“…The field of its application to experimental (X-ray diffraction) charge densities is rapidly developing now (e.g. Kapphahn, Tsirelson & Ozerov, 1988;Craven & Stewart, 1990;Gatti, Bianchi, Destro & Merati, 1992;Destro & Merati, 1995;Iversen, Larsen, Souhassou & Takata, 1995;Tsirelson & Ozerov, 1996) owing to the possibility of analytically representing experimental p by various multipole techniques (Hirshfeld, 1971;Stewart, 1976;Hansen & Coppens, 1978), allowing analytical evaluation of the Laplacian, V2p, and gradient vector field of p. Although it seems that there is no fundamental restriction on the applicability of topological analysis to experimental p (Tsirelson, 1996), the numerical results in this case are affected by the thermal atomic motions in crystals (Tsirelson, 1996;Gatti, Bianchi, Destro & Merati, 1992) and in fact refer to the mean thermal nuclear distribution. In this situation, the only way to evaluate indirectly to what extent quantitative values of topological characteristics depend on the atomic thermal motion effects may be a comparative topological analysis of theoretical and experimental charge densities simultaneously.…”