Greinar (Vísindafélag Íslendinga) - 01.01.1977, Blaðsíða 103

101
tained the argon. This indicates that the magma is in the nano-
crystalline state — without the twinning-like bonds — on its
way to the surface. But at the latest, during macroscopic
crystallization the nanocrystalline structure is lost.
A direct test of our inference concerning the deep sea pillows
could be of great interest.
When two conjugate fractures cross each other, infinitesimal
gliding on one of them, automatically implies gliding on the
other. The produced magma is thrust towards the crossing
point, and so is led to form a shieldvolcano.
As to linear eruptions the early rise of magma into open-
ed fracture corresponds to lubrication, and the shear stress
then leads to such a sudden jump, that forms echelon fractures
in the brittle upper layers, and provides the easiest ways for
the rising magma. Only in special cases (e.g. Þrengslaborgir),
when the lava flows to one side away from the orifices, it is
possible to see, that the lava came indeed up along echelon
fractures. Also during the Surtsey eruption of 1963—67 the
shift beween echelon fractures was evident.
The time of magma formation found for Etna and Strom-
boli, is here understood to mean, that the stress is of such
strength that the expansion in the reservoirs is capable of
opening the fracture overhead after 30—40 years. This time
must depend on the rapidity of magma formation, which will
mean the rate of frictional energy production.
We could well imagine such a weak stress that it just suf-
fices to produce a thin magma reservoir which may not deve-
lop further, because the produced energy is lost by conduction
and radiation. If then the stress becomes a little stronger the
full development to an eruption may need only a short time.
This case may correspond to the rapid succession of small-
scale eruptions in Iceland after 1960. By a very strong stress
field, the energy production is rapid and the magma produc-
tion and expansive force may develop in a short time, The
magma formation time, or “incubation”-time must therefore
be very different. It might be so short that the olivine has
not had time to sink to the bottom, in which case we would