model are now favored for the

Torlesse (e.g., Howell, 1980; George, 1992; Mazen-

garb and Harris, 1994).

It is widely accepted that the Torlesse sub-terranes

formed along Gondwana’s Panthalassan plate margin,

which was one of the most extensive orogenic belts in

Earth history. Originally established during Neopro-

terozoic rifting, the margin was the site of long-lived

subduction during early Paleozoic convergent tecto-

nism (e.g., Floettmann et al., 1993). Subsequent

accretion of arc–trench systems, and possibly mi-

cro-continental fragments, formed a series of east-

ward-migrating orogenic belts: the formerly

continuous Neoproterozoic–Ordovician Ross and

Delamerian orogens (e.g., Stump et al., 1986; Coney

et al., 1990; Floettmann et al., 1993), the Cambrian–

Carboniferous Lachlan and Thompson Orogens (e.g.,

Coney et al., 1990; Foster and Gray, 2000) and the

Silurian–Cretaceous New England Fold Belt (NEFB;

e.g., Powell and Li, 1994). Remnants of these oro-

genic belts are preserved in the formerly continuous

Gondwana fragments of Australia, Antarctica and

New Zealand (Gibson and Ireland, 1996; Fig. 2).

Mesozoic accretionary sandstone and mudstone

sequences incorporated in this margin is an important

requirement for the understanding of the paleogeog-

raphy and evolution of the Panthalassan margin.

However, original relative positions in the New Zea-

land sector of the margin have been disrupted by the

opening of the Tasman Sea and the Southern Ocean

since 80 Ma (Wood, 1994; see also Fig. 2), followed

by further disruption and re-configuration of the NZ

microcontinent up to the present (Sutherland, 1999).

Paleo-geographic reconstructions of this portion of the

margin have therefore been particularly difficult.

2. Basement geology

New Zealand basement geology is described in

terms of batholiths, suites and tectonostratigraphic

terranes (Coombs et al., 1976; Bishop et al., 1983;

Bradshaw, 1989), which are grouped into three prov-

inces: the Western Province, the Median Tectonic

Zone (MTZ) and the Eastern Province (Fig. 1). The

Western Province comprises two terranes and is large-

ly made up of Lower Paleozoic metasedimentary rocks

cut by Devonian, Carboniferous and Early Cretaceous

granitoids (Muir et al., 1994a, 1997; Waight et al.,

1997), with minor volcanic and metamorphic rocks of

Cambrian age (Cooper, 1989; Gibson and Ireland,

1996; Munker and Cooper, 1997). It had attained its

continental thickness and its gross structure by the end

of the Carboniferous and represents a fragment of the

Gondwana Paleozoic margin. The Eastern Province

consists of arc, forearc and accretionary complex rocks

that relate to Permian to Cretaceous plate convergence.

The Median Tectonic Zone separates the Western and

Eastern Provinces and consists of suites of Carboni-

ferous to Early Cretaceous subduction-related calc-

Fig. 1. Tectonostratigraphic terrane map of New Zealand showing distribution of conglomerate locations (filled cirlces). Also shown are

localities of sandstone samples used for U–Pb zircon age (SHRIMP) determination (open circles).

A.M. Wandres et al. / Sedimentary Geology 168 (2004) 193–226 195

alkaline plutons with subordinate volcanic and sedi-

mentary rocks (Kimbrough et al., 1994; Muir et al.,

1998; Mortimer et al., 1999).

3. Provenance of the Torlesse

The provenance of the Rakaia sub-terrane has

occupied New Zealand geologists for many years

(Andrews et al., 1976; Coombs et al., 1976; Howell,

1980; MacKinnon, 1983; Korsch and Wellman, 1988;

Bradshaw, 1989). Sandstone petrography, sedimentary

geochemistry, detrital mineral geochronology and

isotope geochemical studies of sediments have all

been employed to address this problem (e.g., Mac-

Kinnon, 1983; Roser, 1986; Roser and Korsch, 1988;

Frost and Coombs, 1989; Ireland, 1992; Adams and

Graham, 1996; Cawood et al., 1999; Pickard et al.,

Fig. 2. Present-day distribution of orogenic/mobile belts around the southwest Pacific margin. AP=Amundsen Province, RP=Ross Province.

Insert New Zealand micro-continent indicating the main features discussed in the text. MTZ=Median Tectonic Zone. Adapted from Sutherland

(1999) and Moore et al. (1982).


2000). These studies have broadly established a hetero-

geneous cratonic arc provenance and various sediment

sources have been proposed for the Rakaia sub-

terrane, including a provenance east of present-day

New Zealand (Andrews et al., 1976), Marie Byrd

Land in Antarctica (Korsch and Wellman, 1988) and

the New England Fold Belt in eastern Australia

(Adams and Kelley, 1998; Pickard et al., 2000).

Many Torlesse provenance studies to date have

focused on fine-grained heavy minerals, or fine-

grained rocks in general. Fine sand and shale-size

sediment may be transported over thousands of kilo-

metres (Pell et al., 1997; Sircombe, 1999), with

potential problems for provenance studies (e.g., Bas-

sett, 2000). In this study, we have focused on coarser-

grained rocks, notably conglomerates, which imply

much shorter transport distances (tens to hundreds of

kilometres; Kodama, 1994; Ferguson et al., 1996) and

therefore may be used to trace proximal sources. In

particular, we found that igneous conglomerate clasts

are found to be capable of placing exceptionally tight

constraints on Torlesse provenance and on its Meso-

zoic tectonic setting within the Southwest Pacific

margin of Gondwana. This paper reports provenance

data from conglomerates in the Rakaia sub-terrane.

Provenance constraints for the younger Pahau sub-

terrane are discussed in a separate paper (Wandres et

al., 2004).


4. Samples and analytical methods

A total of 140 igneous clasts were collected from

four conglomerates. Criteria for sampling were: rep-

resentation of lithological diversity, degree of alter-

ation and size (>10 cm). Major and trace element

compositions were determined by X-ray fluorescence

(XRF) at the University of Canterbury following the

methods of Weaver et al. (1990). All elemental con-

centrations have been renormalized to 100% anhy-

drous. Rare earth elements (REE) were analysed by

instrumental neutron activation (INAA) at the Univer-

sity of Massachusetts, Lowell, using techniques sim-

ilar to those described by Eby (1984). Representative

clasts were submitted for Sr–Nd isotopic analysis and

SHRIMP U–Pb zircon dating.

Nd–Sr isotope work was carried out at La Trobe

University, Melbourne, following the methods of

Waight et al. (1998). Measured Nd and Sr isotope

ratios were normalized to 146Nd/144Nd = 0.7219 and88Sr/86Sr = 8.37521. Average values for the La Jolla

Nd and SRM987 Sr standard were 0.511860F 15 (2

S.D.) and 0.710230F 30 (2 S.D.), respectively. Sr–

Nd blanks were < 300 pg and are negligible. Rb/Sr

and Sm/Nd ratios for age corrections were derived

from XRF and INAA, respectively. Precision for

initial 87Sr/86Sr is estimated to be F 0.0001 (2 S.D.)

by propagating of errors in Rb/SrXRF (Rb = 1.995%,

Sr = 0.493%, Weaver et al., 1990) and 87Sr/86SrTIMS

(F 0.000076 for granitic rock powders and about

twice the external precision based on the SRM987

standard). This error covers all possible sources of

error including sample powder heterogeneity, but

excludes error in the age. A similar calculation for

initial eNd (F 10% for Sm and Nd from INAA;

F 0.004% for 143Nd/144NdTIMS) produces a combined

2 S.D. error of 0.5 units.

U–Pb crystallisation ages of igneous clasts and

detrital zircons of Rakaia and Pahau sandstones were

obtained on SHRIMP RG and SHRIMP I at Stanford

University and at the Australian National University,

Canberra. Methods are described in detail by Muir et

al. (1997). The ANU standard SL13 was used to

normalize the U concentration (238 ppm). The Th

concentration was then derived from the Th/U ratio.

Duluth standards AS3 and FC1 were used to calibrate

U/Pb ratios after correction of Pb+/U+ for correlated

variation with UO+/U+. The Temora standard (417

Ma) was included in one mount as a secondary

standard. The zircons are dominated by Phanerozoic

material and so emphasis was placed on determining

the 206Pb/238U age. The zircons from each clast were

targeted for the most suitable zircon and spots were

placed so as to avoid cracks, inclusions and other

heterogeneities. Cathodoluminescence images were

used to optimize spot locations. Detailed descriptions

of individual igneous clasts and analysed sandstones

are available as an electronic document from Elsevier.

The AGSO Phanerozoic Time Scale (Jones, 1995) is

used.

5. Results from igneous clasts

5.1. Lithology and petrography

Conglomerates are rare and constitute < 1% of the

Torlesse lithologies. Sandstone sequences are structur-

ally complex with large-scale imbrications and a

general lack of marker beds making it impossible to

define continuous stratigraphic sections because of

structural complexity. However, there are enough

diagnostic fossil occurrences to define several fossil

zones (Campbell and Warren, 1965; Speden, 1974):

Late Permian, Middle Triassic, Lower Late Triassic

and Late Triassic. In addition, there are Late Carbon-

iferous and Permian tectonic enclaves that may repre-

sent separate micro-terranes (Akatarawa and Kakahu,

Fig. 1; Bishop et al., 1985; Hada and Landis, 1995).

The four Rakaia conglomerates at Boundary Creek,

Te Moana, McKenzie Pass and Lake Hill (Fig. 1) are

dominated by beds of poorly to moderately sorted,

matrix- to clast-supported conglomerates interbedded

with massive to thinly bedded, medium to coarse

grained sandstones from graded turbidites. All con-

glomerates are dominated by rounded to well-rounded

pebbles and cobbles but boulders and large angular

blocks (sandstones rip-ups) also occur. Clast popula-

tions in conglomerates are dominated by sandstones

(McKenzie Pass and Lake Hill) and volcanics

(Boundary Creek and Te Moana). Although no statis-

tical counts were made for the Boundary Creek

conglomerate visual estimates indicate that volcanic

clasts are the predominant igneous lithotype. Litho-

type proportions for the conglomerates are shown in

Table 1.

Table 1

Lithotype distribution from Rakaia terrane conglomerates

Lithotype Boundary

Creek, %*

Te Moana,

%

McKenzie

Pass, %

Lake

Hill, %

Granitoid 15 7 13 22

Volcanic 30 41 19 5

Arenite 40 29 40 49

Lutite 10 15 15 14

Quartz, chert

and others

5 7 13 9

Estimated size

range (mm)

40–50* 35–45 40–50 50–60

An area on the outcrop was marked out in order to include at least

500 clasts, which is the minimum number needed to obtain

lithotypes within 5% of true values at the 95% confidence level

(Van Der Plas and Tobi, 1965). A 20� 20 cm grid was

superimposed on the conglomerate to assist in clast counting.

*At Boundary Creek, no statistical counts were made, but visual

estimates indicate that volcanic clasts are the predominant igneous

lithotype. Furthermore, all clasts at Boundary Creek are strongly

deformed and initial grain size is estimated.


All the sequences in which the conglomerates

occur have undergone regional metamorphism. The

broad area of non-schistose Torlesse rocks (textural

zone 1, Bishop, 1972) are zeolite and prehnite–

pumpellyite facies, within which a few semi-schists

and schists occur. The transition from non-schistose to

semi-schistose rocks (textural zone IIA) corresponds

approximately to the transition from prehnite–pum-

pellyite to pumpellyite–actinolite facies rocks. The

conglomerate at Boundary Creek is located within the

chlorite zone of the greenschist facies, whereas all

other Rakaia sub-terrane conglomerates described are

from the prehnite–pumpellyite facies. Igneous and

metamorphic recrystallisation is encountered in all

clast thin sections. Albitisation of plagioclase is com-

mon and chlorite is the most widespread secondary

mineral which, together with epidote and titanite,

partly or completely replaces mafic silicate minerals.

Granitoid clasts at Boundary Creek range in modal

composition from meta-diorite to meta-leucomonzog-

ranite, and the volcanic clasts from meta-dacite to

meta-rhyolite. In most samples, layers and domains

with original igneous crystals and texture are recog-

nisable, and meta-volcanic clasts still display a glom-

erophyric texture. At Te Moana, spherulitic and

felsitic volcanic clasts of rhyolitic composition dom-

inate. Subsolvus monzogranites and a hypersolvus

granodiorite are present. The McKenzie Pass clast

population consists of predominantly rhyolitic clasts,

but clasts of intermediate and mafic compositions are

present. Hypersolvus clasts, which dominate over

weakly foliated subsolvus clasts, are typified by

granophyric intergrowths that often surround discrete

subsolvus mineral grains. Subsolvus clasts at Lake

Hill range in modal composition from tonalite to

leucomonzogranite and are dominated by two-mica

granitoids. These clasts display a full range of defor-

mational features from undeformed to mylonitic and

gneissic. Unlike the other conglomerates, volcanic

clasts are not common at Lake Hill and those that

are present exhibit no pyroclastic characteristics.

5.2. U–Pb zircon ages

5.2.1. Igneous clasts

SHRIMP U–Pb data for 17 igneous clasts are

summarised in Table 2 and examples are shown on

Tera and Wasserburg (1972) diagrams in Fig. 3. In

order to expedite analysis of the large number of

clasts, a relatively small number of analyses (6–9)

were performed on each clast. If the data agreed

within analytical error, the likelihood is that this

represents the igneous age of the clast. If the data

were scattered because of the presence of inherited

zircons, the likelihood is that many analyses would be

required to ascertain the zircon U–Pb characteristics

and even then that result might still be equivocal.

Three age groups have been identified: (1) Permo-

Triassic, with 10 clasts giving ages from c. 292 to 243

Ma, with two minor groups, one ranging from 258 to

243 Ma and the second from 292 to 277 Ma; (2)

Carboniferous, with four clasts from Boundary Creek

ranging from 325F 5 to 356F 5 Ma; and (3) Cam-

brian, with two clasts from the Te Moana (497F 8

Ma) and Lake Hill (c. 517 Ma) conglomerates. Only a

small fraction of the Rakaia sub-terrane clast popula-

tion was dated and clasts of other ages might be

present. Nevertheless, the dated clasts are deemed to

represent the age distribution within the collected clast

population, and the three age groups are discussed in

this order.

5.2.2. Detrital U–Pb zircon ages

Zircon U–Pb ages for two Rakaia sub-terrane

sandstones (Kurow, Balmacaan localities, Fig. 1) were

obtained and have been discussed in detail by Wan-

Table 2

Summary of SHRIMP U–Pb zircon ages for igneous clasts from four Rakaia terrane conglomerates

Location UC no. Rock type Chemistry Texture Age F MSWD Comments

Early Permian–

Early Triassic

Lake Hill 30668 monzogranite I/S-type hypidiomorphic c. 243 3/7 (scattered,

Inh, PbL)

30667 rhyolite S-type moderately

porphyritic

244.1 3.8 0.14 6/6

30659 2-mica

granodiorite

S-type hypidiomorphic

(adakitic)

256.9 5.1 1.64 4/6 (weakly foliated,

adakitic)

30673 monzogranite I/S-type hypidiomorphic c. 279 4/7 (scattered, Inh,

PbL, adakitic)

McKenzie

Pass

30909 trachydacite I/S-type moderately

porphyritic

253.4 4.2 1.85 6/8 (PbL)

30895 monzogranite I-type granophyric 254.3 3.8 1.67 6/6

30902 rhyolite I-type moderately

porphyritic

258.2 4.5 0.75 6/6

30901 rhyolite I/S-type moderately

porphyritic

c. 263 3/6 (scattered,

Inh, PbL)

30914 rhyolite I-type weakly

porphyritic

292.2 5.1 0.57 6/6

Te Moana 31813 rhyolite I-type moderately

porphyritic

276.8 4.6 2.14 5/9 (bimodal,

Inh, PbL)

31818 monzogranite I/S-type hypidiomorphic c. 290 3/6 (bimodal, Inh,

PbL, foliated)

Carboniferous Boundary

Creek

31747 meta-rhyolite I-type mylonitic

rhyolite

325.2 4.4 0.82 6/6 (mylonitic/gneissic)

31775 meta-monzogranite I-type mylonitic

granodiorite

330.1 4.9 1.89 5/6 (Inh)

(mylonitic/gneissic)

31749 meta-granodiorite I-type mylonitic

granodiorite

334.9 4.9 1.52 6/6 (mylonitic/gneissic)

31776 meta-dacite I-type mylonitic dacite 355.6 5.0 1.55 6/6 (mylonitic/gneissic)

Cambrian Te Moana 31816 monzogranite I/S-type hypidiomorphic 496.9 7.7 1.60 5/7 (foliated)

Lake Hill 30644 dacite I/S-type moderately

porphyritic

c. 517 4/6 (Inh)

MSWD=mean square of weighted deviates; 5/6, etc. = proportion of analyses used to calculate mean age (errors are 2r mean and include

uncertainty for Pb/U in standard); Inh = inherited component (age>mean); PbL=Pb loss (age <mean).


dres and Bradshaw (in press). Data for the individual

detrital samples are presented in Fig. 4 as age histo-

grams with a calculated interval of 20 Ma. Sample age

distributions are summaries in terms of major age

components (usually 5%), but some minor groups are

also identified when a peak is noted. The groups

identified in each data set are defined by natural

breaks in the data. The age histograms are overlain

by cumulative probability curves to further highlight

the principal age components.

5.2.2.1. Kurow. This Permian sample is a dark

grey, bedded, medium to coarse grained sandstone

collected from the slope of Kurow Hill (Grid refer-

ence NZMS260 I40/093 060). Of the 60 analysed

zircon grains, 70% are represented by the age group

with ages between 237 and 349 Ma (Fig. 4a). The

group is a complex cluster of three distinct peaks at c.

270, 300 and 340 Ma with the main peak making up

28% of the data. The group between 260 and 240 Ma

(10%) has a peak at c. 255 Ma and is regarded as the

most reliable estimate of the maximum stratigraphic

age limit. A minor Middle Devonian–Silurian group

(6%) at 420–380 Ma has a peak at c. 420 Ma and an

Ordovician–Early Cambrian group (540–480 Ma)

makes up 12% of the data with a broad peak at c.

510 Ma. The remaining six analyses range from 1499

to 630 Ma and make up 10% of the data with a minor

peak (6%) at 640 Ma and insignificant peaks (3%) at

966, 1050 and 1500 Ma. Cores of the two youngest

zircons from this sandstone returned Middle Triassic

(Anisian) 238F 8 and 239F 6 Ma ages.

Fig. 3. Representative U–Pb zircon concordia plots (Tera and Wasserburg, 1972) for igneous clasts from the Rakaia terrane conglomerates.

These plots use the measured Pb isotopic composition (uncorrected for common Pb). The dashed line represents common Pb regression and

mean age. The filled dots represent data included in the regression. Error arcs and bars are 1r.


Fig. 4. (a) U–Pb detrital zircon age histogram for the Permian Kurow Hill sandstone. The age histogram is calculated at 50-Ma intervals with

counts shown on the left axis. The cumulative probability curve is calculated at 1-Ma intervals with the vertical scale on the right axis. Insert

shows histogram for the last 600 million years calculated at 20-Ma intervals, with the cumulative probability curve calculated at 1-Ma intervals.

(b) U–Pb detrital zircon age histogram for the Middle Triassic Balmacaan Steam sandstone.


5.2.2.2. Balmacaan Stream. The Triassic (Anisian)

sample is a dark grey to grey and fine to medium

grained sandstone and was collected from an exposure

in the river bank of the Balmacaan Stream (Grid

reference NZ MS260 J36/518 283). A total of 64

zircons were analysed from this sandstone sample

(Fig. 4b). The major Late Triassic to Carboniferous

(Namurian) group comprises 51% of the whole data,


with the main peak (16%) between 280 and 260 Ma at

c. 265 Ma and two minor younger peaks at c. 250 Ma

(13%) and 230 Ma (8%). The youngest peak is

regarded as the most reliable estimate of the maxi-

mum stratigraphic age. The group between 320 and

280 Ma make up 15% of all the analyses and there are

two possible peaks at 285 Ma (9%) and 305 Ma (6%).

There are insignificant peaks ( < 5%) between the

Carboniferous and the Ordovician (480–320 Ma)

and one minor group between 520 and 480 Ma

(6%) with a peak at 495 Ma. The remaining analyses

(28%) extend to 1848 Ma with minor groups at 929–

870 Ma (5%) and 1085–1004 Ma (9%) with a peak at

c. 1020 Ma. The pattern for the Middle Triassic

(Anisian) Balmacaan Stream sandstone agrees well

Table 3

Chemical analyses of representative igneous clasts from four Rakaia terra

Lake Hill McKenzie Pass

UCno. 30668 30667 30659 30673 30644 30909 30895 30902 3

SiO2 76.15 72.71 73.10 72.61 63.37 66.47 77.20 76.26

Ti2O 0.04 0.38 0.18 0.10 0.58 0.63 0.14 0.17

Al2O3 14.03 14.65 14.99 15.79 16.62 15.69 12.25 12.77

Fe2O3 0.22 1.88 1.47 0.74 4.57 3.62 1.19 1.07

MnO 0.01 0.04 0.04 0.02 0.09 0.07 0.02 0.03

MgO 0.08 0.61 0.49 0.21 1.60 1.17 0.13 0.13

CaO 1.12 1.54 1.32 1.79 4.00 2.40 0.66 0.97

Na2O 3.89 4.33 4.64 5.23 3.65 3.62 4.21 4.38

K2O 4.38 3.08 2.88 2.94 1.85 4.74 3.87 3.87

P2O5 0.04 0.09 0.07 0.03 0.10 0.14 0.02 0.03

LOI 0.34 0.86 0.55 0.54 2.71 0.70 0.53 0.43

Total 100.30 100.17 99.73 100.00 99.14 99.25 100.22 100.11 1

V 9 38 20 12 176 50 6 8

Cr 12 22 19 10 37 7 10 10

Ni 3 7 5 4 6 3 4 3

Zn 7 41 34 15 58 52 47 20

Zr 67 200 96 67 110 308 207 156

Nb 2 11 8 4 5 12 28 20

Ba 733 766 872 1164 564 2083 412 776 2

La 5 43 14 5 16 30 62 43

Ce 19 75 26 9 39 81 117 73

Nd 10 36 11 10 13 31 50 29

Ga 14 16 16 17 16 19 18 15

Pb 36 24 34 31 1 18 22 18

Rb 89 104 61 53 66 144 85 107 1

Sr 423 389 624 852 321 532 93 89

Th 1 15 2 1 1 14 24 19

Y 7 21 12 5 15 28 40 18

Sc < 2 5 3 24 7 6 2

A/CNK 1.06 1.1 1.14 1.05 1.09 1.01 0.99 0.97

A/NK 1.26 1.4 1.39 1.34 2.07 1.41 1.1 1.12

with that of another Middle Triassic sandstone col-

lected nearby by Pickard al. (PUD1, 2000, main peak

at c. 260 Ma), but has a slightly older main peak at

265 Ma.

5.3. Chemistry and Sr–Nd isotopes

Element mobility during deformation and recrys-

tallisation has been reported for mylonitic rocks

which show significant enrichment in SiO2, CaO,

Fe2O3 and Sr, and depletion in Na2O, K2O, MgO,

Ba and Rb with increasing stages of deformation

(Tobisch et al., 1991). Furthermore, mobility of

alkalis, and Ca in particular, may be expected given

the widespread albitisation of plagioclase observed

ne conglomerates

Te Moana Boundary Creek

0901 30914 31813 31816 31818 31747 31775 31749 31776

76.60 69.49 77.65 76.94 76.31 72.36 73.89 72.21 66.64

0.02 0.44 0.16 0.07 0.09 0.25 0.35 0.16 0.50

13.56 14.51 12.78 12.62 13.04 12.25 12.37 11.63 14.62

0.57 3.73 0.51 0.87 0.75 2.37 2.89 1.04 4.23

0.06 0.08 0.04 0.03 0.03 0.04 0.06 0.04 0.08

0.05 1.02 0.15 0.11 0.17 0.86 1.39 0.35 2.15

0.47 2.28 0.53 0.72 0.68 3.17 1.94 4.55 3.63

4.41 5.26 2.90 3.68 3.88 6.19 5.53 6.38 4.56

4.02 1.84 4.74 4.56 4.59 0.16 0.65 0.19 2.02

0.02 0.08 0.02 0.01 0.02 0.08 0.11 0.04 0.14

0.38 0.94 0.65 0.42 0.43 2.77 1.24 3.84 1.83

00.16 99.67 100.13 100.03 99.99 100.50 100.42 100.43 100.40

7 43 10 10 8 23 53 8 90

8 14 10 7 7 10 38 19 16

3 11 4 3 3 3 13 3 11

25 51 9 9 8 29 33 17 50

40 254 101 57 59 149 85 114 140

13 10 21 9 13 8 6 13 8

50 502 276 91 443 77 193 75 594

9 38 44 17 24 19 16 27 21

18 65 47 22 24 40 15 47 33

12 38 15 18 12 23 22 26 27

17 16 15 12 12 8 8 9 15

18 13 16 32 23 16 8 18 11

48 60 192 209 125 1 21 3 67

65 542 69 43 168 110 117 129 264

5 9 30 23 18 13 12 7 10

13 34 15 17 8 22 13 17 21

< 2 < 2 5 < 2 5 6 4 11

1.09 0.98 1.17 1.02 1.03 0.76 0.93 0.61 0.9

1.17 1.36 1.29 1.15 1.15 1.18 1.26 1.09 1.51

A.M. Wandres et al. / Sedimentary G

in thin sections (Crawford et al., 1992, and refer-

ences therein) and has undoubtedly contributed to

data dispersion in graphical chemistry plots result-

ing in outliers.

Fig. 5. Comparison of Permo-Triassic Rakaia terrane igneous clasts from th

with data from potential SW Pacific igneous source provinces (references

are given in brackets): New England Supersuite, Bundarra Suite, Clarence

(1981), Hensel et al. (1985), Bryant et al. (1997), Allen et al. (1998), and C

(1998) and Mortimer et al. (1999); Hekeia Gabbro from Mortimer et al. (199

Thurston Island from Pankhurst et al. (1993) and Leat et al. (1993). (a) Alum

S-type and weakly peraluminous (I/S-type) igneous clasts. (b) Alkali – sili

5.3.1. Permo-Triassic clasts

Dated igneous clasts of this age group occur at Te

Moana, McKenzie Pass and Lake Hill. For simplicity,

clasts from these conglomerates (except the Cambrian

eology 168 (2004) 193–226 203

e Te Moana, McKenzie Pass and Lake Hill conglomerates compared

are listed in Fig. 10). Data for source provinces (number of samples

River Supersuite and Early Permian Volcanics from Shaw and Flood

aprarelli and Leitch (1998); Median Tectonic Zone from Muir et al.

9); Amundsen Province from Pankhurst et al. (1998b); and those for

inium Saturation Indices. Shaded fields show distribution of I-type,

ca diagram.


clasts) are pooled. Representative analyses are listed

in Table 3 and representative geochemical plots are

presented in Figs. 5 and 6. Most of the potential

source provinces along the Gondwana margin are

classified according to the I/S/A classification scheme

of Chappell and White (1974), and this is also applied

here to the clasts. However, the alphabetic classifica-

tion scheme should be used with caution as metal-

uminous granitoids can become weakly peraluminous

to peraluminous with extended fractionation. For

example, a fractionated rhyolitic clast from Te Moana

(UC31813) is classified as peraluminous (ASI = 1.17),

Fig. 6. Mantle-normalised (McDonough et al., 1992) multi-element diagram

for SW Pacific igneous source province data are given in Fig. 5. (a) Ca

(Nakamura, 1974) average REE content of adakitic clasts from Lake H

comparing igneous conglomerate clasts with igneous provinces from the S

volcanic-arc-granite (VAG), within-plate-granite (WPG) and ocean-ridge-g

but has the Sr–Nd isotope characteristics of a typical

I-type rock (Chappell, 1994). This rhyolite is grouped

with the other I-type clasts.

The clasts encompass a broad spectrum of compo-

sitions with SiO2 values ranging from 51% to 81%.

The major concentration is between 73% and 79%

SiO2, with the main peak corresponding broadly to a

monzogranitic composition. The pooled clast popula-

tion is dominated by slightly peraluminous clasts

(68%; referred to as I/S-type), the classification of

which is ambiguous using the scheme of Chappell and

White (1974). Minor groups of I-type (12%), S-type

s of representative igneous clasts and source provinces. References

lc-alkaline clasts and (b) adakitic clasts. (c) Chondrite-normalised

ill. (d) Tectonic discrimination diagram after Pearce et al. (1984)

W Pacific margin. Fields are for syn-collision-granite (syn-COLG),

ranite (ORG). Abbreviations for source provinces are given in Fig. 5.

Table 4

Sr–Nd isotope data for igneous clasts from four Rakaia terrane conglomerates. Maximum propagated errors (2 S.D.) are F 0.0001 for 87Sr/86Sri and F 0.5 units for eNdiUCno. Rock Type Chemistry Age Rb,

ppm

Sr,

ppm

87Rb/86Sr 87Sr/86Sr(m)87Sr/86Sri Sm,

ppm

Nd,

ppm

147Sm/144Nd 143/144Nd(m) eNd(0) eNd(i) TDM

Permian–Triassic

Lake Hill 30668 monzogranite I/S-Type 243 92 424 0.63 0.7094 0.70722 4.67 0.97 0.1255 0.512296 � 6.7 � 4.5 1.48

30667 rhyolite S-type 244 106 389 0.79 0.7104 0.70768 5.07 30.16 0.1016 0.512297 � 6.7 � 3.7 1.16

30673 monzogranite I/S-Type 280 55 864 0.18 0.7065 0.70578 3.27 0.63 0.1158 0.512452 � 3.6 � 0.7 1.09

McKenzie

Pass

20909 trachydacite I/S-Type 253 143 537 0.77 0.7089 0.70610 6.62 36.35 0.1102 0.512441 � 3.8 � 1.1 1.05

30895 monzogranite I-Type 254 84 95 2.57 0.7134 0.70415 10.16 53.05 0.1157 0.512444 � 3.8 � 1.2 1.1

30902 rhyolite I-Type 258 108 91 3.42 0.7168 0.70425 4.20 27.31 0.0930 0.512504 � 2.6 0.8 0.82

30914 rhyolite I-Type 292 61 551 0.32 0.7076 0.70623 6.05 32.22 0.1134 0.512508 � 2.5 0.6 0.98

Te Moana 31813 rhyolite I-Type 277 195 70 8.15 0.7353 0.70320 3.31 20.54 0.0975 0.512475 � 3.2 0.3 0.89

31818 monzogranite I/S-Type 290 125 169 2.14 0.7140 0.70518 2.20 12.60 0.1058 0.512457 � 3.5 � 0.2 0.98

Carboniferous

Boundary

Creek

31747 meta-rhyolite I-Type 325 4 112 0.11 0.7073 0.70675 4.28 23.78 0.1089 0.512424 � 4.2 � 0.5 1.06

31775 meta-monzogranite I-Type 331 20 120 0.49 0.7078 0.70551 2.75 14.57 0.1142 0.512423 � 4.2 � 0.7 1.12

31749 meta-granodiorite I-Type 334 5 131 0.11 0.7072 0.70668 4.02 23.77 0.1022 0.512323 � 6.1 � 2.1 1.13

31776 meta-dacite I-Type 355 67 273 0.71 0.7083 0.70473 4.15 20.82 0.1204 0.512545 � 1.8 1.6 0.99

Cambrian

Te Moana 31816 monzogranite I/S-Type 497 209 43 14.36 0.8058 0.70405 2.09 9.26 0.1368 0.512346 � 5.7 � 1.9 1.59

Lake Hill 30644 dacite I/S-Type 517 66 329 0.58 0.7112 0.70685 2.85 15.05 0.1146 0.512399 � 4.7 0.8 1.16

A.M

.Wandres

etal./Sedimentary

Geology168(2004)193–226

205


(13%) and adakitic (7%, at Lake Hill only with I/S-

and S-type affinities) clasts also occur (Fig. 5a). The

broadly calc-alkaline nature of all igneous clasts is

shown in Fig. 5b. TiO2, Al2O3, Fe2O3, MgO, CaO and

Fig. 7. Comparison of Carboniferous calc-alkaline igneous clasts from th

Devonian) igneous provinces from the dispersed Gondwana margin of the

Hillgrove Suite from Shaw and Flood (1981) and Hensel et al. (1985), tho

Malloch (1999), Beresford et al. (1996), Muir et al. (1998), Mortimer et al.

al. (1986, 1987), Pankhurst et al. (1998b), Weaver et al. (1991, and person

Indices. (b) AFM diagram. Fields for calc-alkaline and tholeiitic series ar

M=(MgO).

P2O5 of the igneous clasts all show a negative

correlation with SiO2 with Na2O remaining relatively

constant. Adakitic samples from Lake Hill generally

have higher Al2O3 and Na2O content but no clear

e Boundary Creek conglomerate with relevant Carboniferous (and

SW Pacific (number of samples are given in brackets). Data for the

se for the New Zealand sector of Gondwana from Muir et al. (1996),

(1997) and those for the Antarctic sector of Gondwana from Borg et

al communication) and Leat et al. (1993). (a) Aluminium Saturation

e from Irvine and Baragar (1971); A=(Na2O+K2O), F=(FeO total),

A.M. Wandres et al. / Sedimentary G

distinction can be made from the other samples. Some

minor scatter is observed within the trace elements,

however. V, Cr, Ni, Zn, Ba, Zr and Sr generally

decrease and Rb increases with increasing SiO2.

Adakite samples are depleted in Y and enriched in

Sr; the Sr/Y ratio serves to separate these clasts from

most other rocks.

On multi-element variation diagrams, all the rocks

are enriched in large ion lithophile (LIL) elements and

display a distinct spiked pattern with pronounced

negative anomalies at Nb, all characteristics of sub-

duction-derived magmas (Fig. 6a). Adakitic clasts are

enriched in LIL elements, have positive Ba and Sr

anomalies and are strongly depleted in Y (Fig. 6b).

Rare earth element patterns indicate that adakitic

clasts are lightly to strongly fractionated (LaN/

YbN = 4.47–15.44; mean = 9.32) and have generally

flat HREE (GdN/YbN = 1.23–2.10; mean = 1.65). The

rocks have no or positive Eu anomalies (Eu/

Eu* = 1.00–2.16; mean = 1.47) indicating that no ma-

jor feldspar retention took place in the source (Fig.

6c). Nb–Y systematics (Fig. 6d) suggest that all calc-

alkaline, metaluminous to peraluminous igneous

clasts have volcanic-arc signatures, with some minor

spread into the within-plate and ocean-ridge fields of

Pearce et al. (1984). This spread is not considered to

have any tectonic significance.

Sr–Nd isotope results are listed in Table 4. Initial Sr

isotope ratios (87Sr/86Sri) for the I-type clasts range

from 0.70415 to 0.70623 and have initial eNd (eNdi)ratios ranging from + 0.6 to � 1.3. I/S-type clasts have87Sr/86Sri ratios in the range from 0.70320 to 0.70722

with corresponding eNdi from + 0.3 to � 4.5 and show

an overall negative correlation with 87Sr/86Sri. An S-

type clast is similar to the most radiogenic I/S-type

clast (87Sr/86Sri = 0.70768, eNdI =� 3.7) while an ada-

kitic clast is similar to the compositions displayed by

the I-type and other I/S-type clasts (87Sr/86Sri =

0.70578, eNdI =� 0.7).

Fig. 8. (a and b) Mantle-normalised (McDonough et al., 1992)

multi-element diagrams of representative Carboniferous igneous

clasts and source provinces. References for SW Pacific igneous

source province data are given in Fig. 7. (c) Tectonic discrimination

diagram after Pearce et al. (1984). Fields are for syn-collision-

granite (syn-COLG), volcanic-arc-granite (VAG), within-plate-

granite (WPG) and ocean-ridge-granite (ORG). Abbreviations for

source provinces are given in Fig. 7.

5.3.2. Carboniferous clasts

Dated igneous clasts of Carboniferous age are

unique to the Boundary Creek conglomerate, which

eology 168 (2004) 193–226 207


is dominated by calc-alkaline, metaluminous I-type

clasts (Fig. 7a and b). Major and trace element

variation plots show that TiO2, Al2O3, Fe2O3,

MgO, CaO and P2O5 all have negative correlations

with SiO2. Minor scatter is observed within the

trace elements, but V, Cr, Ni, Zn, Ba and Zr

generally decrease with increasing SiO2, whereas

Sr remains relatively constant. Rb is strongly de-

pleted due to metamorphism. Strong Rb and K

mobility is also shown on a multi-element variation

Fig. 9. Major and trace element diagrams for the two Cambrian Rakaia ig

provinces from the dispersed Gondwana margin of the SW Pacific (nu

Henderson (1986), Withnall et al. (1991) and Crawford et al. (1992), those

Weaver et al. (1984), Borg et al. (1986, 1987), Cox et al. (2000), Wareham

Indices. (b) Mantle-normalised (McDonough et al., 1992) multi-elemen

provinces. (c) Tectonic discrimination diagram after Pearce et al. (1984). F

(VAG), within-plate-granite (WPG) and ocean-ridge-granite (ORG).

diagram for clasts with >70 wt.% SiO2. LIL element

mobility is less pronounced in igneous clasts with

60–70 wt.% SiO2, but all clasts display character-

istics of subduction-derived magmas (Fig. 8a and b).

On a Nb–Y tectonic discrimination diagram all

clasts plot well within the field for volcanic arc

granite (Fig. 8c). Initial Sr isotope ratios for the I-

type clasts range from 0.70473 to 0.70675, with

initial eNdi ratios in the range from + 1.6 to –2.1

(Table 4).

neous clasts compared with relevant penecontemporaneous igneous

mber of samples are given in brackets). Data for Australia from

for New Zealand from Munker (2000) and those for Antarctica from

et al. (2001) and Pankhurst et al. (1998b). (a) Aluminium Saturation

t diagrams for the two igneous clasts and representative source

ields are for syn-collision-granite (syn-COLG), volcanic-arc-granite

Fig. 10. U–Pb ages of igneous clasts from the Rakaia terrane conglomerates compared with ages of representative igneous provinces from the dispersed Gondwana margin of the SW

Pacific. Also shown at left are the detrital zircon age distribution histograms of the Kurow and Balmacaan sandstones calculated at 20-Ma intervals (Wandres and Bradshaw, in press).

A.M

.Wandres

etal./Sedimentary

Geology168(2004)193–226

209

tary Geology 168 (2004) 193–226

5.3.3. Cambrian clasts

The two dated clasts from Lake Hill (dacite) and Te

Moana (monzogranite) are calc-alkaline and weakly

peraluminous (Fig. 9a). TiO2, Al2O3, Fe2O3, MgO,

CaO and P2O5 concentrations in these clasts are neg-

atively correlated with SiO2. Sr, V, Cr, Ni, Zn, Ba and

Zr decrease and Rb increases with increasing SiO2. On

a multi-element variation diagram (Fig. 9b), the plu-

tonic clast from Te Moana is strongly enriched in LIL

elements, whereas the Lake Hill dacite displays a less

inclined trend. However, both rocks display subduc-

tion-related characteristics. On a Nb–Y discriminant

diagram the two clasts plot within the field for volcanic

arc granites (Fig. 9c). The isotope composition of the

monzogranitic clast (87Sr/86Sri = 0.70405 eNdi =� 1.9)

is similar to that of the dacite (87Sr/86Sri = 0.70685

and eNd= + 0.8; Table 4).

A.M. Wandres et al. / Sedimen210

i

6. Igneous source provinces

6.1. Paleozoic and Mesozoic igneous provinces of the

SW Pacific

Igneous provinces of the now dispersed Gondwana

margin (Fig. 2) are briefly summarised in order to

compare and correlate their major and trace element

geochemistry (Figs. 5–9), geochronology (Fig. 10)

Fig. 11. eNd versus87Sr/86Sr and eNd versus

147Sm/144Nd diagrams for Rak

Pacific (number of samples are given in brackets). Igneous clasts are recalc

provinces have been recalculated so that they cover the range of isotopic c

mean age is 260F 20 Ma, therefore, a lower (240 Ma) and an upper (280

calculated and plotted. Errors are F 0.5 for eNd and 0.0001 for Sr iso147Sm/144Nd ratios for Permo-Triassic calc-alkaline igneous clasts from th

New England, Bundarra and Clarence River Supersuites are from Hensel et

from Caprarelli and Leitch (1998); Median Tectonic Zone from Muir et al.

al. (1999); Amundsen Province (MBL) from Pankhurst et al. (1998a); and

are Rakaia sandstone data (grey field) from Frost and Coombs (1989) and W87Sr/86Sr(340) diagram and (d) eNd(340) and

147Sm/144Nd ratios for the Carb

Australia are from Hensel et al. (1985) and McCulloch and Chappell (198

MTZ Muir et al. (1996, 1998), Foulwind Supersuite and Windy Point Gran

for Antarctica from Borg et al. (1987) and Pankhurst et al. (1993, 1998b

expanded to include the isotopic compositions over the age range ca

abbreviations are as in Fig. 6. (e) eNd(500) versus87Sr/86Sr(500) diagram and (

Moana and Lake Hill conglomerates compared with rocks from Tasman

(Munker, 2000), northern Victoria Land (Borg et al., 1987), southern Vi

Mountains (Borg et al., 1990; Wareham et al., 2001) and Marie Byrd Lan

expanded to include the isotopic compositions over the age range of the ign

Wareham et al. (2001), the Central Transantarctic Mountains (CTAM, gray

Glacier and Liv Volcanics.

and their isotope chemistry (Fig. 11) with those of the

Cambrian to Middle Triassic clasts.

Extensive Permo-Triassic igneous complexes are

exposed in the New England Orogen of eastern

Queensland (e.g., Shaw and Flood, 1981; Gust et

al., 1993; Bryant et al., 1997; Allen et al., 1998;

Sivell and McCulloch, 2001) and northeastern New

South Wales, Australia (e.g., Hensel et al., 1985;

Chappell, 1994; Caprarelli and Leitch, 1998).

Permo-Triassic magmatism in New Zealand has been

reported from the Brook Street and Maitai terrane

(Mortimer et al., 1999; Sivell and McCulloch, 2000)

and the Median Tectonic Zone (Kimbrough et al.,

1994; Muir et al., 1998). In Antarctica, penecontem-

poraneous magmatism in Marie Byrd Land is con-

fined to the Amundsen Province (Pankhurst et al.,

1993, 1998b; Mukasa and Dalziel, 2000). The MTZ

and Amundsen Province form the Median Tectonic

Zone/Amundsen Province Igneous Belt (MAIB, Wan-

dres et al., 2004).

Carboniferous magmatism in Australia occurred

mainly in the New England Fold Belt contemporane-

ous with minor intrusions in the Lachlan Fold Belt

(Shaw and Flood, 1981, 1993; Hensel et al., 1985;

Bryant et al., 1997; Caprarelli and Leitch, 1998). In

New Zealand, Carboniferous magmatism is confined to

theWestern Province (Cooper and Tulloch, 1992; Muir

et al., 1994b; Mortimer et al., 1997; Malloch, 1999;

aia terrane igneous clasts compared with source rocks from the SW

ulated to their approximate mean age. Values for the potential source

ompositions calculated from the clasts, e.g., the Permo-Triassic clast

Ma) value for a specific Permo-Triassic source province has been

topes. (a) eNd(260) versus87Sr/86Sr(260) diagram and (b) eNd(260) and

e Te Moana, McKenzie Pass and Lake Hill conglomerates. Data for

al. (1985) and Bryant et al. (1997); Early Permian Volcanics (NEFB)

(1998) and Mortimer et al. (1999); Hekeia Gabbro from Mortimer et

those for Thurston Island from Pankhurst et al. (1993). Also plotted

andres et al. (2004). Abbreviations are as in Fig. 5. (c) eNd(340) versusoniferous clasts from the Boundary Creek conglomerate. Data for

2), for the New Zealand Karamea Batholith, Riwaka Complex and

ite Malloch (1999) and Waight (personal communication), and those

, personal communication). The fields for the source provinces are

lculated from the clasts (340F 15 Ma). If not otherwise stated,

f) eNd(500) and147Sm/144Nd ratios for the Cambrian clasts from the Te

ia (Read Volcanics, Whitford et al., 1990), Devil River Volcanics

ctoria Land (Dry Valleys, Cox et al., 2000), Central Transantarctic

d (Pankhurst et al., 1998b). The fields for the source provinces are

eous clasts (497–517 Ma). Abbreviations are as in Fig. 7. Following

shaded) are subdivided into Gabbro Hill, Beardmore Glacier, Miller


Ewing, 2003) and the Median Tectonic Zone (Kim-

brough et al., 1993, 1994; Beresford et al., 1996; Muir

et al., 1998). In Antarctica, Carboniferous plutons were

emplaced in Northern Victoria Land (Laird et al., 1974;

Stump, 1995) and Ross and Amundsen Province

(Adams, 1987; Pankhurst et al., 1993, 1998b).

Extensive Silurian to Devonian magmatism oc-

curred in the Lachlan Fold Belt (Richards and Single-

ton, 1981; Chappell, 1994), Tasmania (Brooks and

Compston, 1965; McDougall and Leggo, 1965; Cock-

er, 1982; McClenaghan, 1984; Sawka et al., 1990),

Western Province of New Zealand (e.g., Karamea


Batholith, Muir et al., 1994a), Northern Victoria Land

(e.g., Admiralty Intrusives, Borg et al., 1987) and

Ross Province (e.g., Ford Range, Weaver et al., 1991).

Based on U–Pb SHRIMP ages and radiogenic isotope

compositions, many of these are generally too old to

be suitable sources for the igneous clasts from the

Boundary Creek conglomerate. However, the Mid-

dle–Late Devonian Karamea Batholith, the Admiralty

Intrusives and the Ford Range granodiorites are used

here for comparison.

Cambrian magmatism in eastern Australia occurred

in northeastern Queensland in the Charters Tower area

(Henderson, 1986; Withnall et al., 1991; Draper and

Bain, 1997), in central Queensland at the western side

of the New England Fold Belt (Aitchison et al., 1992),

Adelaide and Kanmantoo Fold Belt (e.g., Turner et

al., 1993) and Tasmania (e.g., Crawford and Berry,

1992; Whitford and Crawford, 1993). The Cambrian

igneous rocks of the Takaka terrane, New Zealand

(Cooper, 1989; Munker and Cooper, 1999), have

correlatives in the Bowers terrane in northern Victoria

Land (Weaver et al., 1984; Bradshaw et al., 1985;

Munker and Cooper, 1995). Gibson and Ireland

(1996) suggest that the c. 500 Ma old granitoids in

Fiordland may represent Delamerian/Ross orogen

rocks in New Zealand. Other Late Proterozoic to

Early Paleozoic rocks are exposed in southern Victo-

ria Land (Encarnacion and Grunow, 1996; Cox et al.,

2000), Marie Byrd Land (Pankhurst et al., 1998b, Kay

Peak at Mt. Murphy and Gerish Peak on Bear Penin-

sula), the Central Transantarctic Mountains (CTAM)

(Borg et al., 1990) and the Queen Maud Mountains

(Wareham et al., 2001).

7. Comparing igneous clasts with potential sources

7.1. Permo-Triassic clasts

7.1.1. Geochronology

The dated clasts from the conglomerates at Lake

Hill, McKenzie Pass and Te Moana are plotted to-

gether with igneous provinces in Fig. 10, where all

igneous provinces are referenced in detail. Ten calc-

alkaline clasts give ages from c. 292 to 243 Ma, with

two minor groups recognisable, one ranging in age

from 258 to 243 Ma (Group 1) and the second from

292 to 277 Ma (Group 2). The Group 1 igneous clasts

broadly correlate with plutons from the Median Tec-

tonic Zone (Hekeia Gabbro and Pourakino Trondhje-

mite) and calc-alkaline granitoids from the New

England Fold Belt, namely the New England Super-

suite and the Clarence River Supersuite. The single

adakite granodiorite (UC30659) is similar in age to

the Pourakino and Duncan Creek trondhjemites. In

Marie Byrd Land, the Kinsey Ridge granites are

similar in age to the youngest dated clasts from Group

1, whereas the Triassic intrusives from Thurston

Island are considered to be too young to be a source.

The two-mica Mount Murphy Syenogranite is also too

young to be considered as a source for the single dated

S-type clast from Group 1 (UC30667).

Two of the Early Permian clasts from Group 2

(UC30914 and 31818) correlate well with plutons

from the Western Province (Pomona Island and Man-

apouri granites), and with plutons and volcanics from

the NEFB (Bundarra Suite and Early Permian Vol-

canics). They are also similar in age to the mafic to

intermediate intrusives from Thurston Island (Morgan

Inlet). A second adakitic monzogranite (UC30673) is

similar in age to adakitic Wilbanks diorites reported

from the Kohler Range in Marie Byrd Land. The

Bundarra Suite is too old to be a suitable source for

the S-type clast from Te Moana (UC31813).

The ‘age-gap’ in the Late Permian between Group 1

and 2 igneous clasts is also reflected in the age record of

the igneous provinces from Australia and New Zea-

land. However, the Kohler Range granitoids overlap in

age with both clast groups, and their ages coincide

strongly with the main peaks of the detrital zircon age

distribution of the Rakaia sandstones (Fig. 10).

7.1.1.1. Chemistry. All source suites display calc-

alkaline trends (Fig. 5) with Al2O3, Fe2O3, MgO, CaO

negatively correlated with SiO2. Na2O remains rela-

tively constant and shows a strong overlap between

the compared rocks. Major element concentrations of

igneous clasts correlate well with the felsic members

of all the igneous provinces. Adakitic clasts broadly

correlate with the source adakites but are depleted in

CaO. A similar correlation trend between clasts and

source rocks is displayed by trace elements. Sr content

generally decreases with increasing SiO2 content,

whereas Rb increases. No REE element data are

available for the source adakites, limiting a more

detailed comparison of these rocks. On a multi-ele-


ment variation diagram, the calc-alkaline I-type igne-

ous clasts (>70 wt.% SiO2) correlate well with aver-

aged trends of representative source provinces (Fig.

6a). All the rocks are enriched in LIL elements and

display a distinct spiked pattern with pronounced

negative anomalies at Nb, all characteristics of sub-

duction-derived magmas. The multi-element trends of

the calc-alkaline source-adakites broadly mimic the

trend displayed by the adakites from Lake Hill (Fig.

6b). Using Nb–Y systematics (Pearce et al., 1984),

virtually all clasts and the source province rocks plot

within the field of volcanic-arc-granite (Fig. 6d), with

some minor spread into the within-plate and ocean-

ridge fields (Early Permian Volcanics). The clasts

display a strong affinity with plutons from the Clar-

ence River Supersuite, the Median Tectonic Zone and

the Amundsen Province. Some of the adakitic clast

population is indistinguishable in composition from

that of the source-adakites.

7.1.1.2. Sr–Nd isotope. Isotope data for igneous

clasts and from source provinces are shown in Fig.

11a. To compare the isotopes with the various igneous

source rocks, all the initial 87Sr/86Sri and eNdi valueshave been recalculated to the approximate mean age (c.

260F 20 Ma) of the dated Permo-Triassic clast popu-

lation (Fig. 11a). This age broadly corresponds to the

crystallisation age of most of the penecontemporane-

ous source rocks. The 87Sr/86Sri values for an igneous

clast with a high Rb/Sr ratio (>2; UC31813) and source

plutons with high Rb/Sr ratios have been excluded

from the comparison, as they give unrealistic recalcu-

lated 87Sr/86Sr(260) values. Values for the potential

source provinces have been recalculated so that they

cover the range of isotopic compositions calculated

from the clasts, i.e., a lower (240Ma) and an upper (280

Ma) value for a specific source is plotted. Similar

procedures are applied when comparing the Carbonif-

erous and the Cambrian clasts with source provinces.

On a broad scale, all the Permo-Triassic clasts are

constrained with 87Sr/86Sr(260) between 0.70320 to

0.70750 and eNd(260) values between � 4.3 and + 0.8

(Fig. 9a). I-type clasts range in 87Sr/86Sr(260) from

0.70393 to 0.70638 and in eNd(260) concentrations from� 1.1 to + 0.8. The clasts designated I/S-type have87Sr/86Sr(260) ratios ranging from 0.70320 to 0.70707

and eNd(260) values from � 4.3 to + 0.1. The adakitic

clast has a 87Sr/86Sr(260) ratio of 0.70583 and a

corresponding eNd(260) value of � 0.9, whereas the

single analysed S-type clast from Lake Hill has a87Sr/86Sr(260) ratio of 0.70750 and a eNd(260) value of

� 3.5.

The Nundle Suite and the Early Permian Volcanics

of the New England Fold Belt, as well as rocks from

the Median Tectonic Zone (Hekeia Gabbro) and from

Marie Byrd Land (Kinsey Ridge), are isotopically

more primitive than all the igneous clasts. Isotope

compositions of the I-type and the I/S-type clasts

broadly correlate with radiogenically evolved end-

members of the New England Supersuite, the Clar-

ence River Supersuite, and the Median Tectonic Zone.

These clasts also show good correlation with plutons

from the Bundarra Suite, Amundsen Province and

Thurston Island. The adakitic clast is indistinguishable

from the compositions displayed by plutons from the

Antarctic margin and the Bundarra Suite but strongly

differs in isotopic composition from the isotopically

primitive Pourakino Trondhjemite. The most radio-

genic clasts (UC30668 and 30667) have affinities with

the most radiogenic members of Thurston Island and

point to the involvement of a crustal component

during petrogenesis.

The Sm/Nd ratio (Fig. 11b) is here used as an

additional provenance tracer (McLennan et al., 1990;

Fletcher et al., 1992). All igneous clasts display147Sm/144Nd ratios similar to the crustal average

(0.11, Goldstein et al., 1984). Isotope compositions

of the majority of the source rocks plot well above the

field of the igneous clasts. The more radiogenic

igneous clasts correlate not only with source plutons

from Marie Byrd Land, Thurston Island, the New

England Supersuite and the Bundarra Suite, but they

are also very similar in composition to Rakaia sand-

stones (Frost and Coombs, 1989; Wandres and Brad-

shaw, in press). The sandstones display uniform

continental characteristics, suggesting derivation from

a single source or a mixture of sources whose weighted

isotopic average is similar to that of the sandstones.

7.1.1.3. Clast provenance. Geochronological, Sr–

Nd isotope, and major and trace element data are

incapable of defining a distinct source province for the

Permo-Triassic igneous clasts. Although the tholeiitic

to calc-alkaline Early Permian Volcanics from the

New England Fold Belt are indistinguishable in age

from the Permian clasts, their primitive Sr–Nd iso-

A.M. Wandres et al. / Sedimentary214

tope composition precludes them from being a clast

source. Similarly, the Hekeia Gabbro is geochemically

and isotopically too primitive to be the source for the

clasts. Felsic and intermediate igneous clasts broadly

correlate with penecontemporaneous intermediate to

felsic members of the Clarence River Supersuite, the

Nundle and Bundarra suites and the MTZ Intrusives,

but all these suites are overall isotopically more

primitive than the igneous clasts. Although penecon-

temporaneous adakites occur within the Australian to

Antarctic sectors of Gondwana, their geochemistry

and isotopic compositions correlate poorly with those

displayed by the adakitic clasts from the Lake Hill

conglomerate (Fig. 11b).

Crystallisation ages, geochemistry and Sr–Nd iso-

tope compositions of most of the clasts correlate well

with felsic members of the New England Supersuite

and felsic plutons from the Amundsen Province.

Based on the available data presented here these two

provinces are considered to be possible sources for the

Permo-Triassic igneous clasts. Nd isotope composi-

tions of the more radiogenic igneous clasts, the

Amundsen Province and Thurston Island plutons are

indistinguishable from the Nd isotope compositions of

the Rakaia sub-terrane sandstones (Fig. 11b).

7.2. Carboniferous clasts

The four dated calc-alkaline metaluminous clasts

from the Boundary Creek conglomerate range in age

from 325F 5 to 356F 5 Ma (Fig. 10). They correlate

well with calc-alkaline plutons from the Western

Province of New Zealand (Cape Foulwind Supersuite,

Paringa Tonalite, Kakapo and Hauroko granites) and

plutons from the Fiordland section of the Median

Tectonic Zone (Pomona Island Diorite, Roxburgh

Tonalite, Poteriteri Granite). The Salamander Intru-

sives of northern Victoria Land are similar in age to the

youngest dated clast. The Bruner Hill Syenogranite

and the Mount McCoy Granodiorite (Ross Province)

are indistinguishable in age from the younger igneous

clasts from Boundary Creek. The Chester Mountain

and Ford granitoids, together with the Admiralty

Intrusives and the Karamea Batholith, are too old,

whereas the calc-alkaline, peraluminous Hillgrove

Suite, the Bear Peninsula granites (Amundsen Prov-

ince) and Morgan Inlet orthogneisses (Thurston Is-

land) are too young to be a source for the dated clasts.

7.2.1. Geochemistry

The representative igneous provinces and the igne-

ous clasts from Boundary Creek all display calc-

alkaline trends, and classification by the aluminium

saturation index of Zen (1986) indicates that the Foul-

wind Supersuite, Hillgrove Suite and Karamea Bath-

olith are predominantly peraluminous, whereas the

remaining source provinces contain plutons of both

metaluminous and peraluminous character (Fig. 7a and

b). Major and trace element concentrations of felsic

clasts (>70 wt.% SiO2) correlate well with felsic

members of the igneous provinces. Clasts with < 70

wt.% SiO2 show chemical affinities similar to those

displayed by source province plutons of similar SiO2

content. Most of the clasts are depleted in Rb but are

indistinguishable in Sr content when compared to the

source rocks. On a multi-element diagram representa-

tive source plutons mimic the subduction-related trend

displayed by felsic clasts (Fig. 8a). The A-type Cape

Foulwind Granite has pronounced negative Ba, Sr and

Ti anomalies, reflecting its fractionated nature. The

pronounced negative Nb anomaly suggests either a

subduction-related tectonic setting or a source inheri-

tance. LIL element mobility is less pronounced in

igneous clasts with SiO2 content ranging from 60 to

70 wt.%, and the trend of these clasts is broadly

mirrored by trends of representative source plutons

(Fig. 8b). Median Tectonic Zone, Western Province,

Amundsen and Ross Province plutons plot, like the

clasts, well within the field for volcanic arc granite of

Pearce et al. (1984, Fig. 8c). In contrast, the Cape

Foulwind Supersuite, the Toropuihi Granite, the Kar-

amea Batholith, and the Ford Range intrusives spread

well into the within-plate field, which is thought to

equate with A-type granitoids. Such discriminant dia-

grams should be interpreted with caution as they do not

take into account the fractionation of the discriminat-

ing elements. For example, the calc-alkaline, metal-

uminous to peraluminous I-/S-type Karamea Batholith

plots in both the A-type (least fractionated) and I-/S-

type fields (most fractionated) despite its clearly sub-

duction-related origin (Muir et al., 1996).

7.2.1.1. Isotopes. Radiogenic Sr and Nd isotope

compositions of the Carboniferous igneous clasts

and representative Devonian to Carboniferous igneous

source provinces are compared in Fig. 11c and d.

Initial 87Sr/86Sr ratios and eNd values for igneous

Geology 168 (2004) 193–226

i


clasts and representative source provinces have been

recalculated to the approximate mean age (340F 15

Ma) of the dated Carboniferous clast population.

The clasts have 87Sr/86Sr(340) ratios between

0.70488 and 0.70673 and eNd(340) values from � 2 to

+ 1.5, radiogenic isotope compositions typical of sub-

duction-related magmas. These compositions are sim-

ilar to those of most of the plotted provinces, with three

of the clasts correlating well with rocks from the Ross

Province. The isotopically most primitive clast corre-

lates also with the Hillgrove Suite, whereas the isoto-

pically most evolved clast plots in the field defined by

the Ford granitoids and overlaps with the isotopically

most primitive endmembers of the Admiralty Intru-

sives. Compared with the igneous clasts, the Cape

Foulwind granitoids have eNd(340) values (� 1.9 to 3.3)

indistinguishable from those of the clasts; but they have

more radiogenic 87Sr/86Sr(340) ratios (0.70700 to

0.71073, with Rb/Sr ratios ranging from 0.8 to 6.7;

Windy Point Suite 0.71774 to 0.72546 and Rb/Sr

between 2 and 3), most likely caused by the disturbance

of their Rb/Sr system during Early Cretaceous tectonic

events in the Western Province of New Zealand (T.

Waight, personal communication). A similar correla-

tion pattern is observed when eNd(340) versus147Sm/

144Nd are plotted (Fig. 11d). Isotopic compositions of

two clasts correspond broadly with those displayed by

plutons from the Ross Province. One clast is indistin-

guishable from the compositional ranges displayed by

the Amundsen Province and Cape Foulwind plutons

whereas the most radiogenic clast has compositional

affinities similar to the Ford granitoids.

7.2.1.2. Provenance. No distinct source for the

Carboniferous clasts can be determined. However,

certain igneous provinces can be excluded as prove-

nance contenders. The peraluminous, S-type Hill-

grove Suite is too young in age and overall

isotopically too primitive to be a clast source. The

Cape Foulwind Supersuite and the Toropuihi Granite

correlate well in age with the clasts, however, their

distinct weakly peraluminous to peraluminous A-type

chemistry excludes them as a source. The Western

Province granites (Paringa, Kakapo and Hauroko) are

of similar age and display similar geochemical major

and trace element characteristics as the clasts. Isotope

data are available for the Kakapo Granite only and

show that this pluton is more radiogenic than the

clasts. The high-K, calc-alkaline plutons of the Devo-

nian Karamea Batholith (and rocks of the Riwaka

Complex) are too old and isotopically too radiogenic

(or too primitive in the case of Riwaka) to be a source

contender. Major and trace elements of penecontem-

poraneous plutons exposed in the MTZ broadly cor-

relate with those of the clasts, but the plutons are

isotopically too primitive to be a source for the clasts.

The Admiralty Intrusives are too old in age and are

overall isotopically too radiogenic. The oldest mem-

bers of the Salamander Intrusives broadly correlate in

age with the youngest Carboniferous clast, but the

lack of geochemical data for this province precludes a

more rigorous comparison. Felsic plutons from the

Amundsen Province (Bear Peninsula granites) and

Thurston Island (Morgan Inlet Orthogneiss) broadly

correlate in major and trace element chemistry with

the igneous clasts, but these plutons are too young and

are overall isotopically too primitive to be a clast

source. The Devonian to earliest Carboniferous gran-

itoids exposed in the Ross Province (Ford, Chester,

Hermann Nunatak granitoids) are similar in age to the

oldest Carboniferous clast, but they are too radiogenic

and composed of material overall too old to be a

viable clast source.

The younger Carboniferous plutons from the Ross

Province (Bruner Hill Granite, Mount McCoy Grano-

diorite) correlate well in geochronology and isotopic

composition with the younger Carboniferous igneous

clasts. Based on the available data presented in this

section these Ross Province plutons are the most

likely source for the younger Carboniferous igneous

clasts.

7.3. Cambrian clasts

7.3.1. Geochronology

The Te Moana clast returned an age of 497F 8 Ma

and the Lake Hill clast an imprecise age of c. 517 Ma

(Fig. 10). Nevertheless, the imprecise age is consid-

ered to broadly represent the crystallisation age of this

clast and is used for comparison. The monzogranitic

clast from Te Moana correlates well in age with rocks

from the Mount Windsor Province and rocks now

exposed in northern Victoria Land (Granite Harbour

Intrusives), southern Victoria Land and the Central

Transantarctic Mountains. It is also similar in age to

the orthogneisses in Marie Byrd Land (Ross and


Amundsen provinces). The dacitic clast from Lake

Hill is indistinguishable in age from the Read, Devil

River and Glasgow volcanics, from the Granite Har-

bour Intrusives and the granitoids in southern Victoria

Land. The clast broadly correlates with the Liv

Volcanics and the Marie Byrd Land orthogneisses.

7.3.2. Geochemistry

The igneous provinces contain a wide range of

rock compositions, ranging from gabbro/basalt to

granitoids/rhyolites. These potential source rocks are

dominated by calc-alkaline to high-K calc-alkaline

volcanics, but tholeiitic volcanic members are present.

The aluminium saturation index for the weakly per-

aluminous dacite correlates with weakly peraluminous

members of the Liv and Read volcanics, whereas the

weakly peraluminous monzogranite correlates with

the Granite Harbour Intrusives and the orthogneisses

of Marie Byrd Land (Fig. 9a). TiO2, Al2O3, Fe2O,

MgO, CaO and P2O5 of the representative source

provinces show a general negative correlation with

SiO2, whereas Na2O defines no clear trend. The

dacitic clast has a slightly higher Al2O3 content than

most of the volcanic source rocks but overall is

indistinguishable from volcanic province members

that have similar intermediate SiO2 contents. The

monzogranite shows compositional affinities with

the most felsic plutons from the Mount Windsor

Province, the Granite Harbour Intrusives and the

orthogneisses from Marie Byrd Land. Trace elements

of source rocks show considerably more scatter. Rb

and Sr concentrations of the dacitic clast correlate well

with intermediate members of the Read Volcanics but

they are distinctly higher than concentrations dis-

played by the Glasgow Volcanics. The Liv Volcanics

have similar Rb but lower Sr concentrations. The

monzogranitic clast shows some affinities with the

Granite Harbour Intrusives and the Liv Volcanics. On

a multi-element variation diagram (Fig. 9b), the dacite

follows the similar subduction-derived trend dis-

played by rocks from the Liv volcanics ( < 70 wt.%

SiO2), whereas the fractionated monzogranite is

strongly depleted in Pb, Ba, Sr and Ti but broadly

follows the trend displayed by the Marie Bryd Land

orthogneisses (Kay Peak and Gerrish Peak). On a

discriminant plot (Fig. 9c) of Pearce et al. (1984) all

the calc-alkaline, metaluminous to peraluminous

source provinces plot predominantly within the field

for the volcanic-arc-granites, with only the Liv Vol-

canics spreading well into the syn-collision granite

field. Both igneous clasts plot within the fields de-

fined by the orthogneisses from Marie Byrd Land and

the Benson Volcanics. The monzogranite also corre-

lates with rocks from the Read Volcanics and adakites

from the Dry Valleys, whereas the dacite shows

similarities with the Granite Harbour Intrusives. A

more rigorous geochemical comparison of the two

igneous clasts with source provinces is hampered by

an often incomplete geochemical data record for these

provinces, in particular, trace elements.

7.3.3. Isotopes

All 87Sr/86Sri and eNd(i) values plotted in Fig. 11e

were recalculated to an approximate common age of

500 Ma. The monzogranitic clast has a high Rb/Sr

ratio (4.9) and therefore the 87Sr/86Sri ratio is plotted.

The isotope composition of the dacitic clast (87Sr/86Sr(500) = 0.70699 and eNd(500)= + 0.6) correlates well

with members of the Liv and Benson volcanics and

shows compositional similarities to the granitoids from

Gabbro Hill in the Transantarctic Mountains. The

Read and Heath volcanics are generally isotopically

distinctly more primitive, whereas isotope values for

the intrusive/extrusive rocks from all other provinces

from the Antarctic Gondwana sector indicate various

degrees of crustal involvement during magma petro-

genesis. The monzogranite has an isotope composition

(87Sr/86Sri 0.70405 and eNd(500) � 1.9) that is indistin-

guishable from that of the orthogneisses exposed on

Mount Murphy in Marie Byrd Land. A similar trend is

observed when eNd(500) versus147Sm/144Nd is plotted

(Fig. 11f). The monzogranitic clast corresponds well

with the Mount Murphy orthogneisses and shows

compositional affinities with the Beardmore Glacier

granitoids. The dacite from Lake Hill has a generally

lower 147Sm/144Nd ratio than members of the Liv and

Benson volcanics that have similar eNd(500) values.

7.3.4. Provenance

Available comparative geochronological and geo-

chemical data from the representative Cambrian igne-

ous source provinces indicate that the most favourable

sources for the two igneous clasts are provinces within

the Antarctic sector of the pre-break-up Gondwana

margin. The dacitic clast from Lake Hill correlates

well in age, major, trace and isotope chemistry with


members of the Read Volcanics of Tasmania. How-

ever, its distal location from the Rakaia sedimentary

basin (see Fig. 12) makes the Read Volcanics a less

likely source: this detritus would have to bypass the

Tasmania sedimentary basin (active in the Permo-

Triassic), the Challenger Plateau of the Western Prov-

ince, and the Median Tectonic Zone, to reach the

Rakaia sedimentary basin. The dacitic clast shows

isotopic similarities with the Benson Volcanics but

their intra-oceanic island arc geochemistry differs

from that of the igneous clast. Gabbro Hill plutons

have isotopic similarities but are too old to be a

suitable clast source. The dacitic clast is indistinguish-

able in age, geochemical and Sr–Nd isotopic compo-

sition from the Liv Volcanics, in particular from the

Levett Formation. This igneous province is favoured

as a source for the dacitic clast. Tectonostratigraphic

constraints and detrital contribution from the Trans-

antarctic Mountains to the Rakaia basins are discussed

further below.

Age, geochemistry and isotope data of the calc-

alkaline, weakly peraluminous monzogranitic clast

from Te Moana correlate well with the orthogneisses

Fig. 12. Late Triassic reconstruction of the Gondwana margin indicating th

crustal block configuration are from (Lawver et al., 1999). LHR=Lord How

Byrd Land; AP=Antarctic Peninsula; EWM=Ellsworth–Whitmore Moun

exposed at Kay Peak (Mount Murphy), which are

favoured as the provenance for the monzogranite.

The Bear Peninsula orthogneisses correlate well in

age with the monzogranitic clast but are isotopically

more primitive compared with the clast. Geochrono-

logical and geochemical data for other Cambrian

igneous provinces indicate that these provinces are

either too old (Upper Bingara) or too young in age

(Dry Valleys adakites), radiogenically too primitive

(Read Volcanics) or too radiogenic (Granite Harbour

Intrusives, Miller Glacier plutons, Beardmore Glacier

plutons, Dry Valleys plutons), or were emplaced in an

island arc setting (Devil River Volcanics, Glasgow

Volcanics).

8. Provenance of the Rakaia terrane

8.1. Rakaia terrane provenance

Rakaia sub-terrane sediments are medium to coarse

grained and show a marked compositional uniformity

from the Permian to the Late Triassic (e.g., MacK-

e main features discussed in the text. South Pole, 60j meridian and

e Rise; EMBL=Eastern Marie Byrd Land; WMBL=Western Marie

tains; SA = South America.


innon, 1983; Roser and Korsch, 1999) indicating

either rapid deposition by large energetic rivers that

drained a relatively uniform continental arc source, or

that an efficient mixing system operated during trans-

port and deposition. Clast data have helped to broadly

characterise the age, chemical and isotopic composi-

tion, and petrogenesis of the sources. Assuming that

the clasts in a conglomerate broadly reflect the lith-

otypes exposed in the source area, then the petrogra-

phy of the collected clasts reflects the structural levels

of the source that were available for erosion. Chemical

alteration indices of Roser and Korsch (1999) show

that the Rakaia source was not intensely weathered

and the paucity of mafic clasts in all conglomerates

may be ascribed to the rapid breakdown during

transportation (e.g., Kodama, 1994), or of course an

absence in the source.

Clast petrography indicates that detritus provided

by a magmatic source to the (Permian?) Boundary

Creek conglomerate was mainly from volcanic edifi-

ces and the erosion to subvolcanic levels. A similar

conclusion can be drawn for the Kazanian Te Moana

and the Dorashamian McKenzie Pass conglomerates,

where volcanic and hypersolvus clasts dominate over

subsolvus clasts. At Lake Hill (Carnian), mylonitic and

gneissic clasts indicate that deeper levels of the con-

tinental arc were exposed and eroded, and that volca-

nic lithologies were minor. The exposure of gneissic

structural levels indicates that substantial erosion of

the upper levels of the source took place before

Carnian times. Although the petrogenesis of the ada-

kitic rocks is not uniquely constrained, they probably

attest to the presence of a mature crust, given that one

petrogenetic model for adakites involves the melting

of newly underplated mafic magma beneath a thick-

ened continental crust (Muir et al., 1995; Petford and

Atherton, 1996; Wareham et al., 1997). The apparent

trend shown by geochronological data and petrograph-

ic observations indicates the progressive unroofing of

a continental volcanic/plutonic arc source that experi-

enced continuous magmatism from (at least) the Car-

boniferous to the Middle Triassic. The trend observed

in the igneous clast source is paralleled by the pro-

gressive evolution of the Rakaia accretionary wedge

from the Permian to the Late Triassic (e.g., MacKin-

non, 1983; Roser and Korsch, 1999).

Detrital zircon ages of Permo-Triassic sandstones

from the Western Province (Parapara Peak Group,

Wysoczanski et al., 1997) are very similar to those

of the Rakaia sandstones (Pickard et al., 2000; Wan-

dres and Bradshaw, in press), suggesting a common

source and a possible close proximity of the two New

Zealand provinces in Permo-Triassic times (Wysoc-

zanski et al., 1997). Cambrian magmatism was con-

fined to the Western Province and its Australian and

Antarctic correlatives, and the presence of Cambrian

cobble-sized clasts supports a linkage between the

Rakaia sub-terrane and the Gondwana margin, at least

in Kazanian times. Although the provenance of the

Rakaia sub-terrane clearly indicates derivation from

Gondwana, the lateral continuity of rock units along

the Panthalassan margin of the supercontinent makes

definition of the exact location of the source difficult.

Detrital zircon age distributions from Rakaia sand-

stones identify a Permo-Triassic source as the main

contributor of detritus (Pickard et al., 2000). However,

comparison of igneous conglomerate clasts of that age

with the potential source provinces is ambiguous and

inconclusive, as both the New England Supersuite and

felsic plutons from the Amundsen Province were

identified as likely sources for the Permo-Triassic

igneous clasts. Combining geochronological, geo-

chemical and isotopic data of the Rakaia igneous

clasts, the igneous source for the clasts can be

envisaged as a volcanic/plutonic continental arc that

experienced continuous magmatism into the early

Middle Triassic, and had Cambrian and Carboniferous

plutons and volcanics exposed at the time of erosion

(Fig. 12).

8.2. Australian source

The detrital zircon age populations of our two

Rakaia sandstones are dominated by a distinct 300–

200 Ma age group (Figs. 4 and 10) and are similar to

U–Pb detrital zircon and 40Ar/39Ar muscovite mineral

ages reported by other researchers (e.g., Adams and

Kelley, 1998; Pickard et al., 2000). These age data are

highly characteristic and clearly point to a New

England Fold Belt (NEFB) provenance for the Rakaia

sandstones (Pickard et al., 2000). These authors pro-

posed a model whereby the depocentres were subse-

quently tectonically transported by strike-slip motion

more than 2500 km over 60 Ma, southwards around

the Panthalassan margin of Gondwana into a high-

latitude Cretaceous position. This model is also fa-


vored by Veevers (2000) who noted that the detrital

zircon age population (and provenance) of the Tor-

lesse sandstones is very similar to that of the Early

Triassic Terrigal Formation of the Sydney Basin and

modern beach sands in northern New South Wales

(Hummock Hill Island, Sircombe, 1999), both of

which he considers to be derived from the NEFB.

Based on this observation, Veevers (2000) supports

the model whereby the Torlesse sandstones were

deposited adjacent to the NEFB (Pickard et al., 2000).

However, the Hummock Hill Island beach sand has

a pronounced peak between 400 and 300 Ma (Sir-

combe, 1999, his Fig. 2) and the same age group is the

dominant age group in the Terrigal Formation (Sir-

combe, 1999, his Fig. 4). This 400–300 Ma peak is

almost completely absent from the Triassic Torlesse

muscovite data of Adams and Kelley (1998) and

Adams et al. (1998). 40Ar/39Ar single crystal mineral

ages of detrital muscovites from Rakaia sandstones

show an age population dominated by a major Permo-

Triassic (290–210 Ma) peak and a minor mid-Paleo-

zoic (460–410 Ma) peak. The pronounced Devonian

to Carboniferous peak in the Terrigal Formation and

Hummock Hill Island beach sand differs markedly

from the zircon age distribution observed in the

Triassic Rakaia sandstones from this (Fig. 4b) and

other studies (e.g., Pickard et al., 2000, their Fig. 4f

and g), in which this peak is poorly developed or

absent, and thus detrital zircon mineral ages argue

against rather than for a NEFB origin for the Torlesse

sandstones.

Rb–Sr whole-rock isochron dating of Rakaia

sandstones yields initial 87Sr/86Sr ratios at the time

of metamorphism as low as 0.7065 (Adams and

Graham, 1996; Adams et al., 2002). These authors

believe that the low 87Sr/86Sr initial ratios of the

sandstones at the time of their Permo-Triassic depo-

sition indicate that the detritus was derived from a

dominantly I-type granitoid terrane, with only sub-

ordinate contributions from older and more radio-

genic country rock. In combining their 40Ar/39Ar

detrital mineral age and bulk-rock initial strontium

isotope data, Adams et al. concluded that these

characteristics point uniquely to a provenance in

the NEFB and immediately adjacent to older terranes

of eastern Queensland.

However, I-type granitoids have a distinct metal-

uminous mineralogy in which hornblende, biotite and

titanite are the common mafic mineral phases and

muscovite is conspicuous by its absence. In contrast,

S-type granitoids are characterised by a peraluminous

mineralogy that is marked by the presence of musco-

vite, often accompanied by cordierite and/or other

aluminosilicates. It is therefore questionable whether

peraluminous minerals should or can be used to date

metaluminous I-type magmatism.

Furthermore, muscovite occurs over a wide range

of metamorphic grades, from lower greenschist to

upper amphibolite and blueschist facies (Deer et al.,

1992), but Adams et al. do not distinguish between

igneous and metamorphic muscovite. The NEFB

provenance model for the Torlesse sandstones, based

on combining the 87Sr/86Sr initial ratios and 39Ar/40Ar

muscovite ages, as proposed by Adams et al., is

therefore highly questionable.

In the Permian, the Bowen Basin formed behind the

arc in a back-arc extensional to foreland-basin setting,

between the NEFB and the cratonic block to the west

(Roser and Korsch, 1999, and references therein).

Deposition continued through the earliest Late Triassic

time, and the Bowen Basin fill thus represents the

back-arc deposits of the proposed Rakaia source.

Petrographic data from the Bowen Basin differ signif-

icantly from data from the Rakaia sandstones; and

Roser and Korsch (1999) noted that the compositional

contrast between the Rakaia and Bowen sediments

requires a sorting mechanism that sheds quartzofeld-

spathic detritus into the forearc basin and volcanolithic

to quartzose detritus into the backarc basin. Roser and

Korsch (1999) argued that the Rakaia and the Bowen

sandstones could only have been derived from the

same arc if a major contrast in lithotype (volcanic

versus granitoid) was present across it.

The Permo-Triassic igneous clast population of

the Rakaia sub-terrane shows a good correlation with

New England Supersuite rocks, and a NEFB prove-

nance for the penecontemporaneous Rakaia detritus

is therefore feasible. However, Famennian to Visean

magmatism (Boundary Creek clasts) in eastern Aus-

tralia was restricted to the backarc basin of the

NEFB (both sides along the Sydney Basin) and to

small pockets in north Queensland. Cambrian mag-

matism was, apart from the Upper Bingara Plagiog-

ranite, restricted to Tasmania and north Queensland

(Scheibner and Veevers, 2000, their Fig. 219 and

215). Therefore, the NEFB clearly lacks the older


igneous rocks required to explain the presence of

these older cobble-sized igneous clasts in the Rakaia

conglomerates.

8.3. New Zealand/Antarctica source

The Carboniferous to Triassic granitoids from the

MAIB represent a remnant volcanic/plutonic arc that

was active at the time of Rakaia sandstone deposition.

The close proximity between the Rakaia sub-terrane

and the Western Province of New Zealand indicates

that other Carboniferous igneous rocks (Ross and

Western Province correlatives) were in close proxim-

ity to the MAIB at the time of Permo-Triassic Rakaia

sedimentation. In addition, Cambrian orthogneisses,

now exposed at Mount Murphy and along the Wall-

green Coast and in the Western Province, were also in

the vicinity of the MAIB (Fig. 12).

Other Cambrian rocks in Antarctica are mainly

confined to the Transantarctic Mountains that are

now separated from West Antarctica (Ross Province)

by the West Antarctic rift system (WARS). Two major

extensional phases throughout the Mesozoic and Ce-

nozoic were identified (Trey et al., 1999; Hamilton et

al., 2001) that post-date the Permo-Triassic Rakaia

sedimentation, implying that at the time of sedimen-

tation West Antarctica must have occupied a position

closer or adjacent to the TAM. However, in the

Permo-Triassic, the TAM were part of a depositional

basin and were covered by the Devonian to Triassic

Beacon Supergroup (e.g., Collinson et al., 1994, and

references therein). Detrital contribution from the

TAM to the Rakaia basin is therefore considered to

be minor. Nevertheless, the dacitic clast from the Lake

Hill conglomerate correlates well with the Liv Vol-

canics of the Central TAM. The presence of a TAM-

derived clast in the Rakaia depocentre may indicate

that prior to the opening of the WARS, and at the time

of the Rakaia sedimentation, the TAM rocks may have

been exposed to erosion along the Panthalassan edge

of the Central Transantarctic Sedimentary Basin. The

edge of the sedimentary basin is now covered by the

Ross Ice Shelf and any contribution of Cambrian

TAM detritus to the Rakaia basin remains speculative.

The Cambrian monzogranite from Te Moana cor-

relates best with basement rocks of West Antarctica

(Ross Province), which is favoured as the Cambrian

protosource for the Rakaia detritus. The subdued

Cambrian detrital zircon age peaks in both Rakaia

sandstones from this study (and other studies) are

ascribed to the detrital contribution of the Amundsen/

Ross Province basement.

Magmatism in the source and sedimentation of

the Rakaia sandstones were penecontemporaneous

and the sedimentary basins received boulder/cobble/

pebble/sand detritus from this source. The presence

of boulder- and cobble-sized clasts suggests a short

travel distance for the coarse detritus (see above).

Age signatures of detrital zircons from Kurow Hill

and Balmacaan Stream correlate well with igneous

clast ages, indicating that the depositional basins

were positioned adjacent to the volcanic/plutonic

arc segment from which the clasts were derived.

The MAIB (including the older provinces) com-

prised plutons and volcanics of ages appropriate to

explain the igneous clast population sampled in the

Rakaia conglomerates and to account for the detrital

zircon age signatures in the Rakaia sandstones. In

stark contrast to eastern Australia, there are very

few rocks exposed in West and East Antarctica

( < 1% of the total area, Stump, 1995), and inter-

pretations offered in this study are based inevitably

on data from a few representative outcrops, but

geophysical evidence indicates that the inferred

exposed source rocks are regionally extensive under

the ice in the proposed source area (e.g., Trey et

al., 1999; Hamilton et al., 2001). Nevertheless,

geochronology and geochemistry of the MAIB

(and its older inboard provinces) correlate broadly

with igneous clasts from the four Rakaia conglom-

erates. The Amundsen Province and Ross Province

are postulated as likely sources for the Permo-

Triassic, Carboniferous and Cambrian igneous

clasts, respectively.

The Permian detrital zircon age peaks for the

Kurow Hill and Balmacaan Stream sandstones (and

other Rakaia sandstones) support the conclusion that

granitoids from the Kohler Range were a major

source for the Triassic sediments. Isotope data pre-

sented in Fig. 11b indicate that the simple erosion of

Amundsen Province plutons could produce at least

part of the Rakaia sub-terrane detritus. The minor

Carboniferous age component in the Kurow Hill

sample resembles that of the Ross Province magma-

tism. The same peak is present but rather subdued in

the Balmacaan Stream sample. The exposure of a


Cambrian protosource next to or within the MAIB is

also shown in both sandstones, with both having a

subdued Cambrian peak.

Apart from the Early Permian fusulinid fauna

within the Akatarawa terrane (Fig. 1), the inferred

cold-water affinities of the Rakaia faunal assemblages

support deposition of the Rakaia detritus in a high-

latitude environment (Shi and Grunt, 2000; Cawood et

al., 2002). Low chemical alteration indices from 52.5

to 54.2 (Roser and Korsch, 1999) indicate that the

erosion of the Rakaia source was not accompanied by

intense weathering and also support a high latitude

position of the Rakaia source. In addition, paleomag-

netic studies have closely constrained the apparent

polar-wander path (APWP) of the Australian conti-

nent and its Gondwana neighbours in the Late Paleo-

zoic, Mesozoic and Cenozoic (Schmidt and Clark,

2000, and references therein). The Late Carbonifer-

ous–Permian (327–250 Ma) pole path indicates a

polar latitude up to the end of the Permian for the

Australian part of Gondwana. The Triassic to Early

Jurassic group of poles indicates mid to high southern

latitudes for Australia before its return to higher

latitudes (Schmidt and Clark, 2000), with Antarctica

and the Western Province of New Zealand generally

depicted as being in the vicinity of the South Pole

(Powell and Li, 1994; Mukasa and Dalziel, 2000) This

further supports an Antarctic source for the Rakaia

sub-terrane as postulated here.

8.4. Conclusions

A detailed sampling program of cobble- to boul-

der-sized igneous clasts from four Rakaia sub-terrane

conglomerates and tectonostratigraphic constraints

have helped to broadly characterise the igneous source

of the Rakaia sub-terrane.

Geochronology, petrography, geochemistry and

Sr–Nd isotopes of these clasts point towards a

progressive unroofing of a continental volcanic/plu-

tonic arc source that experienced continuous magma-

tism from the Carboniferous to the latest Middle

Triassic. Source magmatism and deposition of the

Rakaia sediments were penecontemporaneous, as in-

dicated by the presence of a late Early Permian

volcanic igneous clast in the early Late Permian Te

Moana conglomerate and further supported by detrital

zircon ages of the sandstones. The presence of ada-

kitic rocks in the source area may attest to the

presence of a mature crust, whereas the exposure of

gneissic structural levels at the time of Rakaia sub-

terrane sedimentation indicates that erosion of the

upper levels of the source took place before Carnian

times.

The occurrence of Cambrian igneous clasts in the

Kazanian Te Moana and the Carnian Lake Hill con-

glomerates indicates that Cambrian plutons and vol-

canics were a source that provided first-cycle cobble-

to boulder-sized detritus to the Rakaia depocentres.

Cambrian magmatism was confined to the New Zea-

land Western Province and its Australian and Antarc-

tic correlatives as well as the TAM and their

Australian correlatives. The Cambrian clasts together

with very similar detrital zircon age signatures of the

Rakaia sandstones and the Parapara Peak Group

therefore attest to a close proximity of the Eastern

and Western provinces of New Zealand in Kazanian

times.

Detrital zircon ages of Rakaia sandstones and

igneous clasts ages broadly correlate with crystallisa-

tion ages of plutons and volcanics from the Amundsen

Province and Ross Province, and the Antarctic sector

of the Panthalassan Gondwana margin is postulated as

the likely source for the Permian to Triassic, Carbon-

iferous and Cambrian igneous clasts and the Rakaia

sediments.

The conspicuous absence of Cambrian and Car-

boniferous magmatism along the NEFB sector of the

Panthalassan Gondwana margin, together with the

discrepancy between the detrital mineral age data

from the Rakaia sandstones and the NEFB igneous

data, do not support a NEFB provenance for the

Rakaia sandstones.

Acknowledgements

This paper represents part of a PhD thesis by the

first author, funded by a scholarship at the University

of Canterbury. The costs of field and laboratory work

were partially funded by the Mason Trust and a

University of Canterbury research grant. We thank

Barry Roser and Andrew Morton for providing

constructive and helpful journal review. Mo Turnbull

is thanked for field assistance at the Boundary Creek

conglomerate.


References

Adams, C.J., 1987. Geochronology of granite terranes in the Ford

Ranges, Marie Byrd Land, West Antarctica. New Zealand Jour-

nal of Geology and Geophysics 30, 54–72.

Adams, C.J., Graham, I.J., 1996. Metamorphic and tectonic geochro-

nology of the Torlesse terrane, Wellington, New Zealand. New

Zealand Journal of Geology and Geophysics 39 (2), 157–180.

Adams, C.J., Kelley, S., 1998. Provenance of Permian–Triassic and

Ordovician metagraywacke terranes in New Zealand: evidence

from Ar-40/Ar-39 dating of detrital micas. Geological Society of

America Bulletin 110 (4), 422–432.

Adams, C.J., Barley, M.E., Fletcher, I.R., Pickard, A.L., 1998. Ev-

idence from U–Pb zircon and 40Ar/39Ar muscovite detrital min-

eral ages in metasandstones for movement of the Torlesse suspect

terrane around the eastern margin of Gondwanaland. Terra Nova,

The European Journal of Geosciences 10 (4), 183–189.

Adams, C.J., Barley, M.E., Maas, R., Doyle, M.G., 2002. Prove-

nance of Permian–Triassic volcaniclastic sedimentary terranes

in New Zealand: evidence from their radiogenic isotope charac-

teristics and detrital mineral age patterns. New Zealand Journal

of Geology and Geophysics 45, 221–242.

Aitchison, J.C., Ireland, T.R., Blake Jr., M.C., Flood, P.G., 1992. 530

Ma zircon age for ophiolite from New England Orogen: oldest

rocks known from Eastern Australia. Geology 20, 125–128.

Allen, C.M., Williams, I.S., Stephens, C.J., Fielding, C.R., 1998.

Granite genesis and basin formation in an extensional setting:

the magmatic history of the northernmost New England Orogen.

Australian Journal of Earth Sciences 45 (6), 875–888.

Andrews, P.B., Speden, I.G., Bradshaw, J.D., 1976. Lithological

and paleontological content of the Carboniferous–Jurassic Can-

terbury Suite, South Island, New Zealand. New Zealand Journal

of Geology and Geophysics 19 (6), 791–819.

Bassett, K.N., 2000. Isotopic provenance analysis and terrane tec-

tonics: a warning about sediment transport distances. In: Skil-

beck, C., Hubble, T.C. (Eds.), Understanding Planet Earth:

Searching for a Sustainable Future. Abstracts of the 15th Aus-

tralian Geological Convention. University of Technology, Syd-

ney, NSW, Australia, p. 24.

Beresford, S.W., Bradshaw, J.D., Weaver, S.D., Muir, R.J., 1996.

Echinus granite and pepin group of Pepin Island, Northeast

Nelson, New Zealand—Drumduan terrane basement or exotic

fragment in the Median Tectonic Zone. New Zealand Journal of

Geology and Geophysics 39 (2), 265–270.

Bishop, D.G., 1972. Progressive metamorphism from prehnite –

pumpellyite to greenschist facies in the Dansey Pass area,

Otago, New Zealand. Geological Society of America Bulletin

83 (11), 3177–3197.

Bishop, D.G., Bradshaw, J.D., Landis, C.A., 1983. Provisional ter-

rane map of South Island, New Zealand. In: Howell, G., Jones,

D., Cox, A., Nur, A. (Eds.), Proceedings of the Circum-Pacific

Terrane Conference. Stanford University Publications. Geolog-

ical Sciences. Stanford University, School of Earth Sciences,

Stanford, CA, United States, pp. 24–31.

Bishop, D.G., Bradshaw, J.D., Landis, C.A., 1985. Provisional ter-

rane map of South Island, New Zealand. In: Howell, D.G. (Ed.),

Tectonostratigraphic terranes of the Circum-Pacific region. Cir-

cum-Pacific Council for Energy and Mineral Resources, Huston,

Texas, United States, pp. 515–521.

Borg, S.G., Stump, E., Holloway, J.R., 1986. Granitoids of North-

ern Victoria Land Antarctica: a reconnaissance study of field

relations, petrography and geochemistry. Antarctic Research Se-

ries 46, 115–188.

Borg, S.G., Stump, E., Chappell, B.W., McCulloch, M.T., Wyborn,

D., Armstrong, R.L., Holloway, J.R., 1987. Granitoids of north-

ern Victoria Land, Antarctica; implications of chemical and iso-

topic variations to regional crustal structure and tectonics.

American Journal of Science 287 (2), 127–169.

Borg, S.G., DePaolo, D.J., Smith, B.M., 1990. Isotopic structure

and tectonics of the Central Transantarctic Mountains. Journal

of Geophysical Research, B Solid Earth and Planets 95 (B5),

6647–6667.

Bradshaw, J.D., 1973. Allochthonous Mesozoic fossil localities in

melange in the Torlesse Group of North Canterbury. Journal of

the Royal Society of New Zealand 3 (2), 161–167.

Bradshaw, J.D., 1989. Cretaceous geotectonic pattern in the New

Zealand Region. Tectonics 8, 803–820.

Bradshaw, J.D., Weaver, S.D., Laird, M.G., 1985. Suspect terranes

in North Victoria Land, Antarctica. In: Howell, G., Jones, D.,

Cox, A., Nur, A. (Eds.), Proceedings of the Circum-Pacific

Terrane Conference. Stanford University Publications. Geolog-

ical Sciences. Stanford University, School of Earth Sciences,

Stanford, CA, United States, pp. 36–39.

Brooks, C., Compston, W., 1965. The age and initial 87Sr/86Sr of

the Heemskirk granite, western Tasmania. Journal of Geophys-

ical Research, B Solid Earth and Planets 70 (24), 6249–6262.

Bryant, C.J., Arculus, R.J., Chappell, B.W., 1997. Clarence river

supersuite—250 Ma cordilleran tonalitic I-type intrusions in

Eastern Australia. Journal of Petrology 38 (8), 975–1001.

Campbell, J.D., Warren, G., 1965. Fossil localities of the Torlesse

group in the South Island. Transactions of the Royal Society of

New Zealand. Geology 3 (8), 99–137.

Caprarelli, G., Leitch, E.C., 1998. Magmatic changes during the

stabilisation of a cordilleran fold belt: the Late Carbonifer-

ous–Triassic igneous history of eastern New South Wales, Aus-

tralia. Lithos 45, 413–430.

Cawood, P.A., Nemchin, A.A., Leverenz, A., Saeed, A., Bal-

lance, P.F., 1999. U/Pb dating of detrital zircons: implications

for the provenance record of Gondwana margin terranes. Geo-

logical Society of America Bulletin 111 (8), 1107–1119.

Cawood, P.A., Landis, C.A., Nemchin, A.A., Hada, S., 2002. Perm-

ian fragmentation, accretion and subsequent translation of a

low-latitude Tethyan seamount to the high-latitude East Gond-

wana margin; evidence from detrital zircon age data. Geological

Magazine 139 (2), 131–144.

Chappell, B.W., 1994. Lachlan and New England fold belts of

contrasting magmatic and tectonic development; discussion.

Journal and Proceedings of the Royal Society of New South

Wales 127, 47–59.

Chappell, B.W., White, A.J.R., 1974. Two contrasting granite types.

Pacific Geology 8, 173–174.

Cocker, J.D., 1982. Rb–Sr geochronology and Sr isotopic compo-

sition of Devonian granitoids, eastern Tasmania. Journal of the

Geological Society of Australia 29, 139–158.


Collinson, J.W., Isbell, J.L., Elliot, D.H., Miller, M.F., Miller,

J.M.G., Veevers, J.J., 1994. Permian–Triassic transantarctic

basins. In: Veevers, J., Powell, M. (Eds.), Permian–Triassic

Pangean Basins and Foldbelts Along the Panthalassan Mar-

gin of Gondwanaland. Memoir—Geological Society of Amer-

ica. Geological Society of America (GSA), Boulder, CO, US,

pp. 173–222.

Coney, P.J., Edwards, A., Hine, R., Morrison, F., Windrim, D.,

1990. The regional tectonics of the Tasman orogenic system,

eastern Australia. Journal of Structural Geology 12, 519–543.

Coombs, D.S., Landis, C.A., Norris, R.J., Sinton, J.M., Borns, D.J.,

Craw, D., 1976. The Dun Mountain Ophiolite Belt, New Zea-

land, its tectonic setting, constitution and origin, with special

reference to the southern portion. American Journal of Science

276, 561–603.

Cooper, R.A., 1989. Early Paleozoic terranes of New Zealand.

Journal of the Royal Society of New Zealand 19 (1), 73–112.

Cooper, R.A., Tulloch, A.J., 1992. Early Palaeozoic terranes in

New Zealand and their relationship to the Lachlan fold belt.

Tectonophysics 214, 129–144.

Cox, S.C., Parkinson, D.L., Allibone, A.H., Cooper, A.F., 2000.

Isotopic character of Cambro-Ordovician plutonism, southern

Victoria Land, Antarctica. New Zealand Journal of Geology

and Geophysics 43, 501–520.

Crawford, A.J., Berry, R.F., 1992. Tectonic implications of Late

Proterozoic–Early Paleozoic igneous rock associations in west-

ern Tasmania. Tectonophysics 214, 129–144.

Crawford, A.J., Corbett, K.D., Everard, J.L., 1992. Geochemistry of

the Cambrian volcanic-hosted massive sulfide-rich Mount Read

Volcanics, Tasmania, and some tectonic implications. In: Large,

R.R. (Ed.), A Special Issue Devoted to Australian Volcanic-

Hosted Massive Sulfide (VHMS) Deposits and Their Volcanic

Environment. Economic Geology and the Bulletin of the Soci-

ety of Economic Geologists. Economic Geology Publishing,

Lancaster, PA, US, pp. 597–619.

Deer, W.A., Howie, R.A., Zussman, J., 1992. An Introduction to

the Rock Forming Minerals, 2nd ed. Longman, Harlow, Essex.

695 pp.

Draper, J.J., Bain, J.H.C., 1997. A brief geological hystory of North

Queensland Geology. In: Bain, J.H., Draper, J.J. (Eds.), North

Queensland Geology, AGSO Bulletin 240, Queensland Geology,

vol. 9. Australian Geological Survey Organisation, Canberra,

Australia.

Eby, G.N., 1984. Monteregian Hills: 1. Petrography, major and

trace element chemistry of the western intrusions: Mounts Roy-

al, St. Bruno and Johnson. Journal of Petrology 25, 421–452.

Encarnacion, J., Grunow, A., 1996. Changing magmatic and tec-

tonic styles along the paleo-Pacific margin of Gondwana and the

onset of early Paleozoic magmatism in Antarctica. Tectonics 15

(6), 1325–1341.

Ewing, T., 2003. Provenance of the Loch Burn formation, Eastern

Fiordland: implications for the Median Tectonic Zone. Unpub-

lished MSc Thesis, University of Canterbury, Christchurch,

New Zealand. 267 pp.

Ferguson, R., Hoey, T., Wathen, S., Werritty, A., 1996. Field evi-

dence for rapid downstream fining of river gravels through se-

lective transport. Geology 24, 179–182.

Fletcher, I.R., Paul, D.K., Trendall, A.F., 1992. Sm–Nd and geo-

chemical characteristics of metasedimentary rocks at Mt.

Narryer, Western Australia. Australian Journal of Earth Sciences

39, 67–78.

Floettmann, T., Gibson, G.M., Kleinschmidt, G., 1993. Structural

continuity of the Ross and Delamerian orogens of Antarctica

and Australia along the margin of the paleo-Pacific. Geology 21

(4), 319–322.

Foster, D.A., Gray, D.R., 2000. Evolution and stucture of the

Lachlan fold belt (orogen) of eastern Australia. Annual Review

of Earth and Planetary Sciences 28, 47–80.

Frost, C.D., Coombs, D.S., 1989. Nd isotope character of New

Zealand sediments: implications for terrane concepts and crustal

evolution. American Journal of Science 289, 744–770 (June).

George, A.D., 1992. Deposition and deformation of an Early Cre-

taceous trench-slope basin deposit, Torlesse terrane, New Zea-

land. Geological Society of America Bulletin 104 (5), 570–580.

Gibson, G.M., Ireland, T.R., 1996. Extension of Delamerian (Ross)

orogen into western New Zealand—evidence from zircon ages

and implications for crustal growth along the Pacific margin of

Gondwana. Geology 24 (12), 1087–1090.

Goldstein, S.L., O’Nions, R.K., Hamilton, P.J., 1984. A Sm–Nd

isotope study of atmospheric dusts and particulates from major

river systems. Earth and Planetary Science Letters 70, 221–236.

Gust, D.A., Stephens, C.J., Grenfell, A.T., 1993. Granitoids of the

northern NEO: their distribution in time and space and their

tectonic implication. In: Flood, P.G., Aitchison, J.C. (Eds.),

New England Orogen, Eastern Australia. Department of Geol-

ogy and Geophysics, University of New England, Armidale,

Australia, pp. 565–572.

Hada, S., Landis, C.A., 1995. Te Akatarawa Formation: an exotic

oceanic–continental margin terrane within the Torlesse–Haast

Schist transition zone. New Zealand Journal of Geology and

Geophysics 38 (3), 349–359.

Hamilton, R.J., Luyendyk, B.P., Sorlien, C.C., Bartek, L.R., 2001.

Cenozoic tectonics of the Cape Roberts rift basin and Trans-

antarctic Mountain front, southwestern Ross Sea, Antarctica.

Tectonics 20, 325–342.

Henderson, R.A., 1986. Geology of the Mount Windsor subpro-

vince—a lower Paleozoic volcano-sedimentary terrane in the

northern Tasman orogenic zone. Australian Journal of Earth

Sciences 33, 343–364.

Hensel, H.D., McCulloch, M.T., Chappell, B.W., 1985. The New

England Batholith: constraints on its derivation from Nd and Sr

isotopic studies of granitoids and country rocks. Geochimica et

Cosmochimica Acta 49 (2), 369–384.

Howell, D.G., 1980. Mesozoic accretion of exotic terranes along

the New Zealand segment of Gondwanaland. Geology 8 (10),

487–491.

Ireland, T.R., 1992. Crustal evolution of New Zealand: evidence

from age distributions of detrital zircons in Western Province

paragneisses and Torlesse greywacke. Geochimica et Cosmochi-

mica Acta 56, 911–920.

Irvine, T.N., Baragar, W.R.A., 1971. A guide to the chemical clas-

sification of the common volcanic rocks. Canadian Journal of

Earth Sciences 8, 523–548.

Jones, P.J., 1995. AGSO Phanerozoic Timescale: Wallchart and


Explanatory Notes. Oxford Univ. Press, Melbourne, Australia.

32 pp.

Kimbrough, D.L., Tulloch, A.J., Geary, E., Coombs, D.S., Landis,

C.A., 1993. Isotopic ages from the Nelson region of South

Island, New Zealand: crustal structure and definition of the

Median Tectonic Zone. Tectonophysics 225 (4), 433–448.

Kimbrough, D.L., Tulloch, A.J., Coombs, D.S., Landis, C.A., John-

ston, M.R., Mattinson, J.M., 1994. Uranium– lead zircon ages

from the Median Tectonic Zone, South Island, New Zealand.

New Zealand Journal of Geology and Geophysics 37 (4),

393–419.

Kodama, Y., 1994. Downstream changes in the lithology and grain

size of fluvial gravels, the Watarase River, Japan: evidence of

the role of abrasion in downstream fining. Journal of Sedimen-

tary Research A64, 68–75.

Korsch, R.J., Wellman, H.W., 1988. The geological evolution of

New Zealand and the New Zealand region. In: Nairn, A.E.M.,

Stehli, F.G., Uyeda, S. (Eds.), The Oceanic Basins and Mar-

gins Volume 7B: The Pacific Ocean. Plenum Press, New York,

pp. 411–482.

Laird, M.G., Andrews, P.B., Kyle, P.R., 1974. Geology of the

Northern Evans Neve, Victoria Land, Antarctica. New Zealand

Journal of Geology and Geophysics 17, 587–601.

Lawver, L.A., Dalziel, I.W.D., Gahagan, L.M., 1999. Antarctica

Keystone of Gondwana, http://www.ig.utexas.edu/research/

plates. Institute for Geophysics, University of Texas, Austin,

Texas, USA.

Leat, P.T., Storey, B.C., Pankhurst, R.J., 1993. Geochemistry of

Paleozoic–Mesozoic Pacific rim orogenic magmatism, Thur-

ston Island area, West Antarctica. Antarctic Science 5 (3),

281–296.

MacKinnon, T., 1983. Origin of the Torlesse terrane and coeval

rocks, South Island, New Zealand. Geological Society of Amer-

ica Bulletin 94, 967–985.

Malloch, K.R., 1999. The petrography and geochemistry of the

Carboniferous A-Type granites, Western Province, New Zea-

land. Unpublished MSc Thesis. University of Canterbury,

Christchurch, New Zealand. 201 pp.

Mazengarb, C., Harris, D.H.M., 1994. Cretaceous stratigraphic and

structural relations of the Raukumara Peninsula, New Zealand:

stratigraphic patterns associated the migration of a thrust system.

Annales Tectonicae VIII, 100–118.

McClenaghan, M.P., 1984. Granitoids of northeastern Tasmania.

Journal of Australian Geology and Geophysics 9, 303–312.

McCulloch, M.T., Chappell, B.W., 1982. Nd isotopic characteristics

of S- and I-type granites. Earth and Planetary Science Letters 58

(1), 51–64.

McDonough, W.F., Sun, S.S., Ringwood, A.E., Jagoutz, E., Hof-

mann, A.W., 1992. Potassium, rubidium, and cesium in the

Earth and Moon and the evolution of the mantle of the Earth.

In: McLennan, M.S., Rudnick, R. (Eds.), The Taylor Colloqui-

um: Origin and Evolution of Planetary Crusts. Pergamon, Ox-

ford, pp. 1001–1012.

McDougall, I., Leggo, P.G., 1965. Isotopic age determinations of

granitic rocks from Tasmania. Journal of the Geological Society

of Australia 12, 295–332.

McLennan, S.M., Taylor, S.R., McCulloch, M.T., Maynard, J.B.,

1990. Geochemical and Nd–Sr isotopic composition of deep-

sea turbidites: crustal evolution and plate tectonic associations.

Geochimica et Cosmochimica Acta 54 (7), 2015–2050.

Moore, G.W., Drummond, K.J., Chikao, N., Corvalan, J., Doutch,

H.F., Cradock, C., 1982. Plate-tectonic map of the circum-Pa-

cific region. Circum-Pacific Map Project. The American Asso-

ciation of Petroleum Geologists, Tulsa, OK 74101, USA,

p. 274.

Mortimer, N., Tulloch, A.J., Ireland, T.R., 1997. Basement geology

of Taranaki and Wanganui Basins, New Zealand. New Zealand

Journal of Geology and Geophysics 40 (2), 223–236.

Mortimer, N., Gans, P., Calvert, A., Walker, N., 1999. Geology

and thermochronometry of the east edge of the Median Bath-

olith (Median Tectonic Zone): a new perspective on Permian to

Cretaceous crustal growth of New Zealand. Island Arc 8 (3),

404–425.

Muir, R.J., Ireland, T.R., Weaver, S.D., Bradshaw, J.D., 1994a. Ion

microprobe U–Pb zircon geochronology of granitic magmatism

in the Western Province of the South Island, New Zealand.

Chemical Geology 113 (1–2), 171–189.

Muir, R.J., Ireland, T.R., Weaver, S.D., Bradshaw, J.D., 1994b.

Geochronology of Paleozoic magmatism in the Western Prov-

ince, South Island, New Zealand; implications for correlations

with SE Australia, Tasmania and Antarctica. In: Neil, H., Gil-

lespie, J.L., Moon, V., Briggs, R. (Eds.), Geological Society of

New Zealand 1994 Annual Conference; Programme and

Abstracts. Geological Society of New Zealand Miscellaneous

Publication. Geological Society of New Zealand, Christchurch,

New Zealand, p. 139.

Muir, R.J., Weaver, S.D., Bradshaw, J.D., Eby, G.N., Evans, J.A.,

1995. The Cretaceous Separation Point Batholith, New Zea-

land: granitoid magmas formed by melting of mafic litho-

sphere. Journal of the Geological Society (London) 152 (Part

4), 689–701.

Muir, R.J., Weaver, S.D., Bradshaw, J.D., Eby, G.N., Evans, J.A.,

Ireland, T.R., 1996. Geochemistry of the Karamea Batholith,

New Zealand and comparisons with the Lachlan Fold Belt gran-

ites of southeast Australia. Lithos 39 (1–2), 1–20.

Muir, R.J., Ireland, T.R., Weaver, S.D., Bradshaw, J.D., Waight,

T.E., Jongens, R., Eby, G.N., 1997. SHRIMP U–Pb geochro-

nology of Cretaceous magmatism in northwest Nelson-West-

land, South Island, New Zealand. New Zealand Journal of

Geology and Geophysics 40, 453–463.

Muir, R.J., Ireland, T.R., Weaver, S.D., Bradshaw, J.D., Evans,

J.A., Eby, G.N., Shelley, D., 1998. Geochronology and geo-

chemistry of a Mesozoic magmatic arc system, Fiordland,

New Zealand. Journal of the Geological Society (London) 155

(Part 6), 1037–1052.

Mukasa, S.B., Dalziel, I.W.D., 2000. Marie Byrd Land, West Ant-

arctica: evolution of Gondwana’s Pacific margin constrained by

zircon U–Pb geochronology and feldspar common-Pb isotopic

compositions. Geological Society of America Bulletin 112,

611–627.

Munker, C., 2000. The isotope and trace element budget of the

Cambrian Devil River arc system, New Zealand: identifica-

tion of four source components. Journal of Petrology 41 (6),

759–788.

http:\\www.ig.utexas.edu\research\plates


Munker, C., Cooper, R.A., 1995. The island arc setting of a New

Zealand Cambrian volcano-sedimentary sequence—implica-

tions for the evolution of the SW Pacific Gondwana fragments.

Journal of Geology 103 (6), 687–700.

Munker, C., Cooper, R.A., 1997. The early Paleozoic Takaka Ter-

rane, New Zealand, as part of the Australian/Antarctic Gond-

wana margin: new insights from the trace element and isotope

geochemistry of Cambrian volcanics. In: Bradshaw, J.D., Weav-

er, S.D. (Eds.), Terrane Dynamics 97, International Conference

on Terrane Geology, Christchurch, New Zealand, pp. 124–127.

Munker, C., Cooper, R., 1999. The Cambrian arc complex of the

Takaka Terrane, New Zealand: an integrated stratigraphical, pa-

leontological and geochemical approach. New Zealand Journal


Nakamura, N., 1974. Determination of REE, Ba, Fe, Mg, Na, and K

in carbonaceous and ordinary meteorites. Geochimica et Cos-

mochimica Acta 38, 757–775.

Pankhurst, R.J., Millar, I.L., Grunow, A.M., Storey, B.C., 1993. The

pre-Cenozoic magmatic history of the Thurston Island crustal

block, West Antarctica. Journal of Geophysical Research, B,

Solid Earth and Planets 98 (7), 11,835–11,849.

Pankhurst, R.J., Leat, P.T., Sruoga, P., Rapela, C.W., Marquez, M.,

Storey, B.C., Riley, T.R., 1998a. The Chon Aike province of

Patagonia and related rocks in West Antarctica: a silicic large

igneous province. Journal of Volcanology and Geothermal Re-

search 81 (1–2), 113–136.

Pankhurst, R.J., Weaver, S.D., Bradshaw, J.D., Storey, B.C., Ire-

land, T.R., 1998b. Geochronology and geochemistry of pre-Ju-

rassic superterranes in Marie Byrd Land, Antarctica. Journal of

Geophysical Research, B, Solid Earth and Planets 103 (B2),

2529–2547.

Pearce, J.A., Harris, N.B.W., Tindle, A.G., 1984. Trace element

discrimination diagrams for the tectonic interpretation of granit-

ic rocks. Journal of Petrology 25, 956–983.

Pell, S.D., Williams, I.S., Chivas, A.R., 1997. The use of proto-

lith zircon-age fingerprints in determining the protosource

areas for some Australian dune sands. Sedimentary Geology

109, 233–260.

Petford, N., Atherton, M., 1996. Na-rich partial melts from newly

underplated basaltic crust; the Cordillera Blanca Batholith, Peru.

Journal of Petrology 37 (6), 1491–1521.

Pickard, A.L., Adams, C.J., Barley, M.E., 2000. Provenance of New

Zealand Late Permian to Cretaceous depocentres at the Austra-

lian margin of Gondwanaland: evidence from detrital zircon

SHRIMP dating. Australian Journal of Earth Sciences 47,

987–1007.

Powell, C.M., Li, Z.X., 1994. Reconstruction of the Panthalassan

margin of Gondwanaland. In: Veevers, J., Powell, M. (Eds.),

Permian–Triassic Pangean Basins and Foldbelts Along the Pan-

thalassan Margin of Gondwanaland. Memoir - Geological Soci-

ety of America. Geological Society of America (GSA), Boulder,

CO, United States, pp. 5–9.

Richards, J.R., Singleton, O.P., 1981. Paleozoic Victoria, Australia:

Igneous rocks, ages and their interpretation. Journal of the Geo-

logical Society of Australia 28, 395–421.

Roser, B.P., 1986. Chemistry, provenance and tectonic setting of

New Zealand greywacke terranes. In: Anonymous (Ed.), Geo-

logical Society of New Zealand, 16th Annual Conference;

Programme and Abstracts. Geological Society of New Zealand

Miscellaneous Publication. Geological Society of New Zealand,

Christchurch, New Zealand, p. 88.

Roser, B.P., Korsch, R.J., 1988. Provenance signatures of sand-

stone–mudstone suites determined using discriminant func-

tion analysis of major-element data. Chemical Geology 67,

119–139.

Roser, B.P., Korsch, R.J., 1999. Geochemical characterization,

evolution and source of a Mesozoic accretionary wedge: the

Torlesse terrane, New Zealand. Geological Magazine 136 (5),

493–512.

Sawka, W.N., Heizler, M.T., Kistler, R.W., Chappell, B.W., 1990.

Geochemistry of highly fractionated I- and S-type granites from

the tin– tungsten province of western Tasmania. Special Paper -

Geological Society of America 246, 161–179.

Scheibner, E., Veevers, J.J., 2000. Tasman fold belt system. In:

Veevers, J.J. (Ed.), Billion-Year Earth History of Australia and

Neighbours in Gondwanaland. GEMOG Press, Department of

Earth and Planetary Sciences, Macquarie University, Macquarie,

pp. 154–234.

Schmidt, P.W., Clark, D.A., 2000. Paleomagnetism, apparent polar-

wander path and paleolatitude. In: Veevers, J.J. (Ed.), Billion-

Year Earth History of Australia and Neighbours in Gondwana-

land. GEMOG Press, Department of Earth and Planetary Scien-

ces, Macquarie University, Macquarie, pp. 12–17.

Shaw, S.E., Flood, R.H., 1981. The New England Batholith, eastern

Australia: geochemical variations in time and space, Granites

and rhyolites. American Geophysical Union, Washington, DC,

United States, 10530–10544.

Shaw, S.E., Flood, R.H., 1993. Carboniferous magmatic activity in

the Lachlan and New England Fold Belts. In: Flood, R.H.,

Aitchison, J.C. (Eds.), New England Orogen, Eastern Australia.

Department of Geology and Geophysics, University of New

England, Armidale, Australia, pp. 113–121.

Shi, G.R., Grunt, T.A., 2000. Permian Gondwana—boreal antitro-

picality with special reference to brachiopod faunas. Palaeo-

geography, Palaeoclimatology, Palaeoecology 155 (3–4),

239–263.

Silberling, N.J., Nichols, K.M., Bradshaw, J.D., Blome, C.D., 1988.

Limestone and chert in tectonic blocks from the Esk Head sub-

terrane, South Island, New Zealand. Geological Society of

America Bulletin 100, 1213–1223.

Sircombe, K.N., 1999. Tracing provenance through the isotope ages

of littoral and sedimentary detrital zircons, Eastern Australia.

Sedimentary Geology 124, 47–67.

Sivell, W.J., McCulloch, M.T., 2000. Reassessment of the origin of

the Dun Mountain Ophiolite, New Zealand: Nd-isotopic and

geochemical evolution of magma suites. New Zealand Journal


Sivell, W.J., McCulloch, M.T., 2001. Geochemical and Nd-isotopic

systematics of the Permo-Triassic Gympie Group, southeastern

Queensland. Australian Journal of Earth Sciences 48, 377–393.

Speden, I.G., 1974. Additional fossil localities of the Torlesse rocks

of the South Island (including summations of faunal assemb-

lages and lithological information). New Zealand Geological

Survey Report, vol. 69. Lower Hutt, New Zealand. 37 pp.


Stump, E., 1995. The Ross Orogen of the Transantarctic Mountains.

Cambridge Univ. Press, Cambridge. 284 pp.

Stump, E., White, A.J.R., Borg, S.G., 1986. Reconstruction of Aus-

tralia and Antarctica: evidence from granites and recent map-

ping. Earth and Planetary Science Letters 79 (3–4), 348–360.

Sutherland, R., 1999. Basement geology and tectonic development

of the greater New Zealand region: an interpretation from re-

gional magnetic data. Tectonophysics 308 (3), 341–362.

Tera, F., Wasserburg, G.J., 1972. U–Th–Pb systematics in three

Apollo 14 basalts and the problem of initial Pb in lunar rocks.

Earth and Planetary Science Letters 14, 281–304.

Tobisch, O.T., Barton, M.D., Vernon, R.H., Paterson, S.R., 1991.

Fluid-enhanced deformation, transformation of granitoids to

banded mylonites, western Sierra Nevada, California, and south-

eastern Australia. Journal of Structural Geology 13, 1137–1156.

Trey, H., Cooper, A.K., Pellis, G., della Vedova, B., Cochrane, G.,

Brancolini, G., Makris, J., 1999. Transect across the West Ant-

arctic rift system in the Ross Sea, Antarctica. Tectonophysics

301, 61–74.

Turner, S.P., Foden, J.D., Sandiford, M., Bruce, D., 1993. Sm–Nd

isotopic evidence for the provenance of sediments from the

Adelaide Fold Belt and southeastern Australia with implications

for episodic crustal addition. Geochimica et Cosmochimica Acta

57, 1837–1856.

Van Der Plas, L., Tobi, A.C., 1965. A chart of judging the reli-

ability of point counting results. American Journal of Science

263, 87–90.

Veevers, J.J., 2000. Antarctic Beadmore–Ross and Mirney prove-

nances saturate Paleozoic–Mesozoic East Gondwanaland with

0.6–0.5 Ga zircons. In: Veevers, J.J. (Ed.), Billion-Year Earth

History of Australia and Neighbours in Gondwanaland.

GEMOG Press, Department of Earth and Planetary Sciences,

Macquarie University, Macquarie, pp. 110–130.

Waight, T.E., Weaver, S.D., Ireland, T.R., Maas, R., Muir, R.J.,

Shelley, D., 1997. Field characteristics, petrography, and geo-

chronology of the Hohonu Batholith and the adjacent Granite

Hill Complex, North Westland, New Zealand. New Zealand

Journal of Geology and Geophysics 40 (1), 1–17.

Waight, T.E., Weaver, S.D., Muir, R.J., Maas, R., Eby, G.N., 1998.

The Hohonu Batholith of North Westland, New Zealand; Gran-

itoid compositions controlled by source H2O contents and gen-

erated during tectonic transition. Contributions to Mineralogy

and Petrology 130 (3–4), 225–239.

Wandres, A.M., Bradshaw, J.D., 2004. The recycling of the Rakaia

Terrane, South Island, New Zealand: sandstone clast and sand-

stone petrology, geochemistry and geochronology. New Zealand

Journal of Geology and Geophysics (in press).

Wandres, A.M.C., Bradshaw, J.D., Weaver, S.D., Maas, R., Ire-

land, T.R., Eby, G.N., 2004. Provenance analysis using con-

glomerate clast lithologies: a case study from the Pahau

Terrane of New Zealand. Sedimentary Geology 167 (1–2),

57–89.

Wareham, C.D., Millar, I.L., Vaughan, A.P.M., 1997. The genera-

tion of sodic ranite magmas, Western Palmer Land, Antarctic

Peninsula. Contributions to Mineralogy and Petrology 128 (1),

81–96.

Wareham, C.D., Stump, E., Storey, B.C., Millar, I.L., Riley, T.R.,

2001. Petrogenesis of the Cambrian Liv Group, a bimodal

volcanic rock suite from the Ross Orogen, Transantarctic

Mountains. Geological Society of America Bulletin 113 (3),

360–372.

Weaver, S.D., Bradshaw, J.D., Laird, M.G., 1984. Geochemistry of

Cambrian volcanics of the Bowers Supergroup and implications

for the early Palaeozoic tectonic evolution of northern Victoria

Land, Antarctica. Earth and Planetary Science Letters 68 (1),

128–140.

Weaver, S.D., Gibson, I.L., Houghton, B.F., Wilson, C.J., 1990.

Mobility of rare earth and other elements during crystallisation

of alkaline silicic lavas. Journal of Volcanology and Geothermal

Research 43, 57–70.

Weaver, S.D., Bradshaw, J.D., Adams, C.J., 1991. Granitoids of the

Ford Ranges, Marie Byrd Land, Antarctica. In: Thomson, M.R.,

Crame, J.A., Thomson, J.W. (Eds.), Geological Evolution of

Antarctica; Proceedings of the Fifth International Symposium

on Antarctic Earth Sciences. Cambridge Univ. Press, Cam-

bridge, pp. 345–351.

Whitford, D.J., Crawford, A.J., 1993. The Mount Read Volcanics,

Tasmania; strontium and neodymium isotope geochemistry. In:

Duggan, M.B., Knutson, J. (Eds.), IAVCEI Abstracts; Ancient

Volcanism and Modern Analogues. Australian Geological Sur-

vey Organisation, Australia, p. 122.

Whitford, D., Crawford, T., Korsch, M., Craven, S., 1990. The

Mount Read Volcanics: strontium and neodymium isotope geo-

chemistry. Exploration Research News/CSIRO Division of Ex-

ploration Geoscience 4, 10–11.

Withnall, I.W., Black, L.P., Harvey, K.J., 1991. Geology and geo-

chronology of the Balcooma area: part of a Early Paleozoic

magmatic belt in North Queensland. Australian Journal of Earth

Sciences 38, 15–29.

Wood, R., 1994. Geophysical aspect of the opening of the Tasman

Sea. In: van der Lingen, G.J., Swanson, K.M., Muir, R.J. (Eds.),

Evolution of the Tansman Sea Basin. Balkema, Rotterdam,

pp. 19–34.

Wysoczanski, R.J., Gibson, G.M., Ireland, T.R., 1997. Detrital zir-

con age patterns and provenance in Late Paleozoic Early Meso-

zoic New Zealand; terranes and development of the Paleo-

Pacific Gondwana margin. Geology 25 (10), 939–942.

Zen, E., 1986. Aluminium enrichment in silicate melts by fractional

crystallisation: some mineralogic and petrographic constraints.

Journal of Petrology 27, 1095–1118.

Provenance of the sedimentary Rakaia sub-terrane, Torlesse Terrane, South Island, New Zealand: the...

Documents

Transcript of Provenance of the sedimentary Rakaia sub-terrane, Torlesse Terrane, South Island, New Zealand: the...