FORMATION OF LATE-STAGE MELT-FILLED EXTENSIONAL DOMAINS IN CRYSTAL MUSH OF THE HONEY BROOK ANORTHOSITE

The Honey Brook Anorthosite (HBA) is a Mesoproterozoic pluton composed of anorthosite, dioritic anorthosite, anorthositic diorite, and Fe-rich hornblende gabbro, with an outcrop area ~40 sq. km, in the Piedmont Upland of southeastern Pennsylvania (Crawford and Hoersch, 1984). Garnet in an exposure of dioritic anorthosite is large (1–5 cm), sub- to euhedral and unzoned, with a Mg-rich composition, Alm₅₃Prp₂₉Grs₁₆Sps₀₂. Garnet is surrounded by plagioclase-rich leucocratic halos in elongate polycrystalline clusters that occur in linear arrays on outcrop surfaces, (Higgins, et al., 2017). A metamorphic paragenesis for the garnet is rejected based on the temperature indicated by the Mg-rich garnet composition – isopleths calculated with Theriak-Domino intersect at 950 °C and 700 MPa – and the lack of evidence for similar high metamorphic temperatures in adjacent rock. Thus, garnet and surrounding plagioclase are interpreted to be primary igneous phases. We use rhyolite-MELTS (Asimow and Ghiorso, 1998; Ghiorso, et al., 1995) to test this hypothesis by modeling cooling of anorthositic magma. Using a fractional crystallization model at P = 700 MPa, we find that garnet, with composition nearly identical to that in the rock, begins to crystalize at ~960 °C, with approximately 35 wt.% liquid remaining. At these conditions, ortho- and clinopyroxene crystallization has ceased; only plagioclase and garnet form.

We propose that with this proportion of melt remaining, the crystal mush was strong enough to support formation of discrete vein-like extensional domains in which late-stage melt accumulated and crystalized. The long-dimension orientations of 32 garnet-bearing domains were measured in outcrop. These exhibit a great circle distribution on an equal-area plot, indicating that they lie in the same moderately south-dipping plane. Since extensional features open in the direction of the minimum compressive stress, σ3, we suggest that this plane is the σ1-σ2 plane; the pole to this plane is the σ3 direction. Limited exposure in the Honey Brook Upland limits our ability to assess whether this reflects a local or regional tectonic stress regime at the time of crystallization.