The spontaneous decay of the K ϭ16 ϩ , 31-yr 178 Hf m2 isomer has been investigated with a 15-kBq source placed at the center of a 20-element ␥-ray spectrometer. High-multipolarity M 4 and E5 transitions, which represent the first definitive observation of direct ␥-ray emission from the isomer, have been identified, together with other low-intensity transitions. Branching ratios for these other transitions have elucidated the spin dependence of the mixing between the two known K ϭ8 Ϫ bands. The M 4 and E5 ␥-ray decays are the first strongly K-forbidden transitions to be identified with such high multipolarities, and demonstrate a consistent extension of K-hindrance systematics, with an inhibition factor of approximately 100 per degree of K forbiddenness. Some unplaced transitions are also reported. 178 Hf, has angular momentum and parity I ϭK ϭ16 ϩ (K is the angular-momentum projection on the body-fixed symmetry axis͒ and a half-life of 31 yr ͓2-5͔. Its fourquasiparticle structure is based on a broken neutron () pair and a broken proton () pair, each contributing I ϭK ϭ8 Ϫ . The exceptionally long lifetime of the isomer arises not only because of its high K value and the associated hindrance caused by the K-selection rule, but also because it lies lower in excitation energy than any other states of spin 14 or higher. This yrast trap is forced, therefore, to decay by transitions with both high multipolarity and a large change in K.On account of its long half-life and high excitation energy, the 178 Hf m2 isomer has attracted considerable interest and experimental investigation. For example, enriched samples have been studied by laser hyperfine spectroscopy ͓6͔, and there is evidence that the isomer can be Coulomb excited in a multistep process from the I ϭK ϭ0 ϩ ground state ͓7͔. A more controversial observation is the stimulated decay of the isomer, induced by x rays and synchrotron radiation, which has been reported by Collins et al. ͓8-11͔, but refuted by Ahmad et al. ͓12,13͔.The high-spin level structure of 178Hf has been studied extensively ͓3-5,14͔, but basic knowledge gaps remain. Surprisingly, radiations emitted directly from the isomer itself are not yet well established. The problem arises, in essence, because 99.9% of the isomer decay proceeds through a highly converted, 13-keV E3 transition (16Ϫ ) which has not so far been directly detected. The suggestion by Khoo and Lo "vho "iden ͓4͔ that the isomer is at 2447.4 keV, deduced from its population from higher-energy states, is in conflict with the singles conversion-electron data of van Klinken et al. ͓5͔ for a 0.1%, M 4 decay branch (16Ϫ ) which implies an excitation energy of 2446.0 keV. Although the 1.4-keV difference in proposed isomer energies is small, it is well outside statistical uncertainties, and represents a basic difficulty with the current understanding of the isomer and its decay modes.In the present ␥-␥-coincidence study, the M 4 ␥-ray transition has been clearly identified, confirming the interpretation of van Klinken et al. ͓5͔. We...