Rit (Vísindafélag Íslendinga) - 01.06.1984, Qupperneq 233

mineral chemistry and relationshi ps 229
per cent) to those of the basic Jan Mayen rocks. Mn enriched apatites thus
appear to be primarily, if not solely, of late magmatic origin.
MgO (Fig. 115) was detected in all analysed apatites with a range from
0.04 to 0.54 wt. per cent and an average of 0.22 per cent. The range and the
average is similar in the intermediate rock apatites (0.04—0.39, av= 0.22
per cent) and those of basic rocks (0.08—0.54, av=0.21 per cent). When
looking at extrusive rocks versus the coarse-grained xenoliths, a more
contrasting diíference appears. The extrusive rock apatites show the whole
range reported and an average of 0.30 per cent, while the xenolith apatites
range from 0.06 to 0.30 per cent with an average of only 0.10 per cent. A
similar pattern emerges from the MgO range and averages of thc apatites
reported from elsewhere. The Skaergaard intrusion (Nash, 1976) contains
apatites with a range from 0.02 to 0.16 wt. per cent MgO and an average of
0.08 per cent. The Shonkin Sag laccolith (Nash, 1972) apatites have a range
from 0.09 to 0.33 per cent and an 0.16 per cent average. Ihe Leucite Hills
volcanic rocks (Carmichael, 1967) contain apatites ofO.Ol to 0.33 per cent
range and an 0.22 per cent average. The megacryst apatites of Australian
basalts reported by Irving (1974) contain 0.22 to 0.59 per cent with an
average of 0.36 per cent, which is similar to that of thejan Mayen apatites of
the extrusive rocks. Apatites of extrusive rocks thus seem to have a tendency
to be richer in Mg than apatites of plutonic rocks.
III. Crystallization relationships of the apatites
During the last decade a number of authors have dealt with the cryst-
allization and composition of hydroxy-halogen igneous minerals and the
behaviour of these volatile components during magmatic evolution. As
regards apatite, important papers are those of Nash (1972 & 1976), Car-
Tiichael (1967), Stormer & Carmichael (1971), Carmichael et al. (1974),
Sigvaldason & Óskarsson (1976) and Fuge (1977). trom these it may be
summarized that, due to its low solubility in silicate magmas, C1 has the
tendency to enter early crystallizing hydrous minerals or to escape in
aqueous phases and thus become depleted in the evolved magmas. I, on the
other hand, tends to be concentrated in late magmas and late crystallized
hydrous minerals, as it has a high aflinity for the melt, which is even higher
than water according to Fuge (1977). Sigvaldason & Óskarsson (1976),
however, showed that C1 has a higher solubility in basic silicate magmas
than the other volatiles, while Carmichael et al. (1974) state that in the
presence of OH and F, C1 will be rejected from a crystallizing hydrous
mineral. According to the free energy data (Carmichael et al., op.cit.), micas
and apatites crystallizing from magmas prefer f’ to OH and OH to Gl. What
the volatile composition of the hydrous minerals will be, depends therefore
on the relative amounts of available OH, F and C1 (especially OH and F) in
the crystallizing magma. Absolute amounts of the volatile components of