hrǫnduðr alinbranda ‘distributor of arm-flames [GOLD > GENEROUS MAN]’: This phrase could be an apostrophe (so Skj B), and the Context provides an interlocutor in Sigvatr. However, it seems preferable to take it as a subject in apposition with konungr ‘king’ (cf. NN §2010F; ÓHLeg 1982), especially since sendi is 3rd pers. sg. The base-word hrǫnduðr appears to be an agent noun from hrinda ‘to throw, cast’, though formally derivation from a *hranda would have been expected (Meissner 319; LP: hrǫnduðr (hrandaðr)). The form hrandaðr is suggested by the mss, but normalisation to hrǫnduðr is required since the ‑aðr variants in this class of noun are late analogical forms (ANG §137 Anm. 3, §397), and cf. hrǫnduðr in Anon Þul Sverða 3/1III. The first element of the determinant is problematic, and no ms. has the exact reading alin- adopted here. Mss 73ax, NRA52 and Flat all suggest arin- from arinn m. ‘hearth’, and arinbrandr ‘hearth-flame’ is an intelligible cpd, but hrǫnduðr arinbranda makes little sense as a kenning. (a) Kock emends to alinbranda ‘arm-flames’ (NN §3052B), which gives good sense, and this is adopted here. (b) Finnur Jónsson in Skj B prefers DG8’s reading ǫlun (ms. ‘alun’), which he suggests is a variant form of ǫln ‘(lower) arm’ (LP: ǫlunbrandr), often used in kennings for ‘gold’. This too results in a gold-kenning and thus ‘generous man’. (c) The first element in DG8 is taken in ÓHLeg 1982, 138 as ǫlunn ‘fish, mackerel’, giving ‘fish-flame [GOLD]’. However, this cannot be paralleled, as the usual patterns for gold-kennings are ‘fire of the water’ or ‘land of the serpent’, not ‘fire of the fish’. (d) Of the other ms. readings, 71x’s armbranda gives excellent sense, but leaves the line with only five syllables, not the required six, and Tóm refashions the line to Týr branda handleggjar ‘Týr of the flames of the hand-limb [ARM > GOLD > GENEROUS MAN]’. Both look like scribal attempts to make sense of a garbled text.