[en] It was shown that the 'equatorial' a-emission of ternary fission can be considered as the a-decay of the double giant dipole resonance. The paper shows now that the equatorial ¹⁰Be emission can be interpreted as the ¹⁰Be-decay of the DGDR. But how can the 'polar' a-emission in ternary fission, or the isotropic emission of Be in incomplete fusion reactions such as ²³²Th + ⁴He at 200 MeV, be explained? The paper recalls that the formation of a ¹⁴C cluster in ¹⁴⁶Ba fragments is probably responsible for the recently discovered second mode of fission of ²⁵²Cf, and assumes, in analogy with the 1⁴⁶ Ba case, that less fragile clusters than carbon clusters are formed in other fission fragments, e.g. O-, Ne- (etc.) clusters in Ce-, Nd- (etc.) fragments, from the valence nucleons of their ¹³²Sn core, and that these clusters can be ejected in internal energy-rich collisions of vibrating complex fragment configuration. Several arguments are presented in favour of this new hypothesis. Both modes of cluster emission are new disintegration modes of the fission fragments themselves