Rit (Vísindafélag Íslendinga) - 01.06.1984, Blaðsíða 311

PETROGENETIC RELATIONSHIPS
307
one would be trachytic in composition, as e.g. the sample Jan 77, which in
the field occurs in a nearly phenocryst free glassy state. The ne-normative
crustal magma, produced by the remelting of a high-grade amphibolized
crust, in the Jan Mayen case would be the glass found in the hydrous
mineral xenolith (Jan 329), as its remelting product. Of the rocks of Jan
Mayen this glass most resembles the basaltic tristanites. On the variation
diagrams, the rock suite forms a continous trend, from the most primitive to
the most evolved rocks. There is, however, a major change in the composi-
tion of the rock suite. This is recognized by a change of slope of the trends of
all the major elements and most trace elements as well. This change is far
too sharp and general to be accounted for by the disappearance of one or
two minerals from the crystallization sequence. It is a general break. It
appears at ~4 wt. per cent MgO and the low-Mg basalts to basaltic
tristanites are thus characteristic for the composition where this change
takes place. The basaltic tristanites are generally phenocryst poor rocks.
The phenocryst assemblage (alkali feldspar, biotite, apatite), which is built
up in the subsequent rocks of the rock suite, is quite different from that of the
preceding rocks (clinopyroxene, olivine, plagioclase). The hydrous mineral
xenolith itself is most likely the result of the fractionation process, which
took place in the crustal segment, during the continental break-up discussed
above. These three end member magmas can mix freely, as is indicated by
the mineralogy and textures of mumerous rock samples studied, but they
may reach the surface in pure or nearly pure liquid condition, probably with
the exception of the most primitive mantle magma. As a consequence of the
open nature of the magma system, the primary magmas themselves experi-
ence some diíficulty in reaching the surface in an unmodified state.
The ne-normative and the trachytic liquids produced in the crust are,
according to this modelling, the materials that modify the result of the
crystal fractionation process, which apparently is in operation during the
entire system. The fact that the gabbro xenoliths, which have a basalt
composition and are accumulations of the same minerals as occur in the
basalts as phenocrysts, are the K poorest rocks produced by the magma
system (except for the wehrlite cumulate), may be taken as proof that the
alkali enrichment of the rock suite is a crustal phenomenon and not a
character inherited from the parental mantle material.
No obvious disagreements are found between the petrological observa-
tions and the above modelling for petrogenesis. This model, in fact,
describes a magma system, open for both magma renewal and crystal
transport, contamination and mixing, resulting from a complex geological
evolution.
The compositional similarities between the Jan Mayen rocks and some
East African rocks (Goles, 1975) are more easily understood with reference
to the crustal evolution modelled here. Whether this modelling improves
understanding ofoceanic island evolution in general has not been tested, but