¹⁸O/¹⁶O ratios of well-dated granitoid plutons from NE Nevada and NW Utah provide major constraints on the structure and composition of the underlying crust. Most strikingly, the results demonstrate the systematic involvment of different source rocks during each of the three stages of magmatism in the Jurassic, Cretaceous, and Cenozoic. The age-specific nature of magmatism is particularly evident when oxygen isotope data are combined with previously obtained radiogenic-isotope data on the same plutons. Earlier isotopic studies of plutonic rocks in this region suggest that continentally-derived sediments of the miogeocline were the primary magmatic source during Late Cretaceous time. We confirm that interpretation: the Late Cretaceous (90-70 Ma) granites have elevated whole-rock d¹⁸O (+9.3 to +12.1) and ⁸⁷Sr/⁸⁶Sr_i (0.711 to 0.734), and low eNd (-13 to -23). Surprisingly, the 170-150 Ma late Jurassic plutons in this region span an even larger range of d¹⁸O values (+7.2 to +13.2), despite Sr and Nd isotopic compositions which are much less "crustal" (⁸⁷Sr/⁸⁶Sr_i = 0.705 to 0.711, eNd = -2.5 to -6.5) than the Late Cretaceous values. These results cast doubt on previous interpretations that the Jurassic granitoid magmas in the northeastern Great Basin were mainly formed from a mantle-derived component, instead suggesting the involvement of large quantities of high-d¹⁸O (sedimentary or altered volcanic) mid- to upper crust. The 40-25 Ma Cenozoic plutons (d¹⁸O = +7.0 to +9.7, ⁸⁷Sr/⁸⁶Sr_i  = 0.707 to 0.717, eNd = -13.2 to -26.3) define a third, "continental basement" component, which not at all evident in Jurassic plutons but may have been an important secondary component in the Late Cretaceous suite. These three inferred protolith components imply at least two sources in addition to the upper-crustal Late Cretaceous source: variably depleted ancient continental basement for the Cenozoic granitoids, and enriched  upper-mantle for the lower and moderate-¹⁸O Jurassic granitoids, particularly the most mafic plutons emplaced at relatively shallow depths. Except for the Cenozoic suite, no geographic patterns are observed in the isotopic systematics.  The 40-25 Ma plutons, however, can be subdivided into a higher d¹⁸O (+8.6 to +9.7) southern group and a lower d¹⁸O (+7.0 to +8.2) northern group across a well-defined Crustal Age Boundary (CAB) that extends roughly eastward from Elko, Nevada;  this CAB coincides with Wright and Woodenís (1991) Upper Array - Lower Array isotope boundary, as well as with the approximate southern limit of exposure of Archean basement. The low d¹⁸O values and depleted lead isotope compositions of the Lower Array (northern) samples rule out the possibility that their very low eNd values were derived by mixing an upper-crustal or sedimentary component into a primitive melt.  This also indicates that Archean age-province basement is present beneath a large area of this most northeasterly part of the Great Basin.  The wide range of ¹⁸O/¹⁶O in the Jurassic granitoids strongly suggests either (1) moderate to extensive (typically 20% to 40%) mixing or assimilation of miogeoclinal sedimentary rocks into a melt generated in the upper mantle, or (2) a large, variable component of young altered volcanic or volcano-sedimentary rocks in the source;  the latter interpretation would imply that, after emplacement, the Jurassic suite was detached and transported about 100 km during eastward-directed Cretaceous thrusting. Hypothesis (1) is the most plausible scenario, if it can be proven that the bulk of the metasedimentary section in this area underwent a 2-3 per mil lowering of whole-rock d¹⁸O as a result of fluid-rock interaction sometime between the Jurassic and Late Cretaceous episodes.