The modification of different zirconium propoxide and hafnium propoxide precursors with 2,2,6,6,-tetramethyl-3,5- heptanedione (Hthd) was investigated by characterization of the isolated modified species. The complexes [Zr((OPr)-Pr-n)(3)(thd)](2), [Zr((OPr)-Pr-n)((OPr)-Pr-i)(2)(thd)](2), Zr((OPr)-Pr-i)(thd)(3), [Hf((OPr)-Pr-n)(3)(thd)](2), and Hf((OPr)-Pr-i)(thd)(3) were isolated and characterized. The structure of the n-propoxide analogue of Zr((OPr)-Pr-i)(thd)(3) could not be refined, but its existence was clearly demonstrated by XRD and H-1 NMR. The modification of the propoxide precursors involves mono- and trisubstituted intermediate compounds and does not involve a disubstituted compound; thus, the commercial product that is claimed to be "Zr((OPr)-Pr-i)(2)(thd)(2)" and is most commonly used for the MOCVD preparation of ZrO2 does not exist. No evidence was found for the presence of such a compound in either zirconium- or hafnium-based systems. Formation of the dimeric hydroxo-di-thd-substituted complex, [Hf(OH)((OPr)-Pr-i)(thd)(2)](2), which could be isolated only for hafnium-based systems, occurs on microhydrolysis. All heteroleptic intermediates are eventually transformed to the thermodynamically stable Zr(thd)(4) or Hf(thd)(4). The compounds obtained from isopropoxide precursors showed a higher stability than those with n-propoxide ligands or a combination of both types. In addition, it is important to note that residual alcohol facilitates the transformation and strongly enhances its rate. The unusually low solubility and volatility of M-IV(thd)(4) has been shown to be due to close packing and strong van der Waals interactions in the crystal structures of these compounds.