Neoclassical tearing modes were originally identified experimentally on the TFTR tokamak , and have subsequently been observed in many other tokamaks [12,15,18,23,40]. The flattening of the pressure profile within the magnetic separatrix of a neoclassical tearing mode leads to a degradation of the energy confinement properties of the plasma  that limits the maximum attainable value [see Equation (1.23)] . Consequently, neoclassical tearing modes are nowadays regarded as the main obstacle to obtaining values in tokamak plasmas that are adequate for the achievement of thermonuclear fusion [2,21,32].
The fact that a magnetic island chain can only locally flatten the plasma pressure profile (and, thereby, generate a destabilizing local reduction in the bootstrap current) when its radial width exceeds the critical value leads to the conclusion that neoclassical tearing modes are actually meta-stable . In other words, some sort of seed perturbation must be applied to the relevant rational magnetic flux-surface in order to trigger a neoclassical tearing mode. In practice, the seed perturbation usually takes the form of a transient magnetic perturbation that is resonant at the rational surface . Such perturbations arise naturally in tokamak plasmas as a consequence of plasma instabilities such as internal kink modes and edge localized modes .
Neoclassical tearing modes can be stabilized by driving a parallel (to the magnetic field) current in the vicinity of the rational surface by means of radio frequency electromagnetic waves injected into the plasma; the idea is to replace the missing bootstrap current within the island chain's magnetic separatrix [15,22,25,29,38,39].
The aim of this chapter is to use the nonlinear neoclassical resonant response model derived in the previous chapter to investigate the physics of neoclassical tearing modes.