The nonlinear regression method was used for the evaluation of applicability of the known model equations that describe the ionic strength dependence of the rate of reaction between charged molecules to the electron transfer reaction between myoglobin and cytochrome c. The full and simplified equations of Bronsted-Debay-Hukkel and Wherland-Gray equation, as well as equations, derived from models of complementary interactions and "parallel disks" were used. The experimental dependences of reaction rate on ionic strength at all pH values in the pH interval 5-8 have a negative slope which corresponds to opposite charges of reacting particles and does not correspond (even in sign) to the total charges of Mb and Cyt c. It is found that all models considered, except for those of complementary interactions, permit a satisfactory description of the experimental data. In the case of "parallel disks" and Bronsted-Debay-Hukkel equations, however, the optimal values of Z1Z2 and R1 = R2 = R do not correspond to the real size of the Mb-Cyt c electron transfer complex and to the radii and charges at their contact sites found from the experiment. The Wherland-Gray equation allows the best approximation of the experimental ionic strength dependences assuming that Z1Z2 and R1 = R2 = R are some effective parameters that reflect the protein field distribution in the contact site. This distribution can be approximated as a monopole with parameters not coinciding with the parameters of the protein.