[Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(2)(mu-CO)(2)] reacts with CNMe in refluxing xylene when A-B = R,S-CH(NMe(2))CH(NMe(2)) or CH2C(O) to give [Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(4-n)(CNMe)(n)] (n = 1 or 2). This thermal reaction does not take place when A-B = R,R/S,S-CH(NMe(2))CH(NMe(2)), CH2CH(NMe(2)), CHC(NMe(2)), CH2CH(OH) or CH2CH2, but photolysis allows the preparation of [Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(4-n)(CNMe)(n)] (n = 1-3), and [Fe-2{eta,eta-C5H4CH2CH(OH)C5H4}(CO)(3)(CNMe)] may be prepared by the LiAlH4-reduction of [Fe-2{eta,eta-C5H4CH2C(O)C5H4}(CO)(3)(CNMe)]. All the complexes react with MeI, EtI or MeOSO(2)CF(3) to give derivatives containing one or two mu-CNMe(2)(+) ligands. H-1 NMR studies show that all the neutral complexes are fluxional, with processes taking place within the C5H4-A-B-C5H4 ligands comparable to those observed in the tetracarbonyl precursors as well as CO/CNMe site exchange. This last is more restricted than in the [Fe-2(eta-C5H5)(2)(CO)(4-n)(CNMe)(n)] counterparts, so that two non-interconverting [Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(2)(CNMe)(2)] species are possible, and have been separated in most instances; one has the structure [Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(CNMe)(mu-CO)(mu-CNMe)] and the other is [Fe-2{eta,eta-C5H4-A-B-C5H4}(CO)(2)(mu-CNMe)(2)], which is in equilibrium with [Fe-2{eta,eta-C5H4-A-B-C5H4}(CNMe)(2)(mu-CO)(2)] in solution. Their formation provides unequivocal evidence for the validity of the Cotton-Adams rules on fluxionality in [Fe-2(eta-C5H5)(2)(CO)(4)] derivatives. The structure of R,S-[Fe-2{eta,eta-C5H4CH(NMe(2))CH(NMe(2))C5H4}(CO)(2)(mu-CNMe)(2)] has been determined by an X-ray diffraction study.