The kinetic mechanism of Eg5 is broadly similar to the widely studied kinesin-1 mechanism, but with some distinct features (Waitzman and Rice 2013 ). As with all molecular motors, the most stable state (the ground state) is the apo state, in which the active site is empty and the motor binds strongly (stably) and stereospecifi cally to its microtubule (MT) track. The subsequent steps of MgATP turnover progressively destabilise MT binding. The initial binding of MgATP or MgAMPPNP (a non-hydrolysable analog) into the empty active site drives the motor into a new conformational state, but this new motor·MgADP state binds to MTs only slightly less stably than the apo (empty) state, so that the motor head can still hold substantial force. The subsequent catalytic steps of hydrolysis and Pi release generate a motor·MgADP state that tends to detach rapidly from the MT. Rebinding of this motor·MgADP state to MTs triggers MgADP release, regenerating the apo state. In the motor·MgADP state, spontaneous release of MgADP from the active site is extremely slow, ~0.005 s −1 , and rate-limiting for ATP turnover in the absence of MTs. The weak binding motor·MgADP state is a "slip-state" that slides along the MT almost without friction if force is applied [ 1 ]. Nonetheless, this motor·MgADP slip-state is crucial because it provides access to a transient strong binding motor·MgADP state from which MgADP release is very fast: around three orders of magnitude faster than in the absence of MTs. The MT binding surface of the motor head is on the opposite side of the central beta sheet backbone to the active site, so that much of the structural mechanism of MT activation is necessarily allosteric. As I discuss below, the binding of allosteric inhibitors to the Eg5 motor head adds further stability to the motor·MgADP state and thereby inhibits MT-activated MgADP release. The crucial point therefore is that inhibitors and MTs have