Linear transport theory prescribes that for, e.g. a binary oxide Co1-deltaO, a flow of cations (J1) maybe induced not only by its direct cause which is a gradient of the cationic electrochemical potential (deleta1), but also indirectly by a gradient of the electronic electrochemical potential (deleta2), or J1 = -L11deleta1-L12deleta2, L(ik) being the transport coefficient. Routine practice, however, has been neglecting the cross coefficient, i.e. L12 = 0. We have determined experimentally the charge of transport of the cation alpha1*(=L12/L11) for Co1-deltaO with three different and independent techniques, and concluded that the cross coefficient is not negligible by any means. All the numerical data on alpha*1 up to date are compiled. Consequences of the non-zero cross coefficient and the physicochemical nature of the charge of transport are discussed.