UNSOLVED PETROLOGIC PROBLEMS OF THE KIGLAPAIT LAYERED INTRUSION, LABRADOR

The KI is a story of troctolite to Mg-free ferrosyenite at 1.307-Ga. Successes: The cotectic trace L(Ol,Pl) for the Lower Zone (LZ) has been found experimentally at 5 kb in graphite, for melt compositions that yield the An and Fo values observed in the field. The Aug content of these melts runs from 5% to 24% before Aug+, which occurs at the same Ol:Aug ratio as in the modal data (Morse 1979b JP). This track length is consistent with 81-84 volume percent solidified (PCS) based on rock volumes in the field, and therefore supports multiphase Rayleigh fractionation models for Ol and Pl variation (Morse 1996 JP). The experimental Ol content is greater by ~2% than in the summation model (Morse, 1981 GCA) at the low-Aug (LLZ) end, but identical to it at Aug+, leading to the notion that Ol may be under-represented in the exposed LLZ rocks. The linear-partitioning KD for XAb in Pl/L is 0.524, allowing full modeling of the Pl-L relations. D(K) Fsp/L rises 0.2-1.2 from early Pl to late ternary feldspar. Rocks of the Upper Border Zone (UBZ) contain elevated K in the form of red fluorine oxybiotite. Problems: The summation liquids at 0 PCS have Fo 62, An 60 compared to the equilibrium values Fo 48.4 and 51.5 required by the observed crystals. The low values of D(K) in the LZ require strong evolution of potassium not seen until very late in the UZ rocks. Missing components in summation models: Ol in LLZ; Fa: 0.78*62=Fo 48.4; Ab: 0.86*60=An 51.5; K in later LZ rocks. The multipliers 0.78 (Morse 1996 JP) and 0.86 are mysterious constants. By contrast to these problems, excluded TiMt and Ap are consistent with summation models. All these problems can be satisfied by invoking an Fe-rich pond at the central floor of the intrusion (Morse, 1996 JP), required by drainage of rejected solute down-dip from the solidification of troctolite (Morse 1969 GSA Mem 112), carrying Ol, Fa, Ab, and K (some K rising with F to the UBZ). The residual porosity of the LZ cumulates decreases from 0.14 to 0.03 (Morse 1979b JP), so the ponding activity would decrease over time, ending when violent precipitation of Fe-Ti oxide minerals stirred the magma. A deep drill hole should encounter more mafic and more evolved cumulates at the center of the intrusion than at the exposed limbs. The mysterious constants should relate to the quantities of material transferred to the central pond, if we only knew how.