Seismic tomography studies and plate reconstructions suggest that the Farallon slab broke off shortly before subduction ended off southern Baja California. However, the progress of detachment in time and space and its consequences on the volcanism of central Mexico have not so far been considered. Here I use the Neogene geologic record of central Mexico to propose a lateral propagation of slab detachment beneath the Trans-Mexican volcanic belt during the late Miocene. I suggest that the trace of the detachment is expressed by a short (2–3 m.y.), eastward-migrating pulse of mafic volcanism that took place from ca. 11.5 to ca. 6 Ma to the north of the Pliocene–Quaternary volcanic arc, as hot, subslab material flowing into the slab gap produced a transitory thermal anomaly in the mantle wedge. Slab detachment of the deeper and denser part of the plate was initiated in the southern Gulf of California area by the incoming of progressively younger oceanic lithosphere at the paleotrench that produced an increasing coupling between the Magdalena microplate and the overriding North American plate. The tear in the slab propagated eastward from the Gulf of California to the Gulf of Mexico, paralleling the southern Mexico trench system. The decrease in the Rivera–North America convergence rate between ca. 9 and 7 Ma appears to be related to the loss of slab pull after the detachment. Sparse oceanic-island–type basalts emplaced since the end of the Miocene in the Trans-Mexican volcanic belt are located above a trench-parallel slab window between the inferred detachment trace and the leading edge of the present slab, which has been detected seismically. In this context, the occurrence of these unusual intraplate magmas is easily explained by the infiltration of enriched asthenosphere into the subarc mantle.