Several zones of major thrust faulting exhibit a juxtaposition of rocks of higher metamorphic grade over rocks of lower grade. This configuration may indicate, but does not require, that temperature gradients were temporarily inverted near the fault. We examine the physical conditions under which such a temperature inversion could occur. The overthrusting of hotter rock on colder rock can cause a temporarily inverted gradient both above and below the fault only if the time taken to underthrust rock from the land surface to a given depth, zf, on the fault is less than &pgr; times the characteristic time, zf2/&pgr;2&kgr;, for diffusion of heat through the block above the fault, where &kgr; is thermal diffusivity. An inverted gradient cannot form without heat generation in the fault zone unless V ¿ zf ¿ sin Δ > 100, where V is the rate of underthrusting (in millimeters per year), zf is in kilometers, and Δ is the dip of the fault. This simple criterion is sufficient to demonstrate that several examples of inverted metamorphic gradients cannot be explained simply by the thrusting of hot on cold rock without heat sources in the fault zone. Dissipative heating accompanying deformation can cause an inverted temperature gradient within and beneath the thrust zone, but whereas the overthrusting of hot upon cold rocks cools the fault zone, dissipation heats it. Thus the overthrusting of hot on cold rock and dissipative heating affect the temperature gradient and the maximum temperatures differently.

We show that the magnitudes of the inverted temperature gradient and of the maximum temperature above the inverted gradient yield independent estimates of the rate of dissipative heating. Discrepancy between these estimates implies that some additional process must have occurred, such as the post-thrusting disruption of the isograds. If there is such a discrepancy, the maximum temperature probably provides the more reliable estimate of the rate of heating at the fault. We illustrate this analysis by applying it to reported inverted metamorphic zonation in the Pelona Schist, the St. Anthony Complex, the Mt. Everest region of the Main Central Thrust, and the Olympos Thrust. Petrological inferences of maximum temperatures, depths of metamorphism, and magnitudes of apparent inverted gradients, imply that shear stresses of about 100 MPa accompanied thrust faulting in some of these regions, and that some zones of inverted metamorphism have been tectonically thinned after the metamorphism occurred. ¿ American Geophysical Union 1993