398 TECTONICS/Propagating Rifts and Microplates At Mid-Ocean Ridges The active propagating and failing rift axes overlap by about 20 km and are connected by a broad and anomalously deep zone of distributed shear deformation rather than by a classic transform fault Most of the seismic activity occurs within this ‘non-transform’ zone, where the pre-existing abyssal hill fabric originally created on the doomed rift is sheared and tectonically rotated into new oblique trends Simple equations accurately describe this geometry in terms of ratios of propagation and spreading rates, together with the observed propagating and doomed rift azimuths For example, for the simplest continuous propagation geometry, if u is the spreading half-rate and v is the propagation velocity, the pseudofaults form angles tan 1(u/v) with the propagator axis, and the isochrons and abyssal hill fabric in the zone of transferred lithosphere have been rotated by an angle tan 1(2u/v) The boundaries of the Galapagos high-amplitude magnetic anomaly zone, the ferrobasalt province, and the spreading-centre jumps are all coincident with the pseudofaults bounding the propagating rift lithosphere All of these observations can be explained as mechanical and/or thermal consequences of a new rift and spreading centre breaking through cold lithosphere, with increased viscous head loss and diminished magma supply on the propagating spreading centre close to the propagator tip This leads to an unusually deep axial graben and unusually extensive fractional crystallization and differentiation The 95.5 W propagator tip is also a mantle geochemistry boundary, implying that this rift propagation is associated with plume-related subaxial asthenospheric flow away from the Galapagos hotspot Causes of Rift Propagation One important observation is that many rifts and spreading centres propagate down topographic gradients away from hotspots or shallow ridge axis topography For example, all six known active Galapagos propagators are propagating away from the hotspot Plume-related asthenospheric flow generates gravitational stresses on the shallow spreading-centre segments near the hotspot that promote crack propagation away from the hotspot Flow of asthenosphere into these cracks produces new lithosphere at propagating seafloor-spreading centres Regionally high deviatoric tensile stresses associated with regional uplift provide a quantitatively plausible driving mechanism Crack growth occurs when the stress concentration at the tip, characterized in elastic fracture mechanics by a stress intensity factor, exceeds the resisting stress intensity contribution The spreading-centre propagation rate could be limited by the viscosity of the asthenosphere flowing into the rift, producing viscous suction forces at a local tip depression, or by process zone deformation at the rift tip The overlap/offset ratio of propagating and failing rifts tends to be $1, close to the ratio at which the stress intensity factor is maximized (see Tectonics: Seismic Structure At Mid-Ocean Ridges) Although many rifts appear to propagate in response to hotspot-related stresses, others appear to propagate because of stresses producing changes in plate motion Subduction-related stresses appear to be a common mechanism for producing propagation in the North-east Pacific This probably explains most of the massive reorganizations of the spreading geometry as the Pacific–Farallon ridge neared the Farallon– North America trench, as clearly evident even in the classic Raff–Mason magnetic anomaly data (Figure 3), although some propagation away from the Axial Seamount hotspot has also occurred in the Juan de Fuca area Propagation may be produced in many ways over a wide range of scales, including smallscale propagation of overlapping spreading centres away from local magmatic centres The larger reorganizations sometimes involve transient microplate formation, geometrically similar to the broad overlap zone model of Figure 1C, but on a much larger scale Microplates Microplates are small, mostly rigid areas of lithosphere, located at major plate boundaries but rotating as more or less independent plates They can form in many tectonic settings The two main types along mid-ocean ridges, those formed at triple junctions and those formed away from triple junctions, share many similarities Although it was once thought that stable, growing microplates could eventually grow into major oceanic plates, it now appears that these are transient phenomena resulting from large-scale rift propagation When the overlap zone becomes too big and strong to deform by pervasive bookshelf faulting, it changes mechanical behaviour and accommodates the boundary plate motion shear stresses by beginning to rotate as a separate microplate The most well-studied oceanic microplates are the Easter microplate along the Pacific–Nazca ridge and the Juan Fernandez microplate at the Pacific–Nazca– Antarctica triple junction (Figure 4) Despite their different tectonic settings, they show many striking similarities The scales of the Easter ($500 km diameter) and the Juan Fernandez ($400 km diameter) microplates are similar The eastern and western boundaries of both microplates are active spreading centres, propagating north and south, respectively Both microplates began forming about Ma, and both