AbstractColloid formation is discussed as a possible pathway for the radionuclide release from a nuclear waste repository. An assessment of the colloid relevance on radionuclide migration requires insight into the possible colloid generation mechanisms, their stability and mobility under given groundwater conditions. In various experiments dedicated to the investigation of nuclear waste form behaviour in contact with groundwater, colloidal species are observed mainly for the tri- and tetravalent actinides even in rather high ionic strength solutions where destabilization of colloids is expected. Experimental evidence of laboratory and field studies suggests colloid instability in saline groundwater with time. Groundwater of low ionic strength and high pH enhances colloid stability, as demonstrated in various laboratory and field experiments. The results of an in situ colloid characterization study at the Äspö hard rock laboratory in Sweden are discussed as an example. The mechanism of radionuclide interaction with colloids and notably the reversibility of the radionuclide-colloid binding are other key issues of colloid studies. Kinetic stabilization of the radionuclide binding to colloids may lead to a considerable enhancement of the colloid-mediated radionuclide migration. Substantial kinetic inhibition of actinide dissociation from humic colloids has been established by studying the behaviour of natural humic colloid borne U, Th and rare earth elements. Such behaviour might be explained by the incorporation of these elements into inorganic nanoparticles stabilized by humic coating. Spectroscopic evidence for the occurrence of actinide incorporation into colloidal structures is briefly discussed and the importance of considering the kinetics involved in all kinds of colloidal processes is emphasized. The enhanced migration of tri- and tetravalent actinide ions has been observed recently in various in situ dipole tests at the Grimsel hard rock laboratory in Switzerland. Such experimental findings underline the necessity of further studies on the colloid influence on actinide mobility. Moreover, an improved understanding of colloid-rock interaction mechanisms is required, which is essential for the description and prediction of colloid filtration processes.