Abstract
An explanation is proposed of the maximum at 41 meV in the inelastic spin-flip neutron scattering from YBa2Cu3O7δ, based on the extended-saddle-point model developed by the author in his previous works. It is shown that for appearance of the maximum in the imaginary part of the spin susceptibility a close proximity of the Fermi energy to the extended saddle point is necessary. The energy of the maximum is then close to 2Δmax in agreement with experiment. Theoretical and experimental evidence concerning the energy of the extended saddle point (flat region) is discussed. Different limiting cases are calculated. A proof is given within the present model that interaction in the final state is small, and hence, no collective modes are formed. A general discussion of the experimental situation is presented.