As for the structural elements of polymeric fullerene crystals, these are C60 molecules (monomers). It is conventionally assumed that the singlet C60 molecule has no unpaired electrons. This assumption is quite well substantiated in experiment: no signiœcant electron paramagnetic resonance (EPR) signals from fullerene samples have been observed in either the gas or solid state. Consequently, vanishingly small diamagnetism of the molecule was observed.3 Once not subjected to question, this fact underlies all quantum chemical calculations of the singlet ground state of both C60 molecules and the crystals built of them, which have been performed until recently without allowance for the electron spin, within the framework of the closed-shell approximation. In this approximation, a thorough analysis of the possibility that magnetic electrons appear as a result of polymerization led to a negative result,4-6 from which it has been concluded that the perfect crystals of C60 polymer must be nonmagnetic. Since this statement was at variance with experimental data, several theoretical models have been proposed, which explain the appearance of unpaired electrons by radical7-9 or topological10,11 defects. However, the

presence of such defects was not experimentally conœrmed, the more so that, in order to provide for the magnetic susceptibility in bulk crystals, such defects must either be arranged in a uniform ordered manner or form rather dense structures ensuring percolation spin ordering.