Rit (Vísindafélag Íslendinga) - 01.06.1984, Blaðsíða 220

216
PÁU. IMSI-AND
Or rich alkali íeldspars. The plagioclase phenocrysts primarily occur as
cores in more Or rich feldspar marginal zones. These two feldspars have
thus generally not crystallized simultaneously. Depending on the composi-
tion of the salic magmas, either feldspar may be the first phase to crystallize.
In the Jan Mayen case the plagioclase appears to be íirst in most cases. How
far, regarding temperature diíference, the plagioclase crystallization pre-
ceeds that of the alkali feldspars is not known. In melting experiments on
syenites of the Kungnat complex, southwestern Greenland, which do not
differ much from the Jan Mayen evolved rocks in composition, McDowell &
Wyllie (1971) found alkali feldspar and plagioclase to coexist through a
temperature interval of 120— 140°C.
In Table 34 the results of 1 atm melting experiments on rocks of tristanite
and trachyte compositions are shown. These experiments show that where
plagioclase is the íirst phase it starts to crystallize in the temperature range
of 1130— 1195°C, while where alkali feldspar is the first phase, its crystalliza-
tion starts in the temperature interval of 995—1025°C. If these rocks are
divided into more- and less evolved groups the table further shows that: 1) of
the rocks crystallizing plagioclase first, the less evolved ones do so at higher
temperatures than the more evolved ones, and 2) of the more evolved rocks,
those crystallizing alkali feldspar first are richer in total alkalies than those
crystallizing plagioclase first. This pattern is to be expected and is in
agreement with the general traditional conclusion, that with increasing
evolution of the rocks the role of alkali feldspars increases and the cryst-
allization temperatures fall.
Where plagioclase is the liquidus phase (Table 34) its crystallization
temperature does not differ much from those of the basalts. The early
plagioclases of the evolved rocks of Jan Mayen might thus reflect a similar
history to those of the basalts; i.e. crystallization at depths, upward move-
ments and resorption and finally enclosure in an alkali feldspar envelope.
This would especially be the case in the basaltic tristanites, tristanites and
some of the less evolved trachytes. The more evolved trachytes, on the other
hand, probably did crystallize an alkali feldspar first. Carmichael (1965), in
his article on trachytes and their feldspar phenocrysts, discusses at length
the crystallization relations of the feldspars of a Jan Mayen trachyte (4R).
This trachyte is identical to Jan 88, the most evolved trachyte analysed here,
and the ones described by Carstens (1961). Carmichael concludes that this
sample first crystallized a potassic alkali feldspar which became more sodic
and enriched in anorthite with continued equilibrium cooling and fmally
intersected the two-feldspar surface and precipitated a plagioclase along
with the sanidine. The reacdon of these feldspars with the liquid may have
resulted in resorption of the plagioclase. Such a reaction may result in total
resorption of plagioclase under favourable conditions. Such a resorption
might apparently explain the crystallization temperature difference of
trachytes and syenites of identical compositions.