Vol. 2 pp 52-97, Geol. Surv. of Denmark and Greenland Map Series

Please note that the present html-document is based on an semi-automatic conversion of the original pdf-file.
The conversion from pdf to html is not perfect in all respects, and some typographic errors can occur.
The html-document is meant for general orientation and is not a 100% identical copy of the published version.
Please use the pdf-file for obtaining the exact typography / layout and for printing purposes .
nr2.pdf (~6mb)
PDF-files requires a pdf-reader, Acrobat Reader GSview or similar

Mesoproterozoic-?Neoproterozoic Thule Basin

The Thule Basin overlies the eroded Canadian-Green-
land

shield and straddles northern Baffin Bay and Smith

Sound (Fig. 1). The basin is defined by an unmeta-
morphosed sedimentary-volcanic succession - the
Thule Supergroup - that is at least 6 km thick. The
Thule map sheet covers the eastern and south-east-
ern parts of the basin, with the central part being pre-
dominantly offshore but represented in Greenland by
the thick section on Northumberland Ø (see Dawes
1997, figs 12, 119). The northern margin is defined by
outcrops on both sides of Smith Sound and the west-
ern and south-western margins of the basin are in
coastal Ellesmere Island. These outcrops are featured
on the maps of Frisch (1983, 1984a, b) and Dawes &
Garde (2004).

As mentioned in the Introduction , the reader is re-

ferred to a monograph for a full treatment of Thule
Basin lithostratigraphy (Dawes 1997). This contains
20 geological maps showing the distribution of the
stratal units, 12 of which feature areas within the map
region.

Age of the Thule Supergroup

In the map legend the Thule Supergroup is designated
a Late Proterozoic (Neohelikian-Hadrynian) age, or
in present terminology, middle Mesoproterozoic - late
Neoproterozoic . This is based on both radiometric
dating of basic sills and dykes, and on biostratigraph-
ic ages of microfossils (acritarchs) from the upper part
of the succession that inferred a latest Proterozoic
(Vendian) age (Vidal & Dawes 1980; Dawes & Vidal
1985). Regional unconformities have not been recog-
nised although, as discussed below, the boundary
between the upper groups (Dundas and Narssârssuk
Groups) is unexposed. Nevertheless, despite the some-
what problematic scenario of having an exceptionally
long period of sedimentation - 500 million years from
the middle Mesoproterozoic to late Neoproterozoic -
the age was anchored in what were considered diag-
nostic Vendian (Sinian) acritarch identifications from
the Dundas and Narssârssuk Groups.

The age of the microfossil assemblage is now re-

garded to be of late Mesoproterozoic and/or early Neo-
proterozoic age (Fig. 4; Samuelsson et al . 1999). The
correlation between the Thule succession and those
of other Proterozoic basins in Canada - specifically
the Mesoproterozoic Bylot Supergroup of the Borden
Basin of Baffin Island - suggests that the entire Thule
Supergroup could have been deposited in Mesoprot-

erozoic (Ectasian) time, c. 1300 to 1200 Ma (Knight &
Jackson 1994; Jackson 2000). Since there is nothing in
the make-up of the Thule Basin that conclusively mil-
itates against this restricted evolution, it is certainly
one possibility. However, a categorical Mesoprotero-
zoic age is not adopted for two reasons: (1) the mi-
crofossil assemblage is not specifically age-diagnos-
tic, i.e. Mesoproterozoic or Neoproterozoic (Strother
et al . 1983; Hofmann & Jackson 1996; Samuelsson et
al . 1999) and (2) Mesoproterozoic and Neoprotero-
zoic successions do occur in Greenland farther east in
the Kronprins Christian Land region (Fig. 1; Henrik-
sen et al . 2000).

The Mesoproterozoic-?Neoproterozoic age is based

on the following relationships. Stratigraphically, the
Thule Supergroup is bracketed by the Palaeoprotero-
zoic shield and Lower Cambrian (Atdabanian) strata
of the Franklinian Basin that crop out just beyond the
map region (Fig. 1; see Dawes 1997, figs 8, 14; Dawes
2004, figs 11, 13). Tighter constraints are provided by
two periods of basic dykes that date sedimentation to
between c. 1300 and 730 Ma. The oldest dykes (desig-
nated d₁) cut the shield but not the Thule Basin where-

as the younger dykes (designated d₂) cut the entire

Thule succession. The d₁ and d₂ dykes have yielded

K-Ar whole-rock ages between c. 1670-1310 Ma and
c. 730-630 Ma, respectively (see under Palaeo-, Meso-
and Neoproterozoic basic intrusions, Table 1; Dawes
et al . 1982b; Dawes & Rex 1986).

The youngest pre-Thule Basin age has a large error

(1313 ± 39 Ma) and the most reliable isotopic age of
relevance to the onset of sedimentation history is an
U-Pb age of c . 1270 Ma on a sill indicating that the
lowermost Thule strata are at least middle Mesoprot-
erozoic (Ectasian; LeCheminant & Heaman 1991; see
below under Nares Strait Group ). However, the age
of the upper strata (Baffin Bay, Dundas and Narssârs-
suk Groups) is poorly constrained radiometrically, viz.
the strata pre-date the late Neoproterozoic d₂ dykes,

and the acritarchs of these groups, as mentioned above,
suggest a late Mesoproterozoic and/or early Neoprot-
erozoic age.

Structure and metamorphism

The Thule Basin is a intracratonic fracture basin char-
acterised by block faulting and basin sagging, the prod-
uct of a divergent plate regime. Its central fill is de-
fined by the lower Thule Supergroup (Nares Strait
Group) that thins to the north, east and south-east

from Northumberland Ø. The limit of the basin is de-
fined by the Baffin Bay Group that oversteps the Nares
Strait Group to overlie the shield in the east and south.
The margin on the map stretches from De Dødes Fjord
in the south to the semi-nunatak Nunatarsuaq at the
head of Inglefield Bredning, across Inglefield Bredn-
ing and Hubbard Gletscher, and north to the nunatak
terrain in Prudhoe Land at the heads of McCormick
and Robertson Fjords. Outliers beyond this - too small
to show on the map - range from a coherent sand-
stone and conglomerate veneer into diffuse areas of
rubble with concentrations of quartz pebbles.

Thule strata form predominantly homoclinal, shal-

low-dipping sections, with anomalous inclinations
caused by block faulting, tilting, drag folding and
regional flexuring. The rocks have not been regional-
ly metamorphosed but they are indurated and locally
altered. Quartz impregnation has taken place at cer-
tain levels and such rocks appear as sugary quartz-
ites. Apart from crushing along faults with gouge for-
mation, the main types of alteration seen in the sedi-
ments are contact metamorphic effects and chemical
changes, for example baking of argillaceous lithologies
and bleaching of redbeds, occur adjacent to basic in-
trusions. Bleaching along basic dykes is well illustra-
ted in the Imilik Formation just south of Pituffik (see
Fig. 39) while contact metamorphic effects are com-
mon in the Steensby Land sill complex (see under
Neoproterozoic sills ( s₁) ). Shales adjacent to sills can

be slaty, greenish and chloritised, with a slight waxy
sheen and in which mica is sericitised. Pyritisation
also occurs (see under Economic geology , section Iron-
sulphide mineralisation ( py ) ).

Chemical action by solutions has been widespread

and is seen by ferruginous banding and colour alter-
ation in redbeds. Ferruginous banding in sandstone

occurs on all scales from fine lamination parallel and
sub-parallel to bedding, to discordant liesegang rings
of several generations. Where just of one generation
and regular in form, liesegang rings can readily be
mistaken for bedding; where two generations occur
at a low angle to each other, the pattern resembles
cross-bedding (see Fig. 52; Fernald & Horowitz 1964,
figs 13, 14). Swirl-forms occur, some of which can
simulate folding. Dark brown, highly ferruginous veins
up to 1 cm thick are associated with liesegang rings.
The most intense ferruginous banding occurs in red
to purple quartz arenites in strata not far removed
from shield outcrops suggesting that the unconfor-
mity, as well as faults, have aided the diffusion of
iron-rich solutions. Kurtz & Wales (1951) mention dif-
fusion banding in sandstone associated with solutions
derived from a basic dyke, as well as alteration of a
1 m thick dolomite bed with pyrite adjacent to a basic
sill.

Bleaching and reduction phenomena in redbeds

are common, with the effects well seen along bed-
ding planes, joints and fissures. At bed scale, effects
vary from fish-eye spots and irregular bleach patterns,
to pale, almost fully transformed beds identified by
relict patches of initial red colour (Fig. 33, see also
Fig. 51). On a larger scale, such as in the sea cliffs on
the northern side of Hakluyt Ø, purple sandstone tens
of metres thick may pass along strike into an interfin-
gering network of dark and pale beds, and finally in-
to pale, sandstones in which purple colour has been
eliminated. Strong reduction patterns have been
recorded particularly in basal strata, both in the cen-
tral basin (e.g. Northumberland Ø) and in basin mar-
gins (e.g. Wolstenholme Ø), suggesting that the un-
conformity acted as a passageway for the reducing
solutions.

Fig. 33. Severe bleaching of ferruginous
sandstone showing relict redbeds.
Arrows point to a late generation of
reduction spots. Northumberland
Formation, c. 10 m above the crystalline
shield. East of Parish Gletscher, North-
umberland Ø. A slab from similar
bleached redbeds is shown in Fig. 51.

Nature of the unconformity

The shield below the Thule Basin forms a regional
peneplain (see Erosion surfaces under Physiography ).
The actual unconformity is invariably a recessive zone
and often scree-covered but where examined, it is
generally well preserved. Locally, the contact may be
tectonised with minor faults, shearing and cataclastic
effects in strata both above and below the hiatus. The
palaeosurface is planar to slightly undulating with relief
generally below 2 m. Palaeovalleys and topographic
highs have not been unequivocally identified although
several areas with locally dipping beds that do not
seem to be related to faulting, may represent drapes
over more pronounced relief.

A regionally consistent regolith cover has not been

identified and in some localities the underlying crys-
talline rocks appear remarkably fresh. However, in
most areas the rocks on either side of the unconfor-
mity show alteration and anomalous colour, often due
to hematite impregnation. Gneiss and granitic rocks

may be reddened, slightly to moderately weathered,
with hematite as seams along joint surfaces and as
veins up to 3 cm as joint infillings. With a high inten-
sity of joints, red staining may be over a metre deep.
'Softer' supracrustal rocks, such as pelitic schists of
the Prudhoe Land supracrustal complex, are typically
more severely weathered, and often there is a paler
zone of variably friable and sericitised rocks (Fig. 34).

In several localities, basal beds show signs of silici-

fication and platy siltstone or shale can be enriched
in kaolinite, sericite, chlorite and secondary quartz.
Such beds at Magnetitbugt have been considered to
be reworked residual soils with scattered pebbles of
crystalline rock (Kurtz & Wales 1951). On Wolstenhol-
me Ø, the rock package spanning the unconformity
comprises red platy hematite gneiss passing upwards
into similarly platy, fine-grained hematite-rich sand-
stone. The hematite gneiss contains large quantities
of green to yellowish, angular to rounded quartz set
in a sericitic matrix, for which Davies et al. (1963, p.
28) suggested a possible fault gouge origin. However,
this quartz-sericite deposit resembles regolithic mate-
rial, for example that underlying the Borden Basin on
northern Baffin Island (Jackson 1986).

Thule half-graben system

The outcrop pattern of Thule strata is strongly control-

led by faults, mainly WNW-ESE- to NW-SE-trending,
that split the region into tilted blocks of varying stat-
ure. These fault blocks make up the Thule half-graben
system, that is so named here (Fig. 35). Six major half-
grabens dominate, each with the same fundamental
structure: on the north-eastern side, the shield is over-
lain by a normal, south-westerly dipping section that
is bounded in the south-west by a steeply inclined
master fault that juxtaposes the upper Thule Super-
group against the shield of the adjoining block. Move-
ments along the bounding faults are measurable in
kilometres, with the greatest displacements common-
ly in the west, i.e. in the deepest part of the Thule Basin.

Within the half-grabens, smaller fault blocks occur,
including both graben and horst structures, and these
represent small to moderate displacements, which re-
peat stratal levels within the same formation or group,
as well as larger displacements affecting the map out-
crop pattern. Five of the six half-grabens contain suc-
cessions that top in the Dundas Group; the Pituffik
half-graben preserves the Narssârssuk Group.

The Thule half-graben system is schematically

shown in Fig. 35 (see also front cover illustration).
Many small, often closely-spaced intragraben faults,
that do not radically affect the regional map pattern,
have been omitted from the map sheet (see Fig. 37;

Fig. 34. The unconformity below the Thule Basin at Bowdoin
Fjord. Pale orange sandstone of the Northumberland Forma-
tion (Nares Strait Group) overlying highly folded, graphitic schists
of the Prudhoe Land supracrustal complex. Note the bleaching
of the basal sandstone beds and the pale regolithic zone up to
c. 2 m thick.

Dawes 1997, figs 90, 91) and for a more detailed fault
representation the reader is referred to 1:100 000 maps

in Survey archives (Dawes 1988b). For mineralisation
of the faults, see under Economic geology , section
Fault-related mineralisation . The main characteristics
of the six half-grabens are described below, starting
in the north.

Prudhoe half-graben

This half-graben bounded by the Murchison Fault
forms the north-eastern margin of the Thule Basin
stretching from Prudhoe Land to south of Inglefield
Bredning. The overall structure is a series of tilted
fault blocks with downthrow to the south-west, with
sedimentary contacts to the shield preserved in all but
northernmost exposures. North of Diebitsch Gletscher,
and on the adjoining map sheet, a major fault - the
Dodge Gletscher Fault - limits the Thule strata on the
north-east (Dawes 1997, fig. 111). Apart from this fault,
two other master faults have been named, viz. the
Morris Jesup and Diebitsch Gletscher Faults, both with
downthrow on their coastal side (see Dawes 1997,
figs 1, 5B, 90, 111; Dawes 2004, fig. 12). However,
some faults within the half-graben have reversed dis-
placement, such as the Scarlet Heart Fault described
earlier (see under Map revision , section 8; Dawes 1997,
fig. 91). Local horst and graben structures occur (Figs
5, 37). Spacing of faults varies markedly from dense
fault systems to isolated faults.

The Thule strata above the shield are shallow dip-

ping, commonly a few degrees to the south-west, as
for example on both sides of Bowdoin Fjord, but with-
in fault blocks stratal dips reach up to c . 25°, locally
even steeper (see Dawes 1997, figs 53, 74, 85). A note-
worthy feature is the presence of large-wavelength,
WNW-ESE-trending folds that are well seen in the
Dundas strata on the peninsulas between Morris Jesup
Gletscher and McCormick Fjord, where dips of limbs
up to 10° are shown on the map.

The Murchison Fault intersects the south coast of

Inglefield Bredning east of Tikeraasaq and strikes east
along the northern side of a local ice cap. Steep dips
due to drag characterise the Dundas strata adjacent to
the main fault plane. The master fault is crossed by
dislocations of other directions and this complicates
the outcrop pattern of Thule strata and the shield. The

master fault has not been mapped to the south-east of
Kangerdlugssuaq but a number of WNW-ESE-tren-
ding faults strike towards the head of Academy Bugt.
To the west, it is projected offshore along Murchison
Sund parallel to the outer coast of Piulip Nunaa. The
linear form of this coast west of Kap Ackland reflects
the presence of the Kap Cleveland Fault, that passes

onland west of the snout of Fan Gletscher separating
Kap Cleveland off as an uplifted block. This block
forms one side of a graben preserving the Dundas
Group that is bordered to the north-east by the Fan
Gletscher Fault.

Olrik half-graben

Olrik Fjord is illustrated in the literature as an exam-
ple of a fault-controlled fjord and the type of graben
tectonics that affect the Thule Basin (e.g. Dawes 1976b,
fig. 231; 1997, fig. 109). The steep to vertical southern
bounding fault - the Itilleq Fault - separates a 5-km
wide coastal strip of Dundas strata (Olrik Fjord For-
mation) from an inland shield escarpment (see Fig.
48). In detail, the fault is composed of several splays
and at places, tectonic slivers of pale sandstone re-
ferred to the Baffin Bay Group are caught up in the
fault zone. The strata adjacent to the Itilleq Fault may

70°W

72°W

68°W

66°W

78°N

77°N

76°N

100 km

Inland Ice

Kap

York

Basin

Bounding fault
Cretaceous-Palaeogene
Thule Supergroup
Precambrian shield

Fig. 35. The Thule half-graben system composed of six half-
grabens each with its bounding fault on the southern side along
which Thule Supergroup is downdropped against the shield.
Intragraben faults are not shown. From north to south: A , Prud-
hoe half-graben / Murchison Fault; B , Olrik half-graben / Itilleq
Fault; C , Itillersuaq half-graben / Granville Fault; D , Moriusaq
half-graben / Moltke Fault; E , Pituffik half-graben / Narssarssuk
Fault; F , Qeqertarsuaq half-graben / Magnetitbugt Fault. The
Kap York Basin, an offshore half-graben of similar polarity, is
taken from Whittaker et al. (1997).

be highly contorted, sheared and crushed, with drag
folding and steeply dipping sections, as well as the
presence in some places of fine-grained fault gouge.
Iron-sulphide mineralisation has been recorded along
the fault (Gowen & Sheppard 1994; see under Fault-
related mineralisation ).

The eastern part of the Itilleq Fault is not depicted

on the map but, as described earlier, this strikes
towards lake Tasersuaq (see under Map revision , sec-
tion 12). To the west of Itilleq, the steep linear sea
cliffs are a reflection of the master fault concealed by
morainic deposits forming a narrow coastal strip. The
fault strikes into Hvalsund and towards Northumber-
land Ø where its continuation is the Kiatak Fault. This
juxtaposes the Dundas Group with basic sills against
the Nares Strait Group, and farther west, the Baffin
Bay Group against the shield (Dawes 1997, fig. 48).

A number of faults with downthrow to the south-

west dissect the strata within the half-graben. For ex-
ample, the Narsaq Fault that reaches the coast north
of Kangeq and repeats the south-westerly dipping
stratigraphy of the Baffin Bay Group (Dawes 1997,
fig. 95) and farther east, the Gyrfalco fault limiting
Dundas strata against the shield (Dawes 1997, fig. 103).
Westerly projection of the faults cutting the mountain
Qaqqarsuaq at the distinct bend in Olrik Fjord (Fig.
36; marked Bq on the map) along the northern side
of the fjord led to an earlier assumption that the west-
ern part of Olrik Fjord was a full graben. However,
the shield at Qaqqarsuaq is downfaulted on its north-
ern and southern sides and is now interpreted as a
local horst. There is evidence of syn-depositional fault-
ing and it is possible that the Dundas Group may have
been draped over an early manifestation of the horst
prior to fault rejuvenation.

Itillersuaq half-graben

This half-graben occupies the northern part of Steens-
by Land. It is named after the Greenlandic name for
Politiken Bræ that transects the fault block from Itilleq
to its southern boundary, the Granville Fault. This fault
is traceable from Barden Bugt in the west through
heavily ice-covered terrain to the head of Granville
Fjord where it juxtaposes the Dundas Group and the
shield (Dawes 1997, figs 70, 87; see also front cover
illustration). However, there is also downthrow to the
south-west along parallel dislocations to the north, so
that in effect the Dundas Group lies in a narrow graben
that manifests itself as a low, glacier-filled depression.
This continues to the east from the head of Granville
Fjord under a large, unnamed, advancing glacier which
has overrun several outcrops of the Dundas Group
that were exposed in the early 1970s (see under Re-
cent glacial history ). However, the fault remains expo-
sed,with Dundas strata juxtaposed against the Nares

Strait and Baffin Bay Groups. Several fault planes make
up the contact zone, which is characterised by drag
folds, anomalously steep stratal dips, sheared and
crushed rocks, and local fault gouge.

To the east, the fault is concealed by the large ice

cap but it strikes towards the Rødstenbæk Fault of
Gregory (1956) and Fernald & Horowitz (1964), just
south of Dryasbjerg. This fault juxtaposes the lower
strata of the Baffin Bay Group (Wolstenholme Forma-
tion) against the shield and represents a displacement
of c. 100 m that is considerably less than that docu-
mented at Granville Fjord. To the west, the Granville
Fault is surmised to strike south-west of Northumber-
land Ø (Fig. 35).

On the mainland, the Itillersuaq half-graben exposes

Fig. 36. Qaqqarsuaq fault block within
the Olrik half-graben showing down-
drop to the south-west. The mountain is
located at the pronounced bend in
Olrik Fjord marked on the map by
symbol Bq. Ps , Precambrian shield; Bw ,
Wolstenhome Formation; Bq , Qaanaaq
Formation, D , Dundas Group. As shown
on the map sheet, there are also
downdropped strata on the north-
eastern side of the shield outcrop and
the structure is interpreted as a horst.
The Dundas Group, seen in the
distance, may have been draped over
an early manifestation of the horst prior
to renewed faulting. View is towards the
east; height of summit is c. 700 m a.s.l.

a more or less uninterrupted sequence from the shield
through the Nares Strait, Baffin Bay and the Dundas
Groups, with the gradational passage into the latter
group well seen in the small outlier at Kap Powlett.
However, as described above, it is possible that much
of the Dundas strata to the east of this locality has a
fault relationship to the underlying Baffin Bay Group.
Most of the faults seen within the half-graben are par-
allel to the boundary faults with downdropping to
both the north-east and south-west, for example, two
faults of opposing polarity shown on the map cross
southern Northumberland Ø and form a small graben
(see Dawes 1997, fig. 48). The northern part of this
island is characterised by a series of fault blocks down-
dropped to the north and a similar fault pattern char-
acterises the south coast east of Isussik. The fault
blocks of the latter location are not shown on the map

since stratal repetitions are small, but the structures
are persistent along strike and probably correlate with
a series of faults on the mainland north-east of Kap
Powlett that have been utilised by basic dykes ( d₂).

Moriusaq half-graben

This half-graben, named after the village of Moriusaq,
has the Moltke Fault of Davies et al. (1963, fig. 6) as
its southern bounding fault. This fault is inferred be-
neath Harald Moltke Bræ and Wolstenholme Fjord
separating the south-westerly dipping strata (Baffin
Bay Group) north of the glacier from the shield rocks
east from Ulli. The precise position of the fault is un-
known. There is, however, a marked contrast in to-
pography between the two sides of the glacier valley,
and the steep cliffs of the south side compared to the
more gentle slopes of the northern side, suggests a
proximal position of the fault just off the cliffs. Stratal
thickness considerations and extrapolation of the ge-
ology from southern Steensby Land, west of Knud
Rasmussen Gletscher, suggest that the Baffin Bay and
Dundas Groups continue under Wolstenholme Fjord
and thus in fault contact with the shield along the
Uvdle Fault.

The Moriusaq half-graben is 30 km across and thus

the widest of the six half-grabens of the region. Un-
like the three northern half-grabens, it is relatively
uncomplicated by intragraben faults and it preserves
a more or less uninterrupted section from the shield
through the Nares Strait, Baffin Bay and Dundas
Groups. Contacts between the Thule strata and the
shield are well preserved and mainly shallow dipping
up to 15° (Dawes 1997, fig. 97). The faults recorded
are mainly parallel to the boundary faults, for exam-
ple, the one shown on the map striking across Gran-
ville Bugt to reach the outer coast in the bay north of

Kap Leiningen (Dawes 1997, fig. 70). Small faults may
have downdrop to the north-east (e.g. Dawes 1997,
fig. 88). Several cross-faults are known, including the
Ohio fjeld Fault of Fernald & Horowitz (1964), north
of Harald Moltke Bræ, that delimits the main outcrop
of Thule strata to the east.

Pituffik half-graben

The Pituffik half-graben preserves the youngest part
of the Thule Supergroup - the Narssârssuk Group -
downfaulted against the shield along the Narssarssuk
Fault. Unlike the other half-grabens, its north-eastern
boundary to the shield is not a sedimentary contact
but a steep fault - the Uvdle Fault of Davies et al.
(1963, fig. 6). This shield outcrop represents a horst,
being bounded on the south by this fault along which
strata of the Baffin Bay Group (Qaanaaq Formation)
have been displaced at least 500 m, and on the north
by the Moltke Fault that, assuming a normal stratigra-
phy below the ice, represents a much larger displace-
ment (see Moriusaq half-graben above).

The southern bounding fault of the half-graben is

the Narssarssuk Fault. Although the actual fault plane
is poorly exposed, the fault line can traced from the
coast south of Narsaarsuk, through the poorly expo-
sed ground north of Pinorsuit, to the Inland Ice. Pro-
jecting the fault offshore suggests a position west of
Saunders Ø and the straight south-western coast of
the island may well express fault proximity. This po-
sition was favoured by Kurtz & Wales (1951, fig. 1)
but it is at variance with Davies et al. (1963, fig. 6),
who advocate a change in strike direction offshore
to more or less E-W so as to link up with a fracture
of much lesser magnitude in northern Wolstenhol-
me Ø. This latter island is considered by the present
author to be an integral part of the adjoining half-
graben (Fig. 35).

The mainland between the Uvdle and Narssarssuk

Faults is not crossed by visible faults although the
broad topographic depression in which the Pituffik
air base is situated is most probably fault controlled.
Several faults cross Saunders Ø, including the Agpat
and Kulukupaluk Faults shown on the map sheet (see
Dawes 1997, figs 113, 114). The general structure is
thus a broad WNW-ESE-trending asymmetrical syn-
cline, with dips on the northern limb generally less
than 15° and on the southern limb as much as 30°.
The northern limb is formed of the Dundas Group
(Steensby Land Formation) interspersed by basic sills
and the lower part of the Narssârssuk Group (Imilik
and Aorfêrneq Formations) while the trough of the
syncline and its southern limb is formed of upper
Narssârssuk Group (Bylot Sund Formation). This large-

scale syncline also occurs on Saunders Ø but limbs
are shallower as shown by the dips on the map.

On the mainland, the southern limb of the syncline

is truncated by the Narssarssuk Fault, which indicates
a major displacement zone. Assuming that the Nars-
sârssuk Group wholly post-dates the Dundas Group,
the displacement may well be 5 km or more. Mode-
rate to steep dips occur in connection with contor-
tions, folds and tilted blocks up to 2 km away from
the fault. Davies et al . (1963) suggested that a series
of local anticlines and synclines with dips as steep as
45° occur north of the fault. These, as well as the north-

easterly regional inclination of the strata, might well
be due to a massive drag effect along this major move-
ment zone, rather than a regional compressional event.

Qeqertarsuaq half-graben

This half-graben, named after the Greenlandic name
for Wolstenholme Ø, is the southernmost and smallest
of the six half-grabens. The southern bounding fault
is the Magnetitbugt Fault that on the mainland juxta-
poses outliers of the Baffin Bay Group against the
shield (Davies et al . 1963, fig. 6, plate 2; see under
Map revision section 13). To the west on southern
Wolstenholme Ø, the Baffin Bay and Dundas Groups
are downdropped against the shield.

To the east, the Magnetitbugt Fault has not been

traced far inland but it lines up with observed faults
near the Inland Ice, west of Freuchen Nunatak, where
the Baffin Bay Group occupies the northern part of a
semi-nunatak. However, this area has heavy surficial
cover, relief is low and the nature of the southern
contact between Thule strata and the shield was not
determined by the present author during a helicopter
stop. Thus on the map, the boundary is shown as
'inferred or arbitrary' while a fault borders the out-
crop in the west. This outlier of Thule strata appears
on maps in Davies et al . (1963) but information about
its relationship to the shield is contradictory. On their
geological map (op. cit. plate 1), profuse Quaternary
cover is shown at the outcrop in question but a con-
tact between gneiss and Thule strata is shown as a
solid line corresponding to a normal contact. How-
ever, on a sketch map (op. cit. fig. 6), a bold fault line
trending WNW-ESE with downdrop to the north lim-
its the Thule strata. W.E. Davies (personal communi-
cation 1980) confirms he did not locate the unconfor-
mity at this locality and that an 'inferred or concealed
fault' ought to have been shown on plate 1. This in-
terpretation is supported by a Landsat 'Crosta image'
of the Freuchen Nunatak area shown in Krebs et al.
(2003, fig. 2; Fig. 35).

The most complete section in this half-graben oc-

curs on Wolstenholme Ø, with the Baffin Bay and
Dundas Groups preserved. The standard tilted se-
quence characteristic of all the half-grabens is expo-
sed on the Bylot Sund coast, where south-westerly
dips exceed 20° (Dawes 1997, fig. 93). A second, nar-
row half-graben forms the northern part of the island
with the Baffin Bay Group downdropped against
shield, and with a normal contact preserved at the
northern cape (the shield outcrop is just large enough
to be portrayed on the map). Deformation of this fault
block has produced moderate dips of opposing direc-
tions. Normal faults with northerly directions affect
the outcrop pattern of Thule strata in the central part
of the island.

Age of the faulting

The age of the Thule half-graben system is not tightly
constrained. The main faults are considered to have
been initiated in the Proterozoic but their rejuvena-
tion is only bracketed by Quaternary deposits. The
youngest movements may well be part of the late
Phanerozoic tectonism that affected the Baffin Bay
region (see below under Correlation with the offshore ).
The main period of extensional faulting is placed with-
in the Franklin magmatic episode in the mid-Neopro-
terozoic (Cryogenian) being bracketed by the emplace-
ment of a suite of basaltic sills ( s₁) - that are consist-

ently tilted within the fault blocks - and a regional
swarm of basic dykes ( d₂) that cuts the sills (see Table

1; Dawes 1997, fig. 106). Since the sills do not cross
the main faults and there is no trace of magma splays
along the faults, their pre-faulting age is certain.

In contrast, the relationship of the d₂ dykes to the

main fault movements is less certain, made so by the
facts that dykes and faults are regionally parallel
(WNW-ESE-trending) and both are steeply dipping
features. Moreover, deciphering age relationships is
complicated by the fact that at least some faults (prob-
ably all main structures) register more than a single
movement episode. A particularly dense part of the
d₂ dyke swarm crosses the Itillersuaq and Moriusaq

half-grabens where dykes can be seen to have exploi-
ted faults, for example east of Kap Powlett and south
of Olrik Fjord. Other exposures providing cross-cut-
ting information also indicate that the dykes post-

date faulting (see Fig. 37). However, some dykes show
features of brittle deformation, such as brecciation and
crushing, indicative of fault reactivation.

At Asungaaq, south-eastern Northumberland Ø, a

cross-cutting relationship is shown on the map be-
tween a d₂ dyke and the southern branch of the Kia-

tak Fault that at the outer coast juxtaposes redbeds of
the Baffin Bay Group against the Clarence Head For-

mation of the Nares Strait Group (Dawes 1997, fig. 48;
see under Map revision , section 7). This fault zone is
also known for a stream-sediment gold-barium anom-
aly, as well as quartz-baryte-pyrite mineralisation in
clastic rocks adjacent to the basic dyke (Thomassen &
Krebs 2004, figs 15, 16; see also under Economic geol-
ogy , section Fault-related mineralisation ). The sce-
nario favoured to explain the mineral occurrence is
that initial faulting resulted in brittle deformation of the

clastic rocks, after which the emplacement of the ba-
sic dyke caused the contact mineralisation that is thus
regarded as Neoproterozoic in age. The dyke shows
colour variations and crushing that are referred to fault

reactivation of unknown (?Cenozoic) age.

Correlation with the offshore

The structural characterisation, limits and development
of the Thule Basin as a regional depocentre on the
northern margin of the North Atlantic craton have been
summarised in Dawes (1997). Exposures in both
Greenland and Canada disappear seawards in down-
faulted blocks and they represent the outermost frag-
ments of a large sedimentary and volcanic province
preserved under northern Baffin Bay. Gravity, mag-
netic and seismic reflection data collected since the
1970s indicate that the offshore is composed of a fault-
ed sedimentary section, at least 10 km thick but in
places possibly considerably thicker. Recent seismic
data indicate that the section is composed of at least
two sedimentary packages, a late Phanerozoic sec-
tion and an underlying succession that includes Prot-
erozoic strata of the Thule Basin (e.g. Jackson et al.
1992; Reid & Jackson 1997; Whittaker et al . 1997; F.
Tessensohn, personal communication 2003). Although
not yet mapped in detail, there is clear correlation
between onland geology and the offshore, for exam-
ple, the Steensby Basin of Newman (1982, fig. 7) trends
north-west from the Bylot Sund area and is directly
online with the Pituffik half-graben.

An integral element of regional tectonic models is

that the coastal regions of Baffin Bay have been af-
fected by late Phanerozoic rifting and the faults of the
Thule region have been regarded as of similar age
(e.g. Koch 1926; Fernald & Horowitz 1964; Monahan
& Johnson 1982). A system of extensional faults has
been mapped offshore within the south-western quad-
rant of the map sheet (Whittaker et al. 1997). The
straight, cliffed coastline west-north-west of Kap York
suggests fault control and it is parallel to the main
offshore structure, the Kap York Basin (Fig. 35). This
is a half-graben with the same polarity as the onland
half-grabens with a bounding fault on its south-west-
ern side and with the sedimentary fill onlapping a

'basement' to the north-east. The early-rift sediments
are of Early to mid-Cretaceous (Barremian-Cenoma-
nian) age and the basin - like the other offshore struc-
tures to the south-east in Melville Bugt - is regarded
as late Mesozoic to Cenozoic in age. It seems likely
that the tectonic regime that produced these offshore
features also affected the coastal region, as exempli-
fied by the linearity of the Kap York coastline. Thus,
some of the onland faults, as well as the rejuvenation
of the Thule half-graben system, are probably related
to regional, late Phanerozoic tectonic processes.

Thule Supergroup

The published lithostratigraphic subdivision of the
Thule Supergroup of Greenland and Canada into five
groups has been mentioned in the Introduction . In
that publication (Dawes 1997), the tripartite subdivi-
sion of the two youngest groups that are restricted to
Greenland - the Dundas and Narssârssuk Groups -
was not formalised. Four of these formations - the
Olrik Fjord Formation of the Dundas Group and the
Imilik, Aorfêrneq and Bylot Sund Formations of the
Narssârssuk Group - are named units on the map sheet.
This, as well as the fact that the Smith Sound Group,
that represents the northern basin margin, crops out
beyond the map region, determine that the Thule
Supergroup in the map region comprises four groups,
15 formations and nine members.

Nares Strait Group

Name . Dawes (1991, 1997).
Other literature . Jackson (1986), Steenfelt et al .
(2002), Thomassen et al . (2002a, b), Dawes (2004).
Distribution and age . This group represents the old-
est Thule strata overlying the shield in the central basin.
It is conformably overlain by the Baffin Bay Group.
Basal strata are at least 1268 Ma old (middle Mesopro-
terozoic or Ectasian). This is based on the most relia-
ble isotopic age available: a ²⁰⁷Pb/²⁰⁶Pb baddeleyite

age of a basic sill within the Cape Combermere For-
mation from Ellesmere Island, a sequence of tholeiitic
basalt extrusive and intrusive rocks coeval with the
Mackenzie magmatism well known from elsewhere
in northern Greenland and Arctic Canada (LeChemi-
nant & Heaman 1991; Henriksen et al. 2000). This re-
fines the K-Ar age range of 1220-1205 Ma cited on the
map sheet taken from Dawes & Rex (1986).
Composition . The group comprises five formations
- Northumberland, Cape Combermere, Josephine
Headland, Barden Bugt and Clarence Head Forma-
tions - in which eight formal members have been

defined, two of which are restricted to Canada (see
Dawes 1997, fig. 49). Other members are not formally
defined, for example the tripartite subdivision of the
Cape Combermere Formation recognisable in many
parts of the Thule Basin in Greenland and Canada
(Figs 2, 37). The Nares Strait Group has a composite
thickness of up to 1200 m and the thickest section in
the map region is c. 950 m on Northumberland Ø. As
explained earlier under Map revision , the group has a
wider distribution than shown on the map since it is
now known to be present at the base of the succes-
sion throughout Prudhoe Land and at Tikeraasaq on
the south side of Inglefield Bredning.

The products of the intracratonic Thule Basin vol-

canism are discussed later in the section on basic in-
trusions, particularly under Chemical characteristics
and magmatic types . Both effusive and intrusive rocks
of the Nares Strait Group are included in the TiO₂/

mg# plot (see Fig. 40) while seven representative chem-
ical analyses of lavas and sills are given in Table 2
(analyses 6-12).

Nares Strait Group, undivided ( N )

Composition . This map unit corresponds to all for-
mations of the group except the Clarence Head For-
mation, the strata of which are included in the map
unit Baffin Bay Group undivided. The circumstances
surrounding this have been explained earlier under
Map revision, sections 6-10.
Lithology . The map unit comprises, in ascending stra-
tigraphic order, sandstone with subordinate siltstone
and shale, with occasional basic sills (Northumber-
land Formation), a volcanic/redbed sequence of tho-
leiitic lavas with coeval sills, agglomerates, tuffaceous
strata (lithic tuffs, tuff breccias and ash flows) and
interflow clastic sandstone-siltstone-shale packages

(Cape Combermere Formation), and stromatolitic car-
bonates, sandstone and shale with tuffaceous elements
(Josephine Headland and Barden Bugt Formations).
The unit represents shallow-water deposition in mainly
alluvial plain and littoral environments, with one main
interval of terrestrial volcanism including plateau ba-

Fig. 37. Faulted margin of the Thule Basin at head of McCormick Fjord, Prudhoe Land. Blocks of Mesoproterozoic strata and a
Neoproterozoic(?) basic sill ( s ) cut by a late Neoproterozoic basic dyke of the Thule dyke swarm ( d₂). Ps , Precambrian shield; No ,

Northumberland Formation; CC , Cape Combermere Formation; BB , Barden Bugt Formation; CH , Clarence Head Formation; RF ,
Robertson Fjord Formation. d (top right), basic dyke of uncertain age. The basic sill ( s ) is within the Kap Trautwine Formation, the
basal strata of the Baffin Bay Group (see Dawes 1997, fig. 74 for comparative, undisturbed section at Robertson Fjord). Lower and
upper basaltic members separated by semi-recessive volcaniclastic redbeds characterises CC , a tripartite subdivision that is devel-
oped regionally (see Fig. 2). View is to the north-west with plateau surface at c. 700 m a.s.l.

salts. The thickest section of map unit N on Northum-
berland Ø, c . 700 m, thins towards the mainland (Figs
4, 37), pinching out somewhere in the inner part of
Inglefield Bredning and eastern Steensby Land where
the overlying Baffin Bay Group (Wolstenholme For-
mation) oversteps onto the shield (Fig. 36).

Baffin Bay Group

Name . Dawes (1991, 1997).
Other literature . Hofmann & Jackson (1996), Samu-
elsson et al . (1999), Steenfelt et al . (2002), Thomassen
et al . (2002a, b), Dawes (2004).
Distribution and age . Microfossils suggest a late
Mesoproterozoic (Ectasian/Stenian) and/or early Neo-
proterozoic (Tonian) age. The group represents the
most widespread strata of the Thule Basin present in
the central part of the basin, as well as on the eastern
and south-eastern margins. It overlies the Nares Strait
Group in the central basin along an abrupt contact
that represents a change to redbed sedimentation
(Dawes 1997; figs, 77, 78 ) while to the east and south-
east it overlaps onto the shield (op. cit., figs 93, 103).
Its upper contact is conformable and gradational with
the Dundas Group.
Composition . The group comprises five formations,
four of which are present in Greenland - the Kap
Trautwine, Robertson Fjord, Wolstenholme and
Qaanaaq Formations (Figs 4, 36, 37). Three members
have been formally defined. The group ranges in thick-
ness from at least 1300 m in the central basin to less
than 300 m in basin margin sections. Thinner sections
characterise the eastern exposures at the head of In-
glefield Bredning and around lake Tasersuaq, but such
sections are cut by the present erosion surface.

Baffin Bay Group, undivided ( B )

Composition . This map unit corresponds to the Cla-
rence Head Formation (now formalised as the upper-
most strata of the Nares Strait Group, see under Map
revision, section 7) overlain by the lowermost strata
of the Baffin Bay Group within the central part of the
basin, i.e. the Kap Trautwine and Robertson Fjord
Formations. The statement in the map legend that "at
Bowdoin Fjord" the basal part of this unit includes
"strata of the Nares Strait Group" actually refers to
strata below the Clarence Head Formation, in other
words, the Northumberland, Cape Combermere,
Barden Bugt and Josephine Headland Formations (Fig.
5). However, rather than applying specifically to Bow-
doin Fjord, this statement is now known to apply to
all exposures of the map unit between McCormick

Fjord and the Hubbard Gletscher (see under Map re-
vision, sections 6, 8, 9).
Lithology . The unit comprises multicoloured, shal-
low-water to terrestrial siliciclastic strata. Pale, clean
sandstone with conglomerate at the base (Clarence
Head Formation) indicative of deposition in the tidal
zone, are overlain by a redbed succession composed
of highly ferruginous sandstone and conglomerate,
with siltstone and shale (Kap Trautwine Formation)
and interbedded sandstone, siltstone and shale (Kap
Robertson Formation). The redbed succession, with
regolith deposits at the base, marks the incoming of a
strongly oxidising environment; as a whole the suc-
cession represents mixed continental to marine shore-
line environments.

Wolstenholme Formation ( Bw )

Name . Kurtz & Wales (1951), Davies et al. (1963); re-
definition with drastic reduction of stratigraphic range
and distribution by Dawes (1991, 1997).
Composition . This map unit crops out on the east-
ern and south-eastern margins of the basin directly
overlying the shield (Fig. 36). It is conformably over-
lain by the Qaanaaq Formation and varies in thick-
ness from less than 100 to c. 250 m. Easternmost out-
crops are thinner but limited by the present erosion
surface. These cap the plateau surface of the semi-
nunataks at the head of Inglefield Bredning, for ex-
ample south of the Kinginneq, and they vary from
low-relief outliers to veneer and rubble deposits on
the shield. The outcrops south and south-east of Tik-
eraasaq, on the southern side of Inglefield Bredning,
illustrated in Dawes (1997, fig. 95), are now referred
to the Nares Strait Group while the upper part of the
De Dødes Fjord outlier is known to include strata of
the Qaanaaq Formation (see under Map revision , sec-
tions 10, 11).
Lithology . The map unit comprises redbeds dominat-
ed by ferruginous sandstone and conglomerate with
minor siltstone and shale interbeds. It is interpreted
as a fluvial deposit laid down in an overall oxidising
environment.

Qaanaaq Formation ( Bq )

Name . Dawes (1991, 1997).
Composition . This map unit is the thickest and most
widely distributed formation of the Thule Supergroup
being present both in the central basin and on the
eastern and south-eastern margins. Apart from the
outcrops shown on the map, it is now known to be
preserved in the southernmost outcrops in the De

Dødes Fjord outlier (Dawes 1997, fig. 13; see under
Map revision , section 11). It ranges in thickness from
200 m in the interior of Olrik Fjord to perhaps as much
a 1000 m in Prudhoe Land.
Lithology . A rather monotonous succession of pale-
weathering sandstone with conglomerate beds, and
minor shale and siltstone that is regarded as an alluvi-
al plain to marine shoreline deposit. Some redbeds
are present in the upper strata in northernmost expo-
sures. Argillaceous strata increase in abundance up-
wards producing a transitional contact into the Dun-
das Group that is taken to represent a regional regres-
sion of the shoreline (see Dawes 1997, fig. 102).

Dundas Group

Name . Davies et al . (1963); raised to group status by
Dawes (1991, 1997).
Other literature . Munck (1941), Kurtz & Wales (1951),
Vidal & Dawes (1980), Jackson (1986), Dawes & Vidal
(1985), Dawes (1989, 2004), Hofmann & Jackson
(1996),

Samuelsson et al. (1999), Steenfelt (2002), Steen-

felt et al . (2002), Thomassen et al . (2002a, b).
Distribution and age . Microfossils suggest a late
Mesoproterozoic (Ectasian/Stenian) and/or early Neo-
proterozoic (Tonian) age. The group encompasses
thick basinal clastic strata with a wide distribution from
northern Prudhoe Land (also to Sonntag Bugt just
beyond the map sheet) to the head of Olrik Fjord and
south to Wolstenholme Ø. Regionally, it conformably
overlies the Baffin Bay Group along a gradational
boundary (see under Qaanaaq Formation ) but local-
ly, as in the Olrik Fjord area, Dundas strata overlap
nonconformably fault blocks of the Baffin Bay Group.
The upper limit of the group is unknown and its po-

sition in the map legend below the Narssârssuk Group
is based on lithological and structural inferences sug-
gesting an older age (see under Narssârssuk Group ).
The group forms the uppermost strata in five of the
six half-grabens that dissect the Thule Basin and the
strata are characteristically downdropped against the
shield on the north-eastern side of regional NW-SE
and WNW-ESE-trending faults (Fig. 35; see Thule half-
graben system ).
Composition . The group has a somewhat monoto-
nous lithology without regional markers and correla-
tion of sections is not obvious. It is estimated to be at
least 2 km, possibly as much as 3 km, thick. The three
formations recognised - the Steensby Land, Kap Powell
and Olrik Fjord Formations are based on lateral litho-
logical facies and are essentially geographically de-
fined (Fig. 4). The first two formations conformably
overlie the Qaanaaq Formation of the Baffin Bay
Group; the Olrik Fjord Formation is only recognised
in a downfaulted block and its stratal limits are un-
known. However, this formation may well represent
the youngest strata as its position in the map legend
implies (see below under Narssârssuk Group ).

Dundas Group, undivided ( D )

Composition . This map unit covers the majority of
exposures shown on the map sheet comprising the
Kap Powell and Steensby Land Formations that crop
out in two NW-SE-trending belts. In the north within
the Prudhoe half-graben, the Kap Powell Formation
stretches from Kap Chalon throughout coastal Prud-
hoe Land to the Inglefield Bredning area, while the
Steensby Land Formation, characterised by basic sills
( s₁), forms a broader belt from Northumberland Ø and

Fig. 38. Typical lithology of the Prudhoe
Land Formation, Dundas Group:
coarsening-upwards cycles with dark
shale-rich bases and pale sandstone
tops. North of Kap Chalon, Prudhoe
Land, with height of the foreground cliff
c. 150 m.

Herbert Ø through Steensby Land to the type area
around Dundas and to the southernmost exposures
on Wolstenholme Ø.
Lithology . The map unit is composed of sandstone,
siltstone and shale with lesser amounts of carbonate
(dolomite, limestone, arenaceous dolomite), chert and
evaporitic beds. Regionally, the unit shows wide lat-
eral variation in the ratio of sandstone to siltstone-
shale. The Kap Powell Formation contains more sand-
stone than the Steensby Land Formation, which is thin
bedded and dominated by black shale in which car-
bonate beds with stromatolitic reefs occur (Dawes
1997, figs 105, 112; Thomassen et al. 2002a, fig. 7).
The common upwards-coarsening units suggest depo-
sition in an overall deltaic to offshore environment
(Fig. 38). The thick cycles of the Kap Powell Forma-
tion might represent progradation delta front sequen-
ces, the thinner lower energy cycles with some pyrite
development of the Steensby Land Formation possi-
ble delta plain deposition. Characteristic lithologies
are illustrated in Dawes (1997, fig. 110).

Olrik Fjord Formation ( Do )

Name . Dawes (1991, 1997).
Other literature . Samuelsson et al. (1999), Thomas-
sen et al. (2002b).
Composition . This formation crops out on the south
coast of Olrik Fjord restricted to the central part of the
Olrik half-graben (Dawes 1997, fig. 109; Thomassen
et al. 2002b, fig. 19; see under Thule half-graben sys-
tem ). Contacts to other map units are tectonic and the
stratal limits of the formation are unknown. On the
south, the strata are juxtaposed against the shield and
slivers of Baffin Bay Group along the Itilleq Fault; to
the east the formation is faulted against the Baffin
Bay Group (see Fig. 48). Over the main outcrops, stratal
dips are gentle, but adjacent to the Itilleq Fault, con-
tortions and drag folding produce steeply dipping
sections. The thickness of the unit is estimated to be
at least 400 m.
Lithology .A dark-weathering, thin-bedded, clastic se-

quence characterised by lithological cycles with mul-
ticoloured shale units that are variously intercalated
with laminated siltstone, sandstone, thin carbonate
beds and (?)evaporitic beds. An overall deltaic or coast-
al plain environment is favoured for the Dundas Group
but the characteristic features of this formation with
redbeds, may be indicative of progradation of the shore-

line. The well-layered, dominantly fine-grained litho-
logies, resembles in gross character some parts of the
Narssârssuk Group and similarity in depositional en-
vironment is suggested by siliciclastic redbeds top-
ping cyclic sequences with carbonate rocks.

Narssârssuk Group

Name . Davies et al . (1963); raised to group status by
Dawes (1991, 1997).
Other literature . Munck (1941), Kurtz & Wales (1951),
Dawes (1979), Vidal & Dawes (1980), Strother et al .
(1983), Dawes & Vidal (1985), Jackson (1986), Hof-
mann & Jackson (1996), Samuelsson et al. (1999).
Distribution and age . Microfossils suggest a late
Mesoproterozoic (Ectasian/Stenian) and/or early Ne-
oproterozoic (Tonian) age. The group is restricted to
the Pituffik half-graben on the south-eastern margin
of the basin. It composes Saunders Ø and a mainland
belt limited to the south by the Narssarssuk Fault (see
under Thule half-graben system ). The relationship to
the Dundas Group in the north, which is the nearest
unit both geographically and stratigraphically, is hid-
den by surficial deposits filling Pituffik valley (Frontis-

piece). Regional structure suggests that the Narssârs-
suk Group is likely to be all, or in part, younger than the

Dundas Group (Steensby Land Formation). The group

is limited upwards by the present erosion surface.

Similarities to Narssârssuk Group lithologies in the

Dundas Group suggest depositional affinity and im-
plies a similar biostratigraphic age that is supported
by the acritarch taxa (Samuelsson et al . 1999; see under
Age of the Thule Supergroup ). For instance, the upper-
most beds of the Steensby Land Formation on Dun-
das Fjeld contain thin carbonate beds with stromato-
lites, chert and evaporite, while the Olrik Fjord For-
mation has multicoloured cycles including red silici-
clastic rocks and carbonates. The present consensus
is that the Narssârssuk and Dundas Groups are not
separated by a substantial age gap or a major uncon-
formity.
Composition . The group has an unknown but sub-
stantial thickness estimated at between 1.5 and 2.5
km. The strata represent subtidal to supratidal depo-
sition in very shallow water and in a low-energy, arid
or semi-arid environment, in conditions perhaps anal-
ogous to modern coastal sabkhas. Characteristic litho-
logies are illustrated in Dawes (1997, fig. 117); chem-
ical composition of various carbonate rocks are given
in Munck (1941) and Davies et al. (1963).

Tripartite division into the Imilik, Aorfêrneq and

Bylot Sund Formations is established in the sea cliffs
south of Pituffik, where strata are undisturbed by fault-
ing (Fig. 39). In contast, faults cut Saunders Ø and
have displacements in excess of the island's relief (see
Dawes 1997, fig. 114). This, and lateral facies and thick-
ness changes, make stratigraphic correlation between
the mainland and the island problematic. However,
rather than classify Saunders Ø as a fourth map unit
(i.e. Narssârssuk Group undivided), formational cor-
relation between the mainland and Saunders Ø has
been attempted.

Imilik Formation ( Ni )

Name . Dawes (1991, 1997).
Composition . This formation comprises the lower-
most strata of the group. On the mainland the base of
section is covered by surficial deposits; on the south
side of Saunders Ø it is below sea level.
Lithology . The succession has a well-layered, colour-
ful appearance, due to alternating clastic redbeds and
paler carbonates arranged in lithological cycles (Fig.
39; Davies et al. 1963, fig. 9; Dawes 1976b, fig. 234). A
typical cycle has pale limestone and/or dolomite at
the base grading into mixed carbonate-siliciclastic litho-
logies, in places with chert and evaporite, and finally
into red siltstone and sandstone. The cycles are taken
to indicate regular fluctuations of shallow, quiet wa-
ter indicating repeated progradation from intertidal
carbonates to supratidal siliciclastics. An 8 m-thick bed
of "white, light gray or translucent orange gypsum"
occurs in drill core from just south of Pituffik air base
(Davies et al . 1963, p. 30). Such thick homogeneous
evaporite beds have not been recorded in outcrop.

Aorfêrneq Formation ( Na )

Name . Davies et al. (1963) raised to formation status
by Dawes (1991, 1997).
Compositon . This unit composes the middle strata
in the mainland succession reaching the coast north
of Aafeerneq and it forms the western end of Saun-
ders Ø. In contrast to formations below and above, it
is not characterised by redbeds. It gradationally over-
lies the previous unit within a cyclic sequence in which

individual cycles are aborted and lack red siliciclastic
tops (Fig. 39).
Lithology . A carbonate-dominated (mainly dolomite)
cyclic sequence that in many sections is characterised
by evaporite in varying forms, from thin beds, veins
and nodules to the matrix of thick breccia beds. Stro-
matolites and algal mat associations, with chertified
microbiota, are common in the dolomites indicating
deposition on broad tidal flats with the persistence of
warm hypersaline conditions. Siliciclastic rocks are
restricted on the mainland to very sporadic thin beds,
some of which are red, although on Saunders Ø pale
sandstone, commonly calcareous, and arenaceous
dolomite, come in.

Bylot Sund Formation ( Nb )

Name . Dawes (1991, 1997).
Composition . This formation represents the young-
est strata of the group conformably overlying the pre-
vious unit. On the mainland, it crops out north of
Narsaarsuk in a broad syncline the southern limb of
which is truncated by the Narssarssuk Fault (Fig. 35)
while on Saunders Ø it forms much of the eastern
and northern parts of the island.
Lithology . The map unit has a similar appearance and
lithology to the Imilik Formation with siliciclastic
redbeds topping cycles. However, generally there is
lesser siliciclastic material and dolomite, variably arena-
ceous, predominates. Some transgressive cycles exist
in which multicoloured siliciclastic rocks grade up-
wards into dolomites that are variably arenaceous.

Fig. 39. Imilik ( Ni ) and Aorfêrneq ( Na )
Formations of the Narssârssuk Group.
Multicoloured progradational cycles
with basal grey carbonates topped by
red siltstone-sandstone forms the lower
strata (Imilik) overlain by abortive
carbonate-dominated cycles lacking
redbeds (Aorfêrneq). d₂, basic dyke of

the Thule dyke swarm, which has
caused bleaching in a zone several
metres wide. Coast south of Pituffik,
Bylot Sund, with cliff height c . 150 m
a.s.l.

Palaeo-, Meso- and Neoproterozoic basic intrusions

Minor basic intrusions occur throughout the map re-
gion from the land bordering Steenstrup Gletscher in
the south to Kap Chalon in the north. They form con-
spicuous features of the landscape, as they do on parts
of the map. Kap Chalon itself, is a bold buttress etched
from a basic dyke while the celebrated landmark of
North-West Greenland - table mountain Dundas Fjeld,
known internationally as 'Thule mountain' at the site
of Thule (Ummannaaq) - is capped by a subhorizon-
tal sill (Frontispiece; Dawes & Rex 1986, fig. 3; Dawes
1997, fig. 105). Several islands in Melville Bugt owe
their existence to master dykes.

The bodies are undeformed and unmetamorphosed

being composed of rocks of grossly similar appear-
ance, termed in the map legend 'dolerite', and in ear-
lier literature 'diabase' (e.g. Koch 1920; Munck 1941;
Kurtz & Wales 1951; Davies et al. 1963; Fernald &
Horowitz 1964). They are predominantly sills and
dykes, with occasional sheets, characterised by sharp
chilled contacts. Two volcanic necks have been iden-
tified although these are not shown on the map. In
areas where numerous sills are cut by a dense dyke
swarm, as in southern Steensby Land, dolerite forms
more than a minor rock type representing apprecia-
ble vertical and horizontal crustal extension. Only a
selection of intrusions mapped is shown on the map
sheet and for a more complete representation the read-
er is referred to larger scale maps (Dawes 1988b).

Map categories and their age

The basic intrusions fall into three age groups with
respect to their relationship to the Thule Basin: pre-,
syn- and post-sedimentation (Fig. 4; Table 1). These
ages are confirmed by radiometric dating to be late
Palaeoproterozoic to early Mesoproterozoic (Stathe-
rian-Calymmian), Mesoproterozoic (Ectasian-Stenian)
and Neoproterozoic (Tonian-Cryogenian, possibly
Sinian) or in map terminology, Palaeohelikian, Neo-
helikian and Hadrynian (see Introduction ). The three
age groups have distinctive geochemistry (see Fig. 40;
Table 2). Based on field and laboratory data available
at the time of map compilation, an ambitious attempt
was made to distinguish the three intrusion ages on
the map. However, for many intrusions within the
Precambrian shield beyond the limits of the Thule
Basin, assignment to a precise map category proved
problematical. Fifteen years later, this statement is still
true, with the degree of uncertainty ranging from those
intrusions mapped solely from the air or on photo-

graphs, to those only cursorily studied in the field
and to those for which petrological, chemical and iso-
topic data are available.

Five categories are depicted on the map, three of

dykes ( d , d₁ and d₂) and two of sills and sheets ( s and

s₁). The two most obvious regional dyke swarms, viz.

pre- and post-Thule Basin sedimentation, are desig-
nated d₁ and d₂, while a sill complex that cuts the

youngest Thule strata but pre-dates regional faulting
is designated s₁. All other intrusions were placed in

the less specific units d and s . Although at map com-
pilation it was known that several other ages of sills
and sheets existed, for example Mesoproterozoic (syn-
Thule sedimentation and part of the Cape Comber-
mere Formation) and Neoproterozoic (post- s₁ and post-

faulting), the establishment of a more sophisticated
classification with additional map units to cover the
few bodies of these ages that are shown on the map
was not editorially recommended.

Chronology

Cross-cutting relationships between intrusions and
tectonic features such as faults, supported by compar-
ative geochemistry, indicate that each of the five map
units contains more than a single intrusive episode.
Available field and chemical data have been synthe-
sised into the chronology presented in Table 1 that
shows that map unit assignment of some dykes has
been revised since compilation (see earlier under Map
revision , section 14). While it is fully acknowledged
that the definition of magmatic episodes on the basis
of K-Ar isotopic ages is problematical, the twelve
events of Table 1 are positioned on the basis of sup-
plementary field and/or chemical information, details
of which are given in the descriptions of the five map
units.

For example, two of the most conspicuous Neo-

proterozoic basic intrusions of the map sheet - s₁ sills

and d₂ dykes - have K-Ar ages that overlap within

error and thus on this basis they cannot be separated
as distinct magmatic episodes. However, where such
sills and dykes dominate the landscape and have com-
parable chemistry, as in Steensby Land, dykes of the
main swarm (WNW-ESE-trending) consistently cut the
sill complex thus determining their relative positions
in Table 1 (Dawes 1997, fig. 106; see also under Age
of the faulting ). Furthermore, several intrusions map-
ped as s₁ and d₂ elsewhere show chemical variation

and since the youngest K-Ar ages (610 and 530 Ma)

derive from such sills (the dykes are undated), these
intrusions are placed in Table 1 as concluding the
Proterozoic basaltic magmatism.

Given the nature of the Cretaceous-Paleogene tec-

tonism the Baffin Bay region, mafic intrusions of Ceno-
zoic age might be expected to occur (A.V. Okulitch,
personal communication 2005). Since there are hun-
dreds of undeformed mafic intrusions in the map re-
gion, few of which have been isotopically dated (see
below), the possibility of onland late Phanerozoic
magmatic rocks cannot be excluded. In summary, the
chronology put forward in Table 1 is a model to be
tested by new field work and more refined chemical
identification.

Isotopic age determinations

K-Ar whole-rock isotopic work was carried out con-
currently with the regional mapping. Thirty samples
were dated: 24 within the map sheet and 6 to the
north in northernmost Prudhoe Land and western In-

glefield Land (Fig. 1). The samples stem from 29 in-
trusions: 14 sills, 11 dykes, 3 flows and 1 sheet, with
the ages ranging from 1670 to 430 Ma. The 430 Ma
(Silurian) age from an olivine sill within the Cape
Combermere Formation has been discounted as an
expression of a fundamental disturbance of the K-Ar
isotope system (Dawes & Rex 1986). However, other
'young' ages, 610 Ma and 530 Ma, may be indicative
of the waning activity of the Franklin magmatism that
extends in parts of neighbouring Canada into the Cam-
brian (see Okulitch 1988, fig. 13; also below under
Neoproterozoic sills ( s₁) ). It can also be assumed that

in general the ages are 'younger' than the age of in-
trusion. For example, Christie & Fahrig (1983), deal-
ing with Neoproterozoic dykes in adjacent Canada,
suggested a discrepancy of 10-15%.

Nevertheless, even with these limitations, the K-Ar

ages fall into the three age groups mentioned above -
Palaeoproterozoic, Mesoproterozoic and Neoprotero-
zoic (Dawes et al . 1973, 1982b; Dawes & Rex 1986).
Since map compilation, more precise isotopic dates
are available using the U-Pb method on baddeleyite

and three ages are relevant to the geology of the map
region. These are 1629 Ma (Hamilton et al . 2004), 1628
Ma (Denyszyn et al . 2005) and 1268 Ma (LeCheminant
& Heaman 1991) that confirm the K-Ar ages of the
main Palaeoproterozoic and Mesoproterozoic magma-
tism (Table 1; see earlier under Nares Strait Group
and below under Palaeoproterozoic-Mesoproterozoic
dykes ( d₁) ).

General characteristics

Field, petrological and chemical aspects of basic in-
trusions from the map region are found in Callisen
(1929), Munck (1941), Davies et al. (1963), Fernald &
Horowitz (1964), Dawes et al. (1973, 1982b), Dawes
(1975, 1976a, 1989, 1997), Nutman (1979, 1984), Dawes
& Frisch (1981), Dawes & Rex (1986), Nielsen (1987,
1990) and Steenfelt (2002). Survey data are based on
regional field observations and c . 150 rock samples,
with major element chemistry available for about two
thirds of these (see below). Since dolerite of all ages
is of quite uniform aspect, general comments on field
and mineralogical features of the intrusions as a whole
are given here, thus avoiding repetition in the map
unit descriptions.

The rocks vary from black, dark grey to greenish

grey; severely-weathered dykes can have a reddish-
brown hue. A few of the oldest dykes are distinctly
green, altered and veined. Some d₂ dykes are char-

acterised by greenish margins with a reddish core (Fig.
2). Chilled margins are present but in some intrusions,
mainly sills, they are not particularly conspicuous.
Depending on intrusion size, dolerite is fine-, medi-
um- or coarse-grained. Black aphanitic rock charac-
terises chilled margins and thin dykes and dykelets
but the bodies shown on the map are medium- to
coarse-grained dolerite and gabbro. Medium-grained
intrusions above c. 50 m thick generally have a coars-
er central part. Bodies above 100 m thick are gabbroic
except for a marginal zone and such rocks display the
typical speckled appearance of ophitic-textured gab-
bro. Apart from chilled margins, gabbroic centres and
pegmatitic patches, textures are essentially uniform.
Only a faint suggestion of igneous layering in some
sills has been seen; xenoliths are rare.

All intrusions are pyroxene-plagioclase rocks with

a varying amount of opaque minerals, generally in
accessory amounts although Neoproterozoic intrusions
are characterised by essential amounts of ilmenite,
which may reach over 15% vol. Some dolerites are
quartz-bearing, others olivine-bearing, and in some
cases olivine is completely replaced. The main acces-
sories are biotite, hornblende, zircon, sphene and
apatite; garnet, associated with chlorite, is mentioned

by Davies et al . 1963). Rocks of all ages may show
alteration features and the degree of alteration, both
of feldspar and mafic minerals, varies locally. The most
severely altered are often greenish, typically with chlo-
rite, uralitisation of pyroxene, and sericitisation and/
or saussuritisation of feldspar. Some rocks, although
fresh-looking in hand sample, may show intense al-
teration of the feldspar with no fresh laths preserved.
The mineral alteration is considered to be deuteric.

The basaltic intrusions are post-tectonic in a regional

sense so that they appear as undeformed linear, tabu-
lar bodies; a few dykes have sinuous forms. Most
bodies show some degree of jointing and fracturing:
many show dark green discoloration zones and chlo-
rite films, while epidote and calcite are the most com-
mon vein fillings. Joints can be closely spaced and
sills are characterised by vertical joints, and in places
by columnar jointing (e.g. Munck 1941, figs 9, 12;
Dawes 1997, figs 59B, 71).

Chemical characteristics and magmatic
types

Major element chemistry for c. 100 samples of Proter-
ozoic basaltic rocks from North-West Greenland is
available in Survey archives. The majority of analyses
are from intrusions and effusive rocks within the Thule
map region, with eight (7 sills and 1 dyke) from Ingle-

Dyke
Sill, sheet
Flow, agglomerate,
tuff

Group 4: Latest Neoproterozoic

Group 2:
Mesoproterozoic

Group 3:
Neoproterozoic

Group 1: Palaeoproterozoic-
Mesoproterozoic

mg#

TiO

(wt%)

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Fig. 40. TiO₂/mg# plot of 98 basalt samples showing four mag-

matic groups. All samples are from the map region except five
Mesoproterozoic sills from Inglefield Land to the north. The
plot includes the four analyses from the Palaeoproterozoic
Melville Bugt dyke swarm and two from the Neoproterozoic
Thule dyke swarm given in Nielsen (1990, table 1) and Nielsen
(1987, table II), respectively. Group 1 , continental dyke
magmatism; group 2 , intracratonic basin volcanism; groups 3
and 4 , rift-related magmatism. c , cumulative sample. Repre-
sentative analyses of the four groups are given in Table 2.

field Land to the north (Fig. 1). Samples show varying
degrees of alteration and c. 18% of the analyses con-
tain more than 4 wt% (H₂O + CO₂). However, rather

than undertake a screening, 98 samples are shown in
the TiO₂/mg# plot that defines four compositional age-

related groups (Fig. 40). Distinguishing between sill
rock and lava in the field can be difficult and even in
thin section the rocks are very similar; thus the rock
identifications relating to group 2 - indicated by sym-
bols in Fig. 40 - are not definite. Twenty-two chemi-
cal analyses representing these four groups, all with a
LOI values below 4, are given in Table 2. The divid-
ing line between the fields of alkaline and tholeiitic
basalt mentioned below derives from the alkali/silica
diagram of Irvine & Baragar (1971).

Group 1: continental dyke magmatism
(Palaeoproterozoic-Mesoproterozoic)

All but one of the eight dykes fall above the line sepa-
rating alkaline and tholeiitic basalts in the alkali/silica
diagram, and by this definition the rocks are alkaline.
As mentioned later under Palaeoproterozoic-Mesopro-
terozoic dykes ( d₁) , the NW-SE-trending dykes of the

map region are part of the regional Melville Bugt dyke
swarm regarded as Si-saturated to undersaturated tra-
chybasalts and trachyandesites (Nielsen 1990). The
chemical analyses given here (Table 2, analyses 1-3)
are supplemented in the literature by five other anal-
yses from the map region - four in Nielsen (1990,
table 1) and one in Fernald & Horowitz (1964, p. 39).

Group 2: intracratonic basin magmatism
(Mesoproterozoic)

The dykes, and the majority of the sills and volcanics,
fall within the tholeiitic field in the alkali/silica dia-
gram, and thus match the classification of the Meso-
proterozoic basalts from the western part of the Thule
Basin in Ellesmere Island (Frisch & Christie 1982). The
wide scatter of points in an alkali/silica diagram, par-
ticularly of the effusive rocks, bears witness to the
high mobility of alkalis. The rocks are characterised by

being relatively poor in TiO₂ and P₂O₅ (Table 2, anal-

yses 4-13). The ten chemical analyses given here are
supplemented in the literature by eleven from Elles-
mere Island given in Frisch & Christie (1982, table 1,
appendix p. 13).

Groups 3 and 4: rift-related magmatism
(Neoproterozoic)

Most intrusions of this group are quartz tholeiites hav-
ing both hypersthene and quartz in the norm, but
some are alkaline basalts (e.g. Table 2, analysis 20).
All sills fall below the line dividing tholeiitic and alka-
line fields in the alkali/silica plot, while the majority
of dykes also fall within the tholeiitic field. The rocks
are characterised by relatively high TiO₂ and P₂O₅ (Ta-

ble 2, analyses 14-20); for example, the mean TiO₂

contents of four sills and eight dykes from southern
Steensby Land quoted in Dawes (1989) are 5.3 and
4.9 wt%, respectively. Two dykes from the map re-
gion, classified as a Fe- and Ti-rich tholeiite (GGU
212407) and a trachybasalt with 1.76 K₂O (GGU 166161),

are cited in Nielsen (1987, table II; note that the latter

sample is incorrectly given as 116161). Four dykes
and five sills of comparable chemistry are included in
the mean of nine samples given in Steenfelt (2002,
table 3). Dykes with matching composition - both the
high ratio of tholeiitic to alkaline dykes, and major
element chemistry - occur in Ellesmere, Coburg and
Devon Islands in Canada; for example, the eight
analyses given in Frisch (1988, table VIII) have a mean
TiO₂ of 4.7 wt%. The few sills and dykes separated

out as group 4 have lower TiO₂ and very variable K₂O

and CaO (Table 2, analyses 21, 22).

Dolerite dykes

Dykes show a very wide range of direction and fre-
quency. Some are traceable throughout the map sheet
and are part of regional swarms that extend into
Canada (Fig. 1, inset), others represent more local
swarms, while clusters of dykes with no preferential
orientation are presumably controlled by local struc-
tural conditions. Main swarms strike WNW-ESE, NW-
SE and NE-SW; subsidiary directions are northerly,
varying between NNW-SSE, N-S and NNE-SSW. Oth-
er directions can be found on the map, for example
E-W-trending dykes at the head of Inglefield Bredn-
ing described by Nutman (1984). An interesting struc-
tural condition is that the NW-SE to WNE-ESE sector,
in particular, has been favoured by dykes during all
three periods of basic magmatism (Palaeo-, Meso- and
Neoproterozoic) and cases are known where a Palaeo-
proterozoic master dyke has been utilised by a Neo-
proterozoic dyke, as for example on Thom Ø in Mel-
ville Bugt.

The majority of dykes range from a metre-wide to

c .75 m thick; the thickest dykes, some exceeding 200 m

occur along the Lauge Koch Kyst. Portrayal of dykes
on the map is diagrammatic with representation aided
by two line thicknesses that broadly correspond to
widths, below and above 50 m. Dykes below c. 10 m
thick are not shown unless closely spaced in a swarm.

Most dykes are composed of homogeneous dolerit.

Porphyritic varieties have normally randomly orienta-
ted plagioclase laths, occasionally with preferred 'flow'

orientation parallel to contacts. Some dykes in Mel-
ville Bugt have feldspars up to 6 cm long within a
central core. Several dykes that at first sight appear
composite show strong differential weathering, often
with a reddish centre (Fig. 2).

The oldest dykes ( d₁) have a grey colour on the

map. All other dykes are black but not all have been
given a qualifying symbol. This is due to two things:
(1) it is impractical in areas of map detail, for example
in southern Steensby Land, where incessant use of d₂

would clutter and where it is also superfluous in such
a dense swarm where dykes are interrelated and co-
eval, and (2) in shield exposures, it signifies areas
where many dykes have been plotted from aerial ob-
servations and photo-interpretation, and where age
assignment would amount to guesswork.

This cautious approach has proved its worth. For

example, on Nunatarsuaq, north of Harald Moltke Bræ,
where more than two dozen WNW-ESW- to NW-SE-
trending dykes are plotted within the shield mainly
from aerial photographs and a published map (Fer-
nald & Horowitz 1964, plate 1). On face value, these
dykes might be taken for a single swarm. On the map,
age symbols are only given to four dykes: two marked
d and two d₂. This indicates to the map user that the

dykes are thought to be of more than one age but
that the age of individual bodies is uncertain. Some
dykes clearly are part of the Neoproterozoic d₂ swarm

conspicuous through Steensby Land cutting the Thule
Supergroup, and comparative geochemistry shows that
Palaeoproterozoic-Mesoproterozoic ( d₁) dykes (pre-

Thule Basin) also occur. Moreover, Mesoproterozoic
dykes might also be represented since such dykes in
western Melville Bugt strike towards this area (see
below, under map units).

Where the relationship to the Thule Basin strata is

unknown, dyke direction played an important role in
the map unit classification. However, data refinement
has led to reclassification, for example, several dykes
marked d₂ on the map are now thought to be older

(Table 1).

Fig. 41. Two d

basic dykes (pre-Thule Basin) cutting polydeformed, multiphase orthogneisses of the Thule mixed-gneiss complex.

The NW-SE-trending dyke parallel to the glacier and the southern branch of Olrik Fjord is c . 150 m thick and has an age of 1628 Ma
being part of the latest Palaeoproterozoic Melville Bugt dyke swarm (Table 1). The thinner dyke cutting it is NE-SW-trending and of
Mesoproterozoic age. Dashed line accentuates a refolded isoclinal hinge. Note the well-developed plateau surface that coincides
with the Mesoproterozoic (Calymmian) peneplain. North-eastern side of Sermiarsupaluk glacier with height of plateau above glacier
c . 700 m a.s.l.

Palaeoproterozoic-Mesoproterozoic dykes
( d₁)

Name, direction and distribution . This map unit is
composed of dykes of two main directions, viz. NW-
SE (varying to WNW-ESE) and NE-SW (varying to
NNE-SSW) (Table 1). The former dykes represent the
northern part of the Melville Bugt dyke swarm (MBDS)
of Nielsen (1987), that is traceable for at least 1300 km
along the western coast of Greenland (Fig. 1, inset).
The distribution of MBDS shown in Nielsen (1990,
fig. 1) can be extended north into Inglefield Land,
thus increasing the width of the swarm by about 75
km (Dawes 2004, fig. 13).
Age : The K-Ar age range of 1670-1450 Ma given on
the map straddling the Palaeoproterozoic-Mesoprot-
erozoic boundary is based on four dykes: three NW-
SE-trending dykes and one trending NE-SW. The
former dykes are from inner Olrik Fjord, on the north-
east side of Sermiarsupaluk (1667 ± 50 Ma; Fig. 41),
from Josephine Peary Ø at the head of Inglefield Bred-
ning (1563 ± 60 Ma) and on Thom Ø, an island in
Melville Bugt (1450 ± 44 Ma); the NE-SW-trending
dyke is from Balgoni Øer, Melville Bugt (1538 ± 46
Ma) (Dawes et al. 1973; Dawes & Rex 1986). The lat-
est Palaeoproterozoic (Statherian) age of the MBDS is
established by a U-Pb baddeleyite age of 1628 ± 3 Ma
on the Sermiarsupaluk dyke (Denyszyn et al. 2005)
and 1629 ± 1 Ma obtained from a dyke south of the
map region (Hamilton et al. 2004 ) (Table 1). The pres-
ence of dykes older and younger than the MBDS is
based on field relationships. For example at Kivioq
Havn, Melville Bugt, a slightly sheared, greenish-
weathered NNE-SSW-trending dyke is cut by a MBDS
dyke (P. Hyldegaard Jensen, personal communication
1980, cited in Dawes & Frisch 1981, p. 23) while at
Sermiarsupaluk, Olrik Fjord, a NE-SW-trending dyke
cutting a MBDS dyke has a K-Ar whole-rock age of
1313 ± 39 Ma (Fig. 41; Dawes & Rex 1986).

The pre-Thule Basin age is confirmed by field rela-

tionships, for example, the sea cliffs of north-eastern
Wolstenholme Ø, where a vertical, greenish and some-
what altered dyke - of unknown direction and too
small to show on the map sheet - is seen to be trun-
cated by the unconformity below the Wolstenholme
Formation (Dawes 1975; Table 2, analysis 1).
Characteristics . NW-SE-trending dykes (and related
trends) occur throughout the Lauge Koch Kyst and
shield areas in the central part of the region, to Prud-
hoe Land in the north. The northernmost dykes marked
as d₁ are at the head of Inglefield Bredning, on Jose-

phine Peary Ø and west of Hubbard Gletscher but
some dykes in Prudhoe Land (some marked d) , for
example the dyke west of Verhoeff Gletscher, are part
of this swarm. Dyke rock is characteristically dark

weathering, somewhat greenish, and fairly resistant
compared to younger dykes. This is exemplified in
Melville Bugt, where traces of major dykes offshore
are outlined by series of small gabbro islets, for ex-
ample Ajukus Skær and the small skerries to the east-
south-east. Dykes in Melville Bugt represent the thick-
est of the map region with several between 150 and
200 m thick. One master dyke that in places is more
than 225 m thick crosses Levin Ø, Helprin Ø and Bry-
ant Ø.

As discussed earlier (under Dolerite dykes ), some

dykes crossing the semi-nunatak Nunatarsuaq, just
north of Harald Moltke Bræ, are regarded as part of
MBDS although not marked as such on the map. Evi-
dence for this is two-fold: (1) dykes are on direct strike
with the d₁ dykes around Mohn Gletscher in Melville

Bugt (cf. Nielsen 1990, fig. 1), and (2) the one chem-
ical analysis available from Nunatarsuaq (Fernald &
Horowitz 1964, p. 39) matches the composition of other
MBDS dykes (Group 1; see above under Chemical
characteristics and magmatic types ).

Few of the NE-SW-trending dykes (and related

trends) of the region are designated d₁ on the map,

although it is considered likely that the majority are
of this age. Of the three dykes discussed above under
Age , the Balgoni Øer and Kivioq Havn dykes are too
small to show on the map while the Sermiarsupaluk
dyke is part of a swarm designated d₁ by its grey col-

our. The northerly continuation of this swarm is on
the nunatak to the north of Anngiusalipaluk, while
sporadic dykes marked d occur to the west and far-
ther north on Nunatarsuaq. These dykes are consid-
ered the north-eastern part of a swarm that is tracea-
ble to the Baffin Bay coast where dykes marked d
occur between Kap Atholl and Sineriarsua. The inter-
vening region includes the terrain shown on the maps
of Davies et al. (1963, plate 1) and Fernald & Horo-
witz (1964, plate 1) where NE-SW-trending dykes are
sporadically marked d . To the west, the dyke swarm
is concealed by Thule Basin deposits and to the east
by the Inland Ice. The easternmost dykes appear to
be those at Sineriarsua and dykes on the nunataks
bordering De Dødes Fjord, for example the dyke
shown on the map north-north-east of the quartzite
( qt ) exposures.
Composition . Chemical analyses of eight d₁ dykes

define compositional group 1 of alkaline basalt (Fig.
40; Tables 1, 2 and under Chemical characteristics
and magmatic types ). Petrological descriptions of two
olivine-bearing, NW-SE-trending dykes from Melville
Bugt (Bryant Ø and Sundt Ø), referred here to the
MBDS, are given in Callisen (1929) while a NE-SW-
trending dyke is described in Davies et al. (1963, p.
38). The descriptions of Fernald & Horowitz (1964)
embrace both d₁ and d₂ dykes.

Neoproterozoic dykes ( d₂)

Name, direction and distribution . Dykes of this map
unit are mainly WNW-ESE-trending with variation to
E-W and NW-SE. Referred to as the 'Thule WNW dyke
swarm' (Nielsen 1987) or simply the 'Thule dyke
swarm' (TDS; Dawes 1988b; Dawes 2004, fig. 13), they
are an expression of the Franklin magmatic event,
defined from Arctic Canada. The swarm extends both
to the north and south of the map region, and it strikes
west into Ellesmere and Devon Islands (Fig. 1, inset;
T. Frisch, personal communication 1980; Frisch 1984a,
b, 1988; Dawes 1997, fig. 80). Its precise southern limit
is uncertain but d₂ dykes are present on the map of

Escher (1985b). To the north in Inglefield Land, the
swarm disappears beneath the Cambrian strata of the
Franklinian Basin, thus delineating a swarm more than
300 km wide (Dawes 2004, fig. 13).

However, the fact that older dykes of the same trend

occur along the Lauge Koch Kyst makes correlation
on direction alone spurious. For this reason, dykes in
the south-east of the map sheet, for example around
Duneira Bugt, although probably part of this swarm,
are specified as d. The westernmost dyke of the map
sheet on Nordvestø, Carey Øer, also marked d , is re-
garded as part of the TDS (see below under Composi-
tion ). The long-standing correlation across Baffin Bay
based on dyke trend and chemistry has recently been
refined by geophysical signatures, which reveal that
individual dykes can be traced uninterruptedly across
Smith Sound (Fig. 1; Damaske & Oakey 2003; Oakey
2005).
Age . The Hadrynian K-Ar age range of 675-630 Ma

given on the map is based on three dykes: two within
the map region (676 ± 25 and 645 ± 26 Ma) and one
(627 ± 25 Ma) in Inglefield Land to the north. A dyke
south of Pituffik air base has a K-Ar age of 727 ± 30
Ma (Dawes et al. 1972, 1982b; Dawes & Rex 1986).
Dykes of this group post-date Thule Basin sedimenta-
tion, s₁ sills and at least some of the sheets marked s ,

and early extensional faulting (Fig. 4; Table 1; see
under Age of the faulting ). Several dykes mapped as
d₂ are now known to have a chemistry quite different

from Neoproterozoic intrusions, and affinity to Meso-
proterozoic magmatism is suggested (see below, Dykes
of uncertain age ( d ) at time of map compilation ).
Characteristics . The d₂ dykes form the most dense

swarm of the map region and in areas such as Steens-
by Land, they are particularly conspicuous features,
both in the landscape and on the map (see front cov-
er illustration). Most of the dykes are vertical or near-
ly so and the few dips given on the map are steep
(75°) to the north and south. A preference for north-
erly dips is recorded by Davies et al . (1963) for dykes
south of Pituffik. Most dykes are composed of homo-

geneous dolerite but some are porphyritic, and seve-
ral 'big feldspar dykes' have been noted, for example
on the islands in Melville Bugt.

The dykes tend to weather in brownish hues and

their morphological form depends much on the host
rocks. Thus within Thule strata, dykes often form ridges,

whereas within shield outcrops they tend to form
depressions with poor dolerite exposure, or deeper
gullies, for example, the cleft giving the name to Kløft
Ø in Melville Bugt (see Munck 1941, figs 14, 16; Davies
et al . 1963, fig. 15; Dawes 1997, figs 95, 106). Since d₂

dykes are subparallel with the older dykes (both MBDS
and Mesoproterozoic dykes, see Table 1) that tend to
form positive features in the landscape, weathering
characteristics are useful distinguishing features. TDS
dykes are parallel to faults of the Thule half-graben
system and they are located along them, as for example
the Itilleq Fault of the Olrik half-graben. Regionally,
the swarm retains its WNW-ESE trend far to the east
of the present exposures of the Thule Basin.

In the Steensby Land swarm, en echelon patterns

have developed, dyke bifurcations are common and
master dykes locally peter out into dykelets. It is also
apparent that several generations exist shown by cross-
cutting relations of dykes of slightly varying direc-
tion. In detail, dykes have exploited local fracture di-
rections and at least some of those shown on the map
trending towards the NW and NNW, are offshoots from
dykes of regional direction. Similar intrusive features
are described by Nutman (1984) from Smithson Bjerge
from an E-W-trending swarm, marked d on the map
sheet, but now referred to the TDS (Table 1).
Composition . The TDS dykes - predominantly tho-

leiitic with some alkaline basalts - are part of group 3
(Fig. 40; Tables 1, 2, and under Chemical characteris-
tics and magmatic types ). Some dykes are potassic
(e.g. Table 2, analysis 20), often with lower TiO₂, and

these together with a few s₁ sills, define a separate

group (Fig. 40, group 4). Petrological descriptions of
d₂ dykes south of Pituffik are given in Davies et al.

(1963), including the Nunngarutipaluk dyke, north of
Narsaarsuk, also studied by Munck (1941, table 2, fig.
16; see Table 2, analysis 17). The description of the
quartz diabase dyke on Nordvestø, Carey Øer, by
Munck (1941), marked d on the map sheet, suggests
it to be part of TDS.

Dykes of uncertain age ( d ) at time of map
compilation

Direction and distribution . This map unit compris-
es dykes within shield exposures that at the time of
compilation could not be assigned with reasonable
confidence to the pre- and post-Thule Basin swarms

( d₁ and d₂. The dykes have a wide range of direc-

tions, with thin swarms having local preferences to
the WNW, NW, NNW, N, NNE and E. A main concen-
tration is in the region farthest away from Thule Basin,
viz. in the south-eastern part of the map sheet, east of
Docker Smith Gletscher, where small swarms of pref-
erentially oriented dykes trend to the NNW, N and
NNE. Some dykes have sinuous trends. In the same
area, several dykes have trends matching dated region-
al swarms, viz. MBDS and TDS but the reasons for
hesitancy to correlate what seems obvious have been
mentioned earlier (see under Dolerite dykes ).
Age . It is now known that dykes marked d have a
wide range of ages representing the three main peri-
ods of Proterozoic magmatism. As mentioned above
under Map revision and in the d₁ and d₂ map unit

descriptions, several of the dykes marked d on the
map can now be reclassified on the basis of refined
petrological and chemical information (Table 1). The
main examples are: (1) the poorly exposed swarm of
E-W-trending dykes on Smithson Bjerge described by
Nutman (1984) is part of TDS ( d₂, Table 2, analysis

18), (2) several WNW-ESE-trending dykes, such as
on Carey Øer, are regarded as part of TDS ( d₂), (3)

several NW-SE-trending dykes such as north of Har-
ald Molkte Bræ, are referred to MBDS ( d₁), (4) the

NE-SW-trending swarm stretching from the Baffin Bay
coast, south of Pituffik, to the Inland Ice is part of d₁

and (5) several WNW-ESE-trending dykes along the
Lauge Koch Kyst (some also marked d₂ on the map)

are part of a Mesoproterozoic swarm not recognised
on the map sheet (Table 2, analyses 4, 5). Some of
these dykes may well be feeders to the main episode
of Mackenzie volcanism of the Thule Basin (Cape
Combermere Formation; Fig. 4). One such dyke at
Kivioq Havn, Melville Bugt, that cuts a NE-SW-tren-
ding d₁ dyke, has given a K-Ar whole-rock age of

1016 ± 30 Ma, a latest Mesoproterozoic (Stenian) age
not far removed from the age of six sills cutting Thule
strata within and north of the map region, that have a
K-Ar range of 1190-1070 Ma (Dawes et al . 1973, 1982b;
Dawes & Rex 1986).

A wide range of orientations also characterise the

'Hadrynian and ?older' dolerite dykes on the Canadi-
an side of northern Baffin Bay (Frisch 1988). In addi-
tion to the western extension of the TDS (described
as 'easterly' dykes), three favoured directions are:
'northerly', 'northeasterly' and 'northwesterly' match-
ing d swarms mentioned above. Canadian dykes are
characterised by high TiO₂ corresponding to the chem-

istry of the Neoproterozoic d₂ dykes of the map re-

gion (Fig. 40, group 3). Although by no means con-
clusive evidence, this comparative geochemistry
strengthens the assumption from field characteristics
that the majority of the d dykes are 'late' in the dyke

chronology and products of the Franklin magmatic
event.

Dolerite sills and sheets

Two groups of flat-lying to shallow-dipping tabular
intrusions are distinguished on the map by colour:
black ( s ) and blue ( s₁). They vary from bodies a few

metres to c. 100 m thick and the majority are exposed
within Thule Basin strata. The few sheets known are
shallow- to moderately-dipping bodies discordant to
Thule strata, as well as subhorizontal bodies cutting
shield lithologies. The most concentrated and con-
spicuous sills on the map are within the Dundas Group
and are grouped as unit s₁. All others, including sheets

of uncertain age within the shield, are marked s . The
sills are of two main ages with respect to the Thule
Basin: coeval with sedimentation (Mesoproterozoic)
and post-sedimentation (Neoproterozoic). No sills of
pre-Thule Basin age, matching the Palaeoproterozo-
ic-Mesoproterozoic dyke magmatism, have been iden-
tified. The Neoproterozoic bodies represent several
magmatic events: a main event prior to the major exten-
sional faulting ( s₁) and perhaps two post-faulting

events (Table 1; see under Map categories and their
age , section Chronology ).

Many more sills have been mapped than shown on

the map. For example, sills are an integral part of the
Cape Combermere Formation that is not graphically
shown but included in the Nares Strait Group, undi-
vided map unit (see under Map revision, section 9). It
is also difficult to portray flat-lying tabular bodies in
vertical sections that characterise many parts of the
coast. Thickness is not a decisive factor since a thin
sill in low-relief landscape can form appreciable out-
crops whereas a thicker sill in a vertical cliff cannot
be depicted. For more complete representation, the
reader is referred to larger scale maps (Dawes 1988b).

Palaeoproterozoic?, Mesoproterozoic and
Neoproterozoic sills and sheets ( s )

Distribution and host rock . These intrusions have
been seen in the central and northern part of the map
region between Kap York in the south, Diebitsch Glet-
scher in the north and Carey Øer in the west. As dia-
grammatically represented in Fig. 4, they have four
habitats: (1) sheets within the shield, as on Carey Øer
and in the area between Parker Snow Bugt and Kap
York (Fig. 42), (2) sills within the Nares Strait Group
and in equivalent strata as on Piulip Nunaa, Prudhoe
Land (see under Map revision , section 9), (3) sills with-
in the Baffin Bay Group in Prudhoe Land (Figs 5, 37)

and (4) sheets cutting the Dundas Group and later
faults, as on Wolstenholme Ø.
Age . Sills or sheets of pre-Thule Basin age have not
been recognised. However, since dykes of this age
underlie the region, the presence of sills of similar
age cannot be discounted. The sills within the lower
Thule Supergroup have three main stratal levels, from
base upwards: (1) within sandstones of the Northum-
berland Formation, (2) as an integral part of the Cape
Combermere Formation and roughly coeval with ex-
trusive rocks, and (3) at or above the boundary between

the Nares Strait and Baffin Bay Groups (Figs 4, 5, 37;
Dawes 1997, figs 12, 49A, 74, 85, 91). A baddeleyite
age of 1268 Ma on a sill from the Cape Combermere
Formation quoted previously (see under Nares Strait
Group ) refines the middle Mesoproterozoic (Ectasian)
K-Ar age range of 1220-1205 Ma from rocks of the
same formation (Table 1; Dawes & Rex 1986; see
Dawes 1997, fig. 49A). Sills in the Northumberland
Formation are regarded as coeval (see below under
Characteristics and Composition ). The K-Ar age of 1172
± 40 Ma on a sill at Robertson Fjord, Prudhoe Land
compares well with the 1190 ± 40 Ma age from Rad-
cliffe Pynt, 10 km to the north of the map sheet (Dawes
& Rex 1986; Dawes 1997, figs 27, 74).

The sills higher in the succession are of Neoproter-

ozoic (Cryogenian) age as shown by the K-Ar age of
764 ± 30 Ma for the sill at the boundary between the
Nares Strait and Baffin Bay Groups in Robertson Fjord
(Dawes et al. 1973; see Dawes 1997, fig. 74). This
date compares well with the age of 729 ± 22 Ma from
Nordvestø, Carey Øer, which is the only age available
from a sheet (Dawes & Rex 1986). The Wolstenholme
Ø sheet is regarded as still younger in age since it
post-dates major faulting (Fig. 4; Table 1; see below).
Characteristics . Sill and flow units of the Cape Com-
bermere Formation are similar in appearance and dif-
ficult to distinguish apart (includes the sill marked s
on the map between Bowdoin Fjord and McCormick
Fjord; see under Map revision , section 9). Confirmed
sill rock with chilled contacts may contain amygdules
and what appears to be a single body may have a
vesicular 'flow' top, yet retain intrusive features else-
where. These relationships are taken to indicate a near-
surface origin for the sills. Most are below 60 m thick
and many are columnar-jointed suggesting they are
single cooling units. The structurally lower and gen-
erally thinner sills in the Northumberland Formation
are regarded has coeval as suggested by their isotopic
age.

A sill up to 25 m thick occurs higher in the Thule

succession, either at the boundary between the Nares
Strait and Baffin Bay Groups or within the latter (Figs
4, 5, 37; see also Dawes 1997, fig. 12). This sill ex-
tends throughout Prudhoe Land where, on either side

of Diebitsch Gletscher, it coincides with the plateau
surface and forms outcrops fringing ice caps. In steeper
dipping sections, for example on either side of Rob-
ertson Fjord, at McCormick Fjord and Castle Cliff (Kap
Milne), it is not portrayed on the map, neither is it
shown in the steep sea cliffs of western Steensby Land.
It is, however, shown on the northern coast of Ingle-
field Bredning, west of Hubbard Gletscher (see Tho-
massen & Krebs 2004, fig. 5).

Basaltic rocks on Carey Øer have drawn frequent

comment since several hill tops resembling 'skull-caps'
are etched out of dolerite (Wordie 1938, p. 397; Munck
1941, fig. 2; Bendix-Almgreen et al. 1967, fig. 6). These
summits represent the eroded remnants of a body that
has the lower chilled contact preserved. At least two
sheets occur. The topography of the islands is char-
acterised by flat to shallow-sloping palaeosurfaces the
age of which is uncertain (see earlier under Erosion
surfaces ). Although no outcrops of Thule strata exist
(erratic blocks are profuse), the Mesoproterozoic un-
conformity cannot have been far above the present
land surface and the basaltic magma may well have
utilised it as an access route.

The sheets in the Parker Snow Bugt area, as well as

the Wolstenholme Ø sheet, are shallow-dipping (Fig.
42). Davies et al . (1963, p. 37) recorded a dip of 27° to
the south for a body (called a 'dike') on the north-
eastern side of the island that was shown to reach the
coast west of the northern cape. On the Thule map
sheet, this body is depicted as a sheet cropping out
around the north-eastern part of the island, and al-
though also affected by faulting, the critical relation-
ship is that it cuts the main fault juxtaposing Thule
strata and the shield. In the sea cliffs of south-eastern
Wolstenholme Ø, this sheet is cut by a basic dyke
referred to the Thule dyke swarm ( d₂), thus fixing its

Neoproterozoic age and position in Table 1 (see Davies
et al. 1963, plate 1; Dawes 1997, fig. 93).
Composition . Chemical analyses are available of sills
from the lower part of the Thule Basin (e.g. Table 2,
analyses 9-11) and a single sheet within the shield
(Table 2, analysis 13). The majority of Mesoprotero-
zoic rocks are tholeiitic basalts of compositional group
2 (Fig. 40), having comparable chemistry to sills from
Inglefield Land (Table 2, analysis 12) and from Elles-
mere Island, Canada (see Frisch & Christie 1982). The
sills in the Northumberland Formation (Table 2, ana-
lysis 9) have the same chemistry as sills and lavas from
the Cape Combermere Formation (Table 2, analyses
6-8, 10, 11). The solitary sheet has higher silica but
falls within the central part of the group 2 plot, sug-
gesting affinity to Mesoproterozoic magmatism. The
Mesoproterozoic sills are olivine-bearing; this mineral
can be strongly altered or entirely replaced. The Neo-
proterozoic sills and sheets in the Baffin Bay Group

and on Carey Øer are quartz-bearing tholeiites. Petro-
logical descriptions of 'quartz diabase' composing the
sill at Robertson Fjord and the sheet on Carey Øer are
given by Munck (1941, p. 30-31), who refers to "great
quantities" of ore minerals. Her descriptions compare
well with the Fe- and Ti-rich sills, the analyses of which
fall in compositional group 3 (Fig. 40). The sheet on
Wolstenholme Ø, regarded by the present author as a
younger (post-faulting) intrusion, is described in
Davies et al. (1963).

Neoproterozoic sills ( s₁)

Name, distribution and host rock . Sills designated
s

are concentrated within the Dundas Group. Several

thin sills occur in the uppermost strata of the Baffin
Bay Group, one of which is shown on the map on
northern Herbert Ø (see Dawes 1997, fig. 102B). Also
present but not shown in the map are occasional sills
in the lower strata of the Narssârssuk Group (Imilik
Formation) south of Pituffik. One sill is exposed
(Dawes 1997, fig. 115) and several others are recorded
in a 25 m interval of a drill core (Davies et al. 1963, p.
30). On the map, s₁ sills are restricted to a belt stretch-

ing from Northumberland Ø to the Pituffik area. The
thickest stratigraphic section is in the Moriusaq half-
graben where in southern Steensby Land the clastic
strata host about 15 master sills that make up between
30 and 40% of the section (Dawes 1989). This is the
Steensby Land sill complex (see Dawes 1997, fig. 106;
also front cover illustration).
Age . Chronologically, s₁ sills post-date Thule Basin

sedimentation but pre-date extensional faulting and
d₂ dyking (Fig. 4; Table 1). The Hadrynian or latest

Neoproterozoic (Cryogenian) K-Ar whole-rock range
of 705-660 Ma given on the map is based on analyses
of three sills: Dundas Fjeld, main cap and chill, 705 ±
21 Ma and 688 ± 20 Ma, respectively; Northumberland
Ø, 662 ± 20 Ma and Booth Sund, Steensby Land, 661
± 20 Ma (Dawes & Rex 1986). The youngest K-Ar ages
of 610 ± 24 Ma and 532 ± 20 Ma come from samples
of sill rock (possibly from the same sill) stratigraphi-
cally lower than the Dundas Fjeld cap but whose rela-
tionship to d₂ dykes is not established (Dawes et al.

1973). Moreover, the chemistry of this sill is notably
different (see below, under Composition ), adding to
the suspicion that magmatic pulses may have contin-
ued into the latest Neoproterozoic (Sinian) and even
into the Cambrian (Table 1; see earlier, under Isotopic
age determinations ).
Characteristics . The s₁ sills vary from a few metres

to c . 100 m thick, with the majority between 20 and
50 m. They form the largest basalt outcrops of the
map region, being conspicuous in the terrain due to
their frequency and because the predominantly argil-
laceous host rocks weather recessively. Thus, sills pro-
trude in the landscape: in slopes, flat to shallow-dip-
ping sills form buttresses and ledges, and at the up-
per land surface, tablelands and the caps of mesa struc-
tures (e.g. Munck 1941, figs 4, 8 -11; Dawes 1997, figs
105, 106). Within inclined strata, cuestas are common.
Sill rock can be deeply weathered, and on tablelands
where the upper chill margin is eroded away, the gab-
broic core is in a state of disintegration to a coarse sand

(see Davies et al. 1963, fig. 16). In particularly exposed

places, as for example the tops of table mountains
like Dundas Fjeld where wind erosion is significant
(see Frontispiece), irregular surfaces and honeycomb
patterns are common. In well-jointed sills, like expo-

Fig. 42. Lower contact of an unde-
formed, shallow-dipping Neoproterozoic
basic sheet (designated s on the map)
showing apophyses cutting steeply
dipping, foliated quartz metagabbro of
the Kap York meta-igneous complex
that shows compositional banding.
Coast c. 10 km south-east of Parker
Snow Bugt.

sures at Nuulliit, spheroidal weathering is common.
Most sills display columnar jointing to some degree.

Composition . The vast majority of s₁ sills, like d₂

dykes, are high-TiO₂ and -P₂O₅ tholeiitic basalts that

together define a distinct compositional suite (Fig. 40,
group 3). Two sills with lower TiO₂ (Table 2, analysis

22), one of which has given a younger isotopic age
(see above), plot outside this field and together with
two dykes, define a separate compositional group (Fig.
40, group 4). The petrography of sill rocks from the
Wolstenholme Fjord - North Star Bugt area has been
described by Munck (1941) and Davies et al. (1963).
In most, quartz is present, either discrete or intergrown
with feldspar; it forms up to 3 vol.% in three samples
studied by Munck (1941, table 2). Sill rock is particu-
larly rich in opaque minerals (magnetite and ilmenite)
that reach 15% by volume.

Volcanic necks (not shown on the map)

Two basaltic structures have been interpreted as vol-
canic vents. One of Mesoproterozoic age on North-
umberland Ø is a feeder to extrusive basaltand possi-
ble also to sills in the lower part of the Thule Super-
group (Dawes 1997, fig. 61; see above under Palaeo-
proterozoic?, Mesoproterozoic and Neoproterozoic sills
and sheets ( s ) ), the other within the shield is reported
by Fernald & Horowitz (1964, pp. 37-38). This feature
is a poorly-exposed, oval-shaped basalt outcrop with
a brecciated core on the large semi-nunatak Nunatar-
suaq, north of Harald Moltke Bræ. The angular brec-
cia fragments are composed of porphyritic basalt,
quartz and feldspar with a matrix rich in chlorite and
hematite dust. The size and age of the structure are
unknown.

just south of Store Landgletscher, and stations on and
within the Inland Ice.

The satellite base called Camp Tuto (short for 'Thule

take-off') was the 'gateway' to the ice and the support
facility for many scientific programmes organised by
the U.S. Army Polar Research and Development Cen-
ter (for summary, see Fristrup 1966). These were mainly
based on three experimental constructions: (1) an ice
tunnel penetrating the Inland Ice that acted as an
unique cold-environment laboratory, (2) a permafrost
tunnel that penetrated moraine and allowed exami-
nation and testing of the characteristics of permafrost,
and (3) Camp Century - the nuclear-powered 'City
under the Ice' that was devoted to year-round polar
research and manned in the period 1959-1967. This
extraordinary subsurface installation, well known for
its much-publicised ice coring, is located at 77°10 N

and 61°08 W. Many results of the applied research

carried out under the auspices of the U.S. Army were
confidential, at least initially, but much was published
in reports issued by the Corps of Engineers research
agency SIPRE ( Snow, Ice and Permafrost Establish-
ment ) and later CRREL ( Cold Regions Research and En-

gineering Laboratory ) (e.g. Schytt 1955; Bishop 1957;
Rausch 1958; Benson 1959, 1962; Roethlisberger 1959,
1961; Goldthwait 1960, 1971; Griffiths 1960; Nobles 1960;

Corte 1962; Clarke 1966; Davis 1967; Langway 1967).

The first radiocarbon dates from Greenland were

obtained from deposits in the Pituffik region (Suess,

Quaternary

History and status of research

Following cursory observations along the coasts passed
by the early expeditions (e.g. Sutherland 1853a, b),
systematic recording of glaciers and Quaternary geo-
logy was carried out in 1894 and 1895 by T.C. Cham-
berlin and Rollin D. Salisbury who published more
than a dozen articles on the map region, focussed
particularly on the Inglefield Bredning area (e.g. Cham-
berlin 1894-97, 1985a, b; Salisbury 1895, 1896). The
first regional survey was by Koch (1928b) who descri-
bed the entire Thule map region in a well-illustrated
70-page account. This includes several panoramic
sketches showing the Inland Ice margin in Melville
Bugt and elsewhere; invaluable material for compara-
tive studies of recent glacial history.

Following the pre-war visits of Wordie (1938) and

Wright (1939), it was the establishment of various mili-
tary facilities at Pituffik and environs in the late 1940s
and 1950s that fostered a range of geoscientific activi-

ties directed towards surficial deposits, ice and snow
(e.g. Nicols 1953; Krinsley 1954; U.S. Army 1954; Schytt
1956; Washburn 1956; White 1956; Holmes & Colten
1960; Swinzow 1962; Davies et al. 1963; Dansgaard et
al. 1969; Hooke 1970; Fountain et al. 1981). In addi-
tion to the main site (Thule Air Base) at the western
end of Pituffik valley that was (and still is) a natural
staging point for scientific ventures (see Frontispiece),
the facilities included a satellite base at the ice margin

1954; Crane & Griffin 1959; Goldthwait 1960; Meyer
Rubin, in Davies et al. 1963). Incorporation of these
into the field observations of D.B. Krinsley, W.E. Dav-
ies and others, established the Pituffik region as the
type area for glacial stratigraphy and chronology for
North-West Greenland (Davies et al . 1963). Renewed
field work supported by radiometric dating program-
mes in the 1960s and 1970s revisited and re-interpre-
ted sections at Saunders Ø, Narsaarsuk and Wolsten-
holme Fjord, and also addressed Carey Øer and Olrik
Fjord (e.g. Bendix-Almgreen et al . 1967; Blake 1975,
1977, 1987; Weidick 1976, 1978a, b; Kelly 1980, 1986).
Kelly's field work in 1978 included observations and
the first C-14 dates from the little-known Lauge Koch
Kyst between Kap York and Skene Øer.

From the work cited above, it was known that the

map region hosted a complex stratigraphic record that
included several glacial and marine events extending
back beyond the Last Glacial Maximum. However,
many details of the stratigraphic record were lacking,
including precise dating of the main events. Thus, the
NORDQUA 86 expedition was launched in 1986 to
carry out detailed work on the classical localities in
the Pituffik area aided by modern dating techniques
(Funder 1990; see under History of geoscientific inves-
tigations ). This and later work in 1989 by Kelly et al .
(1999), led to the conclusion that the Middle to Late
Quaternary record is the product of three marine events
- Saunders Ø (Saalian or earlier), Qarmat (Eemian)
and Nuna (Holocene) - and three or four glacial events
- Agpat (Saalian or earlier), Narsaarsuk (Saalian), Kap
Abernathy(?) and Wolstenholme Fjord (Weichselian).
The age of the oldest deposits (Agpat) is uncertain
but they may have been laid down prior to 167 ± 16
ka B.P. Eemian non-marine biotas were studied by
Bennike & Böcher (1992), Brodersen & Bennike (2003)
and Hedenås & Bennike (2003).

In the context of the whole of Greenland, the Qua-

ternary geology of the map region is dealt with by
Funder (1989), the postglacial marine limits by Funder
& Hansen (1996) and the deglaciation chronology by
Bennike & Björck (2002).

Quaternary map units

The Thule map sheet is essentially a bedrock map.
However, rather than display the Quaternary geology
in a single 'undifferentiated' map unit, an attempt has
been made to subdivide the deposits into five catego-
ries - in addition to showing primary sites of the Cape
York meteorite shower. No systematic investigation
of the Quaternary of the map region has been under-
taken and the only specific mapping has been in the
so-called 'North Star Bugt area' between Wolstenhol-

me Fjord and Crimson Cliffs of the Kap York penin-
sula. Two maps at 1:100 000 scale of this area are pub-
lished: one, entitled Surficial geology differentiates the
Quaternary deposits into eight map units, the other
displays Glacial and related marine features (Davies
et al. 1963, plates 3, 4). These maps, although only of
a relatively small area, proved useful in the interpreta-
tion of the glacial geology throughout the map region.

The five map units are based on field information

gathered during the bedrock mapping between 1971
and 1980 (see under Data sources, field work and map
quality ), during which shells were collected for C-14
dating (e.g. Weidick 1976, 1978a). These observations
were supplemented by extensive aerial photograph
interpretation that included updating of ice bounda-
ries (see below under Recent glacial history ). The
Quaternary geology is shown in more detail on the
larger scale maps of Dawes (1988b), on which, for
instance, the marine deposits are subdivided and fea-
tures such as fluvial and marine terraces, raised beaches
and high-level lateral moraines are marked.

Cape York meteorite shower

The Cape York iron meteorite shower - the only
known source of meteoritic iron in Greenland - is
depicted on the map by nine primary fall sites. The
present state of corrosion of the pieces recovered sug-
gests that it reached Earth more than 2000 years ago
(Buchwald 1961, 1992; Buchwald & Mosdal 1985).
Place names such as Meteorbugt, Meteoritø and Iron-
stone Fjeld, as well as many local names derived from
the word 'savik' (Greenlandic for iron), pinpoint the
location of the meteorite shower to the north-east of
Kap York.

Meteoritic iron has been worked for generations

by the Thule Inuit and it is known from archaeologi-
cal sites, both as utensils and as unworked fragments.
It is assumed that many fragments have been trans-
ported from fall sites near Kap York.

The eight localities shown on the map are all re-

garded as original landing sites; seven in the Meteor-
bugt area and a single site south-west of Harald Mol-
tke Bræ. The recovered blocks range from c. 8 to 31
000 kg and they are now in museums in USA, Den-
mark and Greenland. The largest piece, called Ahnighi-
to, was removed from Meteoritø in 1894 (Peary 1898);
the last fragment discovered in 1984 was at sea level
on the same stretch of coast. The largest unworked
pieces found at secondary localities in Greenland are
at Dundas, Northumberland Ø and at Nuulliit, Steens-
by Land; pieces have also been found in Ellesmere
Island, Canada.

The nine map sites delimit a NW-SE-elongated fall

pattern that is almost 100 km long; Buchwald's (1992)
conclusion that fragments have been scattered over at
minimum 125 × 20 km must include those found at
Dundas and elsewhere. Based on the local geogra-
phy characterised by relatively small strips of ice-free
land, it is obvious that the recovered material repre-
sents but a fraction of the shower that reached Earth.
Many fragments are assumed lost under the ice and to
the waters of Melville Bugt. Since the Inland Ice is in
retreat, the chances of finding more of the meteoritic
shower increases with time (see under Recent glacial
history ).

Marine deposits, including raised deltas

Included in this map unit are isostatically raised marine
to littoral deposits scattered along the coastline and
forming rather extensive plains, tiered beaches and
delta terraces. The deposits and associated marine
features can form conspicuous elements of coastal
geology, as for example along southern Steensby Land
(Fig. 43), along Olrik Fjord and at the western end of
the broad valley linking McCormick Fjord and Bow-
doin Fjord.

Smaller areas of well-preserved terraced beaches

including ridges and berms, occur in three main set-
tings: (1) at bay heads, for example, Parker Snow Bugt,
the broad bays along Hvalsund (east of Kap Leinin-

gen, east of Kap Powlett, east of Asungaaq on North-
umberland Ø) and at North Star Bugt where beaches
flank Dundas Fjeld, showing its earlier status as an
island (see Frontispiece); (2) in deltas at the mouths
of rivers, such as at Narsaarsuk (see Funder 1990, fig.
7) and in McCormick Fjord, and (3) as cuspate fore-
lands, such as at Umiivik and Inersussat, the north-
eastern and south-western points of Saunders Ø (see
Fig. 45A). Various types of patterned ground charac-
terise the upper surfaces of the deposits, for example
large-scale polygons are common on raised delta ter-
races (see Davies et al. 1963, fig. 19; Fig. 45A).

The deposits vary from grey silt and sand, various-

ly stratified and laminated, to coarse sand and gravel
and to loose cobbles and boulders. Much of the out-
crops shown on the map are of mixed facies being
associated with glacial and glaciofluvial material, as
exemplified by the main outcrops on Saunders Ø,
southern Steensby Land (Iterlak), Dundas and Narsaar-

suk (Davies et al . 1963; Blake 1975; Funder 1990; Kel-

ly et al. 1999). Cobble to boulder beaches draping
emerged bedrock terraces characterise some parts of

the coast, for example on Carey Øer (Wordie 1938;

Blake 1975, fig. 3) and at Qeqertarsuaq, eastern Her-
bert Ø (see Fig. 45B).

Shells can usually readily be recovered from marine

silt and sand, and even from coarser deposits. Recent
accounts of the fauna of the deposits are found in
Funder (1990) and Kelly et al . (1999).

Fig. 43. Uplifted coastal plain of southern Steensby Land. View is to the east where the plain is c. 1 km wide, with the settlement of
Morisuaq situated at the coastal spit. A dolerite sill of the Steensby Land sill complex forms the island and crops out along the beach
in the foreground. Note the black colour of the active sand beaches (including the spit) due to ilmenite derived from the sills within
the Steensby Land Formation (Dundas Group). The active and uplifted beaches have a potential for placer deposits. The summit
level of the hinterland hills exceeds 300 m a.s.l. Photo: 14 September 1975.

Alluvium and deltaic deposits

Alluvial deposits, including both fluvial and glacioflu-
vial material, vary from narrow thin outcrops along
rivers to the thicker and more extensive areas shown
on the map, as for example along braided watercourses
(see Fig. 48), flat-bottomed valleys and as deltas at
the mouths of major rivers. Areas of outwash sands
and gravels occur in front of many glaciers, for example
Scarlet Heart Gletscher, and small inland outwash
plains occur, for example in the valley to the west of
Tuttu Gletscher. Steep-sided valleys, such as Five Gla-
cier Dal (striking north from McCormick Fjord), have
thick alluvium on the valley floor, with coalesced al-
luvial fans covering the lower valley slopes.

The map unit comprises both active occurrences as

well as inactive uplifted terraces that occur in some of
the major coastal deltas. Fluvial terraces are shown on
the larger scale maps of Dawes (1988b). Along several
coastal stretches, for intance south-east coast of Piulip
Nunaa, in McCormick Fjord, at the head of Granville
Fjord and in Olrik Fjord, substantial submarine estu-
aries and deltas occur, and these can be hazardous
for boats. For example, the large broad delta in front
of the unnamed expanded-foot glacier reaching Olrik
Fjord continues into the fjord as an extensive subma-
rine fan so that passage at low water for vessels other
than a small boat is problematic. Similar fans charac-
terise the coast north-west of the town of Qaanaaq.

Ground moraine, glaciofluvial deposits and
colluvium

This map unit comprises the most widespead of all
Quaternary deposits being composed primarily of
ground moraine or glacial till (non-stratified drift) that
is draped over the bedrock as a thin discontinuous
veneer. Only the most extensive areas are shown on
the map. Deposits of purely glacial origin are preserved
on many parts of the upper plateau surface, for exam-
ple the inland area of shield rocks between Pituffik
Gletscher and Inglefield Bredning and particularly on
areas of subdued topography with flat to slightly slop-
ing surfaces. Coarse to medium till and boulder fields
are the main deposits but with gradations into areas
characterised by deeply-weathered bedrock developed
as felsenmeer mixed with glacial erratics. Glacioflu-
vial material is found in the broad river valleys and
lowland plains. The till and glaciofluvial deposits have
been modified by solifluction, periglacial and fluvial
processes, as well as mechanical frost shattering.

Classical colluvial deposits, such as loose and inco-

herent scree and talus accumulations in the lower
reaches of slopes or cliffs, are ubiquitous but rarely
large enough to be depicted on the map (see earlier

under Exposure ). However, also included are a range
of material in the lower reaches of shallower slopes
in which fluvial and solifluction processes have as-
sisted down-hill movement.

Historical moraine

This map unit comprises unvegetated, ice-marginal
moraines that are associated with historically reced-
ing glaciers, such as the recent lateral moraines flank-
ing Harald Moltke Bræ and the coarse till ridges flank-
ing the front of Store Landgletscher (Davies et al. 1963,
fig. 18; Fig. 27). Several moraines form prominent fea-
tures, for example where lateral moraines are left iso-
lated as promontories or spits protruding seawards,
such as the rugged ridge at Pitoraavik or the northern
moraine of Harald Molkte Bræ, the end of which is
now c. 5 km west of the glacier front.

Well-preserved, arc-shaped terminal moraines char-

acterise several glaciers. Some of the most spectacular
are those encroached by the sea along the coast of
Hvalsund, just west of Itilleq, and along the northern
coasts of Northumberland Ø and Herbert Ø (Fig. 44A).
The steep coasts of Herbert Ø characterised by cirques
and scree slopes, display several very prominent mo-
raines, some of which are composed of multiple arcu-
ate ridges. Some of these deposits lack visible glacier
ice and represent impressive rock glaciers (Fig. 44B).

Ice margin deposits and medial moraine

A green dotted line is used on the map to mark two
types of linear morainic deposit: (1) the medial mo-
raines of active glaciers, and (2) older features on
bedrock or ground moraine now isolated from glacier
ice, such as the high-level moraines on nunataks bor-
dering Chamberlin Gletscher and the morainal ridges
in the broad, lake-filled valley east of the head of
McCormick Fjord. The latter locality preserves the re-
cessional positions of a major ice mass that once filled
the low ground between McCormick and Bowdoin
Fjords. The moraine system, shown in more detail on
the 1:100 000 map (sheet 2, Qaanaaq, Dawes 1988b)
is best preserved on the south side of the valley. It
comprises up to eight parallel to subparallel ridges
traceable from just above lake level at c . 50 m to c.
250 m and just below a series of the alluvial fans below
a bedrock escarpment. The ridges are most continu-
ous in the western end of the valley north-east of
Scarlet Heart Gletscher; towards the front of Tuttu
Gletscher, where there are several small morainic lakes,
the ridges have been disturbed by colluvial and solif-
luction processes, and are less distinct.

Marine limits

The highest shell-bearing marine silt and sand in the
map region are c. 60 m a.s.l. Well-developed, terraced
beach systems, such as those at Qeqertarsuaq, east-
ern Herbert Ø and along Hvalsund contain up to a
dozen tiered low-gradient levels and in several, beach
deposits are continuous from the marine limit down to

modern storm-wave beach ridges. Ten levels, some very

conspicuous but others weakly developed, have been
measured by hand-level at Qeqertarsuaq. The round-

ed-off altitudes are: 9 m, 16 m, 23 m, 28 m, 36 m, 41 m,

49 m, 62 m, 69 m and 84 m a.s.l. (Fig. 45B). All levels
are regarded as marking marine events since there is
no evidence of prominent fluvial action and within
the system there are also emerged sea cliffs etched
out of bedrock. The upper level at c. 84 m, that is
partly overriden by talus, is taken as the upper Holo-
cene marine limit.

This level matches the c. 86 m marine limit estab-

lished to the north of the map region around Smith
Sound (Fig. 1; Nichols 1969). Accurate determinations
of the marine limit between Herbert Ø and Inglefield
Land are sparse but to the south, in the well-studied
Carey Øer - Bylot Sund - Inglefield Bredning area, it
is markedly lower, between 35 and 65 m a.s.l. (Kelly
et al. 1999). Farther south along the Lauge Koch Kyst
and south of the map region there are few determina-
tions but these suggest a much lower marine limit,
less than 20 m a.s.l. (Funder & Hansen, fig. 3). Thus,
seen regionally, the Holocene marine limit falls to the
south-east.

Higher altitude shoreline features, such a bench

marks, also occur and in some delta terraces, for ex-
ample those east of Kap Leiningen, Hvalsund, there is
a water-worn level at c. 90 m a.s.l. Since it is now
known that there are at least two pre-Weichselian
marine events, such high-level features might be of
pre-Holocene origin. Another possibility is that they
are derived from an ice-dammed lake when ice blocked
the entrance of Hvalsund.

Glacial erratics and deglaciation

Glacial erratics occur throughout the map region inclu-
ding the outermost islands - Carey Øer in the west
and Sabine Øer in the south - and from sea-level to
the highest plateau elevations of the 'Thule Upland'
that near the Inland Ice margin in eastern Steensby
Land are in excess of 1100 m a.s.l. Along the Lauge
Koch Kyst, erratics have been recorded on ground as
high as c . 1250 m a.s.l. but no information is available
from higher summits such as Haffner Bjerg (see earli-
er under Physiography ). The presence of mainland
erratics of unmetamorphosed Thule strata on the shield
rocks of Carey Øer and the outer islands in Melville
Bugt, indicate the large expansion of the Greenland
ice sheet over the present coast and shelf. Directional
data for ice movements have been summarised by
Kelly et al. (1999).

Apart from the clear provenance shown by the Thule

strata erratics, the most useful shield rocks for trans-
port modelling are anorthosite and related lithologies
that are known in situ only at the head of Inglefield
Bredning (see under Smithson Bjerge magmatic asso-
ciation ). Erratics derived from the Qaqujârssuaq an-

Fig. 44. Historical moraines and permafrost features. A : Termi-
nal moraines of receding glaciers, Politiken Bræ (foreground)
and Berlingske Bræ, south coast of Hvalsund, with Northum-
berland Ø and Herbert Ø in background. Photo: 10 August
1983. B : Active rock glacier composed of arcuate, steep-sided
talus and ridge with a depression behind. Height of talus is c .
100 m, with ice-capped background cliffs at 800 m a.s.l. South-
ern coast of Herbert Ø, west of Qeqertarsuaq village (now aban-
doned). Photo: 19 August 1974.

orthosite occur along the coasts of Inglefield Bredn-
ing - for example, unmistakable metre-size anorth-
osite and leucogabbro boulders can be readily identi-
fied along the beach at Qaanaaq - whereas cobbles
are sporadic along the coast of Herbert Ø and Hval-
sund. The lithologies of the blocks can be readily
matched with in situ outcrops (Fig. 11). A variety of
feldspar and feldspar-rich erratics of smaller size occur
farther afield and at least some of these are deemed
to have the same derivation. These observations pro-
vide convincing evidence for major ice movement from
the head of Inglefield Bredning and through Hval-
sund and possibly Murchison Sund.

Following work by many persons cited above under

History and status of research , the chronology of ice
sheet recession of the Thule region has been discussed
by Funder (1990) and Kelly et al. (1999). The latest
compilation of all radiocarbon dates from Greenland
pertaining to the last deglaciation suggests that the
present ice-free part of the map region was not degla-
ciated until the early Holocene, 11 000 to 9000 years

ago (Bennike & Björck 2002).

Recent glacial history

Throughout western Greenland including the map
region, the margin of the Inland Ice and its outlet
glaciers are in retreat (Weidick 1995). However, one
unnamed glacier in central Steensby Land at the head
of Bowdoin Fjord is anomalous since it is in a state of
advance and has been for at least 50 years (see below).
Based on information from the early visitors to the
region in the late 19th and early 20th centuries, much
of which was summarised by Koch (1928b), this over-
all recession is known to have been in play for at
least 100 years. Evidence of the general retreat inclu-
ding changes in the glaciers at the head of Wolsten-
holme Fjord and the appearance of new nunataks was
collected by Wright (1939). Thus, Davies & Krinsley
(1962), Davies et al. (1963) and Mock (1966) could
document that in the period from 1916, the terminus
position of Harald Moltke Bræ was nearly in continu-
ous retreat interrupted by one slight advance from
1926 to 1932. The frontal recessions of the main gla-
ciers in the map region north of Kap York have been
recorded by Davies & Krinsley (1962), with an update
by Kollmeyer (1980) that included the ice front in
Melville Bugt.

The terminal positions of the ice margin and gla-

ciers shown on the map sheet are compiled from aer-
ial photographs taken in 1985. These ice limits were
plotted on topographic maps constructed from aerial
photography from 1947-49 (see Dawes 1992). Both
positions are shown on the 1:100 000 and 1:200 000

maps that form the base material of the Thule map
sheet (Dawes 1988b). These show that while some
ice fronts have been almost stationary or show only
minor recession in this 35-year period, for example,
some of the glaciers of Prudhoe Land (Bowdoin Glet-
scher and Verhoeff Gletscher), others in the same ar-
ea show a retreat of more than 1 km (Diebitsch Glet-
scher). They also show that the largest retreats have
occurred on glaciers with fronts that are afloat. Thus,
the largest ice wastage in the map region in this period
is shown by the floating tongue of Rink Gletscher in
Melville Bugt that has retreated more than 6 km on a
broad front, closely followed by Tracy Gletscher at
the head of Inglefield Bredning. Immediately south
of the map sheet, the floating ice margin of Hayes
Gletscher also shows appreciable ice wastage with
the appearance of new land as nunataks (Kollmeyer
1980).

The general pattern seen in the 35-year period to

1985 has continued until today. Thus, the floating

Fig. 45. Holocene littoral and marine features. A : Emerged beach
ridges showing large-scale polygons. Cuspate foreland at
Inersussat, Saunders Ø. B : Raised beach system west of the
now abandoned village of Qeqertarsuaq, Herbert Ø, showing
a prominent bedrock terrace. Arrow marks the uppermost
strandline measured by hand-level to c. 84 m a.s.l.

tongue of Tracy Gletscher at the head of Inglefield
Bredning that in 1947 was attached to Josephine Peary
Ø and today has a front just west of the 200 m con-
tour shown on the map sheet, shows a retreat of al-
most 10 km. Since the position of this glacier is also
well documented in the 1890s when its front was at-
tached to the northern coast of Josephine Peary Ø
and was west of Melville Gletscher, the total with-
drawal in the 120 years has been at least 12.5 km. In
stark contrast, the front of Heilprin Gletscher on the
south side of Smithson Bjerge has shown only limited
retreat in the same period.

The overall consequence of this long-lasting reces-

sion for the nature of the coastline is considerable,
particularly along the Lauge Koch Kyst, where new
land is being released from the ice and where nuna-

taks, semi-nunataks and peninsulas are being trans-
formed to semi-nunataks, peninsulas and islands, re-
spectively. One case of massive ice wastage in Side-
briksfjord is illustrated by Fig. 18.

The anomalous unnamed glacier mentioned above

at the head of Granville Fjord forms the western ex-
tent of the 'North Ice Cap' so named in U.S. Army
(1954). Evidence of ice advance along the eastern side
of this ice cap has been described by Goldthwait (1960,
1971). The Granville Fjord glacier has advanced in the
period 1948 to 1985 by more than 2.5 km, a move-
ment that has changed its front from being land-
grounded to a floating tongue. As mentioned in the
description of the Itillersuaq half-graben , the glacier
has now overriden rock exposures that were studied
by the present author in the 1970s.

Economic geology

The commodites named on the map, and the eco-
nomic geology reviewed here, relate to the three
onland geological provinces: Precambrian shield,
Thule Basin and Quaternary/Recent cover. The hydro-
carbon resource potential of the Phanerozoic sedimen-
tary succession in offshore basins (e.g. Carey Basin,
Kap York Basin and Melville Bay Graben) is not dealt
with, and for petroleum geology, the reader is referred
to the literature, for example Whittaker et al . (1997).
Here, the four metallic commodities shown on the
map ( mg , Cu , py , il ), together with the most notable
of recent discoveries, are described. The region has
attracted commercial interest but drill targets have not
been located, and the occurrences are of no immedi-
ate economic interest.

Non-metallic mineral occurrences such as evapor-

ites are not indicated on the map. Gypsum and anhy-
drite occur in the Dundas and Narssârssuk Groups,
with gypsum forming one bed up to 8 m thick (see
under Imilik Formation ( Ni ) ) while the Qaqujârssuaq
anorthosite at the head of Inglefield Bredning - the
largest single anorthosite mass in Greenland - repre-
sents a source of alumina. Raw materials of local hand-
icraft potential, including the two shown on the map
( ag , sp ), have not been sufficiently publicised. This is
corrected here.

Information on mineral occurrences has been add-

ed to in the last 15 years by exploration throughout
much of the region north of Kap York , financed by

the Danish and Greenlandic governments, as well as
by industry (see under History of geoscientific investi-
gations ). Thus, many of the known metallic mineral
showings have been re-investigated (Gowen & Shep-
pard 1994; Gowen & Kelly 1996; Thomassen et al.
2002a, b; Thomassen & Krebs 2004). The exceptions
are the ilmenite placers ( il ) reviewed by Dawes (1989)
and the iron-formation of Lauge Koch Kyst and the
islands of Melville Bugt, that was last visited in the
1970s and 1980 (Dawes 1976a, 1979; Dawes & Frisch
1981) although localities around Kap Seddon were
re-examined in 1998 (Thomassen et al. 1999a, b).

Parts of the region have been covered by geochem-

ical surveys based on stream sediments, which have
revealed anomalous concentrations of gold, copper,
lead, zinc and nickel, with some exploration support-
ed by Landsat studies (Gowen & Sheppard 1994; Go-
wen & Kelly 1996; Steenfelt 2002; Steenfelt et al. 2002;
Krebs et al. 2003). Mineralised samples collected by
the indigenous population for the so-called Green-
land mineral hunt programme, Ujarassiorit (Ujarassiorit
1993, 1995; Dunnells 1995; Olsen 2002) include some
notable anomalous metal concentrations that have
been assessed in terms of the map sheet geology by
Thomassen et al. (2002b) and Thomassen & Krebs
(2004).

An important source for mineral economic occur-

rences is the Survey's database GREENMIN ( Green-
land Min eralisation Data Bank; Lind et al. 1994; Thorn-

ing et al. 2002), that at the time of writing contains 28
entries of mineral occurrences from the map region.
The reader is referred to this database for specific de-
scriptions of the mineral showings and for analytical
results.

Metalliferous commodities on the map

Only one of the four metallic commodities included
on the map - black mineral sands ( il ) - reflects present-
day distribution. Since map publication, new occur-
rences of magnetite-rich rocks ( mg ), copper mineral-
isation ( Cu ) and iron-sulphide mineralisation ( py ) have
become known.

Magnetite ( mg )

Magnetite-rich rocks or ironstones including classical
banded iron-formation (BIF), are marked on the map
within the Thule mixed-gneiss, the Lauge Koch Kyst
supracrustal and the Melville Bugt orthogneisses com-
plexes. The term 'iron-formation' (adopted in these
notes rather than the 'ironstone' of the map legend),
represents the most widespread mineralisation of the
region. Of the 22 localities shown, all except one
(Smithson Bjerge, see below) are located in a WNW-
ESE-trending belt traceable for 350 km from Kap Sed-
don in the south-east throughout the Lauge Koch Kyst
to Magnetitbugt and Wolstenholme Ø.

Although not identified on the map, magnetite-bear-

ing gneiss occurs farther to the west on Nordvestø,
Carey Øer, and the description of a banded 'red schis-

tose gneissic rock' from the same island given by Ben-
dix-Almgreen et al . (1967) strongly suggests the pres-
ence of iron-formation. This rock is composed of
quartz, magnetite and altered feldspar, with the band-
ing caused by alternating dark magnetite-rich and pink
leucocratic bands. This observation extends the iron-
formation belt another 85 km westwards.

The distribution of oxide-facies iron-formation

shown on the map can also be expanded by occur-
rences within large tracts of the Thule mixed-gneiss
complex where magnetite is enriched in paragneiss-
es, for example in quartzitic rocks at Magnetitbjerg,
north of Harald Moltke Bræ (Fernald & Horowitz 1964).
One of the 'new' occurrences, the 'Mount Gyrfalco
showing' in inner Olrik Fjord, has been mentioned
under Map revision (Thomassen et al. 2002a, fig. 5;
2002b, figs 7, 8). The iron-formation within the Lauge
Koch Kyst supracrustal complex occurs both in meta-
sedimentary lithologies (map units ms , qt ,), for ex-
ample at De Dødes Fjord, Docker Smith Gletscher,
Thom Ø and Bushnan Ø and in the amphibolitic rocks
(map unit a₁), as at Sorte Fjeldvæg, Sidebriksfjord and

north of Pituffik Gletscher (Figs 17-19, 25, 46). Iron-
formation occurs in units of varying thickness: less
than a metre at Kap Seddon, c. 3 m at Docker Smith
Gletscher, c. 4 m at Thom Ø, c . 20 m at 'Mount Gyrfal-
co' and up to 40 m in De Dødes Fjord. Exposures of
the supracrustal units containing iron-formation from
the Lauge Koch Kyst supracrustal complex are illu-
strated in Dawes & Schønwandt (1992, fig. 3) and
Schønwandt & Dawes (1993, fig. 3).

Oxide-facies iron-formation forms a very variable

suite of rocks from low-grade disseminated magnetite
in gneiss, amphibolite and metasediments to iron-for-

BIF

Fig. 46. Banded iron-formation ( BIF )
within amphibolite ( a₁) of the Lauge

Koch Kyst supracrustal complex; the
lower amphibolite shows rusty weather-
ing. The lensoid outcrop is part of a
amphibolite unit within the Melville Bugt
orthogneiss complex. Sorte Fjeldvæg, for
location, see Fig. 17. Hammer is c. 35 cm
long.

mation and high-grade massive pure magnetite beds
(Figs 25, 46, 47). Banded iron-formation (BIF) com-
posed of alternating light and dark bands varies from
types in which the quartz or quartz-rich ( ± feldspar ±
magnetite) beds more than 1 cm thick have diffuse to
sharp contacts with magnetite beds, to regularly band-
ed types in which sharp-bordered quartz and magne-
tite bands are of mm-scale. Structurally, iron-forma-
tion varies from laminated, and in places rather slaty
rock, to small-folded types (Fig. 47). Massive layers
and lenses, and monomineralic magnetite units that
lack obvious macrostructure, are up to c. 25 cm thick.

Varying amounts of hematite and iron sulphides

can be present, a factor that determines the degree of
rusty weathering. Apart from iron, no anomalous metal
concentrations have been recorded. The highest Fe
values are from the Lauge Koch Kyst, where five BIF
samples give a mean value of c. 33.5% (max. c. 41%;
M. Lind, personal communication 1992). Chip sam-
ples over 6.5 m at 'Mount Gyrfalco' returned 30.5% Fe
with 2.1% Mn whereas 17 samples from the Thule
mixed-gneiss complex have a mean value of c. 28%
(max. 35.5%; Thomassen et al. 2002b). In the litera-
ture, BIF samples are illustrated in Thomassen et al.

(2002b, figs 6, 9, 10), with analyses in Davies et al.
(1963), Thomassen et al. (2002b) and Thomassen &
Krebs (2004).

The solitary mg symbol on the map outside the

major WNW-ESE-trending belt is on Smithson Bjerge
where subordinate ferruginous garnet-bearing quartz-
ites containing magnetite and iron suphides are inter-
layered with quartzofeldspathic paragneisses. These
meta-quartzites have been regarded as chemical sedi-
ments "perhaps akin to silicate facies banded iron for-
mation" (Nutman 1984, p. 9; Thomassen et al. 2002b,
fig. 11). Similar rocks exposing silicate-facies iron-for-
mation are exposed on the east side of Hubbard Glet-
scher, where rusty-weathering units of garnet quartz-
ites up to 4 m thick intercalated with paragneiss con-
tain disseminated magnetite, pyrrhotite and pyrite with
minor chalcopyrite and sphalerite (Thomassen & Krebs
2004, figs 2-4). Erratic blocks of quartz-garnet (± py-
roxene ± amphibolite) rocks with a variable suite of
iron minerals are common in several localities in the
inner reaches of Inglefield Bredning. The Fe content
of the silicate-facies iron-formation is lower than the
oxide-facies. Thus, ten samples from Smithson Bjerge
have returned an average of c. 18% Fe (max. c. 27%
Fe) while 17 samples from the section east of Hub-
bard Gletscher have yielded a mean of c. 15% Fe (max.
20% Fe).

With its regional strike extent of more about 400

km, the Archaean magnetite province of the Thule
region is spatially the largest in Greenland. It has been
regarded as a correlative of the Algoma-type iron de-
posits of the Mary River Group of the Committee Fold
Belt of northern Baffin Island and adjacent Melville
Peninsula that show anomalous gold and base-metal
values (Fig. 1 inset; Wilson & Underhill 1971; Dawes
1994; Jackson 2000, fig. 114; see below under Gold ).

Copper mineralisation ( Cu )

Three localities characterised by malachite staining are
marked on the map: from north to south, on the south
coast of Olrik Fjord, at Naajat on the north side of
Wolstenholme Fjord and on Salve Ø north-east of Kap
York. The first two within Thule Basin lithologies have
been re-investigated since map compilation but found
to have only modest metal concentrations; the Salve
Ø locality awaits investigation.

The Olrik Fjord locality , known in the Survey's data

bank GREENMIN and in literature as the 'Hill 620
showing', is an isolated, 100 m², bright green show-

ing near a hill top within pale sandstones of the
Qaanaaq Formation just north of the Itilleq Fault, the
bounding fault of the Olrik half-graben (Fig. 48; Stu-
art-Smith and Campbell 1971; Cooke 1978). Nearby

Fig. 47. Cut slabs of oxide-facies iron-formation. A : thin bedded
variety with small-scale folds; B : banded iron-formation with
mm-thick bands of magnetite and quartz. Erratic blocks, Camp
Tuto, Pituffik, GGU 272298 and 272300. Both types form attrac-
tive stones when cut and polished. The magnet is 2.2 cm long.
Photos: Peter K. Warna-Moors.

are several N-S-trending cross faults. It is composed
of cm-dm-sized malachite-disseminated cobbles and
slabs that generate a green tail downslope. Five pri-
mary sulphides have been identified: chalcopyrite,
pyrite, bornite, digenite and covellite. The mineralisa-
tion may be controlled by the faults along which flu-
ids entered the permeable sandstones, but copper
enrichment is low. A composite grab sample shows
0.4% Cu. Several other localities of this supposed 'red-
bed type' mineralisation exist showing malachite stain-
ing of Qaanaaq Formation sandstones, including the
chalcopyrite-pyrite mineralisation near the Kap Cleve-
land Fault at 'Red Cliffs', McCormick Fjord, that has
yielded Cu values up to 1.5% Cu (Thomassen et al.
2002b, fig. 20).

The Naajat showing , discovered by Gill (1975),

involves dark shales and thin carbonate beds of the
Dundas Group and basic intrusions of the Thule dyke
swarm ( d

) and Steensby Land sill complex ( s

). The

malachite staining is caused by veinlets and pods of
Cu- and Fe-sulphides within dolerite and host rocks.
A carbonate sample returned only 0.03% Cu with 1.8%
Zn (Gowen & Sheppard 1994). This mineralisation has
a comparable setting to Nuulliit, where malachite stain-
ing is derived from chalcopyrite associated with py-
rite mineralisation (Cooke 1978; see under Iron-sul-
phide mineralisation ( py ) below).

The Salve Ø locality represents several areas of viv-

id green staining in the sea cliffs that are composed of
gneisses and metasedimentary rocks, the adjacent
exposures to which contain magnetite in more than
anomalous amounts.

The most notable of the occurrences discovered

after map compilation are: (1) quartzite-hosted and
amphibolite-hosted copper mineralisation in the Prud-
hoe Land supracrustal complex (Thomassen et al.
2002b, fig. 17; Thomassen & Krebs 2004, frontispiece);
(2) malachite-stained paragneiss in the Thule mixed-
gneiss complex, east of Quinnisut, Inglefield Bredn-
ing, an area that is the site of a multi-element stream-
sediment anomaly (Cu, Ni, Zn, Pb) pointing to a po-
tential for base-metal mineralisation (Thomassen 2002a,
fig. 6; Thomassen et al . 2002b, fig. 12; Steenfelt et al.
2002) and (3) malachite coatings and blebs on ag-
glomeratic rocks of the Cape Combermere Formation,
studied mainly on Northumberland Ø and from where
one malachite-hematite sample has returned a Cu val-
ue> 10%, with c. 32% Fe (Thomassen & Krebs 2004,
figs 8-13). Such iron-copper mineralisation may well
represent a 'redbed type deposit', possibly associated
with the Kiataq Fault and thus defining a structural
setting comparable to that of the 'Hill 620' copper
showing farther east (Fig. 48; see under Fault-related
mineralisation ).

It should be noted that some of the most promis-

ing metal values stem from the Ujarassiorit mineral
hunt programme, for example chalcopyrite-rich rocks
from Robertson Fjord and Northumberland Ø that re-
turned Cu values> 10% (Dunnells 1995).

Iron-sulphide mineralisation ( py )

Pyrite from southern Steensby Land has been used by
the Thule Inuit for generations for producing fire, with
one well-known 'firestone' locality at Nuulliit (Peary
1898). This and two other localities are marked on
the map near the mouth of Granville Fjord within
Dundas Group that is invaded by the Steensby Land
sill complex. Although py indicates the predominance
of iron sulphides, traces of chalcopyrite, pyrrhotite
and sphalerite occur, with surfaces showing green
malachite staining (Dawes 1975; Cooke 1978; Gowen
& Sheppard 1994). The three localities are character-
ised by rusty shales and subordinate dolomites: mas-
sive pyrite pods and lenses up to 15 cm thick occur in
carbonate rocks, with disseminated pyrite cubes mainly
in the shales. Thin pyrite-rich sulphide veins pene-
trate both sediment and dolerite. The typical position

Fig. 48. 'Hill 620 showing' is the green patch on the foreground
hill composed of disseminated malachite in pale sandstone of
the Qaanaaq Formation ( Bq ). The showing is in the Olrik half-
graben close to two major faults. The Itilleq Fault causing the
escarpment in the background is the bounding fault of the half-
graben juxtaposing Dundas Group ( Do , Olrik Fjord Formation)
against the shield ( Ps ); the fault in the alluvium-floored valley
is a cross fault downdropping the Dundas Group against the
Baffin Bay Group, the upper part of which is unit Bq . View is to
the west with Dundas section above valley floor c. 300 m thick.
Photo: Bjørn Thomassen.

of this type of mineralisation is just beneath a sill sug-
gesting that the sulphides entered the system with
the dolerite.

Many sediment/sill and sediment/dyke contacts

within the Dundas Group are characterised by rusty
weathering. Most seem to be indicative of the same
type of mineralisation as described above that has only
given modest metal values. Variations are: carbonate
veins up to 50 cm wide with up 2% chalcopyrite-pyrite
following fractures in sills, semi-massive pyrrhotite
within a dyke at Moriusaq and galena-baryte mineral-
isation on Northumberland Ø (Gowen & Kelly 1996;
Thomassen et al. 2002a). Minor sphalerite mineralisa-
tion has also been observed in the Dundas Group on
Northumberland Ø with one composite sample return-
ing 2.1% Zn (Thomassen et al. 2002a, fig. 7).

Apart from Archaean iron-formation mentioned

above, the most promising target in the shield for sul-
phide enrichment is the Prudhoe Land supracrustal
complex, where certain metasedimentary tracts gen-
erate conspicuous red and yellow rust zones. In sec-
tions around Bowdoin Fjord, pyrite and pyrrhotite
occur in units of highly graphitic schists tens of me-
tres thick, in which some of the sulphides have been
remobilised into quartz-rich segregations during hy-
drothermal overprinting (Thomassen et al . 2002b, figs
13-16). Base-metal concentrations are low but stream-
sediment geochemistry suggests units with concen-
trations of REE-rich minerals (Steenfelt et al. 2002).

Black sands, mainly ilmenite ( il )

Black heavy mineral sands occur throughout the map
region being collectively referred to as the Thule black
sand province (Dawes 1989). The selected placer de-
posits marked on the map are at Kap Edvard Holm in
Melville Bugt and the entire coastal stretch from Pituffik
north to Kap Parry in western Steensby Land. North-
ernmost occurrences that are not shown on the map
are in Prudhoe Land, for example at Sonntag Bugt
(just beyond the map), at Siorapaluk in Robertson Fjord
and in McCormick Fjord and Bowdoin Fjord. The
southern localities, in Melville Bugt and around Park-
er Snow Bugt, have a hinterland of shield rocks, and
magnetite and/or titanomagnetite form the dominant
opaque fraction.

The sands within the Thule Basin are enriched in

ilmenite, derived from titanium-rich dolerite sills ( s₁)

and dykes ( d₂). The Steensby Land ilmenite showing,

a coastal stretch 80 km long, is the most economically
promising area since extensive uplifted beaches add
prospective tonnage to the placers of the active beach-
es. Around the settlement of Moriusaq, the intertidal
zone of up to 10 m is backed by raised beaches up to

1 km wide along a 20 km coastal stretch (Fig. 43). The
Pituffik-Moriusaq area has attracted commercial in-
terest (Christensen 1985).

The highest grades recorded are from the river flats

at Pituffik, the sands from which contain an opaque
fraction of up to 95%, with c. 73% absolute ilmenite.
At Moriusaq, the grade in the active beaches, which is
constantly higher than that in the older raised materi-
al, attains 60 wt% ilmenite with an average c. 37%.
The uplifted beaches have an average TiO₂ of c. 12%.

The TiO₂ content of the ilmenite concentrates from all

sands is very constant at c. 46%. Elsewhere, active
beach sands are of lower grade, for example at Siora-
paluk they contain 21% TiO₂ (Thomassen et al . 2002b,

p. 13). Potential tonnage of the active beaches is low
but the possibility of offshore placers considerably
increases this potential. The uplifted beaches contain
viable tonnage if sufficient grade could be maintained
(Dawes 1989).

Mineralisation not on the map

Gold

The long-surviving rumour from the 1960s about gold
exploitation at Pituffik from the river Sioqqap Kuua
('Fox Canyon') is still alive. Although consequential
panning of the river produced only two small gold
colours (Cooke 1978), the fact remains that the most
significant geochemical anomaly of the entire map
region comes from that river's alluvium. A heavy mine-
ral concentrate has yielded 2710 ppb Au while 174
ppb is recorded in the river at Narsaarsuk to the south
(Ujarassiorit 1993). The gold source is assumed to be
rocks of the Thule mixed-gneiss complex and one
candidate must be the prominent occurrences of iron-
formation (see under Magnetite ( mg ) ).

Banded iron-formation provinces have a potential

for gold to the extent that BIF in Canada, and else-
where, is often used as an exploration guide for gold
(Kerswill 1996). Many Canadian iron-formation occur-
rences of Algoma-type show anomalous gold and base-
metal values and the iron ore deposits of the Mary
River Group of northern Baffin Island are no excep-
tion (Jackson 1966, 2000; Wilson & Underhill 1971).
The Archaean province of the map region is now gene-
rally accepted to be the eastern extension of the Com-
mittee Fold Belt of Arctic Canada (Fig. 1, inset; Dawes
1994; Jackson 2000), thus adding to the potential for
sedimentary-exhalative gold and lead-zinc deposits in
the map region.

Apart from Pituffik, the highest gold concentration

recorded during geochemical surveys is 97 ppb Au
from west of Hubbard Gletscher, Inglefield Bredning

(Thomassen & Krebs 2004). The gold source of this
anomaly has not been pin-pointed but it could stem
from heavy-mineral horizons in sandstones or possi-
bly from with vein-type mineralisation hosted by faults
(see below).

Fault-related mineralisation

The WNW-ESE- to NW-SE-trending regional faults
cutting both the shield and Thule Basin constitute a
major exploration target (Fig. 35). Fault-controlled
copper mineralisation has been mentioned above in
relation to the bounding fault of the Olrik Fjord half-
graben, i.e. the Itilleq Fault, and its western extension
on Northumberland Ø, the Kiatak Fault.

It transpires that both fault segments are also barium-

anomalous with 5000 ppm in crushed rock, and 4400
and 5400 ppm in stream-sediment samples (Gowen &
Sheppard 1994; Steenfelt et al. 2002; Thomassen et al.
2002b). The mineralisation occurs in strata of the
graben (Dundas Group), as well as in the hanging-
wall gneisses (in Olrik Fjord) and Baffin Bay Group
strata (on Northumberland Ø). It is seen as yellow
clay alteration and pyritisation, as well as quartz-baryte-
pyrite contact mineralisation in association with a ba-
sic dyke ( d₂ (Thomassen & Krebs 2004, fig. 16). In

addition to barium, the Kiatak Fault on Northumber-
land Ø is also registered by stream-sediment samples
to be gold anomalous, with a highest value of 57 ppb
Au. Clay-altered gneisses with pyrite enrichment is
also reported along faults in Prudhoe Land (Gowen &
Kelly 1996).

It should be stressed that systematic geochemical

sampling and mineral prospecting of anomalies and
faults have been carried out only in the northern part
of the map region (Thomassen et al. 2002a, b; Tho-
massen & Krebs 2004). Similar surveys are necessary
from the southern part of the region before a full eco-
nomic appraisal of the map region can be made. Part
of that region is covered by a Landsat study, which
has identified some two dozen anomalies with miner-
alisation potential (Krebs et al . 2003).

Mineral potential: Thule Basin

In addition to the metallic mineral showings discussed
above, any economic description of the map region
should include the potential of at least two other geo-
logical settings. Both relate to the Thule Basin that is
one of several mid-Proterozoic depocentres on the
northern rim of the North American craton that have
comparable development histories. For example, the
basins have thick sandstone and basalt units at lower

levels that are succeeded by carbonate/shale-domi-
nated sequences (Young 1979; Campbell 1981). Two
of these basins are the Athabasca Basin of northern
Saskatchewan and the Borden Basin of northern Baf-
fin Island known for their productive mineralisations,
uranium and lead-zinc, respectively.

Worldwide, mid-Proterozoic rift basins containing

thick continental clastic sediments and with well-pre-
served unconformities above varied crystalline rocks,
are exploration targets for uranium mineralisation (Fer-
guson 1984; Marmont 1987; Pirajno 1992). In Green-
land, in accordance with the political climate, no spe-
cific exploration for such U-deposits has been encour-
aged. Nonetheless, the main elements of the uncon-
formity U-model occur in the map region (and be-
yond in Inglefield Land), where basal strata with per-
meable sandstones overlie a Neoarchaean-Palaeopro-
terozoic basement that includes granites and thick
metasedimentary tracts, including pelitic schists (Fig.
34; see under Prudhoe Land supracrustal complex ).
Alteration including hematite enrichment occurs above
and below and the unconformity has certainly acted
as a passageway for oxidising/reducing solutions.
Regolithic products are locally preserved (see under
Thule Basin , Structure and metamorphism and Na-
ture of the unconformity ).

In Canadian basins, a wide range of mineral depo-

sits have been discovered in the thick successions,
including sediment- and volcanic-hosted copper in
lower formations, with stratigraphically higher carbon-
ate-hosted zinc-lead (Gibbins 1991). Copper mineral-
isation at Thule in connection with sandstones and
the Cape Combermere volcanics has been mentioned
above. The Borden Basin - geographically the closest
basin to the map region - shows many similarities in
structural setting and sedimentary development to the
Thule Basin (Jackson & Iannelli 1981, 1989; Jackson
1986; Dawes 1997).

Different types of metallic mineral showings are

known, the majority of which involve lead-zinc, both
sediment-hosted stratiform-type and Mississippi Val-
ley-type (MVT; Jackson & Sangster 1987). One fault-
controlled lead-zinc-silver deposit was mined until
recently at Nanisivik (Fig. 1, inset; Olson 1984; Suth-
erland & Dumka 1995; Sherlock et al. 2004). The Nan-
isivik MVT deposit is within the Uluksan Group dom-
inated by shallow-water stromatolitic dolostones of
comparable facies to those of the Narssârssuk Group
of the Thule Basin, thus focussing attention on the
base-metal potential of this part of the Thule succes-
sion. Although no mineralisation of this type has been
recorded, a glance at the map shows that much of the
anticipated extent of the Narssârssuk Group is cove-
red by surficial deposits, leaving ample scope for the
presence of subsurface mineralisation.

Handicraft and other raw materials

Several minerals and rocks have potential as raw ma-
terials for local handicraft industries and some have
gained acceptance for lapidary work. These include
agates from Siorapaluk, a semi-precious stone known
in Greenland beyond the Thule region, and pyrite
crystals from dolomite in the neighbourhood of the
settlement Moriusaq. Soapstone has been used for
generations for household items, like blubber lamps
and food vessels, and today it is still collected for
sculpturing. In the 1960s some local interest was fos-
tered for jewellery stone, and the lapidary facilities in
Siorapaluk and Qaanaaq seen by the present author
in the 1970s and early 1980s, have supported modest
industries. Cutting and polishing stones has been a
recreational pastime at Pituffik air base since the 1950s.

Two raw materials are shown on the map sheet:

soapstone ( sp ) from the shield lithologies and agate
( q ) from Thule Basin basalts.

Soapstone ( sp )

Rocks used for carving that fall under the general term
'soapstone' vary markedly in quality, from soapstone
sensu stricto - a massive impure variety of talc - to
much harder talc-poor rocks. At the best end of this
range are pale grey to pale green, homogeneous talc-
rich ultramafic rocks usually containing serpentine and
that can be sawn from their outcrop and cut by a

knife. Used for generations by the Thule Inuit, such
stone has been quarried out from several localities in
the Pituffik Gletscher area. At the other end of the
range, usable stone includes a variety of usually dark-
er rocks which may only contain small amounts of
talc but are dominated by chlorite, muscovite and am-

phibole. When massive, these rocks have to be hacked
or heaved out of the outcrop by a pick or crowbar
and can only be worked by a file. Both types occur in
the map region.

Soapstone is mainly found within ultramafic pods

and lenses associated with amphibolites of map unit
a , as well as in talc-bearing amphibolite (see section
Ultramafic rocks ( u ) and Amphibolite ( a ) ). Some chlo-
rite-muscovite (± hornblende ± quartz) schists within
map unit ms may also carry talc, and some localities,
for example at Parker Snow Bugt, have produced poor-
quality workable stone.

Three sp occurrences are marked on the map but

others are identified on the larger scale maps in Sur-
vey archives (Dawes 1988b). The main localities are
south-east of Bylot Sund where several localities,
shown to the present author by local inhabitants, are
accessible in the sea cliffs between Kap Atholl and
Parker Snow Bugt (Dawes 1975). The sea cliffs 2-5
km north-west of Pituffik Gletscher, aptly known as
Ukkusissaq (Greenlandic for soapstone) have histori-
cally produced much workable stone as described by
Silis (1968) and Olsen (2004). On the Thule map, these
cliffs coincide with a 50° strike-and-dip symbol. Oth-
er soapstone localities not marked are in inner Olrik
Fjord, on its northern shore within the mafic-ultrama-
fic body showing a mineral lineation of 15°, and in
small, vivid green-weathering ultramafic bodies, too
small to be shown on the map, situated farther west
at the 27° dip-and-strike symbol.

Several areas of serpentine between Wolstenholme

Ø and the Inland Ice are shown on the geological map

of Davies et al. (1963, plate 1), although soapstone is
not mentioned. The inland soapstone locality marked
on the Thule map sheet north of Pituffik Gletscher is
associated with serpentine but is of poor quality.

Agate ( q ) and quartz druses

Agates and quartz druses occur in the basalts of the
Cape Combermere Formation (Fig. 49). On the map
sheet, this formation is included within the unit Nares
Strait Group, undivided ( N ) and in basal beds at Tik-
eraasaq, Inglefield Bredning (see under Map revision ,
section 10). Two localities are shown on the map: on
the north-western side of Robertson Fjord, north-east
of Siorapaluk, and on the west side of Robins Glet-
scher, Northumberland Ø. Poorer-quality agates, not

Fig. 49. Siorapaluk agate, c. natural size. Silica-filled vugs in the
Cape Combermere Formation vary from fully concentric agate
to those with a core of chalcedony, to mixed agate-quartz -
like the rather fractured example figured here - to larger druses
lined by brown to smokey quartz crystals. Photo: Peter K. Warna-
Moors.

found in situ , occur in the basalt section at Tiker-
aasaq, Kap Trautwine and Barden Bugt.

The agates from these localities are mainly reddish

brown, with subsidiary grey to pink varieties, and up
to c . 10 cm across. Often an outer agate coat com-
posed of thin white and reddish brown layers gives
way inwards to grey to pale blue chalcedony, with or
without idiomorphic quartz crystals in the centre. Some
vugs are characterised by a mix of cryptocrystalline
agate/chalcedony with crystalline quartz (Fig. 49). The
largest pure agate druses seen by this author are c. 8
cm in diameter. The agate-bearing unit near Siorapaluk
intersects the coast, and agates can be collected at sea
level in weathered basaltic rocks and in beach sands.
When cut and polished the Siorapaluk agates are at-
tractive stones that have been used in local jewellery
production (see illustrations in Secher et al. 1981, p.
119 and Ljungdahl 2004, p. 12).

Quartz druses are commonly only centimetres across

but may reach 15 cm. However, judged by samples
collected by local inhabitants, one of which seen by
the present author measured c . 30 × 15 cm, druses
lined with quartz crystals can reach another dimen-
sion. Some of these slabs with clusters of idiomorphic
brown to smoky quartz crystals up to 8 cm long, are
of museum quality and form attractive and decorative
display objects. One locality in the sea cliffs east of
Kap Trautwine is unfortunately of difficult access.

Ornamental stones

Five rocks are listed here that are deemed to have
potential as ornamental stone in a local handicraft
industry.

1. Banded iron-formation of alternating bands of

black magnetite and white quartz forms an attrac-
tive stone when cut and polished. Many in situ
localities are along the Lauge Koch Kyst (see above
under Magnetite ( mg ) ) but much more accessible
material can be collected as moraine blocks, in
the Pituffiup Kuussua valley for example, in allu-
vial deposits of the main river and near the Inland
Ice margin below the western end of the airstrip
at Camp Tuto (Fig. 47).

2. Manganese dendrites - the so-called 'Thule flow-

ers' - form attractive brown, yellow and black fern-
like patterns on bedding surfaces of pale sand-
stone of the Clarence Head Formation (Fig. 50).
One accessible locality is at Parish Gletscher, North-
umberland Ø, where slabs of fissile, pale mauve
sandstone with dendrites form scree slopes and
can be easily extracted from in situ exposure.

3. The red and purple banded sandstones, popular-

ly known as 'Thule sandstone', are decorative rocks
and even more so when liesegang rings and re-
duction phenonema such as 'fish-eye' spots are
present (Figs 33, 51, 52; see Structure and meta-
morphism under Thule Basin). Such features be-
come more pronounced when cut and varnished
but even as unworked blocks, the sandstones are
attractive when conspicuously banded. Well-pre-
served ripple marks and mud cracks add to their
charm as a potential sales item. The Northumber-
land and Wolstenholme Formations contain the
most decorative rock types and slabs can be ex-
tracted at several coastal outcrops. Erratics are
prominent in many morainic and fluvial deposits.

Fig. 50. Manganese dendrites or so-called 'Thule flowers' on
pale sandstone slab of the Clarence Head Formation, Northum-
berland Ø. GGU 212534; slab is 17 cm across. Photo: Jakob
Lautrup.

Fig. 51. 'Thule sandstone' slab with one large reduction spot
and parts of others, so-called 'fish-eyes'. The main spot is a
perfect circle 8 cm across. Slabs like this, 1 to 2 cm thick, with
bleach patterns and/or ripple marks, have handicraft potential,
for example as table mats. Photo: Jakob Lautrup.

4. Well-rounded pebbles from the fluvial 'pudding-

stone' beds in the Wolstenholme and Qaanaaq For-
mations can be collected at beach level at many
places (see Dawes 1997, figs 96, 101). The peb-
bles are predominantly vein quartz, pale coloured
with some green and pink varieties, but grey to
black chert, red quartzite and reddish granitic and
gneissic pebbles occur. When polished, such peb-
bles are attractive display objects.

5. Well-rounded

cobbles

and pebbles of banded sand-

stone from both raised and active beaches are also
attractive stones, for example, the mauve-striped
cobbles that form the present beaches on Hakluyt
Ø and Northumberland Ø that in summer are in-
cessantly pounded by the high seas of Baffin Bay
(Fig. 53).

Aggregate and road metal

Large areas of frost-free gravel and sand were exploi-
ted for construction purposes during establishment
of the air base at Pituffik in the 1950s. Basalt, the most
resistant rock and available in quantity from sills and
dykes, has been utilised as crushed rock and was, for
example, used in the building of the pier. Today, it is
quarried mainly as road metal. One other formation
in the district, not previously exploited, has a poten-
tial as road metal and as aggregate. This is the Cla-
rence Head Formation of the Nares Strait Group that
in many places is composed of clean, quartz beach

sands, free from impurities and that are locally indu-
rated. The most attractive outcrops at sea level are on
the southern coast of Northumberland Ø, just west of
Asungaaq and at Qangattaat (see Dawes 1997, fig. 48;
also Map revision , section 7).

Acknowledgements

The author recognises the indispensible support from
many persons, including the local inhabitants, for lo-
gistical and practical help in the field. Those who as-
sisted in the geological mapping are named on the
map sheet; others are mentioned in the extensive ac-
knowledgements in Dawes (1997). W.E. (Bill) Davies
(deceased, United States Geological Survey) supplied
much valuable information over the years from his
knowledge of the Thule region and several personal
communications from him are cited. Ole Bennike, Tro-
els F.D. Nielsen, Bjørn Thomassen and Anker Weidick
(Geological Survey of Denmark and Greenland -
GEUS) are thanked for critical reading and discussions
about various sections of the text; T.F.D.N.'s expert
advice proved invaluable in interpreting the chemical
data of Proterozoic magmatism. Thorough reviews by
Andrew V. Okulitch and Thomas Frisch of the Geolo-
gical Survey of Canada, led to improvements of the
text, for which the author is very thankful, and per-
sonal communications to both these referees are giv-
en in the text. Jakob Lautrup, Kristian Rasmussen and
Benny Munk Schark at GEUS are thanked for help
with photographic work and illustrations.

Fig. 52. Block of 'Thule sandstone': a decorative stone either in
this raw state or cut and polished. A potential bookend. The
quartz arenite shows ferruginous liesegang banding and pseudo-
bedding. Bedding is faintly visible with orientation shown by
arrows . Liesegang rings are secondary features that here simu-
late cross-stratification and sedimentary discordancies. Block is
c. 20 cm across. Photo: Peter K. Warna-Moors.

Fig. 53. Well-rounded cobbles of banded 'Thule sandstone'
collected from active beaches. The six shown range from 5 to
8 cm across. Potential paperweights. Photo: Jakob Lautrup.

References

Bendix-Almgreen, S.E., Fristrup, B. & Nichols, R.L. 1967: Notes

on the geology and geomorphology of the Carey Øer, North-
West Greenland. Meddelelser om Grønland 164 (8), 19 pp.

Bennike, O. & Björck, S. 2002: Chronology of the last recession

of the Greenland Ice Sheet. Journal of Quaternary Research
17 (3), 211-219.

Bennike, O. & Böcher, J. 1992: Early Weichselian interstadial land

biotas at Thule, Northwest Greenland. Boreas 21 , 111-117.

Benson, C.S. 1959: Physical investigations on the snow and firn

of northwest Greenland 1952, 1953 and 1954. Research Re-
port Cold Regions Research and Engineering Laboratory 26 ,
62 pp.

Benson, C.S. 1962: Stratigraphic studies in the snow and firn of

the Greenland ice sheet. Research Report Cold Regions Re-
search and Engineering Laboratory 70 , 93 pp.

Bishop, B.C. 1957: Shear moraines in the Thule area, northwest

Greenland. Snow, Ice and Permafrost Research Establishment
Research Report 17 , 46 pp.

Blake, W. Jr. 1975: Glacial geological investigations in northwest-

ern Greenland. Geological Survey of Canada Paper 75-1A ,
435-439.

Blake, W. Jr. 1977: Radiocarbon age determinations from the Carey

Islands, Northwest Greenland. Geological Survey of Canada
Paper 77-1A , 445-454.

Blake, W. Jr. 1987: Geological Survey of Canada radiocarbon

dates XXVI. Geological Survey of Canada Paper 86-7 , 60 pp.

Bøggild, O.B. 1953: The mineralogy of Greenland. Meddelelser

om Grønland 149 (3), 442 pp.

Brodersen, K.P. & Bennike, O. 2003: Interglacial Chironomidae

(Diptera) from Thule, Northwest Greenland: matching mod-
ern analogues to fossil assemblages. Boreas 32 , 560-565.

Buchwald, V.F. 1961: The iron meteorite 'Thule', North Green-

land. Geochimica Cosmochimica Acta 25 , 95-98.

Buchwald, V.F. 1992: On the use of iron by the Eskimos in Green-

land. Material Characterization 29 , 139-176.

Buchwald, V.F. & Mosdal, G. 1985: Meteoritic iron, telluric iron

and wrought iron in Greenland. Meddelelser om Grønland
Man & Society 9 , 49 pp.

Callisen, K. 1929: Petrographische Untersuchung einiger Gesteine

von Nordgrönland. Meddelelser om Grønland 71 (4), 217-255.

Campbell, F.H.A. (ed.): Proterozoic basins of Canada. Geological

Survey of Canada Paper 81-10 , 444 pp.

Chamberlin, T.C. 1894-1897: Glacial studies in Greenland. Jour-

nal of Geology 2 (7) (1894), 649-666, 2 (8) 768-788; 3 (1) (1895),
61-69, 3 (2)198-218, 3 (3) 469-480, 3 (5) 565-582, 3 (6) 668-
681, 3 (7) 833-843; 4 (5) (1896), 582-592; 5 (3) (1897), 229-
240.

Chamberlin, T.C. 1895a: Appendix A. Geology. In: Bryant, H.G.:

The Peary Auxiliary Expedition of 1894. Bulletin of the Geo-
graphical Club of Philadelphia 1 (5), 166-174.

Chamberlin, T.C. 1895b: Recent glacial studies in Greenland.

Bulletin of the Geological Society of America 6 , 199-220.

Christensen, K. 1985: Greenex' prospektering 1985, 3 pp. Un-

published report, Greenex A/S, Copenhagen (in archives of
Geological Survey of Denmark and Greenland, GEUS Report
File 20058 ).

Christie, K.W. & Fahrig, W.F. 1983: Paleomagnetism of the Borden

dykes of Baffin Island and its bearing on the Grenville Loop.
Canadian Journal of Earth Sciences 20 , 275-289.

Clarke, G.K.C. 1966: Seismic survey, northwest Greenland, 1964.

Research Report Cold Regions Research and Engineering Labo-
ratory 191 , 19 pp.

Cooke, H.R. 1978: Mineral reconnaissance of the Thule district,

North-West Greenland. Rapport Grønlands Geologiske Un-
dersøgelse 90 , 17-22.

Corte, A.E. 1962: Relationship between four ground patterns,

structure of the active layer, and type and distribution of ice
in the permafrost. Research Report Cold Regions Research
and Engineering Laboratory 88 , 79 pp.

Crane, H.R. & Griffin, J.B. 1959: University of Michigan radiocar-

bon dates IV. Radiocarbon Supplement, American Journal of
Science 1 , 173-198.

Damaske, D. & Oakey, G.S. 2004: Basement structures and dykes:

aeromagnetic anomalies over southern Kane Basin. Proceed-
ings of the 4th International Conference on Arctic margins,
Abstract vol. 67-68.

Dansgaard, W., Johnsen, S.J., Møller, J. & Langway, C.C. 1969:

One thousand centuries of climatic record from Camp Cen-
tury on the Greenland ice sheet. Science 166 (3903), 377-381.

Davies, W.E. 1954: Bedrock geology of the greater Thule area.

In: Final report. The scientific program (Program B), Opera-
tion Ice Cap 1953, Department of the Army Project number 9-
98-07-002, 441-442. Stanford, California: Stanford Research
Institute, U.S. Army, Fort Eustis, Virginia.

Davies, W.E. & Krinsley, D.B. 1962: The recent regimen of the

ice cap margin in North Greenland. In: Symposium of the
variations of the regime of existing glaciers. International As-
sociation of Scientific Hydrology 58 119-130.

Davies, W.E., Krinsley, D.B. & Nicol, A.H. 1963: Geology of the

North Star Bugt area, Northwest Greenland. Meddelelser om
Grønland 162 (12), 68 pp. + 3 maps.

Davis, R.M. 1967: Ice surface movements on the Tuto ramp in

North Greenland. Technical Report Cold Regions Research
and Engineering Laboratory 164 , 24 pp.

Dawes, P.R. 1972: Precambrian crystalline rocks and younger

sediments of the Thule district, North Greenland. Rapport
Grønlands Geologiske Undersøgelse 45 , 10-15.

Dawes, P.R. 1975: Reconnaissance of the Thule Group and un-

derlying basement rocks between Inglefield Bredning and
Melville Bugt, western North Greenland. Rapport Grønlands
Geologiske Undersøgelse 75 , 34-38.

Dawes, P.R. 1976a: 1:500 000 mapping of the Thule district, North-

West Greenland. Rapport Grønlands Geologiske Undersø-
gelse 80 , 23-28.

Dawes, P.R. 1976b: Precambrian to Tertiary of northern Green-

land. In: Escher, A. & Watt, W.S. (eds): Geology of Greenland,
248-303. Copenhagen: Geological Survey of Greenland.

Dawes, P.R. 1979: Field investigations in the Precambrian terrain

of the Thule district, North-West Greenland. Rapport Grønlands
Geologiske Undersøgelse 95 , 14-22.

Dawes, P.R. 1988a: Etah meta-igneous complex and the Wulff

structure: Proterozoic magmatism and deformation in Ingle-

field Land, North-West Greenland. Rapport Grønlands Geolo-
giske Undersøgelse 139 , 24 pp.

Dawes, P.R. 1988b: Geological map of the Thule district, North-

West Greenland, 1:100 000 sheets 1-6, Siorapaluk, Qaanaaq,
Hvalsund, Olrik Fjord, Granville Fjord, Bylot Sund and
1:200 000 sheets 7-11, Inglefield Bredning, Carey Øer, Kap
York, Savigsivik, Kap Seddon. Unpublished maps, Geological
Survey of Greenland, Copenhagen (in archives of Geological
Survey of Denmark and Greenland).

Dawes, P.R. 1989: The Thule black sand province, North-West

Greenland: investigation status and potential. Open File Se-
ries Grønlands Geologiske Undersøgelse 89/4 , 17 pp.

Dawes, P.R. 1991: Geological map of Greenland 1:500 000, Thule,

Sheet 5. Copenhagen: Geological Survey of Greenland.

Dawes, P.R. 1992: New geological map of the Thule region, North-

West Greenland. Rapport Grønlands Geologiske Undersø-
gelse 155 , 42 -47.

Dawes, P.R. 1994: Themes in the promotion of Greenland's mineral

resource potential. Rapport Grønlands Geologiske Undersø-
gelse 160 , 22-27.

Dawes, P.R. 1997: The Proterozoic Thule Supergroup, Green-

land and Canada: history, lithostratigraphy and development.
Geology of Greenland Survey Bulletin 174 , 120 pp.

Dawes, P.R. 2004: Explanatory notes to the Geological map of

Greenland, 1:500 000, Humboldt Gletscher, Sheet 6. Geologi-
cal Survey of Denmark and Greenland Map Series 1 , 48 pp. +
map.

Dawes, P.R. & Christie, R.L. 1982: History of exploration and

geology in the Nares Strait region. In: Dawes, P.R. & Kerr, J.W.
(eds):

Nares

Strait

and the drift of Greenland: a conflict in plate

tectonics. Meddelelser om Grønland Geoscience 8 , 19-36.

Dawes, P.R. & Christie, R.L. 1991: Geomorphic regions. In: Trettin,

H.P. (ed.) Geology of the Innuitian Orogen and Arctic Plat-
form of Canada and Greenland. Geology of Canada 3 , 29-56.
Ottawa: Geological Survey of Canada (also The geology of
North America E . Geological Society of America).

Dawes, P.R. & Frisch, T. 1981: Geological reconnaissance of the

Greenland Shield in Melville Bugt, North-West Greenland.
Rapport Grønlands Geologiske Undersøgelse 105 , 18-26.

Dawes, P.R. & Garde, A.A. 2004: Geological map of Greenland,

1:500 000, Humboldt Gletscher, Sheet 6. Copenhagen: Geo-
logical Survey of Denmark and Greenland.

Dawes, P.R. & Haller, J. 1979: Historical aspects in the geological

investigation of northern Greenland. Part 1: New maps and
photographs from the 2nd Thule Expedition 1916-1918 and
the Bicentenary Jubilee Expedition 1920-1923. Meddelelser
om Grønland 200 (4), 38 pp.

Dawes, P.R. & Rex, D.C. 1986: Proterozoic basaltic magmatic

periods in North-West Greenland: evidence from K/Ar ages.
Rapport Grønlands Geologiske Undersøgelse 130 , 24-31

Dawes, P.R. & Schønwandt, H.K. 1992: Geological setting of

Precambrian supracrustal belts: a fundamental part of mine-
ral resource evaluation in Greenland. Rapport Grønlands
Geologiske Undersøgelse 155 , 19-23.

Dawes, P.R. & Vidal, G. 1985: Proterozoic age of the Thule Group:

new evidence from microfossils. Rapport Grønlands Geolo-
giske Undersøgelse 125 , 22-28.

Dawes, P.R., Rex,

D.C. & Jepsen, H.F. 1973: K/Ar whole rock ages

of dolerites from the Thule district, western North Greenland.
Rapport Grønlands Geologiske Undersøgelse 55 , 61-66.

Dawes, P.R., Frisch, T. & Christie, R.L. 1982a: The Proterozoic

Thule Basin of Greenland and Ellesmere Island: importance

to the Nares Strait debate. In: Dawes, P.R. & Kerr, J.W. (eds):
Nares Strait and the drift of Greenland: a conflict in plate
tectonics. Meddelelser om Grønland Geoscience 8 , 89-104.

Dawes, P.R., Peel, J.S. & Rex, D.C. 1982b: The Kap Leiper basic

dyke and the age of dolerites of Inglefield Land, North-West
Greenland. Rapport Grønlands Geologiske Undersøgelse 110 ,
14-19.

Dawes, P.R., Larsen, O. & Kalsbeek, F. 1988: Archean and

Proterozoic crust in North-West Greenland: evidence from Rb-
Sr whole-rock age determinations. Canadian Journal of Earth
Sciences 25 , 1365-1373.

Denyszyn, S.W., Halls, H.C. & Davis D.W. 2005: Paleomagnetism

and U-Pb geochronology of the Melville Bugt dyke swarm,
northwestern Greenland, and implications for the Laurentia-
Baltica reconstructions in the late Paleoproterozoic. Geophysi-
cal Research Abstracts 7 , 09597.

Douglas, R.J.W. 1980: Proposals for the time classification and

correlation of Precambrian rocks and events in Canada and
adjacent areas of the Canadian Shield. Part 2: A provisional
standard for correlating Precambrian rocks. Geological Sur-
vey of Canada Paper 80-24 , 19 pp.

Dunnells, D. 1995: Ujarassiorit: 1989 to 1994. A summary report

of years 1-6, 41 pp. Unpublished report, Nunaoil A/S, Nuuk,
Greenland (in archives of Geological Survey of Denmark and
Greenland, GEUS Report File 21421 ).

Escher, J.C. 1985a: Geological map of Greenland 1:500 000,

Upernavik Isfjord, sheet 4. Copenhagen: Geological Survey
of Greenland.

Escher, J.C. 1985b: Geological map of Greenland 1:100 000,

Kuvdlorssuaq 74 V.1 Nord/Syd. Copenhagen: Geological Sur-
vey of Greenland.

Escher, J.C. & Stecher, O. 1978: Precambrian geology of the

Upernavik - Red Head region (72°15´-75°15´N), northern West
Greenland. Rapport Grønlands Geologiske Undersøgelse 90 ,
23-26.

Ferguson, J. (ed.) 1984: Proterozoic unconformity and stratabound

uranium deposits, 338 pp. Vienna, Austria: International Atomic
Energy Agency.

Fernald, A.T. & Horowitz, A.S. 1954: Bedrock geology of the

Nunatarssuak area. In: Final report. The scientific program
(Program B), Operation Ice Cap 1953, Department of the Army
Project number 9-98-07-002, 11-25. Stanford, California:
Stanford Research Institute, U.S. Army, Fort Eustis, Virginia.

Fernald, A.T. & Horowitz, A.S. 1964: Bedrock geology of the

Nunatarssuaq area, Northwest Greenland. Meddelelser om
Grønland 172 (6), 44 pp. + map.

Forsberg, R., Ekholm, S. & Olsen, H. 1994: Gravity measurements

in the Thule area, Greenland 1994, 29 pp. Unpublished Sur-
vey and Processing report, Kort- og Matrikelstyrelsen, Copen-
hagen, Denmark.

Forsberg, R., Ekholm, S., Madsen, B., Jensen, A. & Olsen, H.

1995: Gravity and GPS measurements in Greenland 1995. Nares
Strait ('OP Bouguer 95') and East Greenland regions, 55 pp.
Unpublished Survey and Processing report, Kort- og Matri-
kelstyrelsen, Copenhagen, Denmark.

Fountain, J., Usselman, T.M., Wooden, J. & Langway, C.C. 1981:

Evidence of the bedrock beneath the Greenland Ice sheet
near Camp Century, Greenland. Journal of Glaciology 27 (95),
193-197.

Frarey, M.J. 1981: A provisional standard for correlating the

Precambrian rocks of the Canadian Shield: discussion. Geo-
logical Survey of Canada Paper 81-1C , 83-88.

Frisch, T. 1983: Reconnaissance geology of the Precambrian Shield

of Ellesmere, Devon and Coburg islands, Arctic Archipelago:
a preliminary account. Geological Survey of Canada Paper
82-10 , 11 pp. + 1:750 000 map.

Frisch, T. 1984a: Geology, Prince of Wales Mountains, District of

Franklin, Northwest Territories, Map 1572A , 1:250 000. Otta-
wa: Geological Survey of Canada.

Frisch, T. 1984b: Geology, Makinson Inlet, District of Franklin,

Northwest Territories, Map 1573A , 1:250 000. Ottawa: Geolo-
gical Survey of Canada.

Frisch, T. 1988: Reconnaissance geology of the Precambrian Shield

of Ellesmere, Devon and Coburg Islands, Canadian Arctic Ar-
chipelago. Geological Survey of Canada Memoir 409 , 102 pp.

Frisch, T. & Christie, R.L. 1982: Stratigraphy of the Proterozoic

Thule Group, southeastern Ellesmere Island, Arctic Archi-
pelago. Geological Survey of Canada Paper 81-19 , 13 pp.

Frisch, T. & Dawes, P.R. 1982: The Precambrian Shield of

nothernmost Baffin Bay: correlation across Nares Strait. In:
Dawes, P.R. & Kerr, J.W. (eds): Nares Strait and the drift of
Greenland: a conflict in plate tectonics. Meddelelser om
Grønland Geoscience 8 , 79-88.

Fristrup, B. 1966: The Greenland Ice Cap, 312 pp. Copenhagen,

Denmark: Rhodos, International Science Publishers.

Funder, S. (coordinator) 1989: Quaternary geology of the ice-

free areas and adjacent shelves of Greenland. In: Fulton, R.J.
(ed.): Quaternary geology of Canada and Greenland. Geo-
logy of Canada 1 , 742-792. Ottawa: Geological Survey of
Canada (also The geology of North America K-1 , Geological
Society of America).

Funder, S. (ed.) 1990: Late Quaternary stratigraphy and glaciology

in the Thule area, Northwest Greenland. Meddelelser om
Grønland Geoscience 22 , 63 pp.

Funder, S. & Hansen, L. 1996: The Greenland ice sheet - a model

for its culmination and decay during and after the last glacial
maximum. Bulletin of the Geological Survey of Denmark 42 (2),
137-152.

Garde, A.A., Glassley, W.E. & Nutman, A.P. 1984: Two-stage co-

rona growth during Precambrian granulite facies metamor-
phism of Smithson Bjerge, north-west Greenland. Journal of
Metamorphic Geology 2 , 237-247.

Ghisler, M. & Thomsen, B. 1971: The possibility of ilmenite placers

in the Thule district, North Greenland. A preliminary exami-
nation of the heavy fractions of some sands. Rapport Grønlands
Geologiske Undersøgelse 43 , 15 pp.

Ghisler, M. & Thomsen, B. 1972: Short note on the ilmenite sands

from the Thule district, North Greenland, 3 pp. Unpublished
report, Geological Survey of Greenland, Copenhagen, Den-
mark.

Gibbins, W.A. 1991: Economic mineral resources, Arctic Islands.

In: Trettin, H.P. (ed.) 1991: Geology of the Innuitian Orogen
and Arctic Platform of Canada and Greenland. Geology of
Canada 3 , 533-539. Ottawa: Geological Survey of Canada (also
The geology of North America E . Geological Society of
America).

Gill, F.D. 1975: Report on the Melville Bugt reconnaissance project

- 1974, 4 pp. + 7 plates. Unpublished report, Cominco Ltd,
Canada (in archives of Geological Survey of Denmark and
Greenland, GEUS Report File 20255 ).

Goldthwait, R.P. 1960: Study of an ice cliff in Nunatarssuaq, Green-

land. Technical Report Cold Regions Research Engineering
Laboratory 39 , 108 pp.

Goldthwait, R.P. 1971: Restudy of Red Rock ice cliff, Nunatarssuaq,

Greenland. Technical Report Cold Regions Research Engineer-
ing Laboratory 224 , 27 pp.

Gowen, J. & Kelly, J. 1996: Follow-up mineral exploration in the

Thule area, North West Greenland, 1995, 10 pp. + figures &
appendices. Unpublished report, Nunaoil A/S, Nuuk, Green-
land (in archives of Geological Survey of Denmark and Green-
land, GEUS Report File 21449 ).

Gowen, J. & Sheppard, B. 1994: Reconnaissance mineral explo-

ration in the Thule area, North West Greenland, 24 pp. +
tables & figures. Unpublished report, Nunaoil A/S, Nuuk,
Greenland (in archives of Geological Survey of Denmark and
Greenland, GEUS Report File 21418 ).

Gregory, J.E. 1956: Geomorphology of the Red Rock Creek area.

In: Goldthwait, R.P. Study of ice cliff in Nunatarssuaq, Green-
land, 113-125. Ohio State University Research Foundation,
Project 636.

Griffiths, T.M. 1960: Glaciological investigations in the Tuto area

of Greenland. Technical Report Cold Regions Research and
Engineering Laboratory 47 , 63 pp.

Grist, A.M. & Zentilli, M. 2004: The thermal history of the Nares

Strait, Kane Basin and Smith Sound region in Canada and
Greenland: contraints from apatite fission track and (U-Th)/
He dating. Manuscript submitted to Canadian Journal of Earth
Sciences.

Hamilton, M.A., Buchan, K.L., Garde, A.A. & Connelly, J.N. 2004:

U-Pb age and preliminary paleomagnetism of a Melville Bugt
diabase dyke, West Greenland, and implications for Mid-
Proterozoic Laurentia-Baltica reconstructions. Eos, Transac-
tions, American Geophysical Union 85 (17), Joint Assembly
Supplement, Abstract GP31B-03.

Harland, W.B., Armstrong, R.L., Cox, A.V., Craig, L.E., Smith, A.G.

& Smith, D.G. 1990: A geologic time scale 1989, 263 pp. Cam-
bridge, UK: Cambridge University Press.

Haughton, S. 1858: Geological notes and illustrations. In:

M'Clintock, F.L.: Reminiscences of Arctic ice-travel in search
of Sir John Franklin and his companions. Journal of the Royal
Dublin Society 1 , 183-250.

Haughton, S. 1859: Appendix No. IV. Geological account of the

Arctic Archipelago, drawn up principally from the specimens
collected by Captain F.L. M'Clintock, R.N., from 1849 to 1869.
In: M'Clintock, [F.L.]: A narrative of the discovery of the fate of
Sir John Franklin and his companions, 372-399. London: John
Murray.

Hedenås, L. & Bennike, O. 2003: Moss remains from the last

interglacial at Thule, NW Greenland. Lindbergia 28 , 52-58.

Henderson, G. & Pulvertaft, T.C.R. 1987: Geological map of Green-

land, 1:100 000, Mârmorilik 71 V.2 Syd, Nûgâtsiaq 71 V.2 Nord,
Pangnertôq 72 V.2 Syd. Descriptive text, 72 pp. + 3 maps.
Copenhagen: Geological Survey of Greenland.

Henriksen, N., Higgins, A.K., Kalsbeek, F. & Pulvertaft, T.C.R.

2000: Greenland from Archaean to Quaternary. Descriptive
text to the Geological map of Greenland 1:2 500 000. Geo-
logy of Greenland Survey Bulletin 185 , 93 pp. + map.

Hofmann, H.J. & Jackson, G.D. 1996: Notes on the geology and

micropaleontology of the Proterozoic Thule Group, Ellesmere
Island, Canada and North-West Greenland. Geological Sur-
vey of Canada Bulletin 495 , 26 pp.

Holmes, C.D. & Colton, R.B. 1960: Patterned ground near Dundas

(Thule Air Force Base), Greenland. Meddelelser om Grønland
158 (6), 15 pp.

Hooke, R. LeB. 1970: Morphology of the ice-sheet margin near

Thule, Greenland. Journal of Glaciology 9 , 303-324.

Hovey, E.O. 1918: Notes on the geology of the region of Parker

Snow Bay, Greenland. Bulletin of the Geological Society of
America 29 , 98 only.

Irvine, T.N. & Baragar, W.R.A. 1971: A guide to the chemical

classification of common volcanic rocks. Canadian Journal of
Earth Sciences 8 , 523-548.

Jackson, G.D. 1966: Geology and mineral possibilities of the Mary

River region, northern Baffin Island. Canadian Mining Journal
87 (6), 57-61.

Jackson, G.D. 1986: Notes on the Proterozoic Thule Group, north-

ern Baffin Bay. Geological Survey of Canada Paper 86-1A ,
541-552.

Jackson, G.D. 2000: Geology of the Clyde-Cockburn Land map

area, north-central Baffin Island, Nunavut. Geological Survey
of Canada Memoir 440 , 303 pp. + maps.

Jackson, G.D. & Iannelli, T.R. 1981: Rift-related cyclic sedimenta-

tion in the Neohelikian Borden Basin, northern Baffin Island.
In: Campbell, F.H.A. (ed.): Proterozoic basins of Canada. Geo-
logical Survey of Canada Paper 81-10 , 269-302.

Jackson, G.D. & Iannelli, T.R. 1989: Neohelikian reef complexes,

Borden Basin, northwestern Baffin Island. In Geldsetzer, H.H.J.,
James, N.P. & Tebbutt, G.E. (eds): Reefs, Canada and adjacent
area. Canadian Society of Petroleum Geologists Memoir 13 ,
55-63.

Jackson, G.D. & Sangster, D.F. 1987: Geology and resource po-

tential of a proposed National Park, Bylot Island and north-
west Baffin Island, Northwest Territories. Geological Survey
of Canada Paper 87-17 , 31 pp + map.

Jackson, H.R., Loncarevic, B.D. & Blake, W.B. 1992: Shallow seis-

mic and magnetic data from northernmost Baffin Bay: insights
into geology. Geological Survey of Canada Paper 92-1B , 23-
29.

Jackson, H.R. & shipboard party 2003: Cruise 2001 Louis S. St.

Laurent. Passage. Pathway to the Arctic. Seismic survey and
geoscientific experiment, Resolute (Nunavut) to Thule (Green-
land). Geological Survey of Canada Open File 4282 , 152 pp.

Kalsbeek, F. 1981: The northward extent of the Archaean base-

ment of Greenland - a review of Rb-Sr whole-rock ages.
Precambrian Research 14 , 203-219.

Kalsbeek, F. 1986: The tectonic framework of the Precambrian

shield of Greenland. A review of new isotopic evidence. In:
Kalsbeek, F. & Watt, W.S. (eds): Developments in Greenland
geology. Rapport Grønlands Geologiske Undersøgelse 128 ,
55-64.

Kalsbeek, F. & Dawes, P.R. 1980: Rb-Sr whole-rock measure-

ments of the Kap York meta-igneous complex, Thule district,
North Greenland. Rapport Grønlands Geologiske Undersø-
gelse 100 , 30-33.

Kelly,

1980:

Preliminary

investigations

the

Quaternary

Melville

Bugt

and

Dundas,

North-West

Greenland. Rapport

Grønlands Geologiske Undersøgelse

100 ,

33-38.

Kelly, M. 1986: Quaternary, pre-Holocene, marine events of west-

ern Greenland. Rapport Grønlands Geologiske Undersøgelse
131 , 23 pp.

Kelly, M., Funder, S., Houmark-Nielsen, M., Knudsen, K.L.,

Kronborg, C., Landvik, J. & Sorby, L. 1999: Quaternary glacial
and marine environmental history of northwest Greenland: a
review and reappraisal. Quaternary Science Reviews 18 , 373-
392.

Kerswill, J.A. 1996: Iron-formation-hosted stratabound gold. In:

Eckstrand, O.R., Sinclair, W.D. & Thorpe, R.I. (eds): Geology
of Canadian mineral deposit types. Geology of Canada 8 ,
367-382. Ottawa: Geological Survey of Canada. (also The
geology of North America P-1 , Geological Society of North
America).

Knight, R.D. & Jackson, G.D. 1994: Sedimentology and stratigra-

phy of the Mesoproterozoic Elwin Subgroup (Aqigilik and
Sinasiuvik formations), uppermost Bylot Supergroup, Borden
Rift Basin, northern Baffin Island. Geological Survey of Canada
Bulletin 455 , 43 pp.

Koch, L. 1920: Stratigraphy of Northwest Greenland. Meddelelser

Dansk Geologiske Forening 5 (17), 78 pp.

Koch, L. 1925: The geology of North Greenland. American Jour-

nal of Science, 5th Series, 9 , 271-285.

Koch, L. 1926: A new fault zone in Northwest Greenland. Ameri-

can Journal of Science, 5th Series, 12 , 301-310.

Koch, L. 1928a: Physiography of North Greenland. In: Vahl, M. et

al. (eds): Greenland 1 , The discovery of Greenland, explora-
tion and nature of the country, 491-518. Copenhagen: C.A.
Reitzel.

Koch, L. 1928b: Contributions to the glaciology of North Green-

land. Meddelelser om Grønland 65 (15), 183-464.

Koch, L. 1929: Stratigraphy of Greenland. Meddelelser om

Grønland 73 (2/2), 205-320.

Kollmeyer, R.C. 1980: West Greenland outlet glaciers: an inven-

tory of the major iceberg producers. Cold Regions Science
and Technology 1 , 175-181.

Krebs, J.D., Thomassen, B. & Dawes, P.R. 2003: A Landsat study

of the Pituffik region, North-West Greenland. With a sum-
mary of mineral occurrences and potential. Danmarks og
Grønlands Geologiske Undersøgelse 2003/92 , 37 pp.

Krinsley, D.[B.] 1954: Surficial geology of the greater Thule area.

In: Final report. The scientific program (Program B), Opera-
tion Ice Cap 1953, Department of the Army Project number 9-
98-07-002, 437-439. Stanford, California: Stanford Research
Institute, U.S. Army, Fort Eustis, Virginia.

Kurtz, V. & Wales, D.B. 1951: Geology of the Thule area, Green-

land. Proceedings of Oklahoma Academy of Science 31 , 83-
92 (for 1950).

Langway, C.C. 1967: Stratigraphic analysis of a deep ice core

from Greenland. Research Report Cold Regions Research and
Engineering Laboratory 77 , 130 pp.

Larsen, O. & Dawes, P.R. 1974: K/Ar and Rb/Sr age determinations

on Precambrian crystalline rocks in the Inglefield Land - Ingle-
field Bredning region, Thule district, western North Green-
land. Rapport Grønlands Geologiske Undersøgelse 66 , 4-8.

LeCheminant, A.N. & Heaman, L.M. 1991: U-Pb ages for the 1.27

Ga MacKenzie igneous events, Canada: support for the plume
initiation model. GAC/AGC, MAC/AMS, SEG meeting, Toronto.
Program with abstract 16 , A73 only.

Lind, M., Tukiainen, T. & Thomassen, B. 1994: GREENMIN -

Database system for the registration of Greenland mineral
occurrences. Rapport Grønlands Geologiske Undersøgelse
160 , 32-36.

Ljungdahl, M. 2004: Kvarts-familien i Grønland. Ujarak 2004/4 ,

12-16. Sisimiut, Grønland: Grønlands Stenklub.

Low, A.P. 1906: Report on the Dominion Government Expedi-

tion to Hudson Bay and the Arctic Islands on board D.G.S.
Neptune 1903-1904, 355 pp. Ottawa: Government Printing
Bureau.

[M'Culloch, J.] 1819: Appendix No. III, Geological memoranda.

In: Ross, J.: A voyage of discovery, made under the Orders

of the Admiralty, in His Majesty's Ships Isabella and Alexan-
der, for the purpose of exploring Baffin's Bay, and inquiring
into the probability of a North-West Passage, LXVIII-LXXXIX.
London: John Murray.

Marmont, S. 1987: Unconformity-type uranium deposits. Ore

deposit models 13. Geoscience Canada 14 , 219-229.

Mikkelsen, P.S. 2006: Twin Otter. Flyvning og rejser i Grønland,

329 pp. Copenhagen, Denmark: Aschehoug.

Mock, S.J. 1966: Fluctuations of the terminus of the Harald Moltke

Bræ, Greenland. Journal of Glaciology 6 (45), 369-373.

Monahan, D. & Johnson, G.D. 1982: Physiography of Nares Strait:

importance to the origin of the Wegener Fault. In: Dawes, P.R.
& Kerr, J.W. (eds): Nares Strait and the drift of Greenland: a
conflict in plate tectonics. Meddelelser om Grønland Geo-
science 8 , 53-64.

Munck, S. 1941: Geological observations from the Thule

District in the summer of 1936. Meddelelser om Grønland
124 (4), 38 pp.

Myers, J.S. 1985: Stratigraphy and structure of the Fiskenæsset

Complex, southern West Greenland. Bulletin Grønlands Geo-
logiske Undersøgelse 150 , 72 pp.

Nathorst, A.G. 1884: Den Svenska expeditionen till Grønland år

1883. Ymer 4 (1), 15-38.

Newman, P.H. 1982: Marine geophysical study of southern Nares

Strait. In: Dawes, P.R. & Kerr, J.W (eds): Nares Strait and the
drift of Greenland: a conflict in plate tectonics. Meddelelser
om Grønland Geoscience 8 , 255-260.

Nichols, R.L. 1953: Geomorphologic observations at Thule, Green-

land and Resolute Bay, Cornwallis Island, N.W.T. American
Journal of Science, 251 (4), 268-272.

Nichols, R.L. 1969: Geomorphology of Inglefield Land, North

Greenland. Meddelelser om Grønland 188 (1), 109 pp.

Nielsen, T.F.D. 1987: Mafic dyke swarms in Greenland: a review.

In: Halls, H.C. & Fahrig, W.F. (eds): Mafic dyke swarms. Geo-
logical Association of Canada Special Paper 34 , 349-360.

Nielsen, T.F.D. 1990: Melville Bugt dyke swarm: a major 1645 Ma

alkaline magmatic event in West Greenland. In: Parker, A.J.,
Rickwood, P.C. & Tucker, D.H. (eds): Mafic dykes and em-
placement mechanisms, 497-505. Rotterdam: A.A. Balkema.

Nobles, L.H. 1960: Glaciological investigations, Nunatarssuaq ice

ramp, northwestern Greenland. Technical Report Cold Re-
gions Research and Engineering Laboratory 66 , 57 pp.

Nutman, A.[P.] 1979: Field work on Precambrian anorthosite and

contact rocks. Smithson Bjerge, Inglefield Bredning, North-
West Greenland. Rapport Grønlands Geologiske Undersø-
gelse 95 , 22-26.

Nutman, A.P. 1984: Precambrian gneisses and intrusive anorth-

osite

Smithson Bjerge, Thule district, North-West Greenland.

Rapport Grønlands Geologiske Undersøgelse 119 , 31 pp.

Nutman, A.P., Dawes, P.R. & Kalsbeek, F. 2004: Palaeoarchaean,

Neoarchaean and Palaeoproterozoic complexes in northern
Greenland: a new terrane constellation for the High Arctic.
Manuscript to be submitted to Precambrian Research.

Oakey, G.N. 2005: Cenozoic evolution and lithosphere dynam-

ics of the Baffin Bay - Nares Strait region of Arctic Canada
and Greenland, 233 pp. Unpublished Ph.D. thesis, Vrije Uni-
versity, Amsterdam. Netherlands.

Okulitch, A.V. 1988: Proposals for time classification and correla-

tion of Precambrian rocks and events in Canada and adjacent
areas of the Canadian Shield. Part 3: a Precambrian time chart
for the Geological Atlas of Canada. Geological Survey of
Canada Paper 87-23 , 20 pp.

Okulitch, A.V. 1991: Geology of the Canadian Arctic Archipelago,

Northwest Territories and North Greenland, 1:2 000 000. In:
Trettin, H.P. (ed.) 1991: Geology of the Innuitian Orogen and
Arctic Platform of Canada and Greenland. Geology of Canada
3 , Figure 2. Ottawa: Geological Survey of Canada (also The
geology of North America E . Geological Society of America).

Olsen, H.K. 2002: Ujarassiorit 2001. The mineral hunt in Green-

land, 29 pp. Unpublished report, Greenland Resources A/S,
Nuuk, Greenland (in archives of Geological Survey of Den-
mark and Greenland, GEUS Report File 21807 ).

Olsen, H.K. 2004: Kortlægning af fedtstenforekomster i Grønland.

Unpublished report, 65 pp. + 10 maps + 3 appendices. Green-
land Resources A/S, Nuuk, Greenland (in archives of Geolo-
gical Survey of Denmark and Greenland, GEUS Report File
21869 ).

Olson, R.A. 1984: Genesis of paleokarst and strata-bound zinc-

lead sulfide deposits in a Proterozoic dolostone, northern
Baffin Island, Canada. Economic Geology 79 , 1056-1103.

Paterson, T.T. 1951: Physiographic studies in North West Green-

land. Meddelelser om Grønland 151 (4), 60 pp.

Peary, R.E. 1898: Northward over the 'Great Ice'. A narrative of

life and work along the shores and upon the interior ice-cap
of Northern Greenland in the years 1886 and 1891-1897. 1 ,
521 pp.; 2 , 618 pp. New York: Frederick A. Stokes Company.

Pirajno, F. 1992: Hydrothermal mineral deposits. Principles and

fundamental concepts for the exploration geologist, 709 pp.
Berlin, New York, London: Springer-Verlag.

Ponnamperuma, C. 1978: Greenland summer expedition, sum-

mer 1978. In: Annual report, Laboratory of Chemical Evolu-
tion, 117-135. Unpublished report, University of Maryland,
Maryland, USA.

Rausch, D.O. 1958: Ice tunnel Tuto area, Greenland 1956. Snow,

Ice and Permafrost Research Establishment Technical Report
44 , 34 pp.

Reid, I. & Jackson, H.R. 1997: Crustal structure of northern Baffin

Bay: seismic refraction results and tectonic implications. Jour-
nal of Geophysical Research 102 (B1), 523-542.

Roethlisberger, H. 1959: Seismic survey 1957, Thule area, Green-

land. Technical Report Cold Regions Research and Engineer-
ing Laboratory 64 , 13 pp.

Roethlisberger, H. 1961: The applicability of seismic refraction

soundings in permafrost near Thule, Greenland. Technical
Report Cold Regions Research and Engineering Laboratory
81 , 19 pp.

Salisbury, R.D. 1895: The Greenland Expedition of 1895. Journal

of Geology 3 (7), 875-902.

Salisbury, R.D. 1896: Salient points concerning the glacial geo-

logy of North Greenland. Journal of Geology 4 (7), 769-810.

Samuelsson, J., Dawes, P.R. & Vidal, G. 1999: Organic-walled

microfossils from the Proterozoic Thule Supergroup, North-
west Greenland. Precambrian Research 96 (1), 1-23.

Schønwandt, H.K. & Dawes, P.R. 1993: An overview of Green-

land's mineral exploration potential. Rapport Grønlands Geo-
logiske Undersøgelse 159 , 10-16.

Schytt, V. 1955: Glaciological investigations in the Thule Ramp

area. Snow, Ice and Permafrost Research Establishment Tech-
nical Report 28 , 88 pp.

Schytt, V. 1956: Lateral drainage channels along the northern

side of Moltkes Gletscher. Geografiska Annaler 38 (1), 64-77.

Secher, K., Nielsen, B.L. & Knudsen, N.Ø. 1981: Grønlands

smykkesten. Tidsskriftet Grønland 1981/4-5 , 105-152.
Charlottenlund, Denmark: Det grønlandske Selskab.

Sherlock, R.L., Lee, J.K.W. & Cousens, B.L. 2004: Geologic and

geochronologic constraints on the timing of mineralization at
the Nanisivik zinc-lead Mississippi Valley-type deposit, north-
ern Baffin Island, Nunavut, Canada. Economic Geology 99 ,
279-293.

Silis, I. 1968: Fedtsten i Thule. En usædvanlig rejseoplevelse med

Thule-eskimoerne i 1965-67. Tidsskriftet Grønland 1968/2 ,
50-54. Charlottenlund, Denmark: Det grønlandske Selskab.

Steenfelt, A. 2002: Geochemistry of southern Steensby Land,

North-West Greenland. Danmarks og Grønlands Geologiske
Undersøgelse Rapport 2002/56 , 25 pp.

Steenfelt, A., Dawes, P.R., Krebs, J.D., Moberg, E. & Thomassen,

B. 2002: Geochemical mapping of the Qaanaaq region, 77°10

to 78°10

N, North-West Greenland. Danmarks og Grønlands

Geologiske Undersøgelse Rapport 2002/65 , 29 pp. + 48 maps.

Strother, P.K., Knoll, A.H. & Barghoorn, E.S. 1983: Micro-organ-

isms from the Late Precambrian Narssârssuk Formation, north-
western Greenland. Palaeontology 26 (1), 1-32.

Stuart-Smith, J.H. & Campbell, D.L. 1971: The geology of Green-

land north of latitude 74°30

N. Report No. 2, 2 . Mineral pros-

pects of northern Greenland, 62 pp. + 3 map folios. Unpub-
lished report, J.C. Sproule and Associates Ltd, Calgary, Canada
for Greenarctic Consortium (in archives of Geological Survey
of Denmark and Greenland, GEUS Report File 20811 ).

Stuart-Smith, J.H. & Sproule, J.C. 1970: The geology of Green-

land north of latitude 74°30

N. Report No. 1, Field season

1969, 1 , 65 pp. Unpublished report, J.C. Sproule and Associ-
ates Ltd, Calgary, Canada for Greenarctic Consortium (in ar-
chives of Geological Survey of Denmark and Greenland, GEUS
Report File 20802 ).

Suess, H.E. 1954: U.S. Geological Survey radiocarbon dates I.

Science, New series 120 (3117), 467-473.

Sutherland, P.C. 1853a: On the geological and glacial phænomena

of the coasts of Davis' Strait and Baffin's Bay. Quarterly Jour-
nal of the Geological Society of London 9 , 296-312.

Sutherland, P.C. 1853b: Appendix II. A few remarks on the physical

geography, & c., of Davis Strait, and its east and west shores.
In: Inglefield, E.A.: A summer search for Sir John Franklin;
with a peep into the Polar Basin, 145-192. London: Thomas
Harrison.

Sutherland, R.A. & Dumka, D. 1995: Geology of the Nanisivik

mine, N.W.T., Canada. In: Misra, K.C. (ed.): Carbonate-hosted
lead-zinc-fluorite-barite deposits of North America. Society of
Economic Geologists, Guidebook Series 22 , 4-12.

Swinzow, G.K. 1962: Investigation of shear zones in the ice sheet

margin, Thule area, Greenland. Journal of Glaciology 4 (32),
215-229.

Thomassen, B. & Krebs, J.D. 2004: Mineral exploration of se-

lected targets in the Qaanaaq region, North-West Greenland:
follow-up on Qaanaaq 2001 . Danmarks og Grønlands Geo-
logiske Undersøgelse Rapport 2004/42 , 64 pp.

Thomassen, B., Kyed, J., Steenfelt, A. & Tukiainen, T. 1999a:

Upernavik 98: reconnaissance mineral exploration in North-
West Greenland. Geology of Greenland Survey Bulletin 183 ,
39-45.

Thomassen, B., Kyed, J. & Tukiainen, T. 1999b: Upernavik 98:

mineral exploration in the Upernavik - Kap Seddon region,
North-West Greenland. Danmarks og Grønlands Geologiske
Undersøgelse Rapport 1999/35 , 72 pp. + map.

Thomassen, B., Dawes, P.R, Steenfelt, A. & Krebs, J.D. 2002a:

Qaanaaq 2001 : mineral exploration in North-West Greenland.
Geology of Greenland Survey Bulletin 191 , 133-143.

Thomassen, B., Krebs, J.D. & Dawes, P.R. 2002b: Qaanaaq 2001 :

mineral exploration in the Olrik Fjord - Kap Alexander region,
North-West Greenland. Danmarks og Grønlands Geologiske
Undersøgelse Rapport 2002/86 , 72 pp. + map.

Thorning, L., Christensen, L.A., Lind, M., Stendal, H. & Tukiainen,

T. 2000: GREENMIN. Introduction and users' manual.
Danmarks og Grønlands Geologiske Undersøgelse Rapport
2000/5 , 87 pp.

Ujarassiorit 1993: Ujarassiorit 1992. Årsrapport og opfølgning, 6

pp. + appendices. Unpublished report, Nunaoil A/S, Nuuk,
Greenland (in archives of Geological Survey of Denmark and
Greenland, GEUS Report File 21543 ).

Ujarassiorit 1995: Ujarassiorit 1994. Samlet rapport, 73 pp. + plates.

Unpublished report, Nunaoil A/S, Nuuk, Greenland (in ar-
chives of Geological Survey of Denmark and Greenland, GEUS
Report File 21113 ).

U.S. Army 1954: Final report. The scientific program (Program

B), Operation Ice Cap 1953, Department of the Army Project
number 9-98-07-002, 472 pp + 3 maps. Stanford, California:
Stanford Research Institute, U.S. Army, Fort Eustis, Virginia.

Vaughan, R. 1991: Northwest Greenland: a history, 208 pp. Orono,

Maine, USA.: The University of Maine Press.

Vidal, G. & Dawes, P.R. 1980: Acritarchs from the Proterozoic

Thule Group, North-West Greenland. Rapport Grønlands Geo-
logiske Undersøgelse 100 , 24-29.

Washburn, A.L. 1956: Unusual patterned ground in Greenland.

Bulletin of the Geological Society of America 67 , 807-810.

Weidick, A. 1976: C

dating of Survey material carried out in

1975. Rapport Grønlands Geologiske Undersøgelse 80 , 136-
144.

Weidick, A. 1978a: C

dating of Survey material carried out in

1977. Rapport Grønlands Geologiske Undersøgelse 90 , 119-
124.

Weidick, A. 1978b: Comments on radiocarbon dates from north-

ern Greenland made in 1977. Rapport Grønlands Geologiske
Undersøgelse 90 , 124-128.

Weidick, A. 1995: Satellite image atlas of glaciers of the world.

Greenland. United States Geological Survey Professional Pa-
per 1386-C , 141 pp. + map.

White, S.E. 1956: Glaciological studies of two outlet glaciers,

Northwest Greenland, 1953. Meddelelser om Grønland 137 (8),
31 pp.

Whittaker, R.C., Hamann, N.E. & Pulvertaft, T.C.R. 1997: A new

frontier province offshore Northwest Greenland: structure,
basin development, and petroleum potential of the Melville
Bay area. American Association of Petroleum Geologists Bul-
letin 81 (6), 978-998.

Wilson, I.D.H. & Underhill, D.H. 1971: The discovery and geo-

logy of major new iron deposits on Melville Peninsula, East-
ern Arctic. Canadian Mining Journal 92 , 40-48.

Wordie, J.M. 1938: An expedition to North West Greenland and

the Canadian Arctic in 1937. Geographical Journal 92 (5), 385-
420.

Wright, J.W. 1939: Contributions to the glaciology of North-West

Greenland. Meddelelser om Grønland 125 (3), 43 pp.

Young, G.M. 1979: Correlation of middle and upper Proterozoic

strata of the northern rim of the North Atlantic craton. Trans-
actions of the Royal Society of Edinburgh 70 , 323-336.

Appendix 1

Spelling of place names: old and new orthography

New orthography

used in

Old orthography used

these explanatory notes

on the map sheet

Aafeerneq

Aorfêrneq

Ajukus Skær

Ajakos Skær

Anngiusalipaluk

Angiussalipaluk

Asungaaq

Asungâq

Inersussat

Inerssússat

Innaaqqissorsuq

Ivnârqigsorssuaq

Isussik

Isuvssik

Itilleq

Itivdleq

Itillersuaq

Itivdlerssuaq

Kangerlussuaq

Kangerdlugssuaq

Kinginneq

Kingingneq

Moriusaq

Moriussaq

Naajat

Naujat

Narsaarsuk

Narssârssuk

Nallortoq

Navdlortoq

Niaqornarsuaq

Niaqornarssuaq

Niaqornaarsuk

Niaqornârssuk

Nunapalussuaq

Nunapalugssuaq

Nunatarsuaq

Nunatarssuaq

Nunngarutipaluk

Núngarutipaluk

Nuulliit

Nûgdlît

Nuussuaq

Nûgssuaq

Pingorsuit

Pingorssuit

Pitoraavik

Pitorâvik

Pituffik

Pitugfik

Piuffik Gletscher

Pitugfik Gletscher

Piulip Nunaa

Piulip nunâ

Puisilik

Puissilik

Puisiluusarsuaq

Puissitdlûssarssuaq

Qangattaat

Qangâtait

Qaqujaarsuaq

Qaqujârssuaq

Qattarsuit

Qátarssuit

Qeqertarsuaq

Qeqertarssuaq

Quaraatit Nuna

Quarautit nûa

Quinnisut

Quínissut

Savissivik

Savigsivik

Savissuaq Gletscher

Savigssuaq Gletshcer

Sermiarsupaluk

Sermiarssupaluk

Sineriarsua

Sineriarssua

Sioqqap Kuua

Siorqap kûa

Sukkat

Súkat

Tikeraasaq

Tikeraussaq

Tuttu Gletscher

Tugto Gletscher

Tuttulipaluk

Tugtulipaluk

Tuttulissuup Tuttoqarfia

Tugtuligssûp tugtogarfia

Ulli

Uvdle

Umiivik

Umivik

Ummannaaq

Umànaq

Geological Survey of Denmark and Greenland Map Series

Vol. 2 pp 52-97, Explanatory notes to the Geological map of Greenland, 1:500000, Thule, Sheet 5

www.geus.dk > Publications > Geological Survey of Denmark and Greenland Map Series > Volume 2 > This page