Aromaticity is one of the most important concepts in chemistry and refers to planar cyclic hydrocarbon molecules that exhibit delocalized p-bonds and unusual stability, such as benzene. This concept has been extended to metal-substituted organic molecules, [1,2] as well as main group organometallic complexes. [3][4][5][6] The recent discovery of aromaticity in allmetal clusters, for instance [Al 4 ] 2À in [MAl 4 ] À (M = Cu, Li, Na), [7] has led to a flurry of research activities [8][9][10][11][12][13][14][15][16] and predictions of new aromatic metal clusters, [17][18][19][20][21][22][23] among which is a class of interesting cyclic species containing Cu. [20,23] While main group clusters can give rise to s-and paromaticity, transition-metal-containing clusters can exhibit d-orbital aromaticity or, more interestingly, d-aromaticity due to d bonding interactions. However, d-orbital aromaticity requires significant d-d bonding interactions. Unlike valent s or p orbitals, d orbitals are spatially more contracted, and their tendency to participate in chemical bonding depends strongly on the position of the transition metals in the periodic table and their coordination environments. The predicted aromaticity in the cyclic Cu-containing species, [20,23] for example, has not been verified experimentally. More promisingly, d-orbital aromaticity is expected to be found in early or 4d/5d transition metal systems, where strong d-d interactions are known. Here, we report experimental and theoretical evidence of d-orbital aromaticity in two early 4d and 5d transition metal oxide clusters, namely [M 3 O 9 ] À and [M 3 O 9 ] 2À (M = W, Mo).