Rit (Vísindafélag Íslendinga) - 01.06.1984, Blaðsíða 293

PETROGENETIC RELATIONSHIPS
289
magmas than primitive ones, but in disagreement with their low density.
Producing the ankaramitic basalt from Jan 166 by crystal fractionation
leaves a 60 per cent fractionate of wehrlite, composed of 75 per cent
chromian diopside and 25 per cent Fogo olivine. This wehrlite is consider-
ably richer in the clinopyroxene than the wchrlite xenoliths observed. These
are, though, only small fragments and far from representing the 60 per cent
produced by the fractionation. Judged from the phenocryst relations in the
ankaramites, where the wehrlite minerals are present as well, this wehrlite
would not be expected to be a quite homogeneous rock. The clinopyroxene
dominance shown by the ankaramite and basalt fractionation is in agree-
ment with the trend shown by the Murata plot, Fig. 8.
In fractionating from the primitive basalts, produced by this wehrlite
fractionation, to the evolved basalts, i.e. within the compositional range
where the bulk of the rocks fall, 52 per cent of minerals are removed. Still
clinopyroxene dominates the fractionate, but this pyroxene is a titan-salite
of varying composition. Here plagioclase is added to the fractionate as well
as iron-titanium oxides. Fractionating further from the evoived basalts into
the slightly evolved intermediate rocks still produces a gabbroic type
fractionate, but here the plagioclase has taken over as the dominating phase
and this role is kept by the feldspars through the remaining steps of the
fractionation. Apatite enters the fractionate here. Compared to the amount
of wehrlite produced by the fractionation, this gabbroic type fractionate is
quite small.
In fractionating further into the intermediate rocks, producing the trista-
nite liquids, the plagioclase becomes sodic, the olivine becomes more
fayalitic and nearly disappears from the fractionate, which is now only a
very small part of the total system. In the íinal step of the fractionation,
where the trachytic liquids are produced, both orthoclase and biotite appear
in the fractionate, which is now of a syenitic nature and of very small
volume. Nevertheless syenite is abundantly represented in the xenolith suite
found on the island.
The equilibrium conditions between the subtracted minerals and the
„liquid“ to be fractionated in the calculations is far from being exactly
matched in most cases. The total fit obtained in the calculations is in most
cases quite good. Nevertheless, the misfit is surprisingly systematic. As the
equilibrium conditions are poorly fulfilled, and the examples studied are
fairly representative for the whole compositional range of the rock suite, one
would rather expect the misfit to affect the elements randomly than to show
a systematic pattern, if crystal fractionation was the sole process behind the
compositional differences.
The misfit may be simplified and summarized in the following way. There
are two elements that show misfit which are easily explained. These are Cr
in the ankaramites and P in the later steps of the fractionation. The cause of
their misfit is the uneven distribution of chrome-spinel and apatite in the