Rit (Vísindafélag Íslendinga) - 01.06.1984, Side 291

PETROGENETIC RELATIONSHIPS
287
VI. Some general features of the crystal
fractionation calculations
The crystal fractionation described and discussed above and quantita-
tively accounted for in the tables is a selected case in that the rocks used in
all instances and the results obtained are those giving the best results of a
number of possibilities. If other “mothers and daughters” are used, the
results obtained are either worse or the petrological restrictions are more
poorly fulfilled. A simplified generalized scheme of the crystal fractionation,
based on the calculations as presented and discussed above, is given in
Table 51, which shows the amount and constitution of the fractionate and
the amount of liquids produced.
If the initial liquid is of the composition ofjan 166, the fractionation gives
~40 wt. per cent liquid of the Jan 26 composition, i.e. an ankaramitic basalt
liquid. If this liquid is then fractionated further towards the low-Mg basalts,
~20 wt. per cent of the original material is left as a basaltic liquid of low Mg
content, and so on. The basaltic tristanite liquid would thus amount to ~15
per cent and the tristanite liquid to ~5 per cent in the case of Nord-Jan and
~8 per cent in the case of Sör-Jan. Finally this would give ~1.5 per cent
liquid of trachytic composition in case of Nord-Jan but ~4 per cent in case
of Sör-Jan. If a slightly more magnesian or primitive liquid than Jan 166 is
present the final highly evolved liquids produced would decrease somewhat
in amount.
This amount of highly evolved liquids is the result of model calculations
and gives its weight percentage of the system involved. In chapter 5, the
volume of highly evolved rocks (tristanites and trachytes) on Sör-Jan (where
the bulk of these occur) was estimated to be ~ 1 km3. This is ~3 volume per
cent of Sör-Jan and ~0.5 vol. per cent of the whole island above sea level.
The density difference between the basic and intermediate rocks is relatively
small allowing the volume and weight percentages of these rocks to be
equalled in these rough calculations. If this is done the tristanites and
trachytes erupted on the island are an order of magnitude (at least) less in
amount, than these liquids as produced by the calculations. The volume of
the basaltic tristanites on the island is not possible to estimate with any
certainty, but the liquid amount obtained through the calculations in this
case apparently far exceeds that of rocks of these compositions exposed on
the island. There is thus no agreement between the amount of exposed
evolved rocks on the island and the amount of the same rocks produced by
crystal fractionation.
As the amount of exposed evolved rocks is low compared to that of the
produced liquid, this does not argue against crystal fractionation as a
process in the case ofjan Mayen rock evolution, but it may indicate that the
unmodified evolved liquids have difficulties in reaching the surface. This
would be in accordance with the generally higher viscosity of evolved