The use of phenyl 2-pyridyl ketoxime (PhPyCNO)/X(-)"blends" (X(-) = OH(-), Cl(-), ClO(4)(-)) in copper chemistry has yielded trinuclear clusters that have been characterized as inverse-9-metallacrown-3 accommodating one or two anions. The magnetic behaviour has shown a large antiferromagnetic interaction. The discrepancy between the Brillouin curve and the experiment has been assigned to the influence of the antisymmetric interaction. By introducing in the magnetization formula the antisymmetric terms derived from the fitting of the susceptibility data the simulated curve become almost superimposable to the experimental one. From the EPR findings it has been shown that the compound [Cu(3)(PhPyCNO)(3)(OCH(3))(Cl)(ClO(4))] () having isosceles magnetic symmetry or lower (delta not equal 0), the antisymmetric exchange is important (G not equal 0) and DeltaE > hv. The structures of the two complexes have been determined by single-crystal X-ray crystallography.