Greinar (Vísindafélag Íslendinga) - 01.01.1977, Side 90

88
magma mass produced in the individual eruptions, was at a
relatively shallow depth, at least a short time before the
eruptions started (for Iceland 25—30 km; the focus of the
quake at the beginning of the Heimaey eruption in question
was 27 km deep (70)). Only at that depth, and during a certain
incubation time, could the vertical order of chemistry have
taken place, if it was to remain as regular chemistry-time order
during the eruption. Here must be a mechanism which was
capable of creating the vertical order of chemistry. Hence, this
diagram is no evidence for an origin of the magma by partial
melting deeper in the mantle.
Concerning the diagram of Fig. 12, one thing is certain: All
the magmas on which that diagram is based, were produced
in individual eruptions. The differentiation shown, could there-
fore have taken place during incubation at the shallow depth,
and is then no evidence for a much deeper origin of the magma.
There can be no greater mystery about this differentiation
than about that between the main eruptions in Hekla, where
we know the time interval and the extent of differentiation, and
that it must take place in a reservoir. (Cf. also (64a)).
Let us now turn to the increase of differentiation towards
Kverkfjöll, as shown in Fig. 12. Fig. 14 is a smoothed-out map
of the Iceland topographic mound, as explained in the caption.
It is a copy of the map, the present author gave in his paper
on gravity in Iceland (71). This smoothing of topography was
done to ease isostatic reduction for Iceland. Clearly, this
would be a very good approximation to the topography just
for that aim. We applied this map in particular for a Heis-
kanen normal crustal thickness of 30 km (crustal density
2.67 g/cm3, mantle density 3.27 g/cm3), using Heiskanen’s
model and his reduction tables. For a region of height H, we
obtained a root thickness D=5.61XH, so that for H=1 km,
we have D=5.61 km. Kverkfjöll is that chemically studied
area, which is closest to the top of the smoothed Iceland mound,
i.e. closest to the greatest value of D. It follows, that if we
assumed everywhere reservoirs, reaching up to the 30 km depth,