A method and procedures for determining the boundaries of the second stage of the kinetic crack resistance diagram or fracture toughness kinetic diagram, sample formation within the aforementioned boundaries and determination of the parameters С and n of the Paris equation from the sample are presented. The necessity of developing the method is attributed to the lack of rules and procedures for accurate determination of the boundaries of the second stage in the current standards and regulatory document (RD). The proposed method provides a given accuracy of determination of the number of cycles corresponding to the length of the fatigue crack at the upper boundary of the second stage obtained by numerical integration of the Paris equation with the found values of the parameters С and re. The developed method is based on the application of two criteria R2 and %. Statistical criterion R2 characterizes a degree of deviation of the experimental data from the linear fragment of the kinetic fracture toughness diagram. Parametric criterion у specifies the level of accuracy of the parameters С and re of the Paris equation. This level is set through a comparative evaluation of the experimental and calculated crack length I and the number of cycles N, obtained by integration of the Paris equation within the specified lower and upper limits of the interval of the stable growth of fatigue crack. Application of the method is shown by the example of the experimental data obtained when testing samples of VT9 titanium alloy, deformable nickel alloy EI437BU and granular nickel alloy EP741NP (granules up to 140 pm) at room and elevated temperatures. Application of the method indicates that the experimental and calculated curves "I - N" obtained by numerical integration of the Paris equation differ by less than the specified value of the criterion X ^ 3%, in contrast to the results obtained in accordance to the recommendations of the regulatory documents.