JOHNSON, Breck R.,
GLAZNER, Allen F., and COLEMAN, Drew S., Department of Geological
Sciences, Univ of North Carolina at Chapel Hill, CB# 3315, Mitchell Hall,
Chapel Hill, NC 27599-3315, breckj@email.unc.edu
Potassium feldspar (K-spar) crystals in granitic rocks grow to huge
sizes (> 10 cm) in some granites and are termed megacrysts. Many
workers interpret these megacrysts as having grown from relatively few
nuclei in a largely liquid magma, and they infer that alignment and
concentration of megacrysts reflect magma flow, crystal sorting, and
debris avalanching. However, experimental studies indicate that K-spar
crystallizes late in most granites (Swanson, 1979; Clemens and Wall, 1981;
Whitney, 1988). Estimates of the amount of melt present during initial
K-spar growth range from 65-70%, with a majority of K-spar growth
occurring within a few degrees of the solidus. These data indicate that
magmas would be choked with crystals and unable to flow by the time the
K-spar crystals grew to significant size.
To determine the origin of these intriguing megacrysts, we studied
K-spar size distribution in the Tuolumne Intrusive Suite, Yosemite
National Park, California. On a 12 km traverse from the equigranular Half
Dome Granodiorite into the megacrystic Cathedral Peak Granodiorite, the
average area of the 10 largest crystals within a 1 m2 area
steadily increases from 0.2 to 28 cm2. In contrast, bulk rock K2O
and K-spar mode (vol %) are constant across this same transect (at 3.7 ±
0.5 wt % and 22 ± 5 vol % respectively). Backscattered electron images of
megacrysts show highly variable and truncated concentric compositional
zones that correlate with Ba content.
The compositional features and especially the spectacular increase in
K-spar size at constant K-spar vol % are interpreted to reflect growth of
megacrysts at the expense of smaller crystals. These data are consistent
with Higgins's (1999) crystal size distribution measurements in the same
area that show trends indicative of textural coarsening and are
inconsistent with other explanations of Higgins's data as reflecting
crystal accumulation. Instead, we suggest that these textures are the
result of remelting and textural coarsening/crystal aging, consistent with
a protracted cooling history.