There are two Atli poems in the Codex Regius, the Atlakvitha (Lay of Atli) and the Atlamol (Ballad of Atli). The poems are not preserved or quoted in any other old manuscript, but they were extensively used by the compilers of the Volsungasaga. In the manuscript superscription to each of these poems appears the word "Greenland," which has given rise to a large amount of argument. The scribe was by no means infallible, and in this case his statement proves no more than that in the period round 1300 there was a tradition that these two poems originated in the Greenland settlement.

The two Atli poems deal with substantially the same material: the visit of the sons of Gjuki to Atli's court, their deaths, and the subsequent revenge of their sister, Guthrun, Atli's wife, on her husband. The shorter of the two, the Atlakvitha, tells the story with little elaboration; the Atlamol, with about the same narrative basis, adds many details, some of them apparently of the poet's invention, and with a romantic, not to say sentimental, quality quite lacking in the Atlakvitha. Both poems are sharply distinguished from the rest of the collection by their metrical form, which is the Malahattr (used irregularly also in the Harbarthsljoth), employed consistently and smoothly in the Atlamol, and with a considerable mixture of what appear to be Fornyrthislag lines (cf. Introduction) in the Atlakvitha.

It is altogether probable that both poems belong to the eleventh century, the shorter Atlakvitha being generally dated from the first quarter thereof, and the longer Atlamol some fifty years or more later. In each case the poet was apparently a Christian; in the Atlamol (stanza 82) Guthrun expresses her readiness to die and "go into another light," and in the Atlakvitha there is frequent use of mythological names (e.g., Valhall, Hlithskjolf) with an evident lack of understanding of their relation to the older gods. These facts fit the theory of a Greenland origin exceedingly well, for the Greenland settlement grew rapidly after the first explorations of Eirik the Red, which were in 982-985, and its most flourishing period was in the eleventh century. The internal evidence, particularly in the case of the Atlamol, points likewise to an origin remote from Iceland, Norway, and the "Western Isles"; and the two poems are sufficiently alike so that, despite the efforts of Finnur Jonsson and others to separate them, assigning one to Greenland and the other to Norway or else where, it seems probable that the manuscript statement is correct in both instances, and that the two Atli poems did actually originate in Greenland. An interesting account of this Greenland settlement is given in William Hovgaard's Voyages of the Norsemen to America, published by the American-Scandinavian Foundation in 1914, and an extraordinarily vivid picture of the sufferings of the early settlers appears in Maurice Hewlett's Thorgils, taken from the Floamannasaga.

From the standpoint of narrative material there is little that is distinctively Norse in either the Atlakvitha or the Atlamol. The story is the one outlined in the prose Drap Niflunga (largely based on these two poems), representing almost exclusively the southern blending of the Attila and Burgundian legends (cf. introductory note to Gripisspo). In the Atlakvitha, indeed, the word "Burgundians" is actually used. Brynhild is not mentioned in either poem; Sigurth's name appears but once, in the Atlamol. Thus the material goes directly back to its South-Germanic origins, with little of the Northern making-over which resulted in such extensive changes in most parts of the Sigurth story. The general atmosphere, on the other hand, particularly in the Atlamol, is essentially Norse.

As has been said, the Atlakvitha is metrically in a chaotic state, the normal Malahattr lines being frequently interspersed with lines and even stanzas which apparently are of the older Fornyrthislag type. How much of this confusion is due to faulty transmission is uncertain, but it has been suggested that the composer of the Atlakvitha made over in Malahattr an older Atli poem in Fornyrthislag, and this suggestion has much to recommend it. That he worked on the basis of an older poem is, indeed, almost certain, for in oral prose tradition a far larger number of distinctively Norse traits would unquestionably have crept in than are found in the material of the Atlakvitha. As for the Atlamol, here again the poet seems to have used an older poem as his basis, possibly the Atlakvitha itself, although in that case he must have had other material as well, for there are frequent divergences in such matters as proper names. The translation of the Atlakvitha is rendered peculiarly difficult by the irregularity of the metre, by the evident faultiness of the transmission, and above all by the exceptionally large number of words found nowhere else in Old Norse, involving much guesswork as to their meanings. The notes do not attempt to indicate all the varying suggestions made by editors and commentators as to the reconstruction of defective stanzas and the probable meanings of obscure passages; in cases which are purely or largely guesswork the notes merely point out the uncertainty without cataloguing the proposed solutions.