Detailed organic geochemistry has been performed on a large number of Lower Mississippian-Upper Devonian Bakken shales from the North Dakota portion of the Williston Basin, and 28 oils mainly from Mississippian Madison Group rocks from different basinal areas. Here we report results of Rock-Eval pyrolysis and vitrinite reflectance (R_o) analyses. Variable paleoheat flows in the Williston Basin caused the threshold of intense hydrocarbon generation to occur at different depths in different basinal areas. In higher paleogeothermal gradient basinal areas, this event occurred at depths of 7,650-8,000 ft, and at 10,000 ft or deeper in lower paleogeothermal gradient areas of the Basin. Distinct organic metamorphic imprints in Williston Basin sediments were also caused by extreme, but variable, paleoheat flows in the basin, as well as secondary migration of crude oils from deep basinal source areas. The high paleoheat flows are postulated as being due to a Late Cretaceous - Paleocene aborted rift event. Only a small volume of Bakken shales in restricted areas of the Williston Basin was responsible for the oil found reservoired in Mississippian Madison Group rocks. However, this small shale volume has been responsible for a relatively large amount of crude oil.

R_o profiles in the Tertiary through Middle Jurassic rocks in the Williston Basin had steep, linear R_o versus depth gradients, with strong reversals of R_o values occurring in the Lower Jurassic rocks. The lower Mesozoic through Paleozoic rocks of the Basin had strongly suppressed R_o values compared to the values in the Tertiary through Middle Jurassic rocks. This was especially true of the R_o values in the Bakken shales. This R_o suppression was due to a change in organic matter (OM) type from oxygen-rich terrestrially derived OM in the younger rocks, to a hydrogen-rich marine derived OM in the deeper, older rocks. The threshold of intense oil generation (TIHG) occurred in the Bakken shales of the Williston Basin at R_o values somewhere between 0.9 and 1.7% (best estimate 0.9), as would be read in oxygen-rich OM. Much higher burial temperatures (and consequently R_o values) than usually held to be necessary were required for both the TIHG as well as mainstage hydrocarbon generation in the Bakken shales. These results are most likely applicable in general to source rocks with hydrogen-rich OM. The data of this study have major implications to petroleum exploration as well as to petroleum resource assessment.