Wall conditioning of fusion devices involves removal of desorbablehydrogen isotopes and impurities from interior device surfaces to permitreliable plasma operation. Techniques used in present devices include baking, metal film gettering, deposition of thin films of low-Z material, pulsedischarge cleaning, glow discharge cleaning, radio frequency discharge cleaning, andin situ limiter and divertor pumping. Although wall conditioning techniqueshave become increasingly sophisticated, a reactor scale facility willinvolve significant new challenges, including the development of techniques applicable in the presence of a magnetic field and of methods for efficientremoval of tritium incorporated into co-deposited layers on plasma facing componentsand their support structures. The current status of various approaches is reviewed, and the implications for reactor scale devices are summarized.Creation and magnetic control of shaped and vertically unstable elongated plasmas have been mastered in many present tokamaks. The physics of equilibrium control for reactorscale plasmas will rely on the same principles, but will face additionalchallenges, exemplified by the ITER/FDR design. The absolute positioning of outermost flux surface and divertor strike points will have to be precise andreliable in view of the high heat fluxes at the separatrix. Long pulses will requireminimal control actions, to reduce accumulation of AC losses in superconducting PF and TF coils. To this end, more complex feedback controllers areenvisaged, and the experimental validation of the plasma equilibrium response models onwhich such controllers are designed is encouraging. Present simulation codes providean adequate platform on which equilibrium response techniques can be validated.Burning plasmas require kinetic control in addition to traditional magnetic shapeand position control. Kinetic control refers to measures controlling density,rotation and temperature in the plasma core as well as in plasma periphery anddivertor. The planned diagnostics (Chapter 7) serve as sensors forkinetic control, while gas and pellet fuelling, auxiliary power and angular momentuminput, impurity injection, and non-inductive current drive constitute the controlactuators. For example, in an ignited plasma, core density controls fusion poweroutput. Kinetic control algorithms vary according to the plasma state, e.g.H- or L-mode. Generally, present facilities have demonstrated the kinetic controlmethods required for a reactor scale device. Plasma initiation - breakdown,burnthrough and initial current ramp - in reactor scale tokamaks will not involvephysics differing from that found in present day devices. For ITER, the inducedelectric field in the chamber will be ∼0.3V· m-1 - comparable to that requiredby breakdown theory but somewhat smaller than in present devices.Thus, a start-up 3MW electron cyclotron heating system will be employed to assure burnthrough. Simulations show that plasmacurrent ramp up and termination in a reactor scale device can follow proceduresdeveloped to avoid disruption in present devices. In particular, simulations remain inthe stable area of the li-q plane. For design purposes, theresistive V·s consumed during initiation is found, by experiments, to follow theEjima expression, 0.45μ0 RIp. Advanced tokamak control has two distinct goals. First, control of density, auxiliary power, and inductive current rampingto attain reverse shear q profiles and internal transport barriers, which persistuntil dissipated by magnetic flux diffusion. Such internal transport barriers can leadto transient ignition. Second, combined use poloidal field shape control withnon-inductive current drive and NBI angular momentum injection to create and controlsteady state, high bootstrap fraction, reverse shear discharges. Active n = 1 magneticfeedback and/or driven rotation will be required to suppress resistive wall modes forsteady state plasmas that must operate in the wall stabilized regime for reactorlevels of β ⩾ 0.03.