Acta naturalia Islandica - 01.02.1946, Qupperneq 20

14
TRAUSTI EINARSSON
eruption of the sub-glacial volcano Grímsvötn in Vatnajökull in 1934.
It was thought that this eruption might throw some light on the
formation of the Palagonite tuffs and expeditions were undertaken,
among others by Dr. N. Nielsen.
No conclusive results as to the formation of “palagonite breccia”
were obtained on that occasion. But two years later it was assumed by
Nielsen and Noe-Nygaard that no material resembling the “pala-
gonite breccias” or tuffs was formed at the eruption. Nevertheless,
these authors accept Peacock’s view, with some specifications: “Pala-
gonite breccia”, in their opinion, is not formed by violent explo-
sions of the Grímsvötn type, but more probably by “slow” eruptions
beneath the ice-sheet through which they are unable to break.
This new theory is, however, not easily understood. If, as the
authors believe, sideromelan is formed by ultra-rapid cooling, why
should it be more readily formed by a slow eruption than a violent
one ? And as to the arguments for this new version of the theory of
sub-glacial origin of sideromelan a “slow” sub-glacial eruption will
always be difficult to establish and up to the present time no such
eruptions are known with certainty. But these specifications of the
theory are also unnecessary as it was shown by Tom F. W. Barth13)
that the ash thrown out by the violent eruptions of Grímsvötn in 1934
and in 1922 did consist of porphyritic sideromelan. And this is actually
the sole argument for the subglacial origin of the Basic tuffs which
was derived from these eruptions of Grímsvötn. In the last chapter
I shall consider these eruptions more closely.
Nielsen and Noe-Nygaard find that field studies in the Palagonite
Formation of Southern Iceland are suggestive of the (still hypo-
thetical) slow subglacial eruptions, but as yet only a preliminary
report of their work is available.14) In this report a general theory
of the Palagonite Formation is put forward. According to this theory
the formation is the result of glacial and inter-glacial volcanism,
i. e. pyroclastic material and lavas, and the work of the different
agents of denudation during the varying conditions of the multiple
Quaternary Ice Age.
I shall now turn to my own observations. It will be seen in the course
of the discussions that the glacial theory of the formation of sidero-
melan is incapable of explaining a number of observations and
a new theory will be framed which appears to be directly suggested
by the observations.