[1-4]: Line 1 makes sense as it stands in the ms. and, since drífa ‘to drive, pelt’ can be used impersonally with the dat. (LP: 2. drífa 1), it is completed by the dat. sg. hagli þinurs ‘hail of the bow [ARROWS]’ at the end of l. 4. For the rest, it is not feasible to solve the problems of this helmingr without multiple emendations, and all interpretations have been purely tentative (Finnur Jónsson 1886b, 328; Reichardt 1928, 204). (a) This edn adopts vargr for ‘vargi’, gleypti for ‘grim’, and ná for ‘a’ in l. 2; auðfundit for ‘aud kundu’ and virði for ‘virdri’ in l. 3; and valgagls for ‘vagll agls’ and þinurs for ‘tímis’ in l. 4. Varð ‘was, became’ (l. 3) is guaranteed by rhyme and alliteration, and provides an auxiliary to the widely accepted emendation auðfundit ‘easily found’. The subject of ‘was easily found’ is l. 3 virði, which is here construed as nom. sg. of n. virði ‘meal’, though it could possibly be (instr.) dat. sg. of m. verðr ‘meal’ (Finnur Jónsson 1886b, 329; cf. LP: virði). For the kenning virði/verðr valgagls ‘meal of the slaughter-goose [RAVEN/EAGLE > CORPSES]’, cf. Eskál Lv 2a/2 verðr ulfs and ÞTref Hrafn 3/3V (Eb 33) virði ulfs, each signifying ‘meal of the wolf [CORPSES]’. Ms. ‘grim a margan’ in l. 2 then remains to be accounted for. Varð at the beginning of l. 3, as a finite verb, cannot occupy other than position 1 or 2 in its clause (Kuhn 1983, 190-1). Because other solutions, including the otherwise attractive one proposed by Kock (see (c) below), infringe this rule, it seems necessary to assume that l. 2 also contains a finite verb. Since varg- is guaranteed by rhyme and alliteration and cannot be a verb, the only recourse is to emend grim, and here gleypti ‘swallowed’, from gleypa, is tentatively suggested, as one possibility among several. The noun ná (m. acc. sg. of nár ‘corpse’) is also conjectural, but correctly fills the syntactic, semantic and metrical slot in a line that, as it stands in the ms., is metrically deficient; the adj. margan, if correct, presupposes a m. object. The principal alternatives are as follows. (b) Finnur Jónsson (1886b, 329; Skj B) proposed reading grimmu ‘savage’ for ms. ‘grim a’ and linking it syntactically with hagli ‘hail’ in l. 4: Grimmu þrimu hagli dreif at Viðris veðri ‘A savage hail of battle [ARROWS] pelted in the weather of Viðrir [BATTLE]’. The other clause is read as mǫrgum vargi varð auðfundit valgagls virði ‘the meal of the slaughter-goose [RAVEN/EAGLE > CORPSE] was easily found for many a wolf’. This solution entails the additional emendation of margan to mǫrgum, qualifying vargi, and produces a tripartite division of l. 2, with mǫrgum vargi interrupted by an element from the first clause. (c) The text proposed by Kock (NN §430, cf. §303B) produces virði valgagls varð auðfundit vargi á morgun ‘the meal of the slaughter-goose [RAVEN/EAGLE > CORPSES] was easily found for the wolf in the morning’, in part on the basis that the battle took place in the morning. Kock also reads grimt, emended from grim, but it is unclear whether this is construed as adj. ‘savage’ or adv. ‘savagely’. The emendations are slight (‘grim’ to grimt, ‘margan’ to morgun) and/or generally accepted (‘aud kundu’ to auðfundit, ‘virdri’ to virði), but this solution has the verb in a proscribed position.