The deep crustal rocks exposed in the Ruby-East Humboldt metamorphic core complex, northeastern Nevada, provide a guide for reconstructing Eocene crustal structure ∼50 km to the west near the Carlin trend of gold deposits. The deep crustal rocks, in the footwall of a west-dipping normal-sense shear system, may have underlain the Piñon and Adobe Ranges about 50 km to the west before Tertiary extension, close to or under part of the Carlin trend. Eocene lakes formed on the hanging wall of the fault system during an early phase of extension and may have been linked to a fluid reservoir for hydrothermal circulation. The magnitude and timing of Paleogene extension remain indistinct, but dikes and tilt axes in the upper crust indicate that spreading was east-west to northwest-southeast, perpendicular to a Paleozoic and Mesozoic orogen that the spreading overprinted. High geothermal gradients associated with Eocene or older crustal thinning may have contributed to hydrothermal circulation in the upper crust. Late Eocene eruptions, upper crustal dike intrusion, and gold mineralization approximately coincided temporally with deep intrusion of Eocene sills of granite and quartz diorite and shallower intrusion of the Harrison Pass pluton into the core-complex rocks. Stacked Mesozoic nappes of metamorphosed Paleozoic and Precambrian rocks in the core complex lay at least 13 to 20 km deep in Eocene time, on the basis of geobarometry studies. In the northern part of the complex, the presently exposed rocks had been even deeper in the late Mesozoic, to >30 km depths, before losing part of their cover by Eocene time. Nappes in the core plunge northward beneath the originally thicker Mesozoic tectonic cover in the north part of the core complex. Mesozoic nappes and tectonic wedging likely occupied the thickened midlevel crustal section between the deep crustal core-complex intrusions and nappes and the overlying upper crust. These structures, as well as the subsequent large-displacement Cenozoic extensional faulting and flow in the deep crust, would be expected to blur the expression of any regional structural roots that could correlate with mineral belts. Structural mismatch of the mineralized upper crust and the tectonically complex middle crust suggests that the Carlin trend relates not to subjacent deeply penetrating rooted structures but to favorable upper crustal host rocks aligned within a relatively coherent regional block of upper crust.