“…These are the reasons why this method found limited application; namely, in the symmetric geometry, one can combine a standard two-beam approach for the neighbourhood of a diffraction maximum with a kinematical formula for the parts of a CTR that are further away from the diffraction maximum. In a series of papers by Takahashi and coworkers (Takahashi & Nakatani, 1995;Yashiro et al, 2001;Yashiro et al, 2004), the Darwin method was generalized to an asymmetric diffraction geometry and to a three-beam diffraction case; however, the method is still based on kinematical reflection coefficients of a single subatomic layer. In the present paper, we follow the general concept of Caticha (1994), but we improve it in the following aspects.…”