Agmatite

Agmatite (derived from the Greek agma, meaning "fragment" or "shard") is a coarse-grained, breccia-like structural variety of migmatite in which angular to subrounded blocks of dark, more competent metamorphic country rock (paleosome or mesosome—most commonly amphibolite, hornblende gneiss, metabasalt, or calc-silicate granulite) are enclosed within an interconnected, anastomosing three-dimensional network of light-colored quartzo-feldspathic veins (leucosome of granitic, trondhjemitic, or tonalitic composition). Introduced into petrological nomenclature by the Finnish geologist Jakob Johannes Sederholm in 1923 during his classic mapping of the Precambrian migmatite complexes of southern Finland, agmatite represents the initial brittle-to-semi-brittle structural disruption of competent rock layers in high-grade upper-amphibolite to granulite-facies metamorphism and deep plutonic roof/margin zones.

In the internationally standardized morphological classifications of migmatites established by Karl Richard Mehnert (1968) and the IUGS Subcommission on the Systematics of Metamorphic Rocks (Sawyer, 2008), agmatite (or agmatic migmatite) is classified as a first-order metatexite or transitional diatexite structure characterized by low-to-moderate melt fractions (typically 10–35 vol% leucosome) where the rigid mafic blocks fractured mechanically rather than flowing ductilely.

Jigsaw-Puzzle Fit, Paleosome Blocks, and Leucosome Matrix

An outcrop of classic agmatite consists of three distinct petrographic domains whose geometric relationships reveal how melt interacted with solid rock at temperatures of 650–850 °C and pressures of 0.4–1.0 GPa (15–35 km crustal depth):

1. Paleosome / Competent Clasts (65–90 vol%): Angular, sharp-cornered blocks ranging from a few centimeters to several meters in diameter, composed of dark refractory lithologies with high solidus temperatures (such as hornblende–plagioclase amphibolite, metadolerite dikes, or biotite–garnet paragneiss). In low-strain agmatites, adjacent angular blocks preserve a jigsaw-puzzle fit: matching re-entrant corners, planar foliation traces, and pre-existing quartz veins can be visually fitted back together across the intervening leucosome vein with minimal block rotation.
2. Leucosome Vein Network (10–35 vol%): White, cream, or pale pink veins (1 cm to 50 cm wide) composed primarily of quartz, plagioclase, and K-feldspar with minor garnet, hornblende, or orthopyroxene. Unlike high-strain stromatic (layered) migmatites where leucosomes parallel the metamorphic foliation, agmatite leucosomes form a polygonal, non-foliated or weakly foliated rectilinear net filling Mode I tensile and Mode II shear fractures between the paleosome blocks.
3. Melanosome / Reaction Rims: Along the margins of the angular amphibolite blocks, a dark 1–10 mm rim enriched in coarse hornblende, biotite, or magnetite frequently develops due to either peritectic dehydration-melting reactions or metasomatic K- and SiO₂-exchange between the hydrous granitic melt and the mafic clast.

Position in Mehnert's and Sawyer's Migmatite Classification

As partial melt fraction (Φ) and differential tectonic stress (σ₁ − σ₃) increase in the middle and lower continental crust, agmatite occupies a specific mechanical niche within the broader spectrum of migmatite structures:

When the volume fraction of leucosome melt rises above the rheologically critical melt percentage (RCMP, ~30–40 vol%) or when regional tectonic shear intensifies, the angular corners of the agmatite blocks are corroded and rounded, the jigsaw-puzzle fit is destroyed by rigid-body rotation in the flowing magma, and the rock transitions from an agmatite into a schollen (raft) migmatite.

Melt-Enhanced Embrittlement and Hydraulic Fracturing

Why do amphibolite and gneiss layers fracture into sharp, angular breccia blocks at temperatures of 700–800 °C deep in the ductile lower crust, where rocks normally deform by crystal-plastic creep? Structural and experimental rock mechanics studies identify two primary mechanisms:

1. Hydraulic Fracturing by Overpressured Anatectic Melt (P_melt > σ₃ + T₀): During prograde dehydration melting of fertile metapelites or greywackes (for example, muscovite/biotite + plagioclase + quartz → granitic melt + garnet/sillimanite + K-feldspar), the reaction has a positive volume change (ΔV_rxn > 0). When melt accumulates beneath or around an impermeable, refractory amphibolite layer with a higher solidus temperature, the pore-melt pressure P_melt rises until the effective minimum principal stress σ₃′ = σ₃ − P_melt becomes tensile and exceeds the tensile strength T₀ (~5–15 MPa) of the amphibolite. In accordance with Griffith failure criteria, the competent amphibolite fails abruptly by melt-driven hydraulic fracturing, sucking low-viscosity granitic leucosome into the newly opened angular cracks.
2. Competency Contrast and High Strain Rate during Pluton Emplacement (Contact Agmatites): Along the roof and walls of syntectonic granitic, tonalitic, or dioritic plutons—and inside Archean tonalite–trondhjemite–granodiorite (TTG) domes—pulses of pressurized magma injected at rapid strain rates (10⁻⁸ to 10⁻⁴ s⁻¹) exceed the viscous relaxation rate of the surrounding country rock or earlier chilled mafic border phases, shattering the mafic host into an intrusion breccia (injection agmatite) cemented by tonalitic or pegmatitic leucosome.

In-Situ Partial Melting vs. External Melt Injection

Field geologists distinguish in-situ anatectic agmatites from external injection agmatites using mass-balance and geochemical criteria:

• In an in-situ peritectic agmatite formed by dehydration melting of calc-alkaline amphibolite (hornblende + plagioclase + quartz → tonalitic melt + clinopyroxene/garnet + hornblende-rich melanosome), the leucosome veins contain peritectic hornblende or orthopyroxene crystals and are bordered by complementary silica-depleted melanosome rims.
• In an injection agmatite (the most common variety in Archean and Proterozoic shield terranes), refractory amphibolite paleosome blocks (solidus > 800–850 °C) sit inside K-rich granitic or trondhjemitic leucosome mobilized from adjacent fertile metasediments or underlying plutons, showing razor-sharp fracture walls with little or no in-situ melanosome residue. Compare this low-rate crustal hydraulic brecciation with high-velocity coseismic frictional melting in pseudotachylite.

How does an agmatite differ from a schollen (raft) migmatite?

In an agmatite, the paleosome fragments are angular to subangular, occupy most of the rock volume (65–90%), and exhibit a jigsaw-puzzle fit where neighboring blocks can be visually pieced back together across narrow leucosome veins. In a schollen (raft) migmatite, melt volume exceeds ~35–50%, so the fragments have been rounded by corrosion and rotated out of alignment by magmatic flow, floating like isolated rafts in a sea of granitic diatexite.

How can rocks at 700 °C and 20 km depth fracture into angular blocks instead of bending ductilely?

High pore pressure inside pockets of buoyant partial melt (P_melt) counteracts the confining pressure of the overlying crust (σ₃′ = σ₃ − P_melt). Once the melt pressure exceeds the minimum compressive stress plus the rock's tensile strength, the competent amphibolite layer fails by brittle hydraulic fracturing, just as high-pressure water fractures shale at shallower depths.

Why is the dark paleosome in an agmatite almost always amphibolite or mafic gneiss?

Under upper-amphibolite facies conditions (650–750 °C), surrounding quartz- and mica-rich metasedimentary rocks partially melt and become weak and ductile, whereas mafic amphibolites have a higher melting temperature (solidus > 800 °C under fluid-absent conditions) and a rigid framework of interlocking hornblende and plagioclase crystals, causing the mafic layers to behave as brittle, competent slabs that fracture as the surrounding rocks melt.

1. Sederholm, J. J. (1923). "On Migmatites and Associated Pre-Cambrian Rocks of Southwestern Finland, Part I: The Pellinge Region." Bulletin de la Commission Géologique de Finlande, 58, 1–153. https://tupa.gtk.fi/julkaisu/bulletin/bt_058.pdf
2. Mehnert, K. R. (1968). Migmatites and the Origin of Granitic Rocks. Elsevier Publishing Company. https://doi.org/10.1016/B978-0-444-40679-8.50001-0
3. Sawyer, E. W. (2008). Atlas of Migmatites. The Canadian Mineralogist, Special Publication 9, NRC Research Press. https://doi.org/10.1139/9780660197876
4. Brown, M. (2013). "Granite: From Genesis to Emplacement." Geological Society of America Bulletin, 125(7–8), 1079–1113. https://doi.org/10.1130/B30877.1