It is well known that the predecessor of Old Norse (ON) must have gone through a process of final devoicing at some point in time. As pointed out by O. Grønvik and H. Bjorvand, among other scholars, this follows quite clearly from internal reconstruction of Old Norse. Forms like 3. sg. pret. ON batt ‘bound’ and sprakk ‘sprang’ from *band (cf. Gothic band) and *sprang (cf. Old High German sprang) can be understood only by assuming an intermediate stage *bant, *sprank with final devoicing and subsequent assimilation, i.e. *band > *bant > batt and *sprang > *sprank > sprakk. However, many details about this devoicing process remain unclear: When it happened, which sounds it affected and – most importantly – which conclusions, if any, it allows us to draw for the reconstruction of the phoneme system of Proto-Norse. This is of interest not least because of potential ramifications for the phonetic character of Runic (< *-z). To shed some light on the issue of final devoicing in Proto-Norse, it will be necessary to investigate whether the attested Runic inscriptions contain any indications of it. As for the chronology, it is necessary to look at the evidence from internal reconstruction of Old Norse and attempt to establish a relative chronology with respect to other sound laws. Furthermore, the talk will include a brief general discussion of devoicing, including a comparison with final devoicing in Gothic.