27. Síð ok Víð, Sækin ok Eikin, Svöl ok Gunnþró, Fjörm ok Fimbulþul, Rín ok Rennandi, Gipul ok Göpul, Gömul ok Geirvimul, þær hverfa um hodd goða, Þyn ok Vín, Þöll ok Höll, Gráð ok Gunnþorin.

1996, Carolyne Larrington Translation: 27. Sid and Vid Sækin and Eikin Svol and Gunnthro, Fiorm and Fimbulthul, Rin and Rennandi, Gipul and Gopul, Gomul and Giervimul, they flow round the realm of the gods, Thyn and Vin Tholl and Holl, Graf and Gunnthorin

 

  

In some editions this verse is broken into verses 27 & 28, lines 1-4. If these are broken into separate verses, they appear as:   27. Síð ok Víð, Sækin ok Eikin, Svöl ok Gunnþró, Fjörm ok Fimbulþul, Rín ok Rennandi, Gipul ok Göpul, Gömul ok Geirvimul, þær hverfa um hodd goða,   28. Þyn ok Vín, Þöll ok Höll, Gráð ok Gunnþorin. Vína heitir ein, önnur Vegsvinn, þriðja Þjóðnuma,   When the line þær hverfa um hodd goða is interpreted to mean "these flow round the realm of the gods" as in the Larrington translation above, the translator following Snorri, presumably assumes that these rivers flow upwards from Niflheim to the districts of the Æsir, which Snorri places in Midgard (Gylfaginning 15) or in heaven, if one accepts the commonly held notion that Asgard in Snorri’s Edda refers to a celestial city. However, the verse literally says that these rivers “wind around the hoard of the gods” (þær hverfa um hodd goða). What this hoard is and where it is located are not stated. The phrase is otherwise unknown in Eddic poetry. All other instances of the word in Eddic poetry refer to actual treasure. In Gylfaginning 39, Snorri paraphrases this verse saying: “Sid, Vid, Sekinm Ekinm Svol, Gunnthro, Fiorm, Fimbulthul, Gipul, Gopul, Gomul Giervimul: these flow through where the Æsir live,” (Anthony Faulkes translation). From the duplicate names and order of the rivers, it is clear that Grímnismál 27 is Snorri’s source of this information, and that he interprets the phrase “þær hverfa um hodd goða” as “these flow through where the Æsir live”.   The word hodd is properly interpreted as "treasure", "hoard". The Cleasby/Vigfusson Dictionary translates it as:   HODD, n. pl., -- the m. pl. hoddar, which occurs twice in verses of the 13th century (Sturl.), is a false and late form; [Ulf. huzd = GREEK; A. S. hord; Engl. hoard; O. H. G. hort] :-   "I. a hoard, treasure, only in poetry; hodd blóðrekin, Hkv. 1. 9; hodd Hniflunga, Germ. Niebelungen hort, Akv. 26; hodd (acc. pl.) ok rekna brodda, Fagrsk. (in a verse); góðum hoddum, Fas. ii. 312 (in a verse); granda hoddum, mæra e-n hoddum, Lex. Poët.; kveðja hodda (gen.), Eb. (in a verse); oddar roðnir hoddum, Arnór; halda hoddum fyrir e-m, Ísl. ii. 224 (in a verse). 2. poët. phrases, as hoddum haettr, hodda (gen.) brjótr, njótr, stökkvir, stríðir, þverrir, the breaker ... of gold, a princely man: as also poët. cornpds, hodd-brjótr, -beiðandi, -finnandi, -geymir, -glötuðr, -lestir, -lógandi, -mildingr, -sendir, -skati, -spennir, -stiklandi, -stríðandi, -sveigir, -sviptir, -veitir, -vönuðr, all epithets of a lordly, princely man: so of women, hodd-gefn, -grund; the nouns, hodd-mildr, -örr, = liberal; hodd-dofi, a, m. stinginess; and the mythical pr. names Hodd-mímir, Hodd-dropnir, 'gold-dripping,' Sdm.   "II. a holy place, temple, sanctuary, where the holy things are hoarded; of this sense, which occurs in Heliand (Schmeller), the Gm. 27 is the single instance left on record, see Bugge's note to l.c. in his Edda, p. 81."       Notice that the second definition "a holy place, sanctuary" is derived solely from Grímnismál 27! I would suggest that this is another instance where the dictionary definition of a word in an obscure line of Eddic poetry has been made to conform to Snorri's interpretation of the passage. In all other instances occurring in Eddic poetry, the word refers solely to a "hoard of treasure". The "hoard of the gods" appears to be a reference to the treasures hidden in Hoddmimis holt, Hoard-Mimir's grove, beneath the central root of Yggdrasill. Mimir is well-known as a collector of valuable treasures. Among the 'treasures' kept in his grove are the human beings Lif and Lifthrasir who will repopulate the world after Ragnarök.      1923 Henry Bellows in his The Poetic Edda as “Grimnismol: The Ballad of Grimnir”:     The entire passage from stanza 27 through stanza 35 is confused. The whole thing may well be an interpolation. Bugge calls stanzas 27-30 an interpolation, and editors who have accepted the passage as a whole have rejected various lines. The spelling of the names of the rivers varies greatly in the manuscripts and editions. It is needless here to point out the many attempted emendations of this list. For a passage presenting similar problems, cf. Voluspo, 10-16. Snorri virtually quotes stanzas 27-29 in his prose, though not consecutively. The name Rin, in line 3, is identical with that for the River Rhine which appears frequently in the hero poems, but the similarity is doubt less purely accidental.   Excerpts from:   Christopher Hale,  “The River Names in Grimnismal 27-29”, Edda: A Collection of Essays, 1988.   “It is a well-known fact that Grímnismál as it has been preserved in its present form is defective. Occasionally a line seems to be missing and scattered throughout are interpolations, many of which do not fit into the ljóðaháttr metre of the main part of the poem. One of the longest of the probable interpolations is stanzas 27-29, which contains a list of river names. This is, in fact, the only section of the poem which contains a list of any place names.  In this paper I should like to examine these lines, offer interpretations of the name contained in them and give parallels, where possible, to them in the Scandinavian area. Based on the results found here, I should like to offer some suggestions as to the possible origin of these lines and a relationship between them to later forms of the verse.   …Of the total of forty-three river names (árheiti) mentioned in the mythological poems of the Poetic Edda, thirty-six of them occur in GRM 27-29 alone. In addition, thirty-one of these are mentioned in Gylfaginning and thirty-one also in the nafnaþular in Skaldskaparmal in Snorra Edda. Besides this, a few are noted in other medieval sources.   Four other rivers occur in GRM besides those in stanzas 27-29. Þund is noted in stanza 21 as being difficult to cross, and Kormt and Ormt and Kerlauger in stanza 30 are rivers which Thor must cross each day on his way to the place of judgment by the ash Yggdrasil. It is possible that stanzas 27-29 were interpolated to enlarge the word vötn (water) in stanza 26, or that if stanza 30 is original, the names in it could also have called them forth.   The names in this list have so far been only spottily researched in scholarly writing. While no work has appeared before that treats all of them in any depth, a number of editions of both the Poetic and Snorra Edda, as well as several etymological dictionaries, make a greater or lesser attempt to explain some of them. Furthermore, from time to time, several have been dealt with in articles which have appeared in various journals and collections. Magnus Olsen, for example, has discussed a few in Edda- og skaldekvad, VII, and others have been referred to by different scholars in articles in such journals as Namn och Bygd, Maal og Minne and Arkiv for nordisk filologi in conjunction with treatments of Scandinavian place names. In this connection, mention should be made of place-name studies such as Oluf Rygh's Norske Elvenavne (NE) and Norske Gaardnavne (NG), Per Hovda's Norske elve namn and Elof Hellquist's Studier ofver de svenska sjonamnen which frequently offer parallels to their subject matter in the Eddic material. Also, attempts have been made to identify these rivers with actual ones in the North European area."   The references for the Poetic Edda and Snorra Edda (SnE) are to Finnur Jonsson's editions, De gamle Eddadigte, and Edda Snorra Sturlusonar. SnE I refers to Gylfaginning, and SnE II to Skaldskaparrnal.   The river names are grouped according to the passages in which they are found in Grm, and in each group additional information regarding the rivers they refer to is indicated; as well, mention is made of other medieval sources outside the Poetic Edda where these names are found. The manuscripts have been consulted either in the original or in photographic reproduction, and the forms of the river names in them are given and discussed, where relevant, under the heading of each name. The manuscripts and the abbreviations used for them in the text are:   (1) Codex Regius of the Elder Edda, Gks 2365 4to, ca. 1270 (R). (2) AM 748 I 4to, early 1300s (A). (3) Codex Uppsaliensis, Uppsala DO 11 4to, early 1300s (U). (4) Codex Regius of the Younger Edda, AM 2367 4to, ca. 1325 (r). (5) Reykjabok, AM 468 4to, 1300-1350 (Rb). (6) Codex Wormianus, AM 242 fol., mid-1300s (W). (7) AM 757 4to, ca. 1400 (B). (8) AM 748 II 4to, ca. 1400 (C). (9) Codex Trajectinus, Utrecht nr. 1374, ca. 1600 but copied from a late 13th century ms. (T).   Grímnismál 27   Síð ok Víð, Sækin ok Eikin, Svöl ok Gunnþró, Fjörm ok Fimbulþul, Rín ok Rennandi, Gipul ok Göpul, Gömul ok Geirvimul, þær hverfa um hodd goða, All the rivers listed in Grm 27, except for Rín and Rennandi, flow from the well Hvergelmir in the middle of Niflheimr through the districts of the Æsir, according to SnE I 25, where they are listed in the same order as in Grm 27. Svöl, Gunnþro, Fjorm and Fimbulpul are mentioned in that order also in SnE I 4. (Gylf. 39) Furthermore, all but Gipul and Gopul are found in the nafnaþulur in Skaldskaparmal but in a different order from that in Grm 27, and mixed in with other river names.   Síð:   Grm 27: Síþ R; Sið A. SnE I 25: Sið U; r; W; Sid T. SnE II 89 (479-84): sið A; r; sid B; C; sid T.   See the following name.   Víð:   Grm 27: viþ R; við A. SnE I 4: við-U: vie r; Vid W; uid T. SnE I 25: við U; r; við W; vid T. SnE II 89 (479-84): við A; r; vid B; víd C; uid T.   …It is difficult to determine from the manuscript forms alone whether the vowels here are originally long or short, as the acute accent over Síð and Víð in R may denote stress instead of length, and the accents over the forms in B and C are undoubtedly graphic, as these manuscripts belong to a relatively late period. At any rate, it appears that these two names have in-rhyme, and that, accordingly, the vowels in both should have the same quantity. If the vowel in the first name is short, there is no reasonable interpretation likely for it. Regarding the second name, a short vowel in it could connect it to Old Norse (ON) víð f. 'withy,' 'wand,' but this is also unlikely. I feel, therefore, it is probable that the vowels in both are long and that Síð and Víð are nominalizations of the feminine singular forms of the ON adjectives síðr 'long, hanging' and víðr 'wide' respectively.   There appears to be a number of parallels to Síð. Svante Strandberg interprets the Swedish name Sibro as containing the adjective sid 'low-lying, marshy' and related to the ON adjective síðr,  and Gosta Franzen says that Sidus is probably formed from a river name *Sidha with a similar derivations. A related word seems to be found in the English river name Sid which, "runs in a narrow valley with very steep sides and high ridges on both sides," in the Swedish lake name Sidlangen (Hellquist, pp. 525-26), and possibly also in the Norwegian farm name Sævik (NG, XV, p. 326). The Eddic name may have a similar meaning.   Regarding Víð there are parallels in several Norwegian river names such as Via, a brook in Hjelmeland, Rogaland which spreads out over its banks after a heavy rain (Norsk stadnamnarkiv, or NSA), as well as Barvio (Hovda, 1966, p. 23). The Swedish lake name Vidöstern (Hellquist, pp. 705-06) and the Norwegian Vidflaa could also be mentioned. Via and Vibækken, which Rygh derives from ON vidja f. 'withy,' (NE, p. 296), may also contain the ON adjective víðr. Semantically identical to the Eddic name is the Norwegian river name Breia from the ON adjective breiðr 'broad, wide' (NE, p. 22).   Sœkin:   Grm 27: sökin R; sækin A. SnE I 25: Sækin U; Sekin r; sekin W; Sekin T. SnE II 89 (479-84): sækin A; sekin r; seek/in C; sekin T.   All the manuscripts are later than the merger of ON (œ) and (æ.) The name is probably the feminine singular of a nominalization of an adjectival derivation with the suffix Germanic *-ina of the ON verb sœkja 'to seek, proceed.' 11 Sœkin would thus mean something like 'the forward-rushing one' (Gering, p. 198). One may compare this to etymologically similar Norwegian river names such as Sokna, derived from an ON sókn, and the semantically similar ones formed from *fausa, ganga and skrio- (NE, pp. 238-39, 47, 64, 227). See also Fjorm below.   Eikin (Ekin):   Grm 27: eiken R; ækin A. Sn/25: ækin U; ekinn r; ekin W; T. SnE II 89 (479-84); ekin A; ekin r; eikin C; ekin T.   In the oldest manuscript (R) the name is written with a diphthong, and if this diphthong is original, Eikin is probably a nominalization of the feminine singular of the ON adjective eikinn 'savage' (Gering, p. 198), formed in a way similar to that of Sœkin. In Nordfjord, Norway, there is a river called Eikjola, the first element of which Per Hovda (1966, p. 34) relates to Norwegian dialect eikja 'to quarrel, dispute,' a derivative of ON eikinn, and this would support the above interpretation of the Eddic name. Another possibility is that it is connected to the ON noun eik f. 'oak' which seems to occur, as well as other appellatives for trees, in several river names. However, all the other manuscript forms, except for C, have no diphthong. If Ekin is then to be the original form, it is probably a nominalization of the feminine singular of the past participle of the ON verb aka 'to move, drive.' If this is indeed the case, there are a number of parallels in Nordic river names: for example, *Aka (NE, p. 3) and Akurda (Hovda, 1966, p. 17) in Norway and Agebcek in Denmark.   Svöl:   Grm 27: svavl R; svöl A. SnE I 4: kvol (!) U; Svavl r; Svol W; Suol T. SnE 125: Svoll U; Svavl R; svöl W; Suaul T. SnE II 89 (479-84): svöl A; -svol r; suól B; suaul C; suol T.   Probably this is a nominalization of the feminine singular of the ON adjective svalr 'cool.' A root related to this adjective seems to be found in a number of Norwegian place names, possibly formed from a river name * Svala (NE, pp. 254-55), as well as the lake name Suluvatnet (Indrebø, 1924, pp. 193-94). Another parallel is probably the Swedish Svalen (Hellquist, p. 590). Semantically similar are names such as Kaldbækken (NE, p 124). Cleasby and Vigfusson (p. 780) identify Svö1 as the Swale River in England, but this is unlikely.   Gunnþro:   Grm 27: gvnnþro R; gvnnþro A. SnE 14: gvndro U; gvnnþra r; Gvnsþra W; gunþre T. SnE I 25: gvndro U; Gvnnþro r; gvnnþro W; gunnþro T. SnE II 89 (479-84): gvnnþro A; gvnnþro r; gunn ... B; gunnþor C; gunnþro T.   The first element appears to be the stem form of ON gunnr f. (with non-radical -r) 'war, battle (poet.).' A number of scholars (e.g. Jonsson, 1900, p. 222, and Gering, 1927, p. 198) seem to have considered the final element as a feminine derivation of ON þrá n. 'obstinacy, defiance, (<þró), and this interpretation would explain the manuscript forms written both -þra and -þro. Another possibility would be to connect it to ON þra f. 'yearning, longing.' However, it might also be ON þró f. 'trough' and refer to the course of the river, the few manuscript spellings with -þra then being due to scribal confusion with one of the first two words. Per Hovda has mentioned to me the Norwegian river Trona in Strand, Ryfylke, which flows for a few kilometers through a narrow passage, and this name is possibly etymologically parallel to the final element of the mythological one. Cf. also Trovatn and Trodalen, formed from a river name *þró (NE, p. 275). With the first element being the stem form of ON gunnr f., I suggest Gunnþro could mean something like 'the one which travels swiftly or wildly in its course.' Cf. Fjorm below.   Fjörm:   Grm 27: fiorm R; fiörn A. SnE I 4: fiorni U; form r; fiorm W; form T. SnE I 25: fiorm U; r; W; T. SnE II 89 (479-84): fiörn A; fiörm r; fiórm B; form C; fiorn T.   Oluf Rygh connects the root fjarm- in names such as Fjermedal and Fjermestad with an ON river name * Fjorm, genetive Fjarmar, and relates it to the Norwegian dialect verb fjarma 'to travel quickly, rashly off,' mentioned by Ross. Per Hovda also notes that Fjermeros in Vest Agder, Norway, comes from a similar name. Magnus Olsen says that Fjorm is formally identical to Anglo Sexotv feorm f. 'feeding, provisions,' but that both this word and the Norwegian dialect fjarma are etymologically unclear. Semantically the meaning 'the rushing one' for Fjorm fits in well with other names in the group such as Sœkin (see above) and Rennandi (see below).   Fimbulþul:   Grm 27: fimbvlþvl R; -fimbvlþvl A. SnE I 4: firnbvlþvl U; funbvl. þvl r: firnbul/þul W; funbul þul T. SnE I 25: fimbvlþvl U; fimbvlþvl r; fimbul. bul W; fimbul T. SnE II 89 (479-84): fimbvlþvl A; fimþvl þyl r; fimbulþul B; fimbul þul C; fimbul þul T.   The first element, ON fimbul, appears only in compounds in poetry with the meaning 'mighty, great." Both Jóhannesson (p. 428) and de Vries (p. 626) mention the possibility that the final element -þul is related to Anglo Saxon (AS) gedyll 'breeze.' Hjalmar Falk notes the similarity of the Eddic river name to the Oðinsheiti Fimbulpulr 'the great skald' and suggests that the former was called forth by the latter, but with a slight change in the meaning of the final element. This he relates to ON bylja 'to mumble.' Fimbulþul could thus mean 'the one which makes a great rushing sound. ' Semantically similar Norwegian river names are Humla (NE, p. 110) and Kumra (Hovda, 1966, p. 67).   Rín:   Grm 27: rín R; A. SnE I 89 (479-84): rin A; rin r; rin B; C; rin T.   Rín is the ON form of the river Rhine which is frequently mentioned in other ON sources as well as in a number of the heroic Eddic poems (e.g., Brot af Sigurðarkviða, Sigurðarkviða hin skamma and Atlakviða), Probably here also the same river is referred to.   Rennand :   Grm 27: rennandi R;-ren/nndi A. SnE 1/89 (479-84): rænnandi A; rennandi r; ... de B; rennandi C; renanndi T.     This is a nominalization of the present participle (cf. Migandi and Hyggjandey" of ON renna 'to run, flow.' Cf. the river names in Norway formed from the verb renna and the noun laup (NE, pp. 190, 140). Magnus Olsen (1964, p. 15) interprets Rennandi as perhaps 'the one which is always free of ice.' Cf. also varma in Norwegian river names (NE pp. 291-92).   Gipul:   Grm 27: gipvl R; A. SnE / 25: gipvl U; r; W; gipul T.   Gipul is probably a feminine nominalization of a derivation in *-ula of a root *gip-. Cf. ON geipa 'to talk, talk nonsense,' Norwegian dialect gip 'mouth,' gipa 'to cause to yawn,' from an Indo-European *ghei : *ghi.21 From the same root probably comes the Norwegian river name Gipa (NE, pp. 72, 319), and the first element in the Norwegian lake name Gipetjerni is from the river Gipa (Indrebø, 1924, pp. 70-71). Semantically parallel is Gjeispa (NE, p. 69). See also Göpul below.   Göpul:   Grm 27: gavpvl R; -gopvl A. SnE I 25: giofvl U; gavpvl r; gopvl W; gaupul T.   These are formed in a way similar to that of the previous name from a root *gap- (ON gap n. 'opening,' gapa 'to yawn'). Etymologically parallel is probably the name of the group of Norwegian seters known collectively as Goppollen. O. Rygh (NG, IV, 1, p. 156) derives this name from Norwegian gople f. 'giant bellflower' (Campanula latifolia). However, according to Per Hovda, just west of this area a river flows through a narrow ravine, and this river quite likely once bore the same name as the seters. Gering and Sijmons (p. 198) say that Gipul and Gopul possibly refer to rivers with shores that fall off sharply. They also note that as gapa occurs in the meaning 'to scream,' Gopul may also be interpreted as 'the roaring one.'   Gömul:   Grm 27: gavmvl R; gömvl A. SnE I 25: gomol U; gavmvl r; gomvl W; gaumol T. SnE II 89 (479-84): gomvl A; gomvl r; gomul B; gaurnul C; gó-/ mul T.   The name is most likely a nominalization of the feminine singular of the ON adjective gamall 'old.' Gering and Sijmons (p. 198) say that it might signify an old river bed in contrast to a newly formed one and compare it to the German Alter Rhein. Cf. Gamlelva and similar names several places in Norway (NSA). There is also a possibility that Gomul might be related to ON gemlingr m. 'one-year-old sheep' and refer to a flooding river which swells up once a year.   Geirvimul:   Grm 27: geirvi/mvl R; gæirvimvl A. SnE I 25: geirvmvl U; geirvimvl r; geirrvmvl W; geir-/vimul T. SnE II 89 (479-84): gæirvimvl A; geirvimvl r; geir (vímul) B; geiruimul C; geiruimol T.   The first element is the stem form of ON geirr m. 'spear.' Regarding the final element, both Magnus Olsen and Ingeborg Hoff have discussed the district called Vimar or Vimir which they connect to the mythological river name. According to Olsen, the root in both these names is Vim- which he compares to Old High German uuimi, corresponding to the Latin scatebra (i.e. fluviorum) and vomens (i.e. ebulliens), as well as Icelandic vim, vím n. 'giddiness, a swimming in the head, wavering' and Norwegian dialect vima 'tumble, go as if confused.' Semantically, he feels the Eddic name is closest to vomens. Hoff derives these names from a root *uei 'turn around' and sees them as referring to the 'turning motion the eddies in the current make.' Geirvimul would thus mean something like 'the river which swarms with spears.' Martin Larsen mentions that the concept of rivers filled with pointed weapons has been common among the Nordic people and refers to "Haddingssagnet" in Saxo Grammaticus' Gesta Danorum. (Book 1)     Grímnismál 28   Þyn ok Vín, Þöll ok Höll, Gráð ok Gunnþorin. Vína heitir ein, önnur Vegsvinn, þriðja Þjóðnuma,   Vín flows from Hvergelmir according to SnE I 4, as do all the rest, listed in the same order as above (except that Vína, Vegsvinn and Þjóðnuma are mentioned later than the previous six) according to SnE I 25. Furthermore, all but Gráð and Gunnþorin are found in the nafnabulur (Vína twice) but in a different order from that in Grm 28 and mixed in with other river names.   Þýn (Þyn):   Grm 28: þyn R; þyn A. SnE I, 25: þyn r; W; T. SnE II 89 (479-84): þ yn A; þyn r; ... ýn B; þyn C; T.   The name appears in a spurious verse in Njáls saga (chapter 45) in a kenning for gold-þýnjar logs (written þyníar logs in Rb). Þýn seems to be eliminated from the version of SnE I 25 in U and the word fyri substituted for it after nefndar. Most scholars interpret it as 'the raging, thundering, roaring one' and compare it to AS þunian and the god's name 'Thórr' (þórr). According to Jóhannesson (p. 872), who agrees with this theory, the name is derived from the Indo-European (IE) roots *sten-, *ten- 'to thunder, rage, roar.' If this is correct, the vowel in þyn would be short. Þyn also occurs in the nafnapulur in SnE as a heiti for a coat of mail. In Norway, however, there are several river names such as Tya which appear to go back to an ON *þý with a long vowel (NE, p. 280). Sophus Bugge (NE, p. 339) interprets these names semantically as 'the ones which swell or have swollen up; the big ones' and relates them to Indic táviti 'have power,' tuvi- 'mighty' and Greek Taús 'big.' Cf. the Norwegian rivers Gro and Reisa (NE, pp. 79, 189). If Þýn has a long vowel, then I feel that it may be a derivation in -n of the above-mentioned root. Cleasby and Vigfusson (p. 780) have identified the Eddic name with the English river Tyne, while Ekwall (p. 426) says the latter is Celtic in origin.     Vín (Vin):   Grm 28: vin R; A. SnE I 25: vin- U; vin r; vin W; vin T. SnE II 89 (479-84): vin A; r; vin B; C; uin T.   A number of scholars have identified it with the Russian river Dvina (see Vína below)." This theory presupposes a long vowel. If Vín indeed has a long vowel, it could be connected to the IE root *-uei 'wither, dry (up)' (see Johannessen, p. 111), and related to those names which refer to rivers which dry up during the summer. Cf., for example, Torke (NE, p. 273). If the vowel is short, then the name could be connected to ON vin f. 'meadow' or vinr m. 'friend,' but this seems to me less likely.   Þöll:   Grm 28: þavll R; Þöll A. SnE I 25: -þoll U; þoll r; þöll W; þoll T. SnE II 89 (479-84): Þöll A; þavll r; þaull C; T.   The name is formally identical to ON þöll f. 'young fir tree.' Cf. Tollaaen and Tolga (NE, pp. 272, 337). Gering and Sijmons (p. 198), on the other hand, interpret it as 'the swollen one' (cf., Þýn, [þyn] above), probably relating it to the IE root * tal- 'grow, flourish,' from which also the appellative ON poll seems to be derived (cf., Johannessen, p. 426). Jan de Vries (p. 631), while noting Gering's and Sijmons' theory, mentions also the possibility, according to Holthausen that the name is connected to AS gedyllan 'to quiet.' Cf. the Norwegian rivers Logn and Stilla (NE, p. 147, and Hovda, 1966, p. 107). Cleasby and Vigfusson (p. 780) have identified it as the Scottish river Thuil.   Höll:   Grm 28: havll R; höll A. SnE I 25: holl U; boll (!) r; holl W; T. SnE II 89 (479-84); höll A; -höll r; -hóll B; haull- C; hóll T.     Höll is probably a nominalization of the feminine singular of the ON adjective hallr 'sloping.' O. Rygh mentions the farm name Hallen, derived from ON hallr, as possibly being named after the brook which flows by this place (NE, p. 90 and NO, XIV, p. 360). Cf. also Haldalen (NO, V, p. 157). Hellquist (pp. 212-14) lists several lake names in Sweden which may contain a root related to the same word. Semantically similar names of rivers are Bretta and Leina (NE, pp. 23, 141).   Gráð:   Grm 28: graþ R; gráð A. SnE I 25: grað U; r; W; grad T.   Quite possibly the name is a feminine derivation of ON gráðr m. 'hunger, greed.' Cf. Sylgr below. Another possibility is that it is related to ON gráði m. 'breeze curling the waves,' Modern Icelandic gráð f. Finnur Jonsson connects the name Gráðvik in Iceland to the latter word in the meaning 'bay where there is always a superficial (shallow) movement of the waves.'   Gunnþorin:   Grm 28: gvnnþorin R; -gvnnþorin A. SnE I 25: gvnþro U; gvnnþrainn r; gvnnþorin W; gunnþo- /rm T.     The final element is probably a feminine nominalization of a derivative in *-ina of the verb ON þora 'to dare' (cf., Mod. Ice. þorinn 'daring, bold' and Sœkin above), possibly here with a meaning such as 'to force through (i.e. in one's course).' The first element is likely the stem form of ON gunnr f. (with non-radical -r) 'war, battle (poet.),' and this has led some scholars to interpret the name as 'the one which is eager for or desirous of battle' (e.g., Gering, p. 198). Others have identified it with Gunnþro (e.g., Jonsson, 1900, p. 222). However, it seems to me more likely that Gunnþorin means something like 'the one which forces itself mightily along its course.' Cf. Gunnþro as well as the semantically similar Fjorm and Rennandi above.   Vína:   Grm 28: Vin á R; Vina A. SnE I 25: vina U; r; vína W; vina T. SnE II 89 (479-84): vina, vina A; vina, vina r; vína, vína B; C; vina, uma T.   Gering and Sijmons (p. 198) have interpreted this name, according to the form in R, as 'wine stream' (cf., ON vín n. 'wine' and ON a f. 'river, stream'). However, the form 'Vin a' in R is unique and probably due to a misunderstanding of the text on the part of the manuscript copyist. Most other scholars (e.g., Jonsson, 1900, p. 40, and Wilken, p. 275) have identified the mythological river with the Russian Dvina. Due to the lack of a better explanation, I tend to agree with these latter authors. Furthermore, the ON form of the Dvina River is Vína. Vína may be connected to the name Vín (see above for its possible interpretation).   Vegsvinn:   Grm 28: vegsvinn R; væg/svin A. SnE I 25: veglvn U; vog. svinn r; ueg. suinn W; vog. suinn T. SnE II 89 (479-84): vægsvinn A; vegsvínn r; vegs ... B; vegsuinn C; T.   The first element is the stem form of ON vegr m. 'way.' The final element seems to be a nominalization of the feminine singular of the ON adjective svinnr 'swift, quick.' This latter word, or a root related to it, seems to occur in a number of Norwegian names such as Svindalen (pron. svinn-) in Andebu, Vestfold (NE, pp. 258-9), but there is a possibility also that the first element in some or these is Norwegian svin n., ON svín n. 'pig.’ Furthermore, in several Swedish lake names such as Svinnaren it is quite likely that a river name *Svinn(a) occurs (see Hellquist, pp. 596-97). The ON adjective svinnr in the meaning 'quick, rapid,' however, only occurs in ON in the expression svinn Rín (Atlaqviða 27), Otherwise the word means 'wise'.'   Nevertheless, I would connect the mythological name to the first meaning (not the least because of its use in conjunction with another river) and thus have Vegsvinn mean something like 'the one which flows rapidly in its course,' rather than 'the one which knows how to, or is wise in, finding its way' as some other scholars have suggested (e.g., Jónsson, 1900, p. 234).   þjóðnuma:   Grm 28: þiodnvma R; þioðnvma A. SnE I 25: þioðnvma U; þioðnuma r; þioðnuma W; þiodnuma T. SnE II 89 (479-84): þioonvma A; r; þiodnuma B; þiodnuma C; T.   The final element, in spite of Magnus Olsen's doubts (1964, p. 25), seems to be related to Norwegian dialect nome, ON *numi m. 'a little lake right beside a river, with the water level of which it rises and falls; water container' listed by Ross (p. 548). O. Rygh (NE, pp. 170-71) connects the names listed under the root naum- to Ross' word as does Amund B. Larsen for the farm name Nomeland (NG, VIII, p. 222) and A, Kjeer for Nome (NG, IX, p. 111), The first element would then probably be ON þjóð f., in compounds meaning 'great, powerful.' Accordingly, þjóðnuma could mean something like 'the one which contains a great deal of water.' Other scholars (e.g., Wilken,  p. 276, and Gering, p. 198) have related the final element to the ON verb  nema 'to take by force' with the meaning then something like 'the one which devours people.' If the final element has this latter connotation, then the meaning 'the one which takes with great force' could apply to a river which floods with disastrous consequences. Cf., for example, Bægisd in Oxnadals-hreppur, Eyjafjarðarsýsla, Iceland, from Icelandic bægja, ‘to remove, drive off,’ which regularly floods its banks.