“…Hence for the classical four-step methods, the error increases as the fifth power of G. For the zero phase-lag multiderivative method presented in [41], the error increases as the fifth power of G. For the implicit P-stable multiderivative method presented in [46], the error increases as the sixth power of G. For the trigonometrically-fitted multiderivative method, first class, produced by Shokri [41], the error increases as the fourth power of G. For the trigonometricallyfitted multiderivative method, second class, produced by Shokri [41], the error increases as the fourth power of G. For the multiderivative method with vanished phase-lag and its first derivative produced by Simos [46], method NMI, the error increases as the third power of G. For the multiderivative method with vanished phase-lag and its first derivative 8 produced by Simos [46], method NMII, the error increases as the third power of G. For the multiderivative method with vanished phase-lag and its first derivative produced by Simos [46], method NMIII, the error increases as the second power of G. For the new four-step method with vanished phase-lag and its first, second and third derivatives obtained in this paper, the error increases as the second power of G with little larger coefficients than the method developed in [46]. But the developed method in [46] is multiderivative method and hence it implies that more CPU time than the new method presented in this paper for the numerical computations.…”