The following is an abstract of a talk presented at the 31 Yellowknife Geoscience forum, November 2003.

The 65 km long, north-trending 1270 Ma Muskox layered mafic-ultramafic intrusion in the northwest Canadian Shield has a maximum width and thickness of about 10 km and 2200 m, respectively, near its northern extremity. It narrows to roughly 200 metres at its south end, and then continues further south as a narrow dyke. Northward continuation of the intrusion beneath basaltic volcanics of the Coppermine River Group is suggested by the presence of prominent linear gravity and magnetic highs.

The intrusion comprises marginal peridotites lying above the inward-dipping contacts, and an overlying, flat-lying igneous stratigraphy dominated by dunites and peridotites (lower and mid levels) and gabbros (higher levels). Geological evidence indicates that the intrusion is funnelshaped in cross-section along its length. Gravity and magnetic anomalies associated with the intrusion provide an independent means of examining the third dimension of the intrusion. A conspicuous linear magnetic high runs along the intrusion, locally branching into two or three separate features, and extends approximately 20 km northward across overlying, gently dipping sedimentary and volcanic sequences of the Coppermine River Group, suggesting continuity of the intrusion at depth in this direction. Magnetite produced by serpentinization of ultramafic units is the probable principal source of this magnetic anomaly. The gravity signature is less uniform. A distinct gravity high is observed only over the northern part of the intrusion, where high-density gabbros are exposed. This continues northward following the path of the magnetic high, but extends some 50 km farther north. Most of the intrusion is associated with only a weak positive gravity signature or no gravity response. This, also, reflects serpentinization affecting olivine-rich ultramafic rocks of the intrusion. In the southern part, dominated by dunite, densities have been reduced from ~3.00 g/cm3 to values as low as or lower than those of adjacent country rocks (2.71 g/cm3).

Gravity modelling, more pragmatically applied in the northern part of the intrusion, suggests that the intrusion increases in width from about 10 to 19 km, maintaining a funnel-like geometry in the east, but becoming slab-like (~750 m maximum thickness) in the west. Small satellite maficultramafic bodies within overlying dolomite and sandstone provide convincing evidence for the western extension. In the gravity model, the position of the eastern keel is displaced roughly 2 km from the position portrayed in the geological model, and gabbroic layers are thicker and are gently synformal, as opposed to being flat-lying. A caveat is that serpentinized portions of the intrusion, having low densities, may be invisible to the gravity modelling. Problematically, the match between the magnetic signature of the gravity model and the observed magnetic signature is extremely poor. It is apparent that the sharp peaks of the magnetic profile cannot be generated by gently dipping units. Successful modelling of magnetic profiles crossing the southern part of the intrusion, where dips are steeper and a presumed vertical feeder zone is closer to the surface, support this conclusion. The reconciliation of gravity and magnetic models for the intrusion is examined and debated.