Abstract The nucleation model for the oxidation of carbons proposed by Dziembaj et al., was found to be adequate for describing the kinetics of oxidation of a char prepared from Fingal subbituminous coal but not for those of a char prepared from Yallourn brown coal nor a coke prepared from Wallarah bituminous coal. The formation of branched pores during the oxidation of Yallourn char means that two necessary conditions in the nucleation model cannot be satisfied. Hence, the inadequacy of the model for Yallourn char was not surprising. No explanation is offered for the lack of fit for the case of oxidising Wallarah coke, but the condition does not appear to be due to experimental error. The nucleation model is useful for describing the oxidation kinetics of some carbons, particularly at small values of conversion. It is argued that the model will unconditionally cease being useful for conversions larger than some value, which may be dependent on the carbon source.