The temperature and field dependences of the magnetic moment of polyaniline powder doped by m cresol were measured by SQUID magnetometry in the temperature range 2-300 K at 1000 Oe and in the range 0-50000 Oe at 2 K, respectively. The field dependence is not described by the Brillouin function for spin 1/2, as is expected in the framework of a commonly accepted "metallic" model. Both dependences are quite correctly described by a "triplet" model using a distribution of singlet triplet splitting (E) with the density distribution function having a narrow peak near E = 0.Conducting polymers including polyaniline, poly acetylene, polythiophene, polypyrrole, poly(para phe nylene vinylene), etc. possess unusual physical properties and can be used in various fields. Practically, the empha sis is placed on the studies of their luminescence and conductivity. Investigations of the magnetic properties occupy a special position because these properties are intimately related to the nature of charge carriers and to subtle features of the polymer structure.Often, the experimentally observed linear dependence of the product of the paramagnetic susceptibility of con ducting polymers by the temperature (χT) on temperature(1) allows one to divide the magnetic susceptibility into two components, viz., the temperature independent component χ P and the component obeying the Curie law χ = C/T (see, e.g., the data for polyaniline and its derivatives, 1-4 poly(ethylenedioxythiophene), 1 and polypyrrole 5 ). The origin of these two components is usually explained within the framework of a "metallic" model, which treats doped conducting polymers (in the form of both powders and films) as highly ordered metallic domains "immersed" into amorphous domains. The metallic domains are asso ciated with the temperature independent component (the Pauli susceptibility) while defects in the amorphous domains are responsible for the Curie susceptibility.However, a number of experimental facts do not obey the pattern mentioned above. For instance, it is unclear why the EPR lines of metallic and amorphous domains characterized by different widths do not overlap. In addi tion, it is difficult to explain the frequently observed nonlinear dependences χT-T in the framework of the "metallic" model. We have proposed a "triplet" model for paramagnetic centers in conducting polymers. 6 According to this model, conducting polymers comprise relatively short periodic fragments with close values of the angles between neighboring ring planes. These fragments are separated from one another by abrupt changes in these angles, each fragment can be in the triplet or singlet state, and there is a number of fragment conformations, which leads to variation of the singlet triplet splitting (E) over a wide range. In this case, the magnetic susceptibility can be described by an integral of the Bleaney-Bowers equa tion over the E distribution. For some fragments, triplets lie lower than singlets; these fragments are responsible for the Curie like contribution to the magnetic su...