Plumes from deep mantle are widely conjectured to define an absolute reference frame, inaugurate rifting, drive plates, and profoundly modify oceans and continents. Mantle properties and composition are assumed to be whatever enables plumes. Nevertheless, purported critical evidence for plume speculation is false, and all data are better interpreted without plumes. Plume fantasies are made ever more complex and ad hoc to evade contradictory data, and have no predictive value because plumes do not exist. All plume conjecture derives from Hawaii and the guess that Emperor-Hawaii inflection records a 60¾ change in Pacific plate direction at 45 Ma. Paleomagnetic latitudes and smooth Pacific spreading patterns disprove such a change. Rationales for other fixed plumes collapse when tested, and hypotheses of jumping, splitting, and gyrating plumes are specious. Thermal and physical properties of Hawaiian lithosphere falsify plume predictions. Purported tomographic support elsewhere represents artifacts and misleading presentations. Asthenosphere is everywhere near solidus temperature, so melt needs a tensional setting for egress but not local heat. Gradational and inconsistent contrasts between MORB and OIB are required by depth-varying melt generation and behavior in contrasted settings and do not indicate systematically unlike sources. MORB melts rise, with minimal reaction, through hot asthenosphere, whereas OIB melts react with lithosphere, and lose mass, by crystallizing refractories and retaining and assimilating fusibles. The unfractionated lower mantle of plume conjecture is contrary to cosmologic and thermodynamic data, for mantle below 660 is more refractory than that above.

Subduction, due to density inversion by top-down cooling that forms oceanic lithosphere, drives plate tectonics and upper-mantle circulation. It organizes plate motions and lithosphere stress, which controls plate boundaries and volcanic chains. Hinge rollback is the key to kinematics. Arcs advance and collide, the fast-spreading Pacific shrinks, etc. A fore-arc basin atop an overriding plate shows that hinge and non-shortening plate front there track together: velocities of rollback and advance are equal. Convergence velocity commonly also equals rollback velocity but often is greater. Slabs sinking broadside push upper mantle back under incoming plates and force rapid Pacific spreading, whereas overriding plates flow forward with retreating hinges. Backarc basins open behind island arcs migrating with hinges. Slabs settle on uncrossable 660-km discontinuity. (Contrary tomographic claims reflect sampling and smearing artifacts, notably due to along-slab raypaths.) Plates advance over sunken slabs and mantle displaced rearward by them, and ridges spread where advancing plates pull away. Ridges migrate over asthenosphere, producing geophysical and bathymetric asymmetry, and tap fresh asthenosphere into which slab material is recycled upward. Sluggish deep-mantle circulation is decoupled from rapid upper-mantle circulation, so plate motions can be referenced to semistable lower mantle. Global plate motions make kinematic sense if Antarctica, almost ringed by departing ridges and varying little in Cenozoic paleomagnetic position, is stationary: hinges roll back, ridges migrate, and directions and velocities of plate rotations accord with subduction, including sliding and crowding of oceanic lithosphere toward free edges, as the dominant drive. (The invalid hotspot and no-net-rotation frames minimize motions of hinges and ridges, and their plate motions lack kinematic sense.) Northern Eurasia also is almost stationary, Africa rotates very slowly counterclockwise toward Aegean and Zagros, Pacific plate races toward surface-exit subduction systems, etc.