“…4), these theories have known considerable progress during the last three decades . Recent developments in crystal-and ligand-field theories have been chiefly achieved in five (not independent) directions: (i) the development, in algebraic and diagrammatic forms, of irreducible tensor methods for the strongfield model 11 -3,7,14,16], the weak-field model [5,6,8,15], and the medium-field model [9, 141; (ii) the adaptation of irreducible tensor methods to crystal-field approaches of photoelectron spectroscopy of partly filled shell ions systems [18,191; (iii) the consideration of chains of groups involving both finite and Lie groups [5-171; (iv) the use of Lie-like and second quantization techniques for the classification of state vectors and operators [6,7,10,15,17]; and (v) the introduction of more and more refined Hamiltonians [9,20,21] and more and more refined bases (as for example MO-LCAO bases) [22-241. To be complete, we should also mention the development of additive submodels (angular overlap model [25], superposition model 1261, and point-charge and point-dipole electrostatic model [U]) to interpret the global crystal-field parameters and thus to approach the metal-ligand bond in a semiempirical way. Our interest in this paper is directed towards global (or symmetry-adapted) rather than local (or additive) contributions to the crystal-field Hamiltonian so that we shall not deal with additive submodels.…”