Jökull - 01.12.1981, Blaðsíða 6

TABLE 1
Microprobe analysis of bonding material in discontinuity from tholeiite, Iceland. Age
approximately 2 m. y. Point analysis.
TAFLA 1. Örgreining á fylliefni úr smásprungu í íslensku póleiíti. Aldur u. p. b. 2 m. á.
Punktagreining.
SiO,, TiOT- AiP;i ]+/),* MgO CaO Na20 K.P MnO Total
38.44 00.00 4.23 26.94 6.22 1.67 0.18 0.17 0.52 78.37
+ Ti usually less than 1.00%
* Totai Fe as Fe20.j
TABLE2
X-ray diffraction analyses of 3 samples of bonding material from tholeiite, Iceland.
Peaks in Angstrom (Á)
TAFLA 2. Röntgengreining á premur sýnum af fylliefni úr íslensku póleiíti. Toppar í Angstrom (Á)
Sample Untreated Ethylene glycol treated Heated to 550°C Pretreated with Mg+2 Glycol
1 15.49 15.76 9.92 14.71 15.22
2 14.46 15.22 9.99 14.62 14.96
3 15.22 15.22 9.99 14.62 14.71
culite group (H. Kristmannsdóttir, personal
communication).
In the Irish rocks smectite (saponite) can be
a constituent of the bonding material (Douglas
1972).
The chemicai analysis is of course for the
total material and it is not clear therefore to
what extent either Ca or Fe is part of the clay
mineral structure. On the other hand both the
red and biack phases are structurally the
same, according to the X-ray diffraction ana-
lyses. Figure 3 (f) is a scanning eiectron
micrograph of the black phase on a freshly
opened discontinuity and shows small irregu-
larly shaped flakes. Small microfractures can
also be discerned in the same photograph.
DISC.USSION
Origin of the disconhnuilies
The large scale open joints which form the
prism surfaces have most often been explained
as tension joints formed by thermal stresses set
up on cooling of the lava (e. g. Spry 1961).
The predominantly vertical joints have
formed perpendicular to the upper and lower
cooling surfaces and perpendicular to the
planes of equal tensile stress.
The discontinuities described in this paper
are on a much smaller scale and do not seem to
be genetically identical to the major joints.
Their relationship to petrology indicates that
they are brittle fractures originally. They seem
to be mainly the result of tensile stresses
although small scale shearing is not com-
pletely absent. The intersection angles of 90°
and 120° can be satisfactorily explained by
assuming that the fractures have formed
dynamically i. e. by extension according to
Griffith crack theory (Griffith 1920). An ex-
tending fracture runs at right angles to the
tensile stress. This explains the 90° intersec-
tions and aiso why some fractures bend to
meet others at 90° since the tensile stress is
4 JÖKULL 31.ÁR