Jökull - 01.12.1974, Blaðsíða 54

Hvítá between Refsstadir and Hraunsás. It is
probably of last interglacial age as are also
small outcrops of interglacial lavas just to the
south of Kalmanstunga and on the spur be-
tween the gorges of Lambá and Geitá. Tr. Ein-
arsson (1963) has noted the latter occurrence
and interpreted it as belonging to the Ok-lava
flow. The field relationships and textural dif-
ference rather favour a different source, how-
ever.
During the last glacial stage the tuya1) com-
plexes of the Ok area and the Langjökull
margin were formed as well as the topmost
Strútur hyaloclastite. Ok itself which is a lava
shield overlying the tuyas was formed under ice
free conditions probably during the Alleröd
warm period. It was overidden by the last
glacial advance of the last glaciation.
Finally the lava flows of Geitlandshraun
and Hallmundarhraun inundated the valley
floors of Hvítá and Nordlingafljót in Post-
glacial time, the latter as recently as 1200
years ago (SœmuncLsson, 1967). The two lavas
were erupted from central vents ancl they thus
appear to be closely related to the lava shield-
and tuya volcanism which predominated in this
part of the active volcanic zone during tlie last
glacial stage. All these youngest members are
of olivine tholeiite composition, the rocks of
Ok and Hafrafell being of a highly plagioclase
porphyritic variety.
3. MAGNETIC POLARITY ZONES AND
K/Ar AGE DETERMINATIONS
The magnetic polarity of the basalts was
established by measurements in the field using
a portable fluxgate magnetometer. This method
allowed a distinction to be made between
normally and reversely magnetized rocks. Ambi-
guities were, however, observed in the immedi-
ate vicinity of magnetic reversals, probably due
to transitions. The results are shown in Figs. 8
and 9. Reversely magnetized groups comprise 1)
1) Several terms have been suggested for subglacial equi-
valentsi of lava shields which breached the glacier and
developed a cap of subaerial lava, among them table moun-
tain (Bemmelen ir Rutten, 1955) and stapi (Kjartansson,
1967). The new AGI Glossary of Geology (1972) lists only
tuya.
the porphyritic lavas on top of the lowest tillite
at Giljafoss, 2) the first acid phase rocks and a
few flows of tholeiite below and on top of
them, 3) the third acid phase rocks and lavas
banked up against them in the east, 4) the three
lava flows at the base of Hafrafell, 5) the lava
remnant on top of Búrfell. The rest is of normal
polarity as is also a 330 m thick group of lavas
beneatli the Giljafoss tillite. Before the K/Ar
datings became available we tentatively suggest-
ed (in Schwarzbach and Noll, 1971) that the
lowest reversely magnetized group should be
correlated with the Mammoth event of the
Gauss epoch. The first acid pliase lavas woukl
then correspond to the Kaena event and the
third pliase lavas to the first reverse part of
the Matuyama.
Tlie absolute ages of 7 samples were deter-
mined at the Teledyne Laboratories in the
U.S.A. For this purpose we chose samples of
tholeiites or basaltic andesites. One dacite
sample was also determined. The results are
listed in Table 1. Some of the dated lava flows
are indistinguishable within standard analytical
error. This is especially true of no. 4, 5 and 7.
They however cluster around the Gauss epoch.
The samples derive from a part of the section
where the pattern of magnetic reversals as
determined by field measurements produced
the very distinctive pattern of the Gauss epoch.
Combination of the two sets of data, we belive
fixes the age of the mapped sequence.
Thus as a whole the samples yielded plausible
results which together with the paleomagnetic
resuts allow most of the section to be correlat-
ed with the geopolarity time scale (Fig. 8). One
sample from the lower part of the section which
is a dacite, yielded a dubious result as is
evident from its stratigraphic position above
nos. 7 and 8. Tliis sample is from almost the
same stratigraphic level as no. 5. Perhaps the
high glass content of this rock as compared to
the more crystalline basalts is responsible for
its behaviour.
The data obtained for the Bæjarfell—Kaldi-
dalur group indicate that it corresponds to the
Gilsá event. Earlier assumptions of Gauss age
(.Sœmundsson, 1967) and Jaramillo age (Piper,
1971) are thus incorrect. A sample from the
ancient valley filling at the top of Búrfell in-
dicates post-Jaramillo age at this site.
52 ■ JÖKULL 24. ÁR