Jökull - 01.12.1974, Qupperneq 48

lava flows near Kalmanstunga. A system of SE-
wardly dipping basaltic sheets in Selgil points
to a center in that direction as does the
SE-wardly dipping sheetlike basalt intrusion of
Nónborg, southwest of Kalmanstunga. A few
eastwardly inclined thin sheets and dykes occur
also in Tunga. In both areas there are also a
large number of irregularly orientated dykes
which do not conform to the system of in-
clined sheets. These intrusions are related to
the second acid phase which is evident because
they are of normal polarity and they do not
extend into rocks erupted during the following
reversed polarity interval.
Rhyolite lavas continued to be erupted after
emplacement of the ignimbrite. Two funnel-
shajred plugs (Raudakúla and Fálkaklettur) and
two flows are exposed in the summit of Tunga
and a third to the southwest of Húsafell direct-
ly overlying the ignimbrite.
At this stage the seconcl acid phase began to
decline and the area was inundated by basalts
which will be discussed in the next section. Two
examples are known of rhyolite flows which
were erupted after basaltic volcanism had be-
come dominant. The first is exposed in Selgil
and Selfjall and is connectecl to its feeder dyke
in the southern wall of Selgil. From the feeder
it flowed downslope towards the east and grows
thicker in that direction. Tlie second flow,
which is also the latest belonging to the second
acid phase, forms the highest part of Tunga.
At the base of this flow are found a tillite, a
basaltic hyaloclastite, and at one locality com-
pound type olivine tholeiites. The rhyolite flow
itself is unique in being made up of a very
tliick columnar pitclistone carapace at its top
and margins. We believe that it was extruded
under subglacial conditions, perhaps during the
6th glacial event, as will be discussed below.
2.2.4 The succession betioeen the second and
third acid phases
The second ancl third acid phases are found
to be separated primarily by compound type
olivine tholeiites and liyaloclastites of olivine
tholeiite composition but also three horizons
indicating glaciations. The thickness of tliis
group is about 100—150 m, but more tlian that
where hyaloclastites are present. It thins to al-
most nothing against the slopes of the central
46 JÖKULL 24. ÁR
hillocks of the second acid phase in Tunga
and Bæjarfell. Tlioleiites are important locally,
especially in Strútur wliere they make up the
entire group of which only the upper part is ex-
posed, however. At Kalmanstunga the tholeiites
are probably in fault contact against rhyolites
of the second acid phase. In Fljótstunguháls
tholeiites interdigitate with rhyolites of the
second acid phase giving way in the upper part
of tlie group to olivine tholeiite lavas and
liyaloclastites. Residual dips of the tlioleiites in
Fljótstunguliáls and south west of Húsafell in-
dicate original dips away from the core area
of the volcano.
The best sections of the group are exposed
in the soutliern part of the area between Sel-
fjall and Valagil. The lowest glacial horizon
(no. 5 of Plate 1) is found near the base of tlie
group separated from the Deildargil ignimbrite
by a few tholeiite flows. In Deildargil striated
boulders are present in the tillite and a striat-
ed basement was seen in Hringsgil further east.
The upper part of this layer is conglomeritic
and on approaching the Bæjarfell hyaloclastite
pile, debris fans derived from that source make
up most of the layer. It abounds also with acid
breccia fragments derived probably from tlie
newly erupted ignimbrite. Tliis glacial liorizon
is covered by up to 150 m thick olivine tholeiite
lavas showing the characteristic flow unit struc-
ture typical of lava shields. The flows increase
in number towards west. In the slopes above
Húsafell farm and in Deildargil only two or
three flows make up the entire section, but in
Asgil 9 flows are present. A glacial horizon (no.
6 of Plate 1) is intercalated with the olivine
tholeiites. Like the first glacial horizon of this
group it varies in constitution along the strike.
Near Bæjargil it is again composed of debris
from the nearby hills. In Hringsgil it is an un-
sorted tillite with striated boulders and a gray
silty matrix. In Asgil it is represented by a til-
lite at the base which is overlain by a 15 m
thick fluvial sandstone and conglomerate of
primarily liyaloclastite origin.
Neither of these two tillite horizons were
directly seen to interfinger with volcanic hyalo-
clastites, but it seems likely that a hyaloclastite
pile exposed in tlie east of Fljótstunguliáls
occurs at the same stratigraphic level as the
upper liorizon. Piobably contemporaneous witli