Imaginary part of Feynman amplitude, cutting rules and optical theorem

We discuss the algorithm of the cutting rules of calculating the imaginary part of physical amplitude and the optical theorem. We ameliorate the conventional cutting rules to make it suitable for actual calculation and give the right imaginary part of physical amplitude. The calculations of the imaginary parts of several Feynman diagrams show that the ameliorated cutting rules agrees with the conventional integral algorithm very well. The investigation helps us to find that the optical theorem has severe contradictions and problem, thus isn't right. Besides, the calculation of a physical amplitude show that the ameliorated cutting rules keeps the physical result gauge-parameter independent.