Greinar (Vísindafélag Íslendinga) - 01.01.1977, Blaðsíða 58

56
on a SW-NE fracture only 3 km east of the andesitic Hekla,
and a new very heavy series of earthquakes took place in the
“strong” lowland area, mentioned above, in 1896. The foci
shifted repeatedly from one place to the other over the whole
area.
In 1912 there were strong earthquakes, apparently of the
same category, in a N-S elongated area, but the fractures form-
ed, show clearly the causative SW-NE shear stress (44).
In 1913 there was a new small basaltic eruption of the form-
erly mentioned fracture just east of Hekla. The two mentioned
basaltic eruptions east of the Hekla fracture fit chemically well
with a voluminous prehistoric lava flow from the Hekla region
(Gunnarsholtshraun) (about 6000 years old), and all fit very
well, as mother magma, with the variation diagram of the
andesitic Hekla lavas proper (61).
The just mentioned facts are most significant from two
points of view. The one is the formation of a volcanic fracture
so close to but independent of the great Hekla fracture. The
other point is the fact that so close to Hekla basaltic magma
is formed from a mother material which has remained un-
differentiated for thousands of years while in the eruptive
system of Hekla differentiation is very marked within less
than a century. This fact will be discussed in connection with
the problem of magma formation in Chapter 6.
On the average, the intervals between Hekla eruption since
1104 (06), have been about 50 years. After the 1845-eruption
the next one was due around 1900. By 1918 it was overdue
by about 18 years, no unusual figure, when rather suddenly
a marked amelioration of the climate began, and the actual
dormancy became 102 years, the longest on record.
Katla tells a similar story. Since early historic time it has
erupted twice in a century, and for the last 4 centuries there
has been a strict rule, the alternation of a long and a short
interval. The longer is around 60 years, but rather fluctuat-
ing in length. The short ones have, on the other hand, been
between the very narrow limits of 35-37 years. The long inter-
val between 1860 and 1918 was 58 years. And now the next